Table of Contents

Table of Contents
Table of Contents

Program Description

Overview

The main theme of this year's annual meeting is Ion-beam Radiotherapy in the 21st Century: Accuracy and Efficacy.

In recent years, the number of facilities dedicated to particle radiotherapy has been on the upswing throughout the world. Japan is no exception; today, five of the world's 10 carbon-ion radiotherapy facilities are in operation in the country. Since 1994, one of these, the National Institute of Radiological Sciences (NIRS), has conducted numerous clinical and research studies on the latest advances in the field.

On the first day of the conference, an expedition to NIRS to observe a rotating gantry using high-temperature superconducting magnets was offered. On the last day of the conference, attendees had the opportunity to tour the ion-beam Radiation Oncology Center in Kanagawa (i-ROCK), which specializes in a scanning irradiation method using 4 in-room computed tomography units.

Target Audience

  • Physicians, Radiation Oncologists, Oncologists, Surgeons, Residents, and Fellows

  • Medical Physicists, Dosimetrists, and Radiation Therapists

  • Radiation Biologists, Accelerator Engineers, and Scientists

  • Registered Nurses and Clinical Researchers

  • Health Care Policy Makers, Insurance Executives, Industry Personnel, and Hospital Administrators

Particle Therapy Cooperative Group (PTCOG) 2017 Committees

PTCOG 56 Regional Organizing Committee

Yuko Nakayama, Conference Chair

Chair of Department of Radiation Oncology, Kanagawa Cancer Center

Tadashi Kamada, Conference Co-Chair

Director General of Clinical Research Cluster, National Institute of Radiological Sciences, QST

Shinichi Minohara

Chief of Section of MedicalPhysics and Engineering, KCC

Tetsuo Nonaka

Chief of Department of Radiation Oncology, KCC

Takuma Nomiya

Chief of Department of Radiation Oncology, KCC

Shinsuke Ide

Chief of Division of Radiological Technology, KCC

Nobutaka Mizoguchi

Vice chief of Department of Radiation Oncology, KCC

Yousuke Kusano

Section of MedicalPhysics and Engineering, KCC

Eri Takeshita

Section of MedicalPhysics and Engineering, KCC

Shinichi Yoshino

Division of Radiological Technology, KCC

Hiroshi Tsuji

Director of Dept. of Charged Particle Therapy Research, Clinical Research Cluster, NIRS, QST

Koji Noda

Director of Dept. of Accelerator and Medical Physics, NIRS, QST

Eiichi Takada

Senior Principal Researcher of HIMAC Operation Section, Dept. of Accelerator and Medical Physics, NIRS, QST

Takeshi Murakami

Senior Principal Researcher of Heavy-Ion Radiotherapy Promotion Unit, Research Planning and Promotion Office, NIRS, QST

Atsushi Kitagawa

Leader of Heavy-Ion Radiotherapy Promotion Unit, Research Planning and Promotion Office, NIRS, QST

Nobuyuki Kanematsu

Manager of Medical Physics Section, Hospital,, NIRS, QST

Takashi Fujita

Manager of Management Section, Clinical Research Cluster, NIRS, QST

PTCOG 56 Regional Program Committee

Tetsuo Akimoto

Division head, Vice president of National Cancer Center Hospital East Division of Radiation Oncology and Particle Therapy

Masayuki Araya

Director of proton therapy center of Aizawa hospital Proton therapy center

Yoshio Hishikawa

Director of Medipolis Proton Therapy and Research Center Department of Radiology

Yasuhiro Kikuchi

Director of Southern Tohoku Proton Therapy Center Department of Radiation Oncology

Masao Murakami

Professor of Dokkyo Medical University Radiation Oncology Center

Shigeyuki Murayama

Chief of Shizuoka Cancer Center Radiation and Proton Therapy Center, Proton Therapy Division

Takashi Nakano

Professor of Gunma University Graduate School of Medicine Department of Radiation Oncology

Kenji Nemoto

Director of Yamagata University Department of Radiation Oncology

Kazuhiko Ogawa

Professor and Chairman of Osaka University Department of Radiation Oncology, Graduate School of Medicine

Hiroyuki Ogino

Nagoya Proton Therapy Center Department of Radiation Oncology

Tomoaki Okimoto

Director of Hyogo ion beam medical center Department of Radiology

Hideyuki Sakurai

Professor of University of Tsukuba Department of Radiation Oncology

Yoshiyuki Shioyama

Director of Ion Beam Therapy Center, SAGA-HIMAT Foundation Department of Radiation Oncology

Hiroki Shirato

Professor of Hokkaido University Department of Radiation Medicine

Hiroyasu Tamamura

Director of Fukui Prefectural Hospital Proton Therapy Center

Biological Treatment Planning and Molecular Imaging

PTC17-0023: Carbon Ion Beam Overcomes Tumor Hypoxia: The in Vivo Study Using Hypoxic Molecular Imaging

J. Cheng1, Y. Sheng 2, Y. Zhang3, J. Lu4

1Shanghai Proton and Heavy Ion Center, Nuclear Medicine, Shanghai, China 2Shanghai Proton and Heavy Ion Center, Medical Physics, Shanghai, China3Shanghai Proton and Heavy Ion Center- Fudan University Cancer Hospital, Nuclear Medicine, Shanghai, China4Radiation Oncology, Shanghai Proton and Heavy Ion Center, Shanghai, China

Purpose: The human breast cancer ER(+) xenografts were exposed to low/medium/high dose of carbon-ion and photon beam to observe the short and long re-oxygenation phenomenon.

Materials and Methods: The naturally chronic hypoxia was identified using 18F-MISO animal PET/CT with TBR>1.4 as the threshold. Xenografts were divided into high (15GyE)/medium(10GyE)/low(5GyE) dose of photon group and carbon-ion group and received only once irradiation. The TV (tumor volume) was observed .The animal 18F-MISO PET/CT scans were performed on the 1st/3rd/5th/7th day for short term and on the 5th/10th/15th day for long term.

Results: Only high dose photon and medium/high dose carbon ion sub-group can shrink or inhibited tumor volume evidently. Short-term re-oxygenation status: In high dose photon group, TBR increased on the 3rd day and recovered on the 7th. In high dose carbon-ion groups, the phenomenon of “transient severe hypoxia” appeared on the 1st day and disappeared on the 7th. Long-term re-oxygenation status: In high dose photon group, the TBR curve decline on the 1st day, reached to its maximum on 5th day, restored on 7th day and ascended continually in the rest of observation period. However, in high dose carbon ion group, a continuous decline of TBR was be observed and no rebound until the end of the observation period.

Conclusion: Tumor showed a “transient severe hypoxia”, no re-oxygenation, after carbon ion irradiation. Intra-tumoral oxygenation status during carbon beam therapy had no influence on irradiation sensitivity of tumors. Carbon ion beam could kill tumor cell no matter oxygen exist or not.

PTC17-0043: Oxygen Beams for Therapy: Advanced Physical and Biological Characterization

O. Sokol1, E. Scifoni 2, W. Tinganelli 2, J. Wiedemann1, T. Friedrich1, S. Brons3, M. Durante 2, M. Krämer1

1GSI Helmholtz Centre for Heavy Ion Research, Biophysics department, Darmstadt, Germany 2National Institute of Nuclear Physics, TIFPA Trento Institute for Fundamental Physics and Applications, Trento, Italy3HIT Heidelberg Ion Beam Therapy Center and University Clinic Heidelberg, Department of Radiation Oncology, Heidelberg, Germany

While only protons and carbon ions are currently used in radiotherapy, several other modalities can be considered as potential alternatives. Among those are 16O ions; the interest towards using them is justified by the increased linear energy transfer (LET) values, which is crucial for overcoming tumor hypoxia.

Recently the 16O beam model was included into TRiP98, the GSI in-house treatment planning system. The physical description implied the update of the transport model with recent attenuation measurements and fragmentation data for underlying ions. Biological effects are calculated by means of the Local Effect Model (LEM-IV) for the relative biological effectiveness (RBE) and with our semiempirical model for the oxygen enhancement ratio (OER).

The delivered dose distributions and biological endpoints were measured in the experimental rooms of HIT, Heidelberg, and GSI, Darmstadt. For the absorbed dose, both lateral and beam-eye-view profiles were verified. Biological measurements included the expansion of the currently scarce RBE-LET dependence dataset and the measurements of the OER values. Additionally, the kill-painting approach for restoring the uniform survival on a heterogeneously oxygenated target was tested. For a given target hypoxia configuration, oxygen ions demonstrate a better peak-to-entrance ratio compared to carbon ions, in contrast to the cases of normally oxygenated targets, thus overcoming the drawback of larger nuclear fragmentation in the entrance channel with the better LET profile in the target.

PTC17-0136: Evaluation of Biological Treatment Planning for Pediatric Proton Beam Therapy Using Wobbling Nozzle

B.H. Huang1, Y.C. Lin1, C.K. Tseng 2, H.Y. Tsai1

1Chang Gung University, Department of Medical Imaging and Radiological Sciences, Taoyuan, Taiwan- Province of China 2Chang Gung Memorial Hospital, Department of Radiation Oncology, Taoyuan, Taiwan- Province of China

With the potentials of physical/biological advantages, proton beam therapy is expected to reduce toxicity with contrary to a photon plan, especially for pediatric patients. The objective of current study is to investigate the distributions of the dose-averaged LET (LETD) and RBE-weighted dose in our current plans for pediatric patients, and those will be compared with the corresponding distribution of physical dose.

Four pediatric patients with brain tumor (3 ependymomas, 1 atypical meningioma) were treated. The delivery system used the combination of a wobbling and a single scattering system to form a large uniform field. For simulation, TOPAS 3.0.1 based on Geant4 toolkit is used to obtain LETD distribution for each of pediatric plans. In order to get the RBE-weighted dose, a published RBE model as a function of fractionation dose, LETD, and tissue type (α/β), are assigned to organs within the treatment fields as soon as dose distributions and LETD distributions are calculated.

This work focuses on the feasibility of using a biologically motivated treatment plan for pediatric patients; however, it does not consider influences of different either physical doses or tissue types. Future studies to include more physical and biological input data or even different beam delivery system (e.g. Intensity Modulated Proton Therapy) will be a major concern for optimizing proton plans for pediatric patients.

PTC17-0165: Design of a New In-Beam System for On-Line Proton Range Verification during Proton Therapy: Study of Geant4 Physics Lists

E. Mikhaylova1, S. St. James 2, P. Arce Dubois3, J. Treffert4, S.R. Cherry1

1University of California- Davis, Biomedical Engineering, Davis, USA 2University of Washington, Department of Radiation Oncology, Washington, USA3Centro de Investigaciones Energéticas- Medioambientales y Tecnológicas CIEMAT, Departamento de Tecnología, Madrid, Spain4ProNova Solutions, Research and Development, Knoxville, USA

Modern proton therapy gantries can deliver accurate dose (±1%) to a precise target location (±1–2 mm). The main uncertainties are related to proton range. To ensure that the target is adequately covered, the maximum range is increased by nominally 3.5% of the maximum depth plus 1 mm. The consequence is that the delivered high dose region is less conformal to the target and more healthy tissue may receive full dose. Dose delivery can be monitored in vivo by PET or prompt gamma imaging (PGI) systems. As the proton beam interacts in the patient, positron emitting radioisotopes and prompt gammas will be created in vivo and can be monitored to verify the therapy. We propose developing an in-beam system that can perform prompt gamma imaging and PET simultaneously and on-line. This task requires complex strategies starting from accurate Monte Carlo (MC) simulations to state-of-the-art detector solutions. A correct physics choice is critical for successful MC studies and prototype development. Several Geant4 physics lists (HadronTherapy+QGSP_BIC_EMY, QGSP_BIC_HP, and the new 2015 QGSP_BIC_AllHP) are compared, using GAMOS 5.1.0 software. The new list QGSP_BIC_AllHP includes experimentally measured cross-sections for He3, alpha, protons, deuterons and tritons for the energy range 0 – 200 MeV. For the other two lists, the same cross-sections are calculated approximately from a theoretical model only validated for high-energy particles. The results for 230 MeV proton beam show a ∼79%, ∼48% and ∼68% discrepancy in the number of produced secondary positrons, gammas and neutrons respectively for QGSP_BIC_AllHP and QGSP_BIC_HP lists.

PTC17-0190: Evaluation of Biological Effective Dose Distribution of Passive Scattering Proton Therapy Treatment Plans with a Variable Relative Biological Effectiveness Model

J. Wang1, Y. Li1, X. Zhang1, H. Li1, M. Gillin1, R. Zhu1, N. Sahoo1

1M.D. Anderson Cancer Center, Radiation Physics, Houston, USA

Purpose: To evaluate the effect of variable relative biological effectiveness (RBE) of passive scattering proton beams on the biological effective dose (BED) distribution of treatment plans.

Materials and Methods: The BED distributions (Dose x RBE) in passive scattering proton therapy treatment plans were calculated by using an empirical variable RBE model for spread out Bragg peaks (SOBP). The increase in the RBE on the distal part of the SOBP was modeled as a function of dose. The RBE in the flat and proximal regions of the SOBP was taken to be 1.1. Dose volume histograms (DVH) for whole brain, brainstem and optic chiasm were recalculated using this variable RBE model for 8 brain cancer patients to test this model. Important DVH metrics with variable RBE and fixed RBE model dose distributions were compared.

Results: Following differences between the variable RBE model and fixed RBE model were observed. The whole brain V20CGE increased by 0.1% to 0.5%. The brainstem V60CGE increased by 0 to 2.9 cc. The maximum dose to optic chiasm increased for 4 patients, the largest increase being 66%. The increased BED is seen mostly in plans where OAR located at the distal end of one or more of the proton fields.

Conclusion: Calculation of dose distribution using a variable RBE model was found to be useful to assess the possible increase in the BED for organs at risk located in the distal part of the SOBP of the passive scattering proton beams.

PTC17-0335: Biologically Based Treatment Planning Optimization for Intensity Modulated Proton Therapy in Glioblastoma Patients

W. Cao1, P. Yepes 2, D. Grosshans3, R. Mohan1

1MD Anderson Cancer Center, Radiation Physics, Houston, USA 2Rice University, Physics and Astronomy, Houston, USA3MD Anderson Cancer Center, Radiation Oncology, Houston, USA

Purpose: To investigate the potential of incorporating variable relative biological effectiveness (RBE) or its physical surrogate in intensity modulated proton therapy (IMPT) optimization for treating glioblastoma patients.

Materials and Methods: We implemented three approaches to biologically optimizing an IMPT treatment plan. First, variable RBE weighted dose was employed in the conventional quadratic objective function. Second, biological effect based on the linear quadratic model was used instead of RBE weighted dose. The first two approaches applied a linear scaling RBE model (Wilkens and Oelfke, 2004). Third, dose-averaged linear energy transfer (LET) was used as a physical surrogate of RBE in an optimization model, which maximizes LET in targets while minimizing it in normal tissues. Four glioblastoma patients were retrospectively selected from our institution. For each patient, four IMPT plans were created. Three were from the biological optimization approaches implemented and one was from conventional optimization considering a constant RBE of 1.1.

Results: The first and the second approaches showed advantages in achieving more homogenous variable RBE weighted doses within the targets and less doses in the critical structures than conventional optimization. The third approach illustrated that LET distributions could be improved compared to the other two approaches while its advantages in resulting variable RBE weighted dose distributions were also observed.

Conclusion: We implemented three approaches of biological optimization for IMPT treatment planning and demonstrated potential benefits of these approaches in treating glioblastoma patients.

PTC17-0344: Rectum and Bladder Linear Energy Transfer Distributions in Spot Scanning Proton Therapy of Prostate Cancer Using Different Beam Angle Configurations

J. Pedersen1, J.B. Petersen1, C.H. Stokkevåg 2, K.S. Ytre-Hauge3, O. Casares-Magaz1, N. Mendenhall4, L.P. Muren1

1Aarhus University Hospital, Department of Medical Physics, Aarhus C, Denmark 2Haukeland University Hospital, Dept. of Oncology and Medical Physics, Bergen, Norway3University of Bergen, Dept. of Physics and Technology, Bergen, Norway4University of Florida Proton Therapy Institute, University of Florida Proton Therapy Institute, Gainesville- FL, USA

Purpose: The increased linear energy transfer (LET) at the end of the Bragg peak causes concern for an elevated relative biological effectiveness, often in or close to normal tissues. We therefore investigated dose-averaged LET (LETd) distributions in the rectum and bladder in spot scanning proton therapy (PT) of prostate cancer using different beam angle configurations.

Materials and Methods: Spot scanning plans (3mm spots/1mm spacing) were created in the treatment planning system PyTRiP for seven patients (without rectal balloons/spacers). The CTV-to-PTV margins were 4mm axially and 6mm in the superior/inferior directions. The PTV was planned to receive a dose of 78 Gy(RBE). Dose and LETd distributions were calculated for plans with two opposing beams (90°/270°), as well as for two-beam plans with ‘mirrored' beams (from 110°/250° to 70°/290°, in steps of 2°).

Results: The median LETd in the prostate was 2.3 keV/μm (2.2–2.5 keV/μm) while higher LETd was found outside of the prostate. Anterior field configurations increased the rectum LETd while decreasing the bladder LETd, whereas the opposite was seen for posterior field configurations. For the 70°/290° configuration 15-29% of the rectum had higher LETd than in the target while for the 110°/250° configuration, 32-54% of the bladder had LETd higher than in the target.

Conclusion: High LETd areas outside the target were seen in both rectum and bladder when treating with spot scanning proton therapy. Anterior field configurations increased the rectum LETd while posterior configurations increased the bladder LETd.

Models and Biological Optimization

PTC17-0031: The Radiosensitization of Heavy-Element Nanoparticles under Hadron Irradiation

W. Chen1

1Institute of Modern Physics, Division of Medical Physics, Lanzhou, China

Hadron therapy is considered superior approach for the treatment of tumors located in highly sensitive tissues, pediatric cancers, and tumors that are resistant to radiotherapy. It is treated as the most advanced radiotherapy technology in 21th century. The major limitation of these techniques stems from the radiation effects that remain significant in front of the tumor (at the entrance of the track). It is thus a challenge to enhance the biological effect of the treatment in the tumor while to lower the dose given to the normal tissues. Using radiosensitizers to enhance the radiation effects on tumors is the alternative method to improve the modality of radiotherapy against tumors. Heavy-element nanoparticles have been attracted increasing interest to improve the discrimination between tumor and healthy tissue. It is known that many electrons can be released in Auger cascades in addition to the primary ionization process due to the large number of electrons in heavy atoms. The presence of metallic nanoparticles contributes to the potential additional nanoscopic radial dose, which leads to the dose enhancement effect under hadron irradiation. Here, we present the radiosensitization of gold nanoparticles (AuNPs) and drug-AuNPs nanocomposite. Our studies showed that the presence of 7.5 mg/ml citrate-capped AuNPs led to an increment of 44% in RBE value for Hela cells under 70 keV/mm carbon ion irradiation, and very low loading of drug-modified PEG-AuNPs (in cell) can also contribute to a significant enhancement effect on HepG2 cell damage irradiated by carbon ion beam.

PTC17-0076: Radiobiological Evaluation Based on Tumour Oxygenation and Clonogenic Cell Density in Proton Therapy

E. Lindblom1, A. Ureba1, J. Ödén1,2, A. Dasu3,4, I. Toma-Dasu1,5

1Stockholm University, Medical Radiation Physics- Department of Physics, Stockholm, Sweden 2RaySearch Laboratories, RaySearch Laboratories, Stockholm, Sweden3The Skandion Clinic, The Skandion Clinic, Uppsala, Sweden4Linköping University, Department of Medical and Health Sciences, Linköping, Sweden5Karolinska Institutet, Medical Radiation Physics- Department of Oncology and Pathology, Stockholm, Sweden

The relative biological effectiveness (RBE) of protons is a topic of debate in proton therapy. Regardless of whether a constant or a variable RBE is applied, tumour hypoxia can be expected to affect the outcome of proton therapy substantially, as the linear energy transfer (LET) is not high enough to overcome the induced radioresistance. Furthermore, the heterogeneous character of tumour oxygenation could also lead to interplay effects in areas of lower RBE and high radioresistance when using a variable RBE. Another important factor in determining the treatment outcome is the clonogenic cell density. Pre-treatment functional imaging can provide spatial information on both tumour hypoxia and clonogen density. Radiobiological modelling taking into account such information could represent a valuable tool for evaluating proton therapy plans.

In this study, an IMPT plan was optimized assuming a constant RBE and subsequently applied to a three-dimensional tumour model with heterogeneous oxygenation using both a constant and variable RBE. The tumour control probability (TCP) was thereafter calculated for various assumptions regarding the clonogen density and reoxygenation pattern.

For a moderate hypoxic fraction (<10 mmHg) of about 15%, acceptable TCP could not be achieved for clinically relevant clonogen densities assuming no reoxygenation. The effect of an increasing clonogen density was most pronounced if no reoxygenation was assumed, indicating that even a small number of clonogens in a consistently radioresistant area could compromise the treatment. Pre-treatment knowledge of the presence of hypoxia and spatial information of clonogen density could therefore be crucial in proton therapy plan evaluation.

PTC17-0138: LET-Based Reoptimization of IMPT Plans

J. Unkelbach1, P. Botas 2, D. Giantsoudi 2, B. Gorissen3, H. Paganetti 2

1University Hospital Zurich, Radiation Oncology, Zurich, Switzerland 2Massachusetts General Hospital, Radiation Oncology, Boston, USA3VU University, Department of Mathematics, Amsterdam, Netherlands

Purpose: We describe a treatment plan optimization method for intensity-modulated proton therapy (IMPT) that avoids high values of linear energy transfer (LET) in critical structures located within/near the target volume, while limiting degradation of the best possible physical dose distribution.

Materials and Methods: We suggest a two-step planning approach. First, an initial IMPT plan is optimized based on physical dose, as is current clinical practice. In the second step, a prioritized optimization scheme is used to modify the LET distribution while constraining the physical dose objectives to values close to the initial plan. The LET optimization step is performed based on objective functions evaluated for the product of LET and physical dose (LETxD). To first approximation, LETxD represents a measure of the additional biological dose that is caused by high LET.

Results: The method is effective for treatments where serial critical structures with maximum dose constraints are located within/near the target. We studied patients with intra-cranial tumors (high-grade meningiomas, base-of-skull chordomas, ependymomas) for whom the target volume overlaps the brainstem and optic structures. In all cases, high LETxD in critical structures could be avoided while minimally compromising physical dose planning objectives.

Conclusion: LET-based reoptimization of IMPT plans represents a pragmatic approach to bridge the gap between purely physical dose-based and relative biological effectiveness (RBE)-based planning. The method makes IMPT treatments safer by mitigating a potentially increased risk of side effects due to elevated RBE of proton beams near the end of range.

PTC17-0156: Biological Geometries for the Monte Carlo Toolkit for Radiobiology TOPAS-nBio

A. McNamara1, J. Perl 2, J. Ramos Mendez3, H. Kathryn1, B. Faddegon3, H. Paganetti1

1Massachusetts General Hospital & Harvard Medical School, Radiation Oncology, Boston, USA 2SLAC National Accelerator Laboratory, Geant4, Menlo Park, USA3University of California San Francisco Comprehensive Cancer Center, Radiation Oncology, San Francisco, USA

Further advances in radiation therapy are likely to come from the complex interface of physics, chemistry and biology. Computational simulations, such as Monte Carlo (MC) simulations, offer a powerful tool for quantitatively investigating radiation interactions within the cell and can thus help bridge the gap between physics and biology. MC simulations, however, generally require advanced programming skills. TOPAS, a MC application, wraps and extends the Geant4 simulation toolkit giving non-experts the ability to easily develop their own MC models. A new extension, TOPAS-nBio, extends TOPAS to include radiobiology applications to help bridge the gap between physics and biology. TOPAS-nBio utilizes the physics processes of Geant4-DNA, to model biological damage from low energy secondary electrons. A comprehensive library of specialized cell, organelle and molecular geometries has been designed for the toolkit. These geometries range from the micron-scale (e.g. cells and organelles) to complex nano-scale geometries (e.g. DNA and proteins). Users have the ability to specify the cell parameters (e.g., type, size, organelles) using easy-to-use input files. Users can also model the full nuclear DNA hierarchy: chromosome territories containing chromatin fiber loops each comprised of nucleosomes on a double helix. The chromatin fibers can be arranged using either a fractal or simple flower arrangement model. TOPAS-nBio has been benchmarked by comparing results to other track structure simulation software as well as published experimental measurements.

PTC17-0214: Proton Induced Complex DNA Damage: In Silico Modelling of Damage and Repair Using Geant4-DNA

N. Henthorn1, J. Warmenhoven1, M. Sotiropoulos1, R. Mackay 2, K. Kirkby1, M. Merchant1

1Paterson Institute, Division of Molecular and Clinical Cancer Sciences, Manchester, United Kingdom 2The Christie NHS Foundation Trust, Christie Medical Physics and Engineering, Manchester, United Kingdom

We present a stochastic model to predict ion induced DNA damage and subsequent repair. DNA damage patterns are predicted using nanodosimetric principles applied to track structure simulations within the Monte Carlo based Geant4-DNA toolkit. A section of detailed DNA geometry is irradiated to study specific DNA double strand break structures; building up a library of break models for a given radiation quality. These patterns are then fed into a modified Geant4-DNA simulation where the DNA double strand break ends are explicitly modelled within a simplified cell nucleus. Double strand break ends then progress along the predefined Non-Homologous End Joining repair pathway according to stochastic, time constant based state changes.

We show that break complexity and repair kinetics are dependent on the particle LET and particle type, with more complex breaks becoming more probable for higher LET.

Our simulations predict a greater number of residual DSBs after 24h when higher LET particles are used, which is in good agreement with the literature. We also observe a difference in break complexity for protons and alpha particles at the same LET due to differences in radiation track structure.

The complexity of the biological response caused by different ions of the same LET was found to differ due to the radiation track structure. We suggest that this is as a direct consequence of the complexity of the breaks caused, as similar trends are observed for both repair and break induction. This is of relevance for potential application to LET based treatment plans.

PTC17-0249: TCP and NTCP Estimated for Radiosurgery of Liver Metastases with Photon- or Scanned Proton-Beams

G. Mondlane1, M. Gubanski 2, P.A. Lind3, A. Ureba4, A. Siegbahn4

1Stockholm University, Department of Physics - Medical Radiation Physics, Stockholm, Sweden 2Karolinska University Hospital, Department of Oncology and Pathology, Stockholm, Sweden3Södersjukhuset, Department of Oncology, Stockholm, Sweden4Stockholm University, Department of Medical Radiation Physics, Stockholm, Sweden

It has been demonstrated in several studies, using either conventional or hypofractionated regimens, that proton therapy enables a reduction of the volume of normal tissue irradiated). However, side effects after radiotherapy may still appear. This study aims to estimate the tumour control probability (TCP) and normal-tissue complication probability (NTCP) after photon- and proton-beam based radiotherapy.

Ten patients diagnosed with liver metastases, previously treated with photon beam-based stereotactic body radiation therapy (SBRT), were retrospectively planned with intensity modulated proton therapy (IMPT). A two-field configuration was used in the IMPT planning. The healthy part of the liver was the main risk organ due to its location, surrounding the target volume. The TCP was assessed using a Poisson model. The NTCP was calculated using two distinct radiobiological models, the Lyman-Kutcher-Burman (LKB) model and the relative-seriality model. The prescribed dose per fraction was converted to equivalent 2-Gy per fraction doses using the linear quadratic (LQ) model. A generic value for the relative biological effectiveness of proton beams of 1.1 was assumed. A two-sided Wilcoxon signed-rank test with significance level of 0.05 was then carried out.

TCP values of 100 % were obtained for the SBRT and IMPT plans for all patients. The relative seriality model predicted lower values of NTCP compared to the LKB model for both RT modalities. However, there was no significant difference in the NTCP values obtained with the SBRT and IMPT plans.

The plans prepared for SBRT and IMPT of liver metastases resulted in similar values of TCP and NTCP.

PTC17-0323: Tumor Control and Normal Tissue Complication Probability Analysis in a Series of Head and Neck Cancer Patients Treated with BNCT

H. Koivunoro1,2, S. González3,4, L. Provenzano3, L. Kankaanranta 2, H. Joensuu5

1Neutron Therapeutics, Medical Physics, Danvers, USA 2Helsinki University Hospital and University of Helsinki, Department of Oncology, Helsinki, Finland3Comisión Nacional de Energía Atómica CNEA, Subgerencia Instrumentacio'n y Control, Buenos Aires, Argentina4Consejo de Investigaciones Científicas y Técnicas CONICET, CONICET, Buenos Aires, Argentina5Helsinki University Hospital- and University of Helsinki, Department of Oncology, Helsinki, Finland

In boron neutron capture therapy (BNCT), effective dose has been derived using constant relative biological effectiveness (RBE) factors, or compound biological effectiveness (CBE) factor. Recently, an alternative method, photon iso-effective dose calculation formalism, has been proposed. The formalism takes into account dose rate and cumulative dose per fraction using first-order repair of sub-lethal lesions in the modified linear-quadratic model and considers synergistic interactions between low-LET and high-LET radiation. Iso-effective dose calculation predicts significantly lower tumor doses and predicts melanoma lesion response to BNCT better than the fixed RBE approach. Currently the formalism has been extended, redefining the photon iso-effective dose as the dose that produces the same tumor control or normal tissue complication rates as a given combination of the absorbed dose components of BNCT. For head and neck (HN) cancer, the model parameters are determined for tumor tissue and precancerous mucosal membranes from the in vivo oral cancer model of the hamster cheek pouch. The tumor response and precancerous tissue toxicity curves provide the Tumor Control Probability (TCP) and the Normal Tissue Complication Probability (NTCP) models. In this retrospective analysis, we applied the extended photon iso-effective dose formalism to calculate the mucosal membrane NTCP and the TCP for patients with locally recurrent HN carcinoma who were treated with BNCT in Finland in 2003 to 2011. The calculated complication probabilities are compared with the observed clinical outcome. The TCP is evaluated with respect of the radiological tumor response and overall survival. The NTCP is evaluated using mucositis as the endpoint.

PTC17-0426: A Quantitative Model and Experimental Program to Connect Ionization Patterns to Biological Effects

K. Kinoshita1, E. Merino 2, M. Lamba3

1University of Cincinnati, Physics, Cincinnati, Ohio, USA 2University of Cincinnati, Chemistry, Cincinnati, Ohio, USA3University of Cincinnati, Radiation Oncology, Cincinnati, Ohio, USA

The biological effects of ionizing radiation, quantified as a function of dose, vary among different source particles and energies. These differences are thought to be rooted in differences in ionization patterns deposited in the DNA and cellular environment of the organism.

For this project, we develop a statistical model for ionization patterns from photons and protons and correlate them with several types of radiation-induced DNA damage. Radiation damage is modeled as a spatial distribution of vacancies (“holes”), created when particles transfer energy to electrons in DNA molecules. As the predominant form of lethal damage to DNA is thought to require at least two holes in close temporal and spatial proximity, our first calculation explores the probability as a function of dose that pairs of holes occur with nanoscale separations. We obtain estimates for rates and dose dependences, for ionizing radiation with and without track structure. There is a clear difference between the two in magnitude and dose dependence. We tentatively conclude that the track-based model is aligned with cell survival data for both photons and protons. We have begun to test the model by measuring the rates of several types of lethal lesions in naked DNA irradiated over a range of doses, from photons and protons at the Cincinnati Children's Hospital/UC Health Proton Therapy Center. We will describe our model and present preliminary test results. We envision our model and experimental method as a new tool for providing insights into the nature of biological effects from ionizing radiation.

PTC17-0439: The Impact of Modern Radiobiology on the Modeling of Charge Particle Treatment Planning

M. Vazquez1

1Loma Linda University Medical Center, Radiation Medicine, Loma Linda, USA

Radiobiological models are used in radiotherapy to evaluate the biological effects of different treatment plans or modalities. However, the complexity in translate relevant biological effects to generate accurate mathematical models that link dose and LET spectra to clinical response remains one of the main problems in modeling. The majority of the present employed models ignore the inherent unpredictable nature and complexity of biological systems. Most radiation biologists/oncologists have long accepted the concept/dogma that the biologic effects of radiation principally involve damage to DNA. However, there has been an explosion of discoveries in stress signal transduction, DNA damage repair, cellular senescence, cell-cycle checkpoints, radio-genomics and immunomodulation in recent years. These advances in radiation biology and related fields are pushing for a paradigm shift in radiation oncology. Currently standard models do not account for any intrinsic inter-patient or intra-organ radiosensitivity variations, or for tumor cell proliferation or tissue oxygenation, or 3-D structures vs. 2-d cell layers, or other clinical factors influencing control and complications. A few studies have been conducted to examine how particle irradiation modulates the multilevel interactions among tumor and normal tissues. The use of charge particles is expanding rapidly and optimal clinical application of ion beams requires a more comprehensive model of radiation response based on systems biological principles of integrating the molecular activity with higher order cell/tissue and system response. This report will discuss different biological parameters that need to be taking into consideration in order to enhance the use of radiobiological modeling for charge particle treatment.

PTC17-0441: Contribution of Indirect Action to High-LET Radiation-Induced Cell Lethality and Mutation

M. Obara1, R. Hirayama1, A. Uzawa1, S. Hasegawa1

1QST- NIRS, Department of Basic Medical Sciences for Radiation Damages, chiba-shi, Japan

Radiation induced mutation and cell lethality are resulting from a combination of direct and indirect actions of ionizing radiation. Indirect action is mediated by hydroxyl radicals (OH radicals). The contributions of indirect action to mutation and cell lethality are dependent on the concentration of hydroxyl radical. The contribution of indirect action on both biological effects can be estimated from the maximum degree of protection by OH radical scavenger (DMSO), which suppresses indirect action by quenching OH radicals without affecting the direct action of ionizing radiations. CHO cells were irradiated with 200 kV X-rays and carbon ions with a dose-averaged LET of 14 or 90 keV/mm in the presence or absence of DMSO. Carbon ions were provided by 290 MeV/nucleon beams at the Heavy Ion Medical Accelerator in Chiba (HIMAC) at the National Institute of Radiological Sciences (NIRS). Cell lethality was determined using colony formation assay. Mutation induction was detected as 6-thioguanine resistant colonies at the hprt locus. Those biological endpoints were protected by DMSO. The contributions of indirect action of X-rays on mutation frequency and cell lethality were approximately 80%, respectively. For carbon ions, the contribution of indirect action of low-and high-LET carbon ions on cell lethality were approximately 80% and 50%, respectively. In case of mutation frequencies, both contributions of low-and high-LET were approximately 50%. Those results show that OH-radical mediated indirect action of radiation plays an important role in the radiation induced mutation and cell lethality even at high-LET radiation.

PTC17-0471: Preclinical Investigation into the Combination of Proton Therapy and Targeted Radionuclide Therapy in Tumor-Bearing Mice

R. Perrin1, C. Müller 2, N. van der Meulen 2, C. Umbricht 2, S. Safai1, J. Hrbacek1, L. Placidi1, D. Weber1, R. Schibli 2, A. Lomax1

1Paul Scherrer Institute, Center for Proton Therapy, PSI-Villigen, Switzerland 2Paul Scherrer Institute, Center for Radiopharmaceutical Sciences, PSI-Villigen, Switzerland

The goal of this study was to pre-clinically investigate combined proton therapy (PT) and targeted-radionuclide therapy (TRT) in tumor-bearing mice.

The experimental design included 4 groups of mice bearing prostate-specific-membrane-antigen (PSMA)-positive prostate cancer (PC-3-PIP) tumor xenografts (n=5): 1) control (sham-irradiated), 2) 5Gy-PT, 3) 5MBq-TRT, 4) 5Gy-PT+5MBq-TRT. PT to the subcutaneous tumors (mean diameter: 6.5mm, range:6-9mm) was performed on the ocular beamline (OPTIS2) at PSI, using a collimated, passively-scattered 75MeV SOBP field, delivering 5Gy from skin surface to medial edge of tumor. Previous PET activation studies had demonstrated accurate tumor targeting from PT, with concentrated activation in the tumor and no measurable activity in the surrounding tissues or non-irradiated tumor. Group 4 mice were injected with 177Lu-PSMA-617 one day after the PT. All mice were monitored with weight and tumor size over a period of 48d. Endpoint criteria were defined as weight loss >15% or tumor volume >1cm3.

Reduced post-irradiation tumor growth for combined therapy is shown, whilst after 48-day follow-up, all mice receiving combined therapy (5Gy-PT+5MBq-TRT) were still alive. In contrast, 50% of the mice had to be euthanized within 23, 22 and 36 days post-irradiation for the sham, PT or TRT treatments alone.

These preliminary results suggest that combined PT and TRT hinders tumor growth and improves survival, providing a basis for determining therapeutic dose for combined PT-TRT treatments in future experiments.

PTC17-0494: Synergistic Cytotoxic Effects of Duocarmycin SA and Proton Radiation on Glioblastoma Cells in vitro

M. Vazquez1, K. Boyle 2, K. Winter3, J. Sacket4, T. Teichman1, A. Bertucci1, J. Genstler1

1Loma Linda University Medical Center, Radiation Medicine, Loma Linda, USA 2Loma Linda University, School of Pharmacy, Redlands, USA3Loma Linda University Medical Center, School of Medicine, Loma Linda, USA4Loma Linda University Medical Center, School of Medicine, Redlands, USA

Glioblastoma multiform (GBM) is the most common primary brain tumor in humans with a poor prognosis. Improvements in treatment modalities for GBM are urgently needed. Proton therapy is considered one of the most effective forms of radiation therapy for GBM. Clinical studies have shown that delivery of DNA alkylating agents such as temozolomide (TMZ) during radiotherapy increases survival rates of GBM patients, which suggest that this DNA alkylating agent can enhance the radiosensitivity of GBM. While clinically useful, TMZ is a fairly ineffective compound. The duocarmycin class of antitumor antibiotics, exemplified by duocarmycin SA (DSA), is an exceptionally potent group of agents capable to induce a sequence-selective alkylation of duplex DNA. Experiments were performed to determine the combined effects of proton radiation and DSA against GBM cell viability. To define dose-response relationships for protons and DSA, we treated GBM cells (U138) with 250 MeV protons and DSA. Cell toxicity was evaluated by trypan blue exclusion assays and flow cytometry at 72 hrs post-treatment. Results indicate that single 2 and 3 Gy of proton radiation alone are insufficient to effect GBM cell toxicity. Exposure to radiation significantly increases the effectiveness of a 0.1nM dose of DSA, decreasing the survival from 76% to 62% in GBM cells. After a single 3 Gy of proton irradiation GBM cells have 62% survival. In combination with DSA (0.1 nM) the survival drops to 31%. These preliminary results support the hypothesis that DSA at sub-nanomolar concentrations can effectively enhance the radiosensitivity of GBM cells in vitro.

PTC17-0495: Differential Tumor Cell DNA Repair after High Energy Proton and Photon Irradiation Potentially Impacts the Therapeutic Window

M. Pruschy1, D. Weber1,2, M. Guckenberger1, A. Lomax 2

1University Hospital Zurich, Radiation Oncology, Zurich, Switzerland 2Paul Scherrer Institute, Center for Proton Therapy, Villigen, Switzerland

Purpose: The rapid introduction of proton therapy worldwide contrasts with the scarcity of mechanistic investigations to understand the differential effectiveness on the molecular, cellular and tissue level. Detailed quantitative experiments are performed along the axis of range-modulated proton beams, which link an increasing relative toxicity with increasing LET across the SOBP and ask for the integration of a modulated RBE into treatment planning. At the same time, evidence indicates that a differential genetic background of the irradiated target also co-determines a differential quantitative outcome.

Materials and Methods: Experiments with non-tumorigenic cells have previously identified a differential requirement for the two major DNA double strand break repair machineries for cellular survival in response to photon- and proton-irradiation. These studies have here been extended to tumor cells derived from multiple tumor entities (lung adenocarcinoma, glioblastoma, ovarian carcinoma).

Results: We have demonstrated an enhanced susceptibility of a) homologous-recombination (HR)-deficient tumor cells to proton-irradiation b) wildtype-tumor cells to HR-targeting agents and c) increased sensitivity of photon-irradiated tumor cells to inhibitors of non-homologous end-joining (NHEJ). Biochemical dissection of the DNA damage response (e.g. H2AX-, DNA-PKcs-phosphorylation; RPA32-, RAD51-foci-formation) indicate reduced affinity of proton-induced DNA-damage towards NHEJ as major repair pathway.

Conclusion: Our results demonstrate a preference of proton-induced DNA damage towards HR in human cancer cells, which may become relevant for stratification of patients carrying mutations in this DNA-repair pathway. Modeling studies will be presented how such a personalized stratification for proton- vs photon-irradiation based on the genetic background of the tumor could further impact the therapeutic window.

PTC17-0533: Design of a 3D Printed Immobilization Device for Treating Transplantable Tumors in Mice Using Proton Radiotherapy

M. Platt1, A. Steinmetz1, Y. Zhang 2, E. Janssen3, V. Takiar1, A. Mascia 2, K. Huang1, R. Vatner4, M. Lamba1

1University of Cincinnati College of Medicine, Department of Radiation Oncology, Cincinnati, USA 2University of Cincinnati Health, Department of Radiation Oncology, Cincinnati, USA3Cincinnati Children's Hospital Medical Center, Department of Immunobiology, Cincinnati, USA4University of Cincinnati / Cincinnati Children's Hospital Medical Center, Department of Radiation Oncology, Cincinnati, USA

Purpose: Transplantable murine tumor models are critical for studying the radiobiology of proton therapy, and immobilization devices are essential for accurate tumor (often <1 cm) targeting in these models. We utilized three-dimensional (3D) printing technology to design and produce an immobilization device for this purpose.

Materials and Methods: All procedures were approved by the institutional animal care committee. CT imaging of a C57BL/6 mouse was performed, and the Blender software package was used to generate a surface contour. A clamshell device was designed with a negative impression of the animal surface contour, and a lateral aperture for flank tumors. A brass block was affixed to reduce beam penumbra. The device was printed from polylactic acid polymer using an Ultimaker 2 3D printer and tested on a mouse harboring a subcutaneous flank tumor. Tumor motion was tracked in the x, y, and z planes using video monitoring over 15 minutes.

Results: Animals (n=3) were safely immobilized with no signs of trauma or distress. Minimal tumor movement was observed in the x, y, and z planes. Treatment plans were generated using Eclipse software, demonstrating uniform targeting of the tumor with <10% of the prescribed dose delivered to the abdominal or thoracic cavity.

Conclusion: We present the rapid and cost-effective design and production of an immobilization device using 3D printing for the treatment of experimental murine tumors with proton beam radiotherapy

Relative Biological Effectiveness and Biomarkers

PTC17-0025: Calculation of Cell Survival Curves Using the Multiscale Approach to the Physics of Ion Beam Therapy

E. Surdutovich1, A. Verkhovtsev 2, A. Solov'yov 2

1Oakland University, Physics, Rochester, USA 2MBN Research Center, Physics, Frankfurt am Main, Germany

A multiscale approach to the physics of radiation damage with ions (MSA) has been developed in order to relate the biological damage as a result of irradiation with ions to physical, chemical, and biological effects [1]. In order to understand and quantitatively predict the biological outcome of irradiation with ions, the scenario that leads to biodamage is studied analytically.

Over years, the MSA has addressed a number of effects starting with ion propagation in tissue, features of the depth-dose profile with a Bragg peak, production of secondary electrons as a result of ionization of tissue, transport and energy loss by these electrons along with other reactive species, the radial dose distribution around each ion, formation of wave fronts around the ions' paths and consequent propagation of cylindrical shock waves, etc. [1].

It has become possible to join the whole multiscale scenario in a recipe for calculating the cell survival probability [1]. This recipe has been tested on plasmid DNA and most recently on a number of cell lines [2]. A criterion for lethal damage has been suggested and tested. A variety of experimental results such as survival probabilities of plasmid DNA lesions, cell survival curves, enzyme repair foci, etc., have become the field of either improving the MSA or testing its predictions.

References: [1] E. Surdutovich and A. V. Solov'yov, Eur. Phys. J. D 68, 353 (2014); [2] A.V. Verkhovtsev, E. Surdutovich and A. V. Solov'yov, Sci. Rep. 6, 27654 (2016)

PTC17-0072: Carbon Ion Irradiation Abrogates Lin28B-Induced X-Ray Resistance in Melanoma Cells

S.J. Park1,2, K. Heo 2, C.W. Choi3, K. Yang3, A. Adachi4, H. Okada5, Y. Yoshida1, T. Ohno1, T. Nakano1,4, A. Takahashi1

1Gunma University, Heavy Ion Medical Center, Gunma, Japan 2Dongnam Institute of Radiological & Medical Sciences, Research Center, Busan, Republic of Korea3Dongnam Institute of Radiological & Medical Sciences, Department of Radiation Oncology, Busan, Republic of Korea4Gunma University Graduate School of Medicine, Department of Radiation Oncology, Gunma, Japan5Gunma University, Initiative for Advanced Research, Gunma, Japan

Purpose: The Lin28/let-7 axis plays an important role in the tumor initiation and developmental processes. Upregulation of Lin28B is often observed in a variety of cancer cells and the overexpression of Lin28B enhances cancer cell proliferation and radio-resistance through the suppression of let-7 miRNA expression. In this study, we investigated the role of the Lin28/let7 axis as a target for the radio-sensitization of melanoma cancer cells.

Materials and Methods: Lin28B overexpressed melanoma cell lines were generated using Lipofectamine transfection systems. The effect of Lin28B overexpression on cancer stem cell properties was evaluated using sphere-forming assays. To determine the role of Lin28B on radiosensitization, control or Lin28B-overexpressed melanoma cells were irradiated with X-ray or carbon ion beams, then the survival fractions were evaluated using colony forming assays. ICC was performed to evaluate the DNA damage after X-ray or Carbon ion beam irradiation.

Results: The overexpression of Lin28B reduced mature let-7 microRNA expression in melanoma cell lines and enhanced the sphere-forming ability of melanoma cell lines. Lin28B-overexpressed melanoma cells were more resistant to X-ray irradiation than control cells, and Lin28B-induced radio-resistance was abolished after carbon ion irradiation. Consistent with these results, Lin28B overexpression reduced the number of γH2A.X foci after X-ray irradiation, whereas the number of foci showed no change after carbon ion irradiation.

Conclusion: Our results suggest that the carbon ion beam is more effective than the X-ray beam for killing cancer cells, and this might be due to the elimination of CSC populations.

PTC17-0086: A Study of Energy Distribution and Biological Effectiveness in Charged Particle Therapy Using Geant4-DNA Simulation

P. Thongjerm1, R. Barlow1

1University of Huddersfield, International Institute for Accelerator Applications, Huddersfield, United Kingdom

To improve the understanding of ion beam therapy, an investigation of energy deposition and the pattern of particle tracks has been performed by using Geant4 simulation including the determination of secondary ions, which cause the consequent damage to the cells.

The probability of radiobiological effect to cells depends on details of the linear energy transfer of charged particles. Different ion species at different beam energies may have the same mean LET but may produce different biological outcomes because of the variation of energy distribution along their tracks.

In this study, the frequency of DNA double-strand breaks produced by different charged particles at different energies is compared as a function of LET. By using a Monte Carlo study, it enables us to predict the impact of particle beams on a microscopic level, which may eventually aid in treatment planning.

PTC17-0094: Age-Dependent Response of the Rat Mammary Gland to γ Rays and Fast Neutrons Before and After Puberty

Y. Kamochi1, T. Imaoka1, K. Daino1, A. Hosoki1, M. Nishimura1, Y. Nishimura1, M. Takabatake1, S. Kakinuma1, M. Fukushi 2, Y. Shimada1

1National Institutes for Quantum and Radiological Science and Technology, National Institute of Radiological Sciences, Chiba, Japan 2Graduate University of Human Health Sciences, Tokyo Metropolitan University, Tokyo, Japan

Purpose: Particle radiotherapy may expose normal tissues to neutrons and increase the risk of second cancer. Mammary gland is susceptible to radiation-induced carcinogenesis, especially during the pubertal development. Fast neutrons have potent carcinogenic effects and our previous study suggests age dependence of their relative biological effectiveness on rat mammary gland. The present study thus aims to clarify the response of the gland to γ rays and fast neutrons before and after puberty.

Materials and Methods: Female rats were whole-body irradiated with γ rays (5 Gy) or fast neutrons (1 Gy; mean energy, 2 MeV) at pre- and postpubertal ages. Immunoreactivity for γH2AX, Ki67, p21 and cleaved caspase 3 was assessed in the terminal end buds, a suggested target of carcinogenesis, of mammary gland until 48 h after irradiation.

Results: Reduction of γH2AX-positive cells was faster, and induction of p21 was weaker, after γ irradiation of prepubertal than postpubertal glands, suggesting rapid recovery in the former. Ki67 staining showed markedly suppressed cell proliferation after this initial recovery, which was especially eminent in the prepubertal gland, suggesting induction of senescence. In contrast, responses to neutrons were similar between ages. Apoptosis as indicated by cleaved caspase 3 was only slightly induced irrespective of the age and radiation type.

Conclusion: These results are concordant with past studies indicating the effect of ovarian hormones on the mammary gland response to γ rays and age dependence of γ ray–induced inactivation of the clonogenic activity in transplantation experiments, and further suggest their dependence on radiation type.

PTC17-0112: Secondary Cancer Risk Induced by Delayed DNA Repair Defect under Heavy-Ion Irradiation

T. Zhao1, Q. Li1

1Institute of Modern Physics- Chinese Academy of Sciences, Department of Medical Physics, Lanzhou, China

Individuals with reduced DNA damage response face a greater sensitivity to mutagen challenge. In order to study and evaluate the risks of secondary cancer during heavy-ion cancer therapy, we have studied the long-term DNA damage response (DDR) and expression of DNA repair genes after irradiation with heavy ions.

Non-transformed human liver epithelial THLE-2 cells were irradiated with carbon ions of 30 keV/μm at 2Gy. Eight population doublings post-irradiation, a part of the irradiated cells were stimulated secondly by X-rays of 2 Gy. Then cell cycle, apoptosis, clonogenic survival, and g-H2AX foci of the cells were analyzed, respectively. Besides, the expression levels of 80 DNA repair-related genes were detected by means of the RT 2 Profiler PCR Arrays.

Under the heavy-ion irradiation, the long-term cell apoptotic rate increased, and the clonogenic survival fraction reduced significantly. The secondly-stimulated cells showed low levels of DNA damage repair ability and clonogenic formation, and thus the apoptosis and cellblock were enhanced accordingly. Confirmed by the real-time PCR, there were five expression-upregulated and seven down-regulated genes compared with the secondly-stimulated group.

In sum, heavy ion radiation could induce DNA repair defect in normal THLE-2 cells indeed, which might be associated with secondary cancer risk.

PTC17-0159: Initial Experimental Investigation of RBE Prediction in Helium Ion-Beam Therapy

M. Stewart1,2, T. Thomas 2,3, K. Carmen1, D. Ivana1,4,5, K. Parodi3, H. Thomas4, D. Jürgen1,2,4,5, A. Amir1,2,4,5, M. Andrea4,6

1German Cancer Research Center, DKFZ, Heidelberg, Germany 2Heidelberg University Hospital, Department of Radiation Oncology, Heidelberg, Germany3Department of Medical Physics, Ludwig-Maximilians-Universität München, Munich, Germany4Heidelberg Ion-Beam Therapy Center, HIT, Heidelberg, Germany5National Center for Tumor Diseases, NCT, Heidelberg, Germany6National Center for Oncological Hadrontherapy, CNAO, Pavia, Italy

Purpose: In preparation for upcoming 4He treatments at the Heidelberg Ion-Beam Therapy Center (HIT), evaluation and comparison of predictive relative biological effectiveness (RBE) models is currently underway. Experimental RBE measurements for 4He ion-beams are compared to a data-driven phenomenological model and three biophysical models.

Materials and Methods: Renal adenocarcinoma cells (RENCA, ATCCTM) were cultured in supplemented RPMI-1640 Medium and seeded cells in 96-well plates. X-rays (6MV) were delivered to the samples at various dose levels from 0 to 8 Gy for determination of alpha (αph) and beta (βph) from the Linear Quadratic (LQ) model. 4He beams (E4-He =56.65 MeV/u, Bragg-peak depth = 24 mm) were delivered to separate samples with dose values between 0.25 to 4.0 Gy, positioned at depths within the entrance channel (6mm) and Bragg-peak gradient (21 mm). FLUKA Monte Carlo simulations were performed for LET spectra and dose-average LET (dLET) prediction at two experimental measurement points. Colonies were imaged and surviving fraction (SF) was measured for the single experiment. LQ curve-fitting was applied and examined against the selected models.

Results: Simulated dLETZ=2 values were 6.5(±0.47) keV·μm−1 and 12.5(±0.68) keV·μm−1 at 6 mm and 21 mm depth, respectively. Dose contribution from Z=1 charged particles was <2%. (α/β)ph ratio of 2.2 was measured for the cell type.

Conclusion: Sound dosimetric and biological evaluation of variable RBE modeling is crucial prior to the first 4He ion-beam treatments at HIT. Further in vitro studies of the existing models are ongoing.

PTC17-0183: Genomic and Histological Alterations in Mouse Lung Adenocarcinomas Induced by Gamma-Rays, Carbon Ions or Neutrons

S. Yamazaki1, K.I. Iwata1, Y. Yamada1, T. Morioka1, K. Daino1, M. Kaminishi1, M. Ogawa1, Y. Shimada1, S. Kakinuma1

1National Institute of Radiological Sciences- National Institutes for Quantum and, Radiation Effects Research, Chiba, Japan

Purpose: The high LET heavy ion radiotherapy and intensity modulated radiation therapy (IMRT) have been recently used more frequently for cancer treatment because of their advantage to increase the dose to the target. However, these modalities generate high LET neutrons, and raises concern over radiotherapy-induced second cancer risk. However, little data are available for the mechanism of lung carcinogenesis induced by neutrons. In this study, we examined the genomic and histologic alterations in mice lung adenocarcinomas developed after neutron irradiation, and compared them to those after gamma-rays or carbon-ion irradiation.

Materials and Methods: B6C3F1 mice were exposed whole-body to 0.05 to 4 Gy of gamma-rays (137Cs), carbon ions (13 keV/μm) or neutrons (2 MeV) at 1 or 7 weeks of age. Lung tumors were collected and subjected to histological and molecular analyses.

Results: Whereas spontaneous incidence of lung adenocarcinomas was increased after 800 days of age, interestingly, lung adenocarcinoma developed earlier than 400-500 days after neutron irradiation at 1 or 7 weeks of age. Sequence analysis revealed that Egfr, Braf and Kras mutations were also frequently observed in both spontaneous and radiation-induced lung adenocarcinomas. Immunostaining showed a higher expression level of pERK1/2 protein in neutron and carbon ion-induced lung adenocarcinomas compared with spontaneous carcinomas. These results suggest that although abnormalities in the Egfr/Braf/Kras pathway are commonly observed in mouse lung adenocarcinomas, activation levels of this pathway in lung adenocarcinomas are different by the type of radiation.

PTC17-0197: Comparative Exploration of Biomarkers in Mice Exposed to X-Ray or Fe-Ion Radiation

T. Nakajima1, G. Vares 2, Y. Ninomiya1, B. Wang1, T. Katsube1, K. Tanaka1, C. Liu1, H. Hirakawa1, K. Maruyama1, A. Fujimori1

1National Institute of Radiological Sciences- National Institutes of Quantum and, Department of Radiation Effects Research, Chiba-shi, Japan 2Okinawa Institute of Science and Technology, Advanced Medical Instrumentation Unit, Onna-son, Japan

Biological markers, which indicate radiation effects and are used for predicting late effects in the future, are expected to be identified. The purpose of this study is to explore useful biological markers for evaluating radiation effects. Particularly, low-dose particle radiation may induce bystander effects due to low particle fluence and to compare the changes in molecules after particle irradiation with that after X-rays irradiation might possibly detect particle-radiation specific biomarkers. Metabolic analyses were performed in livers of mice irradiated with X-rays. Dose-dependent or dose-specific alterations were evaluated. In addition to it, investigating differences of biomarkers due to radiation quality, specific alteration in metabolites was compared with that in the case of Fe-ion irradiation. Moreover, we tried to detect inflammatory biomarkers in both X-rays and Fe-ion irradiation. Metabolites in livers of mice (C57BL/6J) were analyzed by capillary electrophoresis-time-of-flight mass spectrometry (CE-TOFMS) one month after X-ray irradiation. Significantly altered metabolites to the control were selected from the irradiation groups and specific trends were evaluated. One month after irradiation, 4, 8 and 16 metabolites were significantly altered in irradiated groups at doses of 0.1Gy, 0.5Gy and 2Gy, respectively, compared to the levels in the control. Significantly altered metabolites included free amino acids. However, changes in free amino acids after Fe-ion irradiation (0, 0.1 and 2 Gy) were not detected. These will be discussed in the viewpoint of amino acid metabolism and usage as biomarkers for evaluating radiation effects with results of cytokine alteration after irradiation.

PTC17-0205: Evaluation of Survival Curve for Solid Tumor Cells after Large Doses of Carbon Ion Beams

Y. Yoshida1, K. Ando1, S. Koike 2, T. Yako1, H. Ikeda1, A. Uzawa 2, A. Takahashi1, T. Kanai1, T. Nakano1

1Gunma University, Heavy Ion Medical Center, Maebashi- Gunma, Japan 2NIRS- QST, Research Center for Charged Particle Therapy, Chiba, Japan

The physical characteristics of Carbon-ion beams (C-ions) may possible to perform a hypofractionated radiation therapy. We previously evaluated the therapeutic gain for C-ion fractionation using mouse model. These experiments led to the assumption that the therapeutic gain would increase when the dose per fraction increased. Here, we evaluated the survival curves for solid tumor cells after large doses of C-ions.

Transplanted fibrosarcoma (NFSa) tumors growing in C3H/He mice were exposed to C-ions (290 MeV/n, 20 keV/μm and 74 keV/μm of SOBP beam) and X-rays (200 kVp, 14.6 mA). Cell survival was defined using the TD50 (cell dose required to produce tumors in 50% of the animals) assay. Parameters alpha and beta were estimated by fitting the survival curves (SCs) to the linear quadratic (LQ) model. The TCD50 assay (50% tumor control radiation dose) was used to determine the tumor control probability (TCP).

The lowest surviving fractions (SFs) were about 10−7 for either radiation quality. SCs showed continuous bending down to 10−7 for X-rays, and were well fitted using the LQ model. SCs for 20 keV/μm C-ions showed continuous bending down, and almost linear when SFs exceed 10−5. Meanwhile, SCs for 74 keV/μm C-ions was almost linear. The alpha values increased with an increase in the LET, while the beta values showed least or no dependence on LET. The RBE for 20 keV/μm and 74 keV/μm C-ions at TCD50 were about 1.2 and 1.8, respectively. This is the first study evaluating the SFs of 10−7 after C-ions irradiation.

PTC17-0211: Genomic Alterations and Subtypes of Rat Mammary Carcinomas Induced by Gamma-Rays or Neutrons

H. Moriyama1,2, K. Daino1, T. Imaoka1, Y. Nishimura1, M. Nishimura1, T. Morioka1, S. Kakinuma1, K. Inoue 2, M. Fukushi 2, Y. Shimada1

1National Institute of Radiological Sciences- National Institutes for Quantum and, Department of Radiation Effects Research, Chiba, Japan 2Tokyo Metropolitan University, Department of Radiology- Faculty of Health Sciences, Tokyo, Japan

We know that neutrons have high RBE. The mammary gland is shown to be one of the most sensitive organs to radiation-induced carcinogenesis by epidemiological studies. There is no information so far on whether features in neutron-induced breast cancer differ from that in gamma-ray–induced cancer. At 7 weeks of age, rats were whole-body irradiated with gamma-rays at an absorbed dose of 0.5–4 Gy or neutrons at 0.0485–0.97 Gy. RBE was calculated to be about 14. Based on this, decision was made to choose mammary cancers arising in rats irradiated with 0.5 Gy of gamma-rays or neutrons for further analysis. As the results of classification into clinically relevant subtypes by immunostaining, no significant difference was found between groups. However, all carcinomas of the neutron-irradiated group were positive for estrogen receptor α, unlike those of the control and gamma-ray–irradiated groups. On the other hand, as the results of the comprehensive search for DNA copy number alterations by the aCGH analysis showed that, the number of aberrations and their length were not significantly different between groups. Nevertheless carcinomas in the irradiated groups tended to have large deletions (larger than 1 Mb) compared with the spontaneous carcinomas. Frequent deletions on 5q32 were found only in radiation-induced carcinomas and included Cdkn2a, Cdkn2b and Mtap genes, all of which have been reported in breast cancer in previous studies. These results suggest that molecular biological features depend on the history of radiation exposure and radiation types.

PTC17-0233: Modelling Direct DNA Damage on Gold Nanoparticle Enhanced Proton Therapy

M. Sotiropoulos1, N.T. Henthorn1, J.W. Warmenhoven1, R.I. Mackay 2, K.J. Kirkby1,3, M. Merchant1,3

1University of Manchester, Division of Molecular & Clinical Cancer Sciences Faculty of Biology Medicine and Health, Manchester, United Kingdom 2The Christie NHS Foundation Trust, Christie Medical Physics and Engineering, Manchester, United Kingdom3The Christie NHS Foundation Trust, Manchester, United Kingdom

Gold nanoparticles have demonstrated a radiosensitization potential under photon and proton irradiation. Most existing studies have attributed the effect to the increased local dose delivered by electrons generated from interactions of the beam protons with the gold nanoparticles. However, the mechanism leading to an increase in the cell killing is yet not clear.

To further understand the underlying mechanisms of the radiosensitization at the cellular level, a cell model with detailed nuclear DNA structure was implemented in the Geant4 Monte Carlo simulation toolkit. A realistic gold nanoparticle distribution was incorporated, allowing for the formation of clusters of vesicles filled with the gold nanoparticles. A clinically relevant gold concentration was simulated for the gold nanoparticle size of 6, 15, and 30 nm. Protons with linear energy transfer values found in a spread out Bragg peak (1.3-4.8 keV/μm) were simulated. The event-by-event models available through the Geant4-DNA were used for accurate calculations of DNA damage. Damage to the DNA inducing either single (SSB) or double strand breaks (DSB) was used for the quantification of the radiosensitization effect, for a dose fraction of 2 Gy. Each case was repeated 100 times to get an average number of SSB or DSB numbers.

For the combinations of gold nanoparticle size and proton energies studied in the present work, no statistically significant increase in the single or double strand break formation was observed. As gold nanoparticles enhanced proton therapy have been proven experimentally, our results allow hypothesizing contribution from alternative mechanisms of radiosensitization.

PTC17-0379: Synergistic Effects by Arsenite in Radiosensitization

Y. Ninomiya1, Y. Dong 2, E. Sekine-Suzuki3, T. Nakajima1

1National Institute of Radiological Sciences- National Institutes for Quantum and Radiological Science and Technology, Department of Radiation Effects Research, Chiba, Japan 2Medical College of Soochow University, School of Radiological Medicine and Protection, Soochow, China3National Institute of Radiological Sciences, National Institutes for Quantum and Radiological Science and Technology, Department of Basic Medical Sciences for Radiation Damages, Chiba, Japan

Glioblastoma is frequently (about 30%) occured and radioresistant brain tumor. Arsenite that penetrates blood-brain-barrier is known to show its synergistic effects with radiation in vitro and in vivo. However, its mechanism remains unclear. As the synergistic radiosensitization has been reported in p53 deficient cancer cells, its radiosensitivity may be related to p53 function. Therefore, we analyzed radiosensitization using p53 deficient cell lines, U87MG-E6 inactivated p53 and HCT116 (p53-/-). Synergistic effects by arsenite in radiosensitivity were observed in both cell lines. Using irradiation with heavy-ion, we also identified synergistic radiosensitization by arsenite in U87MG-E6 cell lines. Effects of arsenite on cells is known to involve generation of reactive oxygen species (ROS). To examine contribution of ROS to the cellular toxicity of arsenite in the radiosensitivity, we studied survival fraction analysis (SF) using N-Acetyl-L-Cysteine (NAC), a ROS scavenger. The radiosensitization by arsenite was disappeared by NAC. On the other hand, as DNA damage induced by arsenite is repaired by BRCA2-dependent homologous recombination, we performed SF analysis using knockdown of BRCA2 with BRCA2 siRNA. The results revealed that the increased radiosensitivity by arsenite is disappeared by the knockdown of BRCA2.Moreover, because it was also demonstrated that suppression of cell growth by arsenite is due to abnormal amplification of centrosomes, we analyze its contribution to the increased radiosensitivity by arsenite. As a result of that, the increased radiosensitization by arsenite was correlated to the abnormal amplification of centrosomes. These results will be discussed referring to relationship between radiosensitization and p53 functions.

PTC17-0417: Evaluation of Cellular Response to Proton Radiation and 5-Fluorouracil Combination Therapy in Esophageal Cancer Cell Lines

H. Hojo1, T. Dohmae 2, K. Hotta1, S. Kageyama1, A. Motegi1, M. Onozawa1, N. Nakamura1, S. Zenda1, K. Tsuchihara3, T. Akimoto1

1National Cancer Center Hospital East, Division of Radiation Oncology and Particle Therapy, Kashiwa, Japan 2High Energy Accelerator Research Organization, High Energy Accelerator Research Organization, Tsukuba, Japan3National Cancer Center Hospital East, Division of Translational Research- EPOC-, Kashiwa, Japan

Purpose: Proton beam therapy (PBT) with concurrent chemotherapy is currently used to treat several locally advanced cancers. Although the relative biological effectiveness (RBE) varies with the depth of the spread-out Bragg Peak (SOBP), the differences in cytotoxicity along the depth of the SOBP have not been investigated. Therefore, this study examined differences in the cellular response to PBT and 5-fluorouracil treatment.

Materials and Methods: OE21 human esophageal squamous cells were irradiated with a 235 MeV proton beam at 4 different SOBP positions. The respective effects of irradiation and 5-FU treatment on cell survival was assessed by clonogenic assays and sensitizer enhancement ratio (SER). Additionally, DNA double-strand breaks were estimated based on phospho-histone H2AX (γH2AX) foci formation at 0.5 and 24 h post-irradiation.

Results: RBE in vehicle control cells tended to increase according to the depth of SOBP. However, no significant differences in SER10 were observed across the various SOBP positions of SOBP in cells treated with PBT with 5-FU. Moreover, cells irradiated at the distal end of SOBP displayed a marked increase in residual γH2AX foci at 24 h post-irradiation when compared those irradiated at the entrance or the proximal of the SOBP.

Conclusion: Our data indicate that the therapeutic effects of 5-FU were not significantly different among the various SOBP positions in OE21 cells irradiated with a 235 MeV proton beam; however, 5-FU-induced cytotoxicity tended to be higher at the distal end than that at other positions.

PTC17-0478: Large RBE and Small OER in Cell Killing Derived from Radiation Actions of High-LET Radiations

R. Hirayama1, M. Obara1, A. Uzawa1, Y. Matsumoto 2, A. Ito3, Y. Furusawa1, S. Hasegawa1

1National Institutes for Quantum and Radiological Science and Technology- National Institute of Radiological Sciences, Department of Basic Medical Sciences for Radiation Damages, Chiba, Japan 2University of Tsukuba, Proton Medical Research Center and Department of Radiation Oncology, Tsukuba, Japan3Tokai University, Department of Nuclear Engineering- School of Engineering, Hiratsuka, Japan

The biological effects originate by a combination of direct and indirect actions. The contribution of indirect action can be estimated by from the degree of protection by radical scavenger. CHO cells under oxic and hypoxic conditions were exposed to high-LET radiations of 14 to 480 keV/μm in the presence or absence of OH radical scavenger (DMSO) and their survival was determined using a colony formation assay. The contribution of OH radical mediated indirect action from X-irradiation on cell killing of CHO cells were 76% and 50% under oxic and hypoxic conditions, respectively. For high-LET radiation, the contributions of indirect action were estimated to be 32% and 22% under oxic and hypoxic conditions, respectively. The RBE determined at the 10% survival level (D10 or LD90) increased with LET, reaching a maximum value of 2.7 and 5.1 under oxic and hypoxic conditions at 200 keV/μm, and decreased thereafter. When the RBE was estimated separately for direct action (RBE-DA) and indirect action (RBE-IA); the RBE-DA was greater than RBE-IA over the ion LET range tested under both oxygen states. When the OER was estimated separately for direct action (OER-DA) and indirect action (OER-IA); the OER-DA was smaller than OER-IA over the ion LET range tested. For example, OER-DA and OER-IA showed the values of 2.0 and 3.8 for X-rays, and 1.0 and 1.6 for high-LET radiation at 480 keV/μm, respectively. Thus, the direct action of high-LET radiation gives a remarkably larger RBE and smaller OER for cell killing than indirect action.

Breast

PTC17-0066: DIBH for Post-Mastectomy Chest Wall Irradiation Using IMPT: A Dosimetric Study

N. Depauw1, S. Patel1, E. Batin1, S. Macdonald1, H.M. Lu1

1Massachusetts General Hospital, Radiation Oncology, Boston, USA

Deep inspiration breath hold technique (DIBH) is commonly employed for conventional post-mastectomy chest wall irradiation (PMRT) using tangent photon fields due to significant improvements in cardiac sparing. Meanwhile, proton pencil beam scanning (IMPT) also allows for cardiac and ipsilateral lung sparing along with great target coverage. The combination of both approaches may thus further improve sparing of cardiac structures.

The DIBH CT scan of 10 left-sided PMRT patients, previously treated at our institution with IMPT at free breathing, were re-contoured by the same physician and replanned by the same planner using clinical guidelines. The plans were generated based on our clinical spot sizes of 8 to 14 mm as a function of energy. Clinically relevant DVH metrics were subsequently collected for each modality and compared.

On average, DIBH slighty improves cardiac sparing: heart mean dose and LAD V5 reduced by 0.27 and 0.38 GyRBE, respectively. Although DIBH might show dosimetric benefits in some cases, such differences present little to no clinical significance in light of errors associated with contouring and inverse planning. Furthermore, the practical implementation of DIBH results in greater delivery uncertainties. As a consequence, the use of a machine with smaller spot sizes would result in greatly improved free-breathing dosimetry and, thus, not warrant the use of DIBH either.

PTC17-0167: Feasibility of Intensity Modulated Proton Therapy (IMPT) in Deep Inspiratory Breath Hold (DIBH) for Breast Cancer with Unfavorable Anatomy

K. Corbin1, S. Park1, N. Remmes1, C. Beltran1, R. Mutter1

1Mayo Clinic, Radiation Oncology, Rochester, USA

Purpose: DIBH is widely used in photon therapy for breast cancer (BC) to reduce cardiac dose. The utility of DIBH in BC proton therapy is not known. Here, we report initial results of DIBH IMPT for patients with complex anatomy.

Materials and Methods: All BC patients treated with protons using DIBH were reviewed. DIBH was accomplished using marker based tracking with visual feedback to the patient to set the gate. Optical surface imaging was captured at simulation and used during patient set up for monitoring consistency. CTV included the breast/chest wall and regional lymph nodes. All patients were treated with two-field IMPT on a Hitachi PROBEAT-V proton therapy system with daily 2D/3D kV imaging. Verification CTs were performed at least once, during the first week of therapy.

Results: Since 6/2015, 7 patients were identified, all left-sided. DIBH intent was cardiac displacement in all 7 patients. Six patients were treated post mastectomy, and one post lumpectomy. All 6 PMRT patients were reconstructed (5 tissue expanders, 1 implant). Average age was 45 (24-62). Four patients had radiographic internal mammary nodal involvement at presentation. Three received an IMN boost. Planning results and indications for DIBH are shown. Average daily set-up and treatment time was <60 minutes. Re-planning was performed in 1 case for improved supraclavicular nodal coverage.

Conclusion: IMPT in DIBH for BC patients with unfavorable anatomy and/or deep nodal boosts is feasible and resulted in low OAR doses and excellent CTV coverage. DIBH should be considered amongst patients with these indications.

PTC17-0491: Assessment of Respiration on Dose-Volume Parameters for Breast Cancer Regional Nodal Irradiation Using Double Scattered Proton Therapy

J. Bradley1, X. Liang1, M. Rutenberg1, L. Bochner1, B. Jyoti1, Z. Li1, N. Mendenhall1, M.W. Ho1

1University of Florida Health Proton Therapy Institute, Radiation Oncology, Jacksonville, USA

Purpose: This study investigates the use of 4D-CT for double-scattered proton therapy (PT) treatment planning in breast cancer.

Materials and Methods: PT was used to treat the regional lymphatics in 14 women (11 left-sided, 3 right-sided). PT was also used to treat the breast/chestwall in 11, while 3 had matched photon tangent fields. Free-breathing 3D-CTs were acquired for simulation and treatment planning in 9 patients, with follow-up 4D-CTs to assess respiratory motion. 4D-CTs were acquired for simulation in 5 women and treatment planning was performed on the average. Heart and lung were contoured on all scans.

Results: Cardiac dose varied minimally between the plans based on 3D-CT applied to the 0 and 50 phases of the 4D-CT; slightly larger differences were noted in lung doses. For the 5 patients with only 4D-CT, cardiac and lung DVH parameters showed minimal variability between the plan based on the 4D-CT average phase and the 4D-CT phase 0 and phase 50.

The variation in breast/chestwall CTV dose and internal mammary node CTV dose between 3D-CT and 4D-CT phases 0 and 50 was assessed in 7 patients.

For all patients, the differences in DVH parameters between the 4D-CT phases 0 and 50 were assessed.

Conclusion: With robust PT treatment planning, target coverage and OAR doses vary minimally during the respiratory cycle in most patients. However, given the wide range, 4D-CT is recommended to capture patients for whom there are clinically meaningful differences in these parameters.

PTC17-0497: Toxicity of Uniform Scanning Proton Therapy for Breast Cancer

Y. Zheng1, K. Prahbu 2, A. Chang 2, G. Larson 2, C. Vargas3

1ProCure, Medical Physics, Oklahoma City, USA 2ProCure, Radiation Oncology, Oklahoma City, USA3Mayo Clinic, Radiation Oncology, Phoenix, USA

Purpose: To analyze the toxicity and patterns of failure of uniform scanning proton therapy for breast cancer patients.

Materials and Methods: We analyzed 100 breast cancer patients treated at our center. Each patient was treated with uniform scanning proton beams, using typically 2-4 anterior oblique and/or lateral fields. The prescription was 45-50.4 Cobalt gray equivalent (CGE) at 1.8 CGE/fraction and followed by a 9-16.2 CGE boost treatment as needed. Toxicity types and frequencies during and post treatment were obtained from the proton collaborative group (PCG) and analyzed. In addition, skin reaction was carefully studied by analyzing photos of treatment surface, which were taken every 5 fractions during proton therapy.

Results: No patient had experienced Grade 4 or 5 toxicity. The most common adverse effect was dermatitis (Grade 1, 31%%; Grade 2, 52%, Grade 3, 6%), followed by skin discomfort, hot flashes, fatigue, cough, breast pain, chest wall pain, lymphedema and nipple deformation. Other toxicities such as esophagitis were infrequent, with an incidence rate of 2% or lower. Post treatment toxicities mainly included hot flashes, arthralgia, lymphedema, nipple deformity, breast pain and brachial plexopathy, with an incidence rate of 3% or lower.

Conclusion: Dermatitis and discomfort are the top two toxicities in terms of frequency and severity. However, no Grade 4 or above was observed, and other toxicities were of Grade 2 or less. Overall, uniform scanning proton therapy is well tolerated for breast cancer treatment.

PTC17-0538: Intensity Modulated Proton Therapy Can Be Used for Post-Mastectomy Patients with Tissue Expander and Metal Port

H. Giap1, R. Lepage1, L. Dong1

1Scripps PTC, Scripps Proton Therapy Center, San Diego, USA

Purpose: Intensity modulated proton therapy (IMPT) has been used exclusively at Scripps Proton Therapy Center over the past 3 years to treat breast cancer patient with tissue expander after mastectomy. This study describes our technique and reports our early experience.

Materials and Methods: Eight patients were identified from EMR. All patients underwent CT based simulation and treatment planning and were set up supine on a breast board. Daily setup and localization was accomplished with 3-6 skin surface fiducial markers tracked with orthogonal X-ray image. The CTV was contoured using the guideline from RADCOMP study. The CTV was divided into two half-shell volume for optimization purpose and was analyzed as two separate CTV. Typical dose of 50-50.4 Gy in 25-28 daily treatments was prescribed to the chest wall and regional lymphatics including internal mammary nodes. Boost was done sequentially. Treatment was delivered using IMPT with 2-3 fields using Multi-Field Optimization (MFO) technique. Plan robustness was analyzed. A total of 5 patients have been treated with this technique.

Results: Most patients experienced grade 2 dermatitis, no patient with grade 3 skin toxicity. Most patients have mean heart dose < 1.5 Gy and mean lung dose < 1.1 Gy. All patients self-reported ‘good to excellent' cosmetic outcomes. No patients had evidence of local failure.

Conclusion: IMPT is a safe and feasible approach for treatment of post-mastectomy patients with tissue expander and metal port. Further validation with more patients and longer follow-up are needed.

PTC17-0539: Intensity Modulated Proton Therapy for Re-Irradiation of Recurrent Cancer in the Breast and Chest Wall: Scripps Experience

F. Giap1, L. Dong 2, R. Lepage 2, H. Giap 2

1University of Texas at Southwestern, Medical School, Dallas, USA 2Scripps PTC, Scripps Proton Therapy Center, San Diego, USA

Purpose: IMPT has been used exclusively at Scripps Proton Therapy Center for re-irradiation of recurrent malignancies in the breast and chest wall over the past 3 years. This study describes our technique and reports our early experience.

Materials and Methods: We identified 10 patients in this category, who all had previous radiation to the breast or chest wall. Two patients needed chest wall radiation after resection of recurrent cancer due to positive margins; seven patients had gross nodal and chest wall recurrence; one patient had post-lumpectomy radiation in previously irradiated area from another malignancy. All patients except one (23.4 Gy) had previous full dose radiation to 50 Gy or higher. Patients, immobilized with a either breast board or Vac-Q-Fix cushion, were set up in the supine position with arms over their head. One to two beams using IMPT with MFO technique was used. Dose was prescribed at 1.8-2 Gy to 50 to 66 Gy daily treatment. Weekly adaptive simulation was done with CT. Photographs were obtained during and after treatment. All cases were reviewed and approved by our weekly physicist and physician treatment planning conference.

Results: All patient had grade 1-2 skin toxicity. There was no Grade 3 or greater acute or late toxicity. One patient developed grade 2 lymphedema. No patients have local failure, and all are still alive.

Conclusion: IMPT is feasible and a safe modality for re-treatment of recurrent cancer in the breast and chest wall. Further validation with more patients and longer follow-up is needed.

PTC17-0542: Intensity Modulated Proton Therapy for Accelerated Partial Breast Irradiation: Scripps Experience

F. Giap1, A. Waldinger 2, R. Lepage 2, L. Dong 2, H. Giap 2

1University of Texas Southwestern, Medical School, Dallas, USA 2Scripps PTC, Scripps Proton Therapy Center, San Diego, USA

Purpose: Intensity Modulated Proton Therapy (IMPT) has been used exclusively at Scripps Proton Therapy Center for accelerated partial breast irradiation over the past 3 years. This study describes our technique and reports our early experience.

Materials and Methods: All 25 patients with ductal carcinoma except for 6 DCIS and one lobular carcinoma. Five patient with breast implant. All patients underwent CT simulation and were set up supine on a breast board or in the prone position. Daily setup was accomplished with 3-6 skin surface fiducial markers tracked with orthogonal x-ray pairs. 40 Gy and 34 Gy in 10 daily treatments was prescribed to the lumpectomy cavity (GTV) and GTV plus 1-1.5 cm (CTV) excluding chest wall and skin. MRI fusion is used in some patients. Treatment was delivered using IMPT with a single enface field with simultaneous integrated boost.

Results: Mean total patient time in treatment room was 14 minutes. Maximum and mean doses are as follows: cardiac 5.5 Gy/ 0.05 Gy, ipsilateral lung 22 Gy/ 0.6 Gy, chest wall 38 Gy/7 Gy and skin 5mm of 40Gy/10 Gy. Most patients experienced grade 1 dermatitis, and 3 with grade 2. With a mean follow up time of 19 months, 3 patients had minor dry skin in the treatment area and no other late toxicities. All patients self-reported ‘good to excellent' cosmetic outcomes. No patients had evidence of local failure.

Conclusion: Using single field IMPT is a safe, feasible and effective approach for APBI. Further validation with more patients and longer follow-up are needed.

Central Nervous System/Skull Base/Eye

PTC17-0030: Outcomes Comparison of Treatment with Stereotactic Radiosurgery and Proton Beam Therapy for Uveal Melanoma

H.W. Yu1

1Taipei Medical University Hospital, Department of Radiation Oncology, Taipei, Taiwan- Province of China

The patients treated with Gamma Knife stereotactic radiosurgery (SRS), or Proton beam therapies (PBT) for uveal melanoma were reviewed. We analyzed post-operative visual outcomes and if visual outcomes varied with proximity to the optic nerve or fovea. The two patients with choroidal melanoma in this study were still alive with no evidence of tumor progression or distant metastasis. In case 1 patient, the fundoscopic after SRS showed hyperpigmentation of tumor surface, distinct borderline of tumor margin, and tumor shrinkage. Significant tumor thickness reduction was observed in one of the choroidal melanoma patients and the metastasis patient. In case 2 and 3 patients, post-treatment ultrasonography after SRS showed serous retinal detachment disappearance. In one melanoma patient, post-gamma knife 3 months, patient complained about blurred vision of right eye. According to the references were compared the visual outcome, patients treated with PBS retain better vision post-operatively however possible confounding factors in study include age, tumor location and systemic morbidities. Sikuade et al. report treatments achieved excellent local control rates with eye retention in 98% of the SRS group and 95% in the PBT group. The SRS group showed a poorer visual prognosis with losing acuity compared to with PBT. A report from Verma et al. showed PBT could reduce ocular toxicities in patients with uveal melanoma. Five-year local control rates exceed 90%, which persist at 10- and 15- years. Five-year overall survivals range from 70–85% using proton beam radiotherapy. Only 7–10% of uveal melanoma patients needed eye removal after treatment.

PTC17-0036: Proton Therapy versus Photodynamic Therapy for Circumscribed Choroidal Hemangiomas

J. Thariat1, J. Herault 2, C. Maschi3, B. Stephanie4, J. Caujolle4

1Centre Lacassagne, Radiation Oncology, Nice, France 2CAL, RT, Nice, France3CHU, Opthalmology, Nice, France4CHU, Ophthalmology, Nice, France

Purpose: To compare the results of low-dose proton therapy (PT) and photodynamic therapy (PDT) for the treatment of circumscribed choroidal hemangiomas (CCH).

Materials and Methods: Forty-eight patients (48 eyes) with CCH were treated between 1994 and 2014. A historical series of 20 patients treated with PT since 1994 was compared to 28 patients treated with PDT predominantly used after 2006. Tumor and functional outcomes were compared.

Results: Compared to PDT patients, patients undergoing PT were younger (49 versus 59 years-old, p=0.038), had paramacular location in 80% versus 32% (p=0.015), thicker CCH (3.3 versus 2.8 mm, p=0.022). Baseline associated findings and visual acuity were similar. Mean follow-up was 49 months (78 and 30 months for PT and PDT, respectively (p<0.001)).There was a higher complication rate (p=0.006) or retreatment with PDT than PT, respectively (p=0.044). Mean final logMAR visual acuity was 0.3 and 0.4 mm in patients treated with PT and PDT, respectively (p=0.087). There was a mean 67% and 32% CCH thickness decrease after PT and PDT respectively (p=0.002).

Conclusion: PT may be a valuable treatment option in CCH. Better visual acuity and lower complication rate were observed after PT than with PDT. Initially promising results with PDT are challenged by PT. PT being more invasive because of the need for clips placement onto the sclera, PT may be proposed at first post-PDT relapse. The better tumor and trend for visual outcomes need to be confirmed with larger patient datasets, ideally within prospective trials to determine the best initial treatment.

PTC17-0051: Proton Beam Therapy for Chordomas and Chondrosarcomas of Skull Base

M. Tseytlina1, Y. Luchin1, A. Agapov1, V. Gaevsky1, E. Kizhaev1, G. Mytsin1, A. Molokanov1, S. Shvidkij1, K. Shipulin1

1Joint Institut for Nuclear Research, Laboratory of Nuclear Problems, Dubna, Russian Federation

Purpose: Estimate efficiency and safety proton beam therapy of chordomas and chondrosarcomas of skull base at the Joint Institute for Nuclear Research.

Materials and Methods: From November 2002 through April 2016, 29 patients with skull base chordomas (n=27) and chondrosarcomas (n=2) were treated by fractionated proton beam therapy. Only 7% of patients had received gross total resection of the tumor. The mean tumor volume was 42 cc (range, 3,9 cc to 154cc). The mean isocenter dose was 73 GCE (range, 63 CGE to 80 CGE). Target volume (PTV) included 80% – 90% isodose line. The mean dose to the surface of the brainstem was 62 CGE (56,6 - 64 CGE). The chiasm of the optic nerves received at mean of 46 CGE (9 - 56 CGE).

Results: The mean follow- up period was 66 months (range, 7-168 months). At the time of analysis 7 (24%) were missed for follow up, nineteen patients (66%) had tumor control, three (10%) patients had local marginal recurrence.

Complications: four patients (17%) had 2 grade RTOG scale toxicity from the mucous membranes and skin in the area of radiation fields. There was no evidence of brainstem and visual toxicities or late radiation damage.

Conclusion: Proton therapy in our experience is effective and relatively safe mode of treatment skull base chordomas and chondrosarcomas.

PTC17-0056: Intensity Modulated Proton Therapy (IMPT) of Patients with Brain and Skull Base Tumors - Acute Toxicity and Early Results

I. Gulidov1, V. Galkin1, K. Gordon1, D. Gogolin1, O. Lepilina1, A. Kaprin1, Y. Mardynsky1, S. Ulianenko 2

1A. Tsyb Medical Radiological Research Centre - branch of the National Medical Research Radiological Centre of the Ministry of Health of the Russian Federation, Radiation Oncology, Obninsk, Russian Federation 2A. Tsyb Medical Radiological Research Centre - branch of the National Medical Research Radiological Centre of the Ministry of Health of the Russian Federation, Radiation Biophysics, Obninsk, Russian Federation

Purpose: To report acute toxicity and the first clinical outcomes of intensity modulated proton therapy of patients with brain and skull base tumors provided on medical proton complex “Prometheus”.

Material and Methods: 39 patients with brain and skull base tumors (meningiomas, hemangiopericytomas, pituitary adenomas, low-grade gliomas, chordomas, chondrosarcomas, brain metastases) were treated with visual-guided intensity-modulated proton therapy between November 2015 and December 2016 on medical proton complex “Prometheus” by specialists of Medical Radiological Research Center. Mean age was 49.3 years old. Most of patients (18 cases) were treated in moderate hypofractionated regimen (46 -57.5Gy/20-25 Fr), 17 patients – by conventional schedule (50.4-63Gy/25-35 Fr), in 4 cases we used radiosurgery and stereotactic radiotherapy (31.8Gy/7Fr, 16.4-21.8Gy/1 Fr). Re-irradiation underwent 5 patients; in 1 case (hemangiopericytoma of left orbit) it was 4th course of radiotherapy. Acute toxicity and short-term results were evaluated.

Results: All patients finished proton therapy without interruptions. Follow-up time is from 1 to 12 months. Time of irradiation was from 1 to 16 minutes per fraction. Acute toxicity was presented by local alopecia (19 patients) and dermatitis (RTOG/EORTC grade 1-2 in 16 cases, grade 3 in 2 patients). There were no accidents of severe acute toxicity. Tumor size was stabilized in 26 (66.7%), in 9 (23.0%) cases it was partial response, in 4 (10.3%) patients complete response was registered. No evidence for late toxicity up to now.

Conclusion: the first clinical outcomes show good tolerance and feasibility of IMPT on “Prometheus”. The early results look promising. However, longer follow-up time is needed.

PTC17-0068: Proton Therapy for Glioblastoma Multiforme in Literature

Y.J. Huang1, H.C. Hsu1, J.T. Ho 2

1Kaohsiung Chang Gung Memorial Hospital, Radiation Oncology, Kaohsiung, Taiwan- Province of China 2Kaohsiung Chang Gung Memorial Hospital, Neurosurgery, Kaohsiung, Taiwan- Province of China

The prognosis of glioblastoma multiforme (GBM) is poor. The standard treatment of GBM is combined radiochemotherapy with temozolomide (TMZ) with 60 Gy using photon, but it usually recurs at its original location. A large prescribed radiation dose would improve local tumor control and patient survival; however, it may increase the frequency and severity of radiogenic impairment in brain. Dose escalation up to 90 cobalt gray equivalent (CGE) with conformal protons and photons prevented central recurrence in almost all cases, and the median survival extended to above 20 months. Although radiogenic impairment of the brain was inevitable, it had been well controlled and patients maintained a stable performance status without hospital care. These reports support and confirm the tolerate and genefit of concomitant boost proton radiotherapy in GBM. For single modality radiation therapy, the dosimetric comparison of intensity-modulated proton therapy and volumetric-modulated arc therapy have showed an essential dose reduction while maintaining equal target volume coverage using proton technique. With the advance of proton therapy, intensity-modulated proton therapy with simultaneous integrated boost for dose escalation may be safe and feasible to improve the local control and patient's survival in GBM.

PTC17-0089: Significant Reduction of Doses to Organs at Risk, Including Hippocampus, Is Made Possible with Proton Therapy for Low Grade Gliomas

P. Witt Nyström1,2, P. Bergström 2,3, A. Flejmer 2,4, T. Herlestam5, E. Tegnelius6, K. Werlenius7

1Uppsala University Hospital, Oncology, Uppsala, Sweden 2The Skandion Clinic, Skandion, Uppsala, Sweden3Umeå University Hospital, Oncology, Umeå, Sweden4Linköping University Hospital, Oncology, Linköping, Sweden5Karolinska University Hospital, Oncology, Stockholm, Sweden6Örebro University Hospital, Oncology, Örebro, Sweden7Sahlgrenska University Hospital, Oncology, Gothenburg, Sweden

Low-grade glioma (LGG) patients with good prognostic markers and expected long-term survival are considered candidates for proton therapy (PT) as an alternative to photon radiotherapy to reduce toxicity. Decision on treatment modality is in Sweden made on a national PT board where comparative plans including photon plans are discussed. The Skandion Clinic (SC) is a national PT facility based on distributed competence, i.e. the patients are prepared for PT at one of the seven university hospitals and thereafter referred to SC, which executes the PT, and then re-referred to their home hospital for follow-up.

The LGG patients were identified and allocated for RT at regional neuro-oncological multi-disciplinary tumor boards. Comparative photon plans were made additionally. The patients were treated with pencil beam scanning PT and assessed prospectively for toxicity with first MRI 3 months after PT. Since clinical start, 36 adult patients with a LGG grade II have been treated and evaluated by MRI post PT. All had pathology-verified WHO grade II glioma and treated to a total dose of 50.4– 54.0 Gy (RBE).

The dominating acute side effects were alopecia, erythema and fatigue grade I-II. At MRI 3 months after PT so far 5 patients have shown a new contrast-enhancing lesion and are under surveillance for progressive disease alternatively pseudo-progression. Comparisons between the PT-plans and the conventional plans showed substantial reductions of radiation dose to several organs at risk, including brain (>30%).

PT of LGG showed a well-tolerated treatment. Noticeable are reports of asymptomatic pseudo-progression on MRI post treatment.

PTC17-0103: Selection Strategy of Patients to Validate the Clinical Benefit of Protons Compared with Photons at the Skandion Clinic

B. Sorcini1, E. Lindbäck1, J. Zimmerman1, M. Gubanski 2

1Karolinska University Hospital, Department of Medical Physics, Stockholm, Sweden 2Karolinska University Hospital, Department of Oncology, Stockholm, Sweden

The Skandion Clinic is a national project that is run jointly by the seven county councils with university hospitals in Sweden. The clinic offers proton therapy, with Pencil Beam Scanning (PBS), to cancer patients from all over Sweden. The decision regarding the clinical benefit of protons compared to photons is made on a national video conference for each individual patient who might be eligible for proton therapy. The translation of observed differences in the dose distributions between protons and photons into clinical benefit is determined by the oncologist based on the following criteria:

  • Target location and the distance to organs at risk (OAR).

  • Tumor stage: Primarily low grade tumors in patients with long expected survival might be beneficial for proton radiotherapy

  • Patient performance status

  • Patient Age: Integral dose and risk of secondary malignancy

  • Dose Volume Histogram (DVH) based comparisons to show the differences between proton versus Volumetric Modulated Arc Therapy (VMAT) photon plans

  • Dose plan robustness evaluations and the OAR toxicity endpoint limitations

  • Target motion and necessity of motion control.

  • Tumor vicinity to air cavities or implants.

Based on the results from numerous dose-planning cases, comparing differences between photon and proton dose distributions, a summary of reasons and analysis for rejected proton therapy patients planned at the Karolinska University Hospital will be presented.

From our clinical experience for patients receiving proton therapy, potential cases and reasons for re-planning during the course of treatment will also be presented.

PTC17-0124: NTCP-Based Prediction of Maculopathy after Proton Therapy or Ruthenium-106 Brachytherapy for Choroidal Melanomas

J. Thariat1, J. Folke Kiilgaard 2, J.P. Caujolle3, C. Maschi3, A. Lindegaard Appelt4, J. Herault5, C. Alfast Espensen 2

1Centre Lacassagne, Radiation Oncology, Nice, France 2Rigshospitalet, Ophthalmology, Copenhagen, Denmark3CHU, Ophthalmology, Nice, France4Leeds Institute of Cancer, Statistics, Leeds, United Kingdom5CAL, Radiation Oncology, Nice, France

Purpose: Normal tissue complication probability (NTCP)-models can be used to predict radiation-induced complications. We built NTCP-models for maculopathy following proton therapy (PT) or Ruthenium-106 brachytherapy (BT) of choroidal melanomas.

Materials and Methods: Consecutive patients treated with PT (52Gy/4 fractions, Eyeplan) or BT (100Gy to tumor apex in 6 days, Plaque Simulator). Patients were included if data on dose to the macula and occurrence of maculopathy were available. Logistic regression was used to fit the NTCP-models using the statistical software R. A secondary PT/BT matched-pair analysis, using age, sex, tumor height, and distance between tumor and macula was performed.

Results: Median follow-up for BT (90 pts, 2005-2008), PT (2207 pts, 1991-2015) and matched-PT cohort (109 pts, 2010-2011) were 7.4, 4.3 and 3.1 years. Median (mean) maximum dose to the macula for BT, PT and matched-PT were 19 (64), 50 (29) and 52 (32) Gy. Maculopathy and severe maculopathy rates were 49% for BT, and 6% or XX% for PT and matched-PT cohorts. A steep increase was seen after 40Gy in PT patients. Maximum dose to macula and maculopathy were significantly correlated in the BT NTCP-model for maculopathy, with odds ratio (OR) for 10Gy-increases of 1.9, with the PT matched PT-cohort models for severe maculopathy with OR 1.2 and 1.1.

Conclusion: Maculopathy rates were different between PT and LDR-BT. The NTCP-model for maculopathy showed marked dose-response relationship with BT and less pronounced response with PT. Study limitations include lack of standardized reporting between ophthalmologists and intrinsically different PT/BT populations. PT maculopathy rates may increase with longer follow-up.

PTC17-0137: Radiation-Induced Lymphopenia in Patients with Glioblastoma Multiforme Treated with Proton versus Intensity-Modulated Photon Radiotherapy

R. Yeung1, J. Rockill1, E. Ford 2, L. Halasz1

1University of Washington, Radiation Oncology, Seattle, USA 2University of Washington, Radiation Oncology- Medical Physics, Seattle, USA

Purpose: Treatment-induced lymphopenia (TIL) has been associated with inferior survival for patients with glioblastoma multiforme (GBM). A previously published model has shown that conventional photon plans deliver potentially lymphotoxic doses to the circulating blood pool. As proton therapy (PBT) is associated with smaller low dose volumes, we compared the difference in TIL compared to intensity-modulated photon radiotherapy (IMRT).

Materials and Methods: We reviewed patients treated with concurrent temozolomide and PBT or IMRT for GBM between 2013-2016. Patients were treated to 54.9-60 GyE in 30-33 fractions with planning treatment volumes (PTV) <600 cc. Total lymphocyte count (TLC) were collected at baseline, weekly during treatment and 1 month post-treatment. Proportions and medians were compared using Fisher exact and Mann-Whitney U tests, respectively.

Results: Nineteen patients were treated with PBT and 32 with IMRT. Age, gender and PTV volume were similar between the groups. Baseline median TLC were also comparable (p=0.29). IMRT patients had a significantly lower median TLC at week 4 compared to PBT patients (1.24 x109/L vs 0.96 x109/L, p=0.04) but this difference did not persist. By week 6, 44.8% of IMRT patients developed grade 2 or greater lymphopenia compared to 25% of PBRT patients, but this was not significant (p=0.22).

Conclusion: GBM patients treated with PBT and IMRT experienced similar rates of TIL. However, PBT patients had a delayed development of TIL compared to IMRT patients, suggesting that radiation modality affects TIL. Modelling studies are currently underway to explore differences in dose delivered to circulating blood pool with these different techniques.

PTC17-0237: Lens Avoidance during Proton Therapy for Ocular Melanomas to Prevent the Risk for Cataract

J. Thariat1, J. Salleron 2, S. Jacob3, J.P. Caujolle4, C. Maschi4, J. Herault5, A. Carnicer5

1Centre Lacassagne, Radiation Oncology, Nice, France 2CAV, Statistics, Nice, France3IRSN, Statistics, Toulouse, France4CHU, Ophthalmology, Nice, France5CAL, RO, Nice, France

Purpose: According to the ICRP, cataracts occur after 0,5 Gy to the lens. Volume-effects doses are irrelevant to photons. Since part of the lens can be spared with PT, the risk for cataractogenesis might be different. We investigated lens dose/volume-effects after ocular PT.

Materials and Methods: Between 1991 and 2015, 1717 ocular melanoma patients without cataracts were consecutively treated with proton therapy. Lens opacifications (=cataracts) were assessed on slit lamp examination and severity assessed on vision impairment (=severe cataracts). Dose-volume relationships for cataracts and severe cataracts were assessed using multivariate Cox model. Results are described with hazard ratio and 95% confidence interval.

Results: The median follow-up was 48 months. At 60 months, the cumulative incidence of de novo cataracts and severe cataracts was 19.5% and 12.8%, respectively. Mean values of mean lens doses and lens periphery doses (potential surrogate for reponse of epithelial cell lining) were 1.1 Gy and 6.5 Gy, respectively. The lens received 0 Gy in 25% of the patients. After adjustment, the risk for severe cataracts was significantly increased for patients with a mean lens dose of 5 to 10 Gy (2.44 [1.04;5.77]) and for patients with a mean lens dose over 10 Gy ( 3.57 [1.80;7.07]) compared to patients with a mean lens dose < 0.5 Gy. A mean lens periphery dose above 10 Gy was also a risk factor for severe cataracts (3.77 [1.59;8.89]).

Conclusion: Sparing some volume of lens from exposure to proton therapy may reduce the risk of lens opacification and severe radiation-induced cataracts.

PTC17-0289: MRI/CT-Based GTV Delineation for Treatment of Ocular Melanoma with Proton Therapy

L. Halasz1, C. Bloch1, S. Bhatia 2, T. Klesert3, A. Skalet4, C. Wells3, A. Fung5, R. Rengan1

1University of Washington/Seattle Cancer Care Alliance Proton Therapy Center, Department of Radiation Oncology, Seattle, USA 2University of Washington/Seattle Cancer Care Alliance, Medical Oncology, Seattle, USA3Vitreoretinal Associates of Washington, Ophthalmology, Seattle, USA4Oregon Health Sciences University, Department of Ophthalmology, Portland, USA5Seattle Cancer Care Alliance Proton Therapy Center, Dosimetry, Seattle, USA

Purpose: Most proton centers currently use model-based planning for eye tumors. However, with advancements in imaging and dosimetry, magnetic resonance imaging (MRI) and computed tomography (CT) based planning may allow for more accurate dose calculation and avoidance of normal structures. Here, we present our initial experience and compare tumor volumes (TV) predicted by ultrasound with gross tumor volume (GTV) delineation based on MRI/CT.

Materials and Methods: We reviewed 62 consecutive patients treated for ocular melanoma in 2015 and 2016. B-scan ultrasounds were performed for transverse (T), longitudinal (L), and height (H) tumor measurements in 55 patients. The ellipsoid formula (Π/6) x T x L x H was used to estimate TV. MRI/CT-based GTV delineation was based on the union of MRI, CT, ultrasound, and funduscopic photographs. The GTV was expanded 2 mm for planning treatment volume (PTV). The least squares method was utilized to calculate a linear equation comparing these values.

Results: Among patients with ultrasound measurements, the median MRI/CT-based GTV size was 0.29 cc (range, 0.05-1.77 cc) and the median PTV was 1.27 cc (range, 0.49-4.36 cc). The median TV estimated by ultrasound was 0.15 cc (range, 0.02-1.40 cc). The linear coefficient of determination (R 2) for the overall volume was 0.83.

Conclusion: Here we present our initial experience utilizing CT/MRI based proton planning for ocular melanoma. Utilizing these imaging modalities, contoured GTVs were generally larger than the TV estimated by ultrasound. There was overall a significant correlation between ultrasound and MRI/CT-based tumor delineation. Analyses of dimension by dimension concordance are ongoing.

PTC17-0297: Norwegian Patients Treated Abroad with Particle Therapy – Experience from Norwegian South-East Health Region, 2014-2016

B.L. Rekstad1, T. Martens 2, C. Ramberg1, T. Furre1, M.E. Evensen 2, P. Brandal 2, H. Magelssen 2, E. Waldeland1, T.P. Hellebust1,3, E. Dale 2

1Oslo University Hospital, Department of Medical physics, Oslo, Norway 2Oslo University Hospital, Department of Oncology, Oslo, Norway3University of Oslo, Department of Physics, Oslo, Norway

Purpose: Proton therapy is planned to be established in Norway. Meanwhile, patients in need of particle therapy according to national guidelines are referred for treatment abroad. The South Eastern region in Norway has highest population density and is by far referring most Norwegian patients for particle treatment. We present number of patients, age, diagnoses, treatment sites and treatment delivered over the last three years, 2014-2016.

Materials and Methods: Treatment plans and records were provided by treating institutions. When possible, anonymized CT for planning, contoured structures, plans and doses were imported into a treatment planning system with research license for protons. Patient data was retrieved and compiled from an anonymized database.

Results: In total 19, 36 and 30 patients were treated abroad with particles in 2014, 2015 and 2016, respectively. Patients were treated at five different institutions. The number of children and youth up to 16 years of age was about the same in this period with 8, 10 and 9 treated in 2014, 2015, and 2016. Young adults, age 17 – 39, increased in number from 4 in 2014 to 14 in both 2015 and 2016. A number of different diagnoses were treated, with base of skull as the most prominent treatment region. Other regions included eye, pelvis and head and neck.

Conclusion: Treatment of patients with particle therapy abroad is well established and ensures proper care for Norwegian patients. Several diagnoses and age groups are treated, and solid documentation of treatment is provided by treating institutions.

PTC17-0381: Feasibility of Spot-Scanning Proton Arc Radiosurgery of a Single Peripheral Brain Metastasis

K. Sura1, P. Kabolizadeh1, I. Grills1, A. Maitz1, X. Li1, J. Zhou1, D. Yan1, C. Stevens1, P. Chinnaiyan1, X. Ding1

1Beaumont Health System, Radiation Oncology Proton Therapy Center, Royal Oak, USA

Purpose: To determine the feasibility of Spot-Scanning Proton Arc (SPArc) radiosurgery using Gamma Knife radiosurgery (GKRS) and intensity modulated proton therapy (IMPT) as a baseline.

Materials and Methods: A previously treated peripherally located brain lesion using GKRS was planned using 3-field robust optimized IMPT and SPArc technique. GKRS was planned according to institutional standards with stereotactic frame setup. Robust optimization parameters were set to ±3.5% range and ±1mm setup uncertainties in x,y,z directions (21 worst case scenario optimizations). Gross Target Volume (GTV) was prescribed to 18 Gy [RBE] to 100% of volume for IMPT and SPArc plans. Multiple parameters were used to compare the plans including conformity Index (CI) [target volume covered by 18Gy/total volume covered by 18Gy], gradient index (GI) [V50%RX/V100%RX], mean Brain dose, V1Gy, V2Gy, V5Gy and V12Gy.

Results: Isodose comparisons and the dose-volume histograms of the three techniques were compared. The GKRS and SPArc techniques had similar parameter comparisons. In general, SPArc had decreased or similar integral brain doses compared to GKRS technique. The 3-field IMPT plan was inferior compared to SParc plan.

Conclusion: SPArc technology offers a potential for proton-based stereotactic radiosurgery. The new finding will help our society to rethink of the current dosimetric ceiling using Pencil Beam Scanning technique.

PTC17-0485: Searching for an Optimized Irradiation Technique for Melanoma of the Conjunctiva

W. Sauerwein1, T. Juliette 2, S. Scholz3, M.L. Peyrichon 2, A. Stang4, A. Wittig5, K.P. Steuhl3, D. Flühs6, L. Brualla1, J. Herault 2

1University Hospital Essen, NCTeam- Department of Radiation Oncology, Essen, Germany 2Centre Antoine-Lacassagne, Centre de Protonthérapie, Nice, France3University Hospital Essen, Department of Ophthalmology, Essen, Germany4University Hospital Essen, Center of Clinical Epidemiology- Institute of Medical Informatics- Biometry and Epidemiology, Essen, Germany5Philipps-Universität Marburg, Klinik für Strahlentherapie und Radioonkologie, Marburg, Germany6University Hospital Essen, Department of Radiation Oncology, Essen, Germany

Purpose: Conjunctival melanoma are rare and a heterogeneous group of tumors with different outcome depending from their origin (primary acquired melanosis with atypia PAM, conjunctival nevi or de novo) and the primary tumor localization. Non-bulbar tumors have a higher risk for local recurrence and regional metastasis and have a lower survival probability than tumors with bulbar location. Furthermore, localized and limbal tumors have a substantially better prognosis than multifocal, diffuse and non-limbal tumors regarding to local recurrence, metastatic disease and overall survival rate. This might also be due to the fact, that the bad prognosis criteria are resulting challenging target volumes for radiation therapy making it difficult to deliver the requested dose at the right place.

Materials and Methods: In order to obtain an objective set of data for selecting the optimal radiotherapy technique in a given situation, dose distributions were calculated for different irradiation modalities: brachytherapy with I-125 and Ru-106 plaques, proton beam irradiation with low-energy beams dedicated for ophthalmic tumors and degraded high energy beams including the use of bolus combined with individual compensators.

Results: Proton beams are superior as compared to brachytherapy techniques with respect of the conformal covering of the clinically defined target at surface. Assuming that a reduction of the irradiated intraocular volume results less side effects, proton therapy using bolus and individual compensators may be optimal for treating advanced and extended melanoma including multifocal disease.

PTC17-0530: Utilization of Cone Beam Computed Tomography (CBCT) for Inter-Fraction Cyst Assessment in Patients Undergoing Proton Radiotherapy for Craniopharyngioma

L. Pater1, R. Vatner1, R. Scott 2, Y. Zhang1, J. Breneman1, A. Mascia1

1University of Cincinnati, Radiation Oncology, Cincinnati, USA 2University of Cincinnati, College of Medicine, Cincinnati, USA

Purpose: The majority of craniopharyngiomas include a cystic component and up to 40% of these will have changes in their cyst size during radiotherapy requiring adaptive re-planning. T2 MRI is the gold standard for re-assessment of cyst dimensions but this introduces costs and complexity, especially if the patient requires anesthesia. We studied the feasibility of CBCT for cyst assessment during proton radiotherapy.

Materials and Methods: Patients undergoing proton radiotherapy for craniopharyngioma were studied with T2 MRI, simulation CT, low-dose CT and CBCT for treatment planning and during their treatment course. All images were co-registered with the simulation CT, and cystic tumor components were contoured on these scans. New volumes were compared to those delineated for treatment planning.

Results: Three patients were assessed with CBCT during proton radiotherapy. Only one patient with extensive peripheral cyst wall calcification had a cyst volume that could be reliably assessed. For this patient, CBCT volumes (21.6cm3-28.1cm3) mirrored increases noted on fan beam CT (23.1cm3-27.5cm3) from planning CT and MRI (23.0cm3) necessitating adaptive replanning.

Conclusion: CBCT as currently implemented is not sufficient for assessing inter-fraction cyst volume changes in the majority of patients undergoing proton radiotherapy for craniopharyngioma. Further optimization will be required before this modality can replace fan beam or MRI.

PTC17-0531: Utility of Upfront Proton Radiotherapy in the Treatment of High-Grade Glioma to Facilitate Re-Irradiation at Recurrence

L. Pater1, Y. Zhang1, R. Vatner1, T. Struve1, J. Breneman1, A. Mascia1

1University of Cincinnati, Radiation Oncology, Cincinnati, USA

Purpose: Malignant gliomas frequently relapse after radiotherapy, and salvage therapy with re-irradiation is limited by dose to normal tissues from initial radiotherapy. This study evaluates the potential advantage of upfront proton therapy for malignant gliomas for facilitating safe re-irradiation at the time of tumor relapse.

Materials and Methods: Patients with malignant glioma who received re-irradiation for recurrent tumor after initial radiotherapy at the University of Cincinnati between October 2012 and December 2016 were identified from a departmental database. All patients received photons for initial and salvage therapy, and retrospectively planned for proton radiotherapy for initial treatment and re-irradiation using the CT simulation datasets. Proton and photon plans were compared to assess differential dose to previously uninvolved areas and critical structures (brainstem, optic apparatus and cochlea).

Results: Twenty-four patients (median age = 58, range 15-76) underwent re-irradiation to a median dose of 30.6 Gy (range 18-69.4). Median time between initial and salvage radiotherapy was 21.0 months (range 3.0-104.3). Preliminary comparison between upfront proton vs. photon plans demonstrated similar dose statistics for critical structures within or adjacent to high-dose target regions. Plans with upfront protons improved the dose profiles to contralateral structures. Proton plans for tumors in close approximation to critical structures attained a decrease in initial radiation exposure.

Conclusion: Upfront treatment of malignant glioma with proton radiotherapy results in improved dosimetry to uninvolved normal structures in approximation to target volumes. Select patients may derive a clinical benefit from upfront proton radiotherapy, especially those with tumors in close proximity to critical structures.

Gastrointestinal Organs

PTC17-0113: Improve Heat Effect of Radiochemotherapy Combined with Deep Regional Hyperthermia in Rectal Cancer

T.C. Lin1, C.H. Chen 2, C.Y. Tsai 2, C.C. Chou 2, C.P. Chen 2, C.Y. Lin 2, S.Y. Wu 2

1Wan Fang Hospital- Taipei Medical University, Cancer Center, Taipei City, Taiwan- Province of China 2Wan Fang Hospital- Taipei Medical University, Department of Radiation Oncology, Taipei City, Taiwan- Province of China

Hyperthermia (HT) is most often used in combination with other therapeutic modalities including chemotherapy and radiotherapy, which has been used for cancer treatment since 1910. HT can increase perfusion and oxygen content to tumor, making legion more sensitive to radiotherapy and increasing cells apoptosis. This study was purpose to found a way to improve heat effect of treatment when patients treated with hyperthermia. When patients were undergoing the HT, the temperature of bladder was always higher than rectum. In this study, we used BSD-2000 system with a Sigma Elipse applicator. Our patients who diagnosed with rectal cancer were treated with radiochemotherapy and hyperthermia in the Wanfang Hospital were selected in this study. We distinguished the position of patients between supine and prone. When patients were in supine and prone position, the average temperature of bladder is 39.7°C and 37.4°C. The average temperature of rectum was 39°C and 39.8°C in supine and prone. The results show that when patients were in prone position, the temperature of bladder is lower than rectum. Patients also reply more comfortable when they change the position supine into prone. If the temperature of bladder is lower, we can increase the temperature of rectum higher and enhance effect of hyperthermia.

PTC17-0196: Concurrent Chemoradiotherapy Using X-Ray and Proton Beam Irradiation for Advanced Esophageal Cancer

M. Sasaki1, H. Tamamura1, Y. Maeda1, Y. Tameshige1, H. Nakazawa1, S. Shibata1, Y. Satou1, M. Maeda 2, K. Kume 2, K. Yamamoto1

1Fukui Prefectural Hospital, Proton Therapy Center, Fukui City, Japan 2The WAKASA WAN Energy Research Center, Proton Medical Research Division- Research & Development Dept., Tsuruga City, Japan

In a radical radiotherapy for the advanced esophageal cancer with a high radioresistance, the prevention irradiation to the broad region including the regional lymph nodes should be considered as well as the local irradiation to the primary tumor. However, the prescription dose to the primary tumor is restricted to 50 - 60 Gy in a standard concurrent chemoradiotherapy using X-ray, because of the severe dose constraint for the surrounding organs as lung, spinal cord, and heart. Therefore, in order to prescribe dose of ∼ 70 GyE to the primary tumor satisfying the dose constraint of the surrounding organs at risk, the concurrent chemoradiotherapy, which changed the boost irradiation in the second half of conventional method for proton beam irradiation with high dose concentration, is being performed in our hospital.

For the treatment plannings of 18 cases, by which the total dose of 66 - 74 GyE (X-ray: 36 Gy and proton beam: 30 - 38 GyE) was prescribed to the primary tumor from July, 2014 to September, 2016, a median and a standard deviation of index V95%(CTVprimary) were 99.7 % and 4.7 %, respectively. On the other hand, V20GyE(lung) and Dmax(spinal cord) were 18.9 ± 6.0 % and 42.0 ± 2.0 GyE, and they were enough lower than the dose restriction values indicated by the guideline of NCCN. Therefore, it is expected that the clinical influence to surrounding organs becomes to be reduced by use of this irradiation method, even if high dose is prescribed to the primary tumor.

PTC17-0244: The Time Performance Evaluation of Real-Time-Image Gated Spot-Scanning Proton Therapy System for Clinical Use

S. Shimizu1, T. Yoshimura 2, N. Katoh3, T. Hashimoto3, H. Tamura 2, Y. Matsuo 2, T. Matsuura4, S. Takao 2, K. Umegaki4, H. Shirato3

1Hokkaido University, Radiation Oncology, Sapporo, Japan 2Hokkaido University Hospital, Proton Beam Therapy Center, Sapporo, Japan3Hokkaido University, Radiation Medicine, Sapporo, Japan4Hokkaido University, Division of Quantum Science and Engineering- Faculty of Engineering, Sapporo, Japan

Purpose: In Hokkaido University Hospital Proton Beam Therapy Center, we treat with real-time-image gated spot-scanning proton therapy (RGPT) system for liver, prostate, pancreas, and lung cancer patients. Treatment reservation time is 30 min per patient basically. In this study, we evaluated the on-time performance rate and the treatment process time for RGPT system for clinical usage with or without using gating treatment.

Materials and Methods: This analysis consists of total of 344 patients who had treated with proton therapy in our institution from October 2016 to December 2016. We have been recording the treatment process time using log system in combination with proton beam treatment system provided and in-house made for this purpose. We calculated the time need for treatment and patient set-up using these log systems.

Results: Sixty-six percent of all cases were treated using RGPT system. The mean and SD treatment time for a patient (walk-in to walk-out) was 30.3 ± 8.8 min for RGPT system and 25.9 ± 8.7 min for non-RGPT system, also the patient set-up time was 14.9 ± 5.6 min for RGPT system and 13.8 ± 8.0 min for non-RGPT system. At least, the mean and SD patient beam delivery time for a patient was 6.9 ± 6.1 min for RGPT system and 4.1 ± 2.1 min for non-RGPT system. Each treatment was completed as scheduled for the 90.1% of the treatment session.

Conclusion: The results of the evaluation for the treatment time including using RGPT system show that it is possible to perform on-time as scheduled.

PTC17-0312: Preliminary Toxicities for a Trial of Dose Escalated Proton Radiotherapy with Elective Nodal Irradiation for Pancreatic Adenocarcinoma (UFHPTI PC04)

R. Nichols1, C. Morris1, Z. Li1, S. Huh1, S. Flampouri1, M. Ho1, M. Rutenberg1

1University of Florida, Radiation Oncology, Jacksonville, USA

Purpose: Review preliminary toxicities for a trial of dose escalated proton radiotherapy with elective nodal irradiation for unresectable, borderline resectable, or medically inoperable pancreatic adenocarcinoma.

Materials and Methods: The UFHPTI PC04 trial was activated March 2016. By December 2016, six patients have completed radiotherapy. Protocol therapy delivers 40.50Gy(RBE) in 18 fractions to an initial PTV volume including: an internal gross tumor volume (iGTV); and an elective nodal volume consisting of a 2cm expansion around the most proximal 1cm of the celiac artery and the most proximal 2.5cm of the superior mesenteric artery. A second volume containing the iGTV receives an additional 22.50 Gy (RBE) in 10 fractions subject to normal tissue constraints. Patients receive capecitabine 1000mgPO BID on radiotherapy treatment days.

Results: Median age 77.5 years (range 71 to 88); 4 males, 2 females; T4 – 3, T3 – 3; There were no grade 3 toxicities. Grade 2 toxicities were experienced by two patients. One experienced abdominal discomfort and weakness related to ascites developing during the third week of treatment. One noted grade 2 dermatitis in the final week of radiotherapy. Two experienced interruptions in treatment. One with ascites required a 10 day break for paracentesis. One was hospitalized with urosepsis unrelated to protocol therapy and completed radiotherapy after a 27-day break. Median weight change from the first to sixth week of treatment was -2.3lbs (range +12.3 to -14.2).

Conclusion: The lack of acute treatment related toxicity is encouraging. We continue to enroll patients. An update will be available at PTCOG56 in May 2016.

PTC17-0318: Postoperative Proton Therapy for Pancreatic Cancer Patients Enrolled on the Proton Collaborative Group (PCG) Registry

R. Nichols1, C. Morris1, K. Prabhu 2, L. McGee3, O. Cahlon4, S. Apisarnthanarax5, C. Vargas3, W. Hartsell 6

1University of Florida, Radiation Oncology, Jacksonville, USA 2Procure Proton Therapy Center, Radiation Oncology, Oklahoma City- Oklahoma, USA3Mayo Clinic, Radiation Oncology, Phoenix- Arizona, USA4Procure Proton Therapy Center, Radiation Oncology, Somerset- New Jersey, USA5Seattle Cancer Care Alliance Proton Therapy Center, Radiation Oncology, Seattle- Washington, USA6Northwestern Medicine Chicago Proton Center, Radiation Oncology, Chicago- Illinois, USA

Purpose: The PCG registry is a multicenter registry for patients receiving proton therapy for various malignancies. The current abstract reviews the outcomes for patients receiving postoperative proton therapy for resected pancreatic cancer.

Materials and Methods: From 2/2013 to 6/2015 12 patients with resected pancreatic adenocarcinoma received postoperative proton therapy. The current study reviews the pretreatment characteristics and outcomes of these patients.

Results: Median age - 72 years (range 52 to 79); Males 8, Females 4; T Stage - T2–3, T3–8, T4–1; N Stage - N1–10, N0–0; Margin Status - Close-5, Positive-3, Negative-4; Surgical approach - open-10, laparoscopic 2; Operations performed - standard pancreaticoduodenectomy-7, pylorus sparing pancreaticoduodenectomy- 4, total pancreatectomy-1; Median lymph nodes taken-19.50 (range 11 to 72); Median lymph nodes positive-2 (range 0 to 7); PNI positive-9, PNI unknown-3; LVI positive-6, LVI negative-3, LVI unknown-3; Median tumor size – 3cm (range 2.2 to 6.2); Median dose delivered – 50.51Gy(RBE) (range 27.88 to 54.00); Median treatment duration – 38 calendar days (range 25 to 48); One patient died during treatment. Only one patient's treatment was protracted by more than 5 days. Median available follow up is 0.9 years (range 0.07 to 1.7 years); 1 year survival 54%

Conclusion: Postoperative proton therapy after pancreatectomy was well tolerated without significant toxicity or treatment interruption.

PTC17-0345: Pencil Beam Scanning (PBS) Proton Radiotherapy for Anal Canal Squamous Cell Cancer, Dosimetry Correlates with Clinical Outcome

P. Vitek1, J. Kubes1, V. Vondracek1, S. Vinakurau1

1Proton Therapy Center Czech, Proton radiotherapy, Praha 8, Czech Republic

Chemoradiotherapy is the standard treatment for SCC of the anus. Pencil beam scanning (PBS) proton radiotherapy is able to significantly decrease bone marrow (BM), bladder and small intestine doses compared to IMRT resulting in less grade 3-4 toxicity, less treatment interruptions and lower failure risk. The dosimetry of PBS proton radiotherapy (PRT) was analysed and correlated with toxicity.

PBS PRT of 8 patients, stage T2N0M0 – T3N1M0 is presented. Doses: PTV-1, tumor + margin, involved nodes - 57,5 CGE (Cobalt Gray Equivalents). PTV-2, regional nodes (incl. inguinal), 45 CGE. Both in 25 fractions simultaneously. Concomitant chemotherapy – CDDP+5-FU or CDDP+capecitabine. BM delineated in the range of both PTV. Dose calculated for BM, bladder and abdominal cavity containing intestinal loops.

Tolerance is good with 6 episodes of neutropenia grade 1-2 and 1 episode of neutropenia gr. 4 (not febrile). Gr. 4 neutropenia doesn't correspond with highest doses in BM. 1 treatment interruption was required, for a period of 2 days. Bowel movements grade 1-2 in 5 patients. No urogenital toxicity occurred. All patients achieved complete regression.

PBS PRT maintained a substantial dose reduction in BM and an acceptable risk of significant myelotoxicity below 10%. A small volume (V10%) receiving higher doses has low impact. The doses in intestine and bladder maintain a low risk of toxicity. The dosimetry of PBS PRT supports the indication for anal SCC.

PTC17-0360: Single Fraction Carbon Ion Radiotherapy for Colorectal Cancer Liver Metastasis

H. Makishima1, S. Yasuda1, Y. Isozaki1, G. Kasuya1, N. Okada1, M. Miyazaki 2, S. Yamada1, H. Tsuji1, T. Kamada1

1National Institutes for Quantum and Radiological Science and Technology, National Institute of Radiological Sciences Hospital, Chiba, Japan 2Chiba University, Graduate School of Medicine, Chiba, Japan

Purpose: The objective of this study is to clarify the optimum dose for colorectal cancer liver metastasis.

Materials and Methods: Study design was a dose escalating single arm prospective intervention study. Patients with liver metastasis from colorectal cancer, without any other known metastasis were enrolled. Patients received carbon ion radiotherapy (CIRT) in a single dose, escalating from 36Gy (RBE) until satisfactory local control is reached in 5 to 10 percent increments. Dose-limiting toxicity was defined as grade 3 or severe toxicity attributed to radiotherapy.

Results: A total of 31 patients were enrolled, 29 received carbon ion treatment. Dose prescribed were as following: 36Gy (RBE) 3 cases, 40Gy (RBE) 2, 44Gy (RBE) 4, 46Gy (RBE) 6, 48Gy (RBE) 3, 53Gy (RBE) 8, 58Gy (RBE) 3. Dose-limiting toxicity was observed in 2 patients at 53Gy (RBE), with grade 3 liver toxicity due to biliary obstruction. Both cases had lesions close to hepatic portal region, and therefore dose was escalated to 58Gy (RBE) limited to peripheral lesions. Two-year accrual overall survival was 86%, MST 65 months. Local control improved significantly at 53Gy (RBE) and over, showing 2-year accrual local control rate of 80%, compared to 34% in lower doses. Compared to 60Gy (RBE)/4Fr treatment done at our facility, 2-year accrual local control showed a tendency for better control favoring single fraction (55% vs 80%, p value= 0.334)

Conclusion: Treatment for colorectal cancer liver metastasis with single fraction CIRT appear to be safe and effective.

PTC17-0376: Results of Carbon Ion Radiotherapy for 80 Years and Older Patients with Hepatocellular Carcinoma

T. Abe1, K. Shibuya1, S. Shiba1, H. Katoh1, Y. Koyama 2, H. Shimada3, T. Ohno3, T. Nakano1

1Gunma University- Graduate School of Medicine, Radiation Oncology-, Maebashi, Japan 2Shibukawa Medical Center, Diagnostic Radiology, Shibukawa, Japan3Gunma University Heavy ion Medical Center, Medical Physics, Maebashi, Japan

Purpose: To report the feasibility and efficacy of carbon ion radiotherapy (C-ion RT) for 80 years and older patients with hepatocellular carcinoma (HCC).

Materials and Methods: Eligibility criteria of this retrospective analysis were: 1) HCC confirmed by histology or typical hallmarks of HCC by imaging techniques of four-phase multidetector-row computed tomography or dynamic contrast-enhanced magnetic resonance imaging; 2) no intrahepatic metastasis or distant metastasis; 3) no findings suggesting direct infiltration of the gastro-intestinal tract; 4) performance status ≤ 2 by Eastern Cooperative Oncology Group classification; 5) Child-Pugh classification A or B; and 6) age (80 years and older). Total dose was 52.8 Gy (RBE) or 60.0 Gy (RBE). Toxicities were recorded using the National Cancer Institute's Common Terminology Criteria for Adverse Events (Version 4.0).

Results: Between March 2011 and November 2015, 31 patients were treated. The median follow-up period of all patients was 23.2 months (range: 8.4-55.3 months). Median age at the time of registration of C-ion RT was 83 years (range: 80-95 years). Child-Pugh grade A and B were 27 patients and 4 patients, respectively. The 2-year estimated overall survival, local control, and progression-free survival rates were 82.3%, 89.2%, and 51.3%, respectively. No patients had Grade 2 or higher acute toxicities (within 3 months after C-ion RT). One patient experienced progression in Child-Pugh classification from A to B within 3 months after C-ion RT. Grade 3 encephalopathy was observed in 3 patients with disease progression.

Conclusion: C-ion RT was well tolerated and effective for 80 years and older patients with HCC.

PTC17-0391: Proton Therapy for Pancreatic and Ampullary Carcinoma

R. Nichols1, C. Morris1, L. Tottenham1, Z. Li1, S. Flampouri1, S. Huh1, M.W. Ho1, W. Mendenhall1, N. Mendenhall1, M. Rutenberg1

1University of Florida, Radiation Oncology, Jacksonville, USA

Purpose: Review outcomes for a series of patients treated with proton therapy for non-metastatic pancreatic and ampullary carcinoma.

Materials and Methods: From 2009 to 2016, 59 patients with non-metastatic pancreatic (58) or ampullary (1) cancer were treated with proton therapy (PT). Patients were enrolled on either the UFHPTI PC01, PC02, PC03, PC04 protocols or the IRB approved Outcomes Tracking Protocol (OTP). Median age was 68.7 years (range 45.7 to 89.4). Treatment indications included: unresectable -32; borderline resectable-6; medically inoperable-3; postoperative-8; planned preoperative-9 ; and salvage-1. Median dose in Gy(RBE) for the 6 groups was: 59.4; 63.0; 60.0; 54; 50.4; 59.4 respectively. Most patients received concomitant chemotherapy of capecitabine 1000mg PO BID on treatment days.

Results: Patients underwent toxicity evaluation using CTCAE vs4.0 grading system weekly during treatment. No patient experienced grade 3+ Gastrointestinal (GI) toxicity. 15 patients (25%) experienced grade 2+ GI toxicity at some point during treatment. Median weight loss from start to end of PT was 3.0lbs (range -18.2 to +12.3). Treatment interruptions were minimal with only 6 (10%) patients experiencing more than a 3 day delay for any reason. Median survival for all patients was 10.8 months. Median survival for patients treated with planned postoperative PT was 32.4 months.

Conclusion: Chemoradiation with protons was well tolerated even in the setting of dose escalation.

PTC17-0396: Single Posterior Field Pencil-Beam Scanning Protons for Esophageal Cancer: Preliminary Toxicity and Outcome Analysis and Comparison with Intensity-Modulated Radiation Therapy

R. Yeung1, A. Rodriguez 2, M.W. Macomber1, B.K. Oelschlager 2, F. Farjah3, V. Shankaran4, J. Zeng1, S. Apisarnthanarax1

1University of Washington, Radiation Oncology, Seattle, USA 2University of Washington, General Surgery, Seattle, USA3University of Washington, Cardiothoracic Surgery, Seattle, USA4University of Washington, Medical Oncology, Seattle, USA

Purpose: We previously published the feasibility and dosimetric superiority in normal tissue sparing with a single posterior field pencil-beam scanning (SPF-PBS) protons technique compared to 2-3 field uniform scanning protons approaches in esophageal cancer patients treated with preoperative chemoradiation. We present preliminary toxicity and outcome data with the SPF-PBS technique and compared them to patients treated with intensity-modulated radiation therapy (IMRT) photons.

Materials and Methods: Retrospective analysis of consecutive patients with locally advanced esophageal cancer (T2-4N0-2M0) treated with trimodality therapy (neoadjuvant chemoradiation and esophagectomy) with SPF-PBS or IMRT from 2010 to 2016. Chi-square or Fisher's exact testing assessed associations between RT modality and post-operative complications (pulmonary, gastrointestinal, cardiac, neurologic, infection), and pathologic complete response (pCR). Kaplan-Meier technique was used to calculate overall survival (OS).

Results: Eighteen patients were treated with SPF-PBS and 23 patients with IMRT. Median follow-up was 16 months. Patient characteristics were similar in both cohorts, except SPF-PBS patients were older (68 vs. 61 yrs, p<0.05) and had higher rates of hyperlipidemia (67% vs. 35%, p=0.04). Median radiation dose was 50.4 GyE in both groups. Overall pCR rate was similar (17%). There were no differences in Clavien-Dindo grade 3 or higher surgical complications between groups. Median OS was not reached. One-year locoregional control was 94% and 95% for SPF-PBS and IMRT, respectively (p=0.75).

Conclusion: SPF-PBS clinical outcomes were comparable to IMRT in this studied patient population. Larger patient numbers and longer follow-up may be needed to detect significant differences in perioperative complications between SPF-PBS and IMRT.

PTC17-0404: Hypofractionated Carbon Ion Radiation Therapy for Hepatocellular Carcinoma

K. Shibuya1,2, H. Katoh1, S. Shiba 2, D. Irie 2, M. Okamoto1,2, S. Kakizaki3, Y. Koyama4, K. Shirabe5, T. Ohno1, T. Nakano1,2

1Gunma University, Heavy ion medical center, Maebashi, Japan 2Gunma University Hospital, Department of Radiation Oncology, Maebashi, Japan3Gunma University Hospital, Department of Hepatology, Maebashi, Japan4National Hospital Organization Shibukawa Medical Center, Department of Radiology, Shibukawa, Japan5Gunma University Hospital, Department of Hepatobiliary and Pancreatic Surgery, Maebashi, Japan

Purpose: To evaluate the safety and efficacy of hypofractionated carbon ion radiation therapy (C-ion RT) for hepatocellular carcinoma (HCC).

Materials and Methods: Eligibility criteria of this retrospective analysis were: 1) pathologically or clinically diagnosed HCC; 2) no intrahepatic or distant metastasis; 3) no direct infiltration of the gastro-intestinal tract; 4) performance status ≤ 2 by Eastern Cooperative Oncology Group classification; and 5) Child-Pugh classification A or B. The dose-fractionation schedules were 52.8 or 60.0 Gy (RBE) in 4 fractions for the tumor which is not adjacent to organs at risk (OAR) and 60.0 or 64.8 Gy (RBE) in 12 fractions for the tumor adjacent to OAR. Local control and overall survival rates, and grade 3 or greater toxicities were evaluated.

Results: From September 2010 to June 2016, 87 patients were included in the analysis. Median age was 75 years (range, 45-90 years). Child-Pugh grade A and B were 76 patients and 11 patients, respectively. Mean maximum diameter of the treated lesions was 3.9 cm (0.9-10.4 cm). The medical follow up period was 28.7 months (range: 5.7-72.8 months). The 1-, 2-, and 3-year local control rates were 97.2%, 85.6%, and 77.1%, respectively. The 1-, 2-, and 3-year overall survival rates were 95.1%, 82.2%, and 71.2%, respectively. Grade 3 encephalopathy and Grade 3 acute cholecystitis (calculous cholecystitis, suspected) were observed in two and one patients. Grade 4 or 5 toxicities were not observed.

Conclusion: Hypofractionated C-ion RT were safe and effective treatment for HCC.

PTC17-0427: Chest Wall Toxicity in Hypofractionated Proton Beam Therapy for Primary Liver Malignancies

R. Yeung1, S.R. Bowen 2, T.D. Mullen1, G. MacLennan3, T.R. Chapman1, S. Apisarnthanarax1

1University of Washington, Radiation Oncology, Seattle, USA 2University of Washington, Radiation Oncology, Medical Physics, Seattle, USA3Seattle Cancer Care Alliance, Medical Dosimetry, Seattle, USA

Purpose: The liver-sparing characteristic of proton beam therapy (PBT) allows for dose escalation in treatment of liver malignancies, but may result in high doses to the chest wall. Limited data exist on chest wall toxicity after PBT in these patients. We present the rates and associated factors for chest wall toxicity after hypofractionated PBT.

Materials and Methods: Retrospective study of 36 consecutive primary liver malignancies patients treated with hypofractionated PBT. Chest wall toxicity was scored using CTCAEv4. Logistic regression was performed to identify patient and dosimetric factors associated with incidence of chest wall toxicity.

Results: Thirty-eight liver lesions were treated with a median dose of 60 GyE (35-67.5 GyE) in 15 fractions (13-20 fractions). Median follow-up was 11 months (1-36 months). Grade ≥2 and grade 3 chest wall pain occurred in 6 (17%) and 3 (8%) of patients, respectively. Median time to onset of pain was 7 months from PBT (1-14 months). Chest wall volumes receiving > 63, 77.5, and 84 Gy EQD2α/β=3 were highly associated with incidence of grade ≥2 chest wall pain as were tumors closer to the chest wall and higher tumor EQD2α/β=10. Chest wall EQD2α/β=3 V63 >17cm3, V77.5 >8cm3, and V84 >5cm3 achieved 100% sensitivity and >79% specificity.

Conclusion: Chest wall toxicity after high-dose hypofractionated PBT for liver malignancies is clinically relevant. For a 15-fraction regimen, V50 >17cm3, V57 >8cm3, and V60 >5cm3 were associated with higher rates of chest wall toxicity. Further investigation of PBT techniques such as PBS to reduce chest wall dose are warranted.

PTC17-0438: Carbon-Ion Radiotherapy for Isolated Lymph Node Recurrence from Esophageal Cancer after Curative Resection

Y. Isozaki1, S. Yasuda 2, S. Yamada 2, S. Kawashiro 2, N. Okada 2, H. Tsuji 2, T. Kamada 2

1Hospital of the National Institute of Quantum and Radiological Science and Techn, 4-9-1- Anagawa- Inage Ward- Chiba city- Chiba- 263-0024- Japan, Chiba, Japan 2Hospital of the National Institute of Quantum and Radiological Science and Technology, Hospital, Chiba, Japan

Purpose: Investigation of the treatment potential therapeutic efficacy of carbon-ion radiation therapy for isolated lymph node recurrence from esophageal cancer after curative resection.

Materials and Methods: Carbon-ion radiation for solitary lymph node recurrence from esophageal cancer was performed in 11 cases from May2000 to October 2016 in our institute. 48.0 Gy (RBE) was delivered with a daily dose of 4.0 Gy(RBE) over 3weeks. We evaluated the anti-tumor effect, adverse events and outcomes.

Results: A total of 11 patients were recruited and their characteristics were as follows: 10 patients postoperative recurrence, one patient recurrence after endoscopic resection; one recurrent site in neck lymph node field, nine in mediastinal, one in abdominal. Four patients (preoperative radiation: 3, prophylactic radiation: 1) had already received radiotherapy for irradiation field. The median follow-up duration for all patients was 27.7 months (range, 3-92.0 months). Local recurrence occurred in only one patient. Recurrence in prophylactic irradiation field occurred in two patients. Regional recurrence out of irradiation field occurred in two patients and salvage carbon-ion re-irradiation were performed in both patients. There were 11 cases (78.6%) who achieved complete or partial response after treatment. The local control rates at 5 years were 91.7%. The overall survival rates at 3 and 5 years were 62.5% and 20.8%, respectively. The median survival period was 45.3 months. There were no toxicities of Grade 3 or higher.

Conclusion: Carbon-ion radiation therapy might be a safe and effective treatment option for isolated lymph node recurrence from esophageal cancer after curative resection, including re-irradiation cases.

PTC17-0467: Proton Beam Therapy for Hepatocellular Carcinoma with Extensive Portal Vein Tumor Thrombosis

K. Miura1, T. Okumura1, N. Fukumitsu1, M. Mizumoto1, H. Ishikawa1, K. Onishi1, T. Aihara1, T. Sakae1, K. Tsuboi1, H. Sakurai1

1University of Tsukuba, Proton Medical Research Center, Amakubo- Tsukuba City, Japan

Purpose: To evaluate the efficacy of proton beam therapy (PBT) for hepatocellular carcinoma with extensive tumor thrombosis in the main trunk (Vp 4) or major branches (Vp 3) of the portal vein.

Materials and Methods: Ninety-seven patients were analyzed who were treated by PBT between August 2009 and March 2015. There were 82 men and 15 women, and the median age was 65 years old (range, 27-88). Thirty-four patients were previously untreated, and 63 had recurrent tumors. Degree of liver function was Child-Pugh class A in 71, and class B in 26 patients. Fifty-four patients had Vp 3 lesion and 43 patients had Vp 4 lesion. The delivered total dose ranged from 70 to 80.47 Gy10 (median: 80.47 Gy10) in terms of equivalent dose in 2 Gy fractions. The most frequently prescribed dose was 72.6GyE with 22 fractions.

Results: Median follow-up time was 14.2 months. Median survival rate for all the patients, Vp 3 patients, and Vp 4 patients were 12.9, 23.2, and 9.7 months, respectively. MST for the patients treated with PTV that encompassed all the detectable lesions was 27.6 months, and MST for the patients who had viable tumor outside of their PTV was 7.8 months. Multivariate analysis revealed existence of viable tumor outside of the PTV, clinical stage, and value of des-gamma-carboxy prothrombin as significant factors affecting the OS.

Conclusion: PBT was effective for patients with extensive portal tumor thrombus, if the PTV encompassed all the detectable lesions.

PTC17-0501: Impact of Iodized Oil on Proton Dose Distribution for Liver Irradiation Post TACE Treatment

J. Zhang1, L. Liu 2,3, C. Sun1, S. Huang1, H.M. Lu 2,3

1Zibo Wanjie Cancer Hospital, Radiation Oncology, zibo, China 2Massachusetts General Hostipal, Radiation Oncology, Boston, USA3Harvard Medical School, Radiation Oncology, Boston, USA

Purpose: Transcatheter arterial chemoembolization (TACE) is frequently used for hepatocellular carcinoma (HCC) treatment, but often with unsatisfactory local controls. Radiotherapy after TACE improves outcomes and proton therapy could spare more liver tissues, which is especially important for patients with cirrhotic conditions. We evaluated the effects of iodized oil deposits in the liver on proton dose distributions and explored mitigation techniques.

Materials and Methods: The relative stopping power (RSP) for the iodized oil was measured by the method of range pullback. CT scans for 10 patients after TACE treatments were acquired, with the regions of iodized oil deposits contoured and the distribution of CT Hounsfield unit (HU) values assessed. Proton treatment plans were designed and dose computations were performed and compared with and without RSP overridden for regions with iodized oil deposits.

Results: The CT HU of pure iodized oil was around 3000, corresponding to a RSP of 2.5 according to the standard HU-RSP conversion. However, the measured RSP was only 1.12. For iodized oil deposits in liver tissue, the CT HU value has broad distributions with values up to 3000.

Conclusion: The actual effects on the proton dose distribution in patients were found to be relatively small and can be well managed by RSP override of contoured regions containing iodized oil deposits.

PTC17-0540: Intensity Modulated Proton Therapy for Re-Treatment of Rectal and Anal Cancer

F. Giap1, R. Lepage 2, A. Waldinger 2, H. Giap 2

1University of Texas at Southwestern, Medical School, Dallas, USA 2Scripps PTC, Scripps Proton Therapy Center, San Diego, USA

Purpose: This retrospective review represents the clinical experience using IMPT technique for the re-treatment of rectal and anal cancer.

Materials and Methods: 7 patients (5 for curative and 2 for palliative) with previously irradiated recurrent rectal or anal cancer underwent IMPT at SPTC. All patients underwent CT and MRI based simulation. PET/CT was fused for planning. Patients were set up in the supine position with rectal balloon. Daily setup was done with orthogonal kV x-rays and CBCT. Daily fraction of 1.8-2 Gy for 50-66 Gy prescribed to the GTV and CTV (tumor seen on CT/MRI/PET plus a 5 mm) was given. No elective nodal volumes were treated. Dose constraints and beam selection were made for maximum robustness and to minimize the dose to previously treated areas while pushing the dose to the CTV. Treatment was delivered with 1-3 beams using IMPT with Multiple Field Optimization (MFO) technique. All patients were put on a low-gas diet and simethicone. Weekly adaptive simulation with CT.

Results: Maximum and mean doses are as follows: rectum 66 Gy/56 Gy, femoral heads 29 Gy/ 12 Gy, bladder 67Gy/18 Gy and bowel 46 Gy/4 Gy. With a mean follow up time of 18 months, all patients are alive. 3 patients have CR, 2 have PR, and 3 with progression of disease. One patient has grade 3 ulceration requiring HBO steroid.

Conclusion: Re-irradiation using MFO IMPT is technically feasible and potentially an effective approach. Further follow-up and more patient will be needed for assessment of late toxicities and local control.

PTC17-0541: Early Experience with Hydrogel Rectal Spacer with Proton Therapy: Stability Assessment with Weekly MRI

B. Giap1, M. Young1, R. Lepage1, C. Rossi1, H. Giap1

1Scripps PTC, Scripps Proton Therapy Center, San Diego, USA

Purpose: This study describes our early experience with hydrogel (SpacerOAR) rectal spacer for the first 26 patients over the past 12 months with weekly assessment of spacer stability with CT and MRI.

Materials and Methods: Two patients had local recurrence from previous radiation, and the rest were de novo stage T1-2. First 2 patients had DuraSeal, and the rest had SpacerOAR. The implant and 3 fiducial markers were implanted by 4 different physicians. All patients underwent simulation with CT & MRI, and no rectal balloon was used. Treatment was delivered with 2 beams (LL & RL) using IMPT. Dose prescription was either 70-77 Gy in 28 fractions or 80-88 Gy in 40 fractions. Weekly adaptive simulation was done with CT and MRI. The measurements were obtained from CT/MRI: spacer volume, spacer thickness at base, apex, and mid-gland. The data were tabulated for each patient and as the whole group, and were mapped over the course of the treatment.

Results: The first 2 patients with DuraSeal had the spacer disappeared within 3 weeks. In the SpacerOAR group, 2 patients had the spacer mostly disappeared within 4 weeks from the time of insertion. 2 patients had material shift around; the rest has relatively stable geometry. No patients have more than grade 1 GI toxicity.

Conclusion: Care must be taken to have a consistent high quality implant, and monitoring of spacer integrity should be checked during the course of treatment. The use of DuraSeal should be avoided. The most benefit of rectal spacer is seen in patients who had previous radiation.

Genitourinary Organs

PTC17-0026: Clinical Outcomes of Carbon-Ion Radiotherapy for Vulvar Malignant Melanoma

Y. Miyasaka1,2, N. Okonogi1, K. Karasawa3, M. Wakatsuki4, H. Kiyohara 2, S. Kato5, D. Kobayashi 2, T. Nakano 2, T. Kamada1, M. Shozu6

1National Institute of Radiological Sciences, Hospital, Chiba, Japan 2Gunma University Graduate School of Medicine, Radiation Oncology, Maebashi, Japan3Tokyo Women's Medical University School of Medicine, Radiation Oncology, Tokyo, Japan4Jichi medical University, Radiology, Tochigi, Japan5Saitama Medical University International Medical Center, Radiation Oncology, Saitama, Japan6Chiba University Graduate School of Medicine, Reproductive Medicine, Chiba, Japan

Purpose: Malignant melanoma (MM) is radioresistant tumor, demonstrating poor regression after photon radiotherapy. Carbon-ion beams have higher biological effectiveness than photon beams. We retrospectively evaluated the clinical outcomes of carbon-ion radiotherapy (C-ion RT) for vulvar MM.

Materials and Methods: Between December 2007 and July 2015, 14 patients with histologically proven vulvar MM without distant metastasis were treated with C-ion RT in our institution. These 14 patients refused surgery or were judged as inoperable. Median age was 70-year olds (range, 61–88), and their stages are listed. In regard to the dose fractionation schedule, the small pelvic space, including the tumor and the metastatic lymph node, was irradiated with up to a total dose of 36 Gy [relative biological effectiveness (RBE)] followed by a tumor boost of up to a total dose of 57.6 Gy (RBE) in 16 fractions.

Results: The median follow-up time was 24 months (range, 5-66 months). The 2-year local control, disease-specific survival, and overall survival rates were 83%, 33%, and 60%, respectively. There was recurrence in 8 patients, 5 of 8 patients developed distant metastases. As for acute toxicities, Grade 3 mucositis were observed in 2 patients and Grade 3 diarrhea was observed in 1 patient. Whereas, no patient developed Grade 3 or worse late toxicity.

Conclusion: C-ion RT may become a less invasive treatment option for vulvar MM. To reduce the incidence of distant metastasis, combination with C-ion RT and systemic therapy is worth considering.

PTC17-0054: The Initial Experience with Heavy Ion Therapy for the Inoperable Ureteral Cancer

A. Adachi1

1Gunma University, Radiation Oncology, Maebashishi, Japan

Purpose: We report 5 cases with medically inoperable ureter cancer treated with carbon-ion radiotherapy (CIRT) to analyze the therapeutic potential.

Materials and Method: Retrospective chart review was performed for 5 patients with medically inoperable ureter cancer treated with radical CIRT between December 2013 and December 2014.

Results: The median age of the 5 patients was 83 years (range, 68-84 years). The reasons for inoperability was advanced age, post-contralateral nephrectomy, alcoholic cirrhosis, both advanced age and contralateral renal function degeneracy, and pneumonia. The median size of tumor was 2.8 cm (range; 2.2-4.0 cm). Neither lymph node metastases nor distant metastases were not pointed out in all cases by using diagnostic imaging. All patients underwent CIRT (52.8 Gy (RBE) / 12 fractions / 3 weeks). The clinical target volume (CTV) was encompassed the growth tumor volume (GTV) with a 5 mm margin in the bilateral directions, and there was a 40 mm margin in the craniocaudally directions but not whole ureter. With the median follow up time of 21 months (range; 18-24 months), all patients were alive. Neither the local recurrence nor regional lymph node metastases were not observed. The secondary bladder tumor was observed in 2 patients. The Grade 1 hematuria was observed in 2 patients and the Grade 3 pyelonephritis was observed in 1 patient as acute toxicity. The ureteral obstruction was observed in 2 patients as late toxicity.

Conclusion: CIRT might be a useful treatment option for inoperable ureter cancer.

PTC17-0060: Dose Volume Histogram (DVH) Comparison between IMPT vs. VMAT for Irradiation of Pelvic Nodes and Prostate in High-Risk Prostate Cancer

R. Choo1, D. Routman1, H. Schultz1, W. Harmsen1, K. Corbin1, B. Stish1, B. Davis1, T. Pisansky1, T. Whitaker1

1Mayo Clinic, Radiation Oncology, Rochester, USA

Purpose: To compare DVH of the organs at risk (OAR) and clinical target volumes (CTV1: prostate and seminal vesicles; CTV2: regional pelvic nodes) between intensity-modulated proton therapy (IMPT) and volumetric-modulated arc therapy (VMAT) for high-risk prostate cancer.

Materials and Methods: A study of moderate hypofractionation proton therapy treating CTV1 and CTV2 is in progress for high-risk prostate cancer since August 2016. CTV1 and CTV2 receive 6750 cGy and 4500 cGy in 25 fractions over 5 weeks, respectively. Seven accrued patients, as of November 2016, were the subjects for DVH comparison between IMPT and VMAT. Two treatment plans (IMPT and VMAT) were prepared for each patient with pre-defined planning objectives for CTVs, PTVs, and OARs. IMPT plans were prepared with 2 lateral beams, and VMAT plans with 2 arcs.

Results: Coverage of CTV1 and CTV2 was adequate for both plans with 99% of CTVs receiving 100.7 - 101.7% of the prescription doses. Mean doses to the bladder, rectum, large bowel, and small bowel were lower with IMPT vs. VMAT. Mean femoral head dose was slightly higher with IMPT. % volumes of rectum and large bowel receiving V10-45 Gy, and % volumes of bladder and small bowel receiving V10-40 Gy were smaller with IMPT vs. VMAT. IMPT did not produce the typical low-dose ‘‘bath'' to the pelvis seen with VMAT.

Conclusion: IMPT can significantly reduce the dose to OARs while providing adequate target coverage, compared with VMAT, when the pelvic nodes, along with prostate and seminal vesicles, are irradiated.

PTC17-0081: Is Pelvic Node Coverage Adequate When Intraprostate Carbon Markers Are Used in the Daily Set-Up of Intensity-Modulated Proton Therapy (IMPT)?

T. Whitaker1, D. Routman1, W. Harmsen1, H. Schultz1, K. Corbin1, B. Stish1, B. Davis1, T. Pisansky1, R. Choo1

1Mayo Clinic, Radiation Oncology, Rochester, USA

Purpose: To evaluate the impact on pelvic nodal coverage when intraprostate markers are used in the daily set-up of IMPT

Materials and Methods: A study of moderate hypofractionation proton therapy treating CTV1 (prostate and seminal vesicles) and CTV2 (regional pelvic nodes) is in progress for high-risk prostate cancer. CTV1 and CTV2 receive 6750 cGy and 4500 cGy in 25 fractions over 5 weeks, respectively. Seven patients were accrued, and the subjects for this analysis. IMPT plans were prepared with two lateral beams, and delivered with on-line daily matching of intraprostatic markers. Weekly verification CT scans (5 scans/patient) were obtained. CTV1 and CTV2 were propagated from the planning CT to the verification CT scans by matching intraprostatic markers and pelvic bones, respectively. Coverage of CTVs was evaluated on the weekly verification CT scans.

Results: On verification CT scans, the average of mean dose of CTV1 was 6921.2 cGy, while that of CTV2 was 4792.1 cGy. Average V100% (%) of CTV2 was 92.2% of prescription dose, while that of CTV1 was 99.6%. Average D95% (%) of CTV2 was 97.1% of prescription dose, while that of CTV1 was 101.5%. The area of CTV2 most frequently missing the dose coverage was the anterior portion of the mid-to-distal external iliac nodal CTV. This is likely related to CTV1 motion being greatest in anterior-posterior direction.

Conclusion: The use of intraprostatic markers for the daily set-up of IMPT for CTV1 and CTV2 yielded a reasonable coverage of pelvic nodes in a 2- lateral beam arrangement.

PTC17-0090: Preliminary Results of Carbon-Ion Radiotherapy for Prostate Cancer: Prospective Observational Study (GUNMA0702)

H. Kawamura1, N. Kubo1, T. Mizukami1, H. Sato1, A. Adachi1, T. Ohno1, H. Matsui1, K. Ito 2, K. Suzuki1, T. Nakano1

1Gunma University, Heavy Ion Medical Center, Maebashi, Japan 2Gunma University Graduate School of Medicine, Urology, Maebashi, Japan

Purpose: Carbon-ion radiotherapy (CIRT) for localized prostate cancer has been reported effective and safe but only single institutional data was available. We perform prospective observational study to confirm the efficacy and toxicity of CIRT.

Materials and Methods: Between March 2010 and August 2013, 304 patients with prostate cancer were treated with CIRT (57.6 Gy RBE in 16 fractions). Androgen deprivation therapy was given to the patients according to the risk group.

Results: The median age of participants was 67 years. The clinical T stage distribution was T1c-T2b in 169, T2c in 60, and T3 in 75. The median PSA was 7.81 ng/mL. The Gleason score distribution was 6 in 20, 7 in 171, and ≥8 in 113. 16 had low-risk, 142 had intermediate-risk, and 146 had high-risk diseases. The median follow-up period was 29.9 months. Acute grade 1 and 2 gastrointestinal (GI) toxicity was seen in 1.3% and 0%. Late grade 1 and 2 GI toxicity was seen in 7.6% and 0.3%. Acute grade 1 and 2 genitourinary toxicity (GU) was seen in 53.0% and 4.3%. Late grade 1 and 2 GU toxicity was seen in 42.1% and 4.6%. There was no patient experienced grade 3 or more toxicity. According to the Phoenix definition for biochemical failure, biochemical failure occurred in 9 (3.0%) patients. No local failure was observed. Distant failure occurred in 4 (1.3%) patients. There were 2 (0.7%) deaths from prostate cancer.

Conclusion: The current prospective study demonstrated that GI and GU toxicities were safe. Further follow-up will be necessary to confirm long-term efficacy and toxicities.

PTC17-0101: Proton Beam Therapy for the Treatment of Prostate Carcinoma at West German Proton Therapy Center Essen (WPE)

D. Geismar1, T. Steinmeier 2, B. Winckler-Saleske1, P.H. Kramer 2, B. Timmermann1

1Clinic for Particle Therapy- University Hospital Essen, West German Proton Therapy Center Essen WPE, Essen, Germany 2University Hospital Essen, West German Proton Therapy Center Essen WPE, Essen, Germany

Purpose: Proton beam therapy (PT) is an attractive part in tumors with close proximity to critical structures.

Materials and Methods: Between January 2014 and October 2016, 41 patients (age 67.0 y (53.2-78.5 y)) with prostate carcinoma (T1c-T4N0M0) were treated at WPE and were prospectively enrolled in the in-house registry. The patients' Gleason-Scores before irradiation were between 6 and 9 (G6 n=6; G7 n=23; G>7 n=12). PSA levels <10 ng/mL were present in 21, 10-20 ng/mL in 13 and >20 ng/mL in 7. Most patients belonged to the high-risk (n=20) and intermediate-risk (n=20) groups (low-risk n=1). The median PT dose was 78.0 Gy (72.0-78.0 Gy) applied in mean 39 fractions (range, 30-39) by primarily using pencil beam scanning (95.1%). Toxicities were classified according to CTCAE V4.0.

Results: The median follow-up (FU) is 0.6 y (0.0-2.4 y). 97.6% of the patients showed disease control; distant failure occurred in one patient. No patient died so far. The mean radiation dose at the rectum was 21.5 Gy ± 5.4 Gy, V70Gy was 9.8% ± 4.9% and V50Gy 20.9% ± 6.3%. The mean dose, V70Gy and V50Gy at the bladder was 21.5 Gy ± 9.3Gy, 11.9% ± 7.5% and 22.4% ± 10.5%, respectively. PT was well-tolerated. No new acute high-grade (≥°3) toxicities occurred. FU-data show one new °3 toxicity, regarding erectile dysfunction. No new °4 or °5 effects occurred.

Conclusion: Current data support safety, tolerance and effectivity of PT in prostate carcinoma close to sensitive tissues. However, long-term FU data is still needed to assess long-term outcomes.

PTC17-0133: Comparative Efficacy of Radiation Therapy with High or Traditional Pelvis Doses at Patients with High Progression Risk Prostate Cancer

E. Khmelevsky1, I. Kancheli 2, A. Kaprin3

1P.A. Herzen Research Oncology Institute- Moscow, Radiation oncology, Moscow, Russian Federation 2SSC RF “Institute for Theoretical and Experimental Physics” SRC “Kurchatov Institute”, Medical physics, Moscow, Russian Federation3P.A. Herzen Research Oncology Institute- Moscow, P.A. Herzen Research Oncology Institute- Moscow, Moscow, Russian Federation

Purpose: To evaluate effects of the еscalation of the BED in small pelvis and prostate at patients with high progression risk prostate cancer.

Materials and Methods: 272 patients were included in the randomized investigation (proton against photon boost). Due to a calibration failure 21 patients, treated from 08.2009 to 02.2010 instead of 2 Gy, received 2.4 Gy of the photon dose per fraction with BED to the whole volume of small pelvis (α/β=3 Gy) 54.9 GyE. The photon boost (15 patients out of 21) was 32.5 GyE. The proton boost BED at last 6 patients was unchanged: 28 Gy(RBE), 3 fx in 5.5 Gy. The control group included 30 pts with the standard dose to the whole pelvis (BED 44 Gy), the photon boost (BED 26-28 Gy at 24 pts) or the proton boost (BED 28 Gy(RBE) – at 6 pts) in prostate, threated 6 mo. before and 8 mo. after incident.

Results. Acute GI and GU reactions in the control and high dose groups did not differ; the damages GI≥3 were absent. Late GI2+ and GU2+ damages were 38.3±9.7% and 25.0±8.5% in the main group against 20.0±8.4% and 10.0±7.2% in control (everywhere p<0.05). 5 years actuarial relapse-free survival was 87.6%±6.6% (high dose group) vs 85.0%±4.3% (control), and OS was 93.3±4.2% vs 94.6±3.9% (everywhere p<0.05).

Conclusion. The increase of the small pelvis dose (44.0 Gy vs 54.9 Gy ) and the prostate dose (72.0 Gy vs 87.4 Gy ) did not show a RFS and OS growth, but tendencies of a late reactions rise.

PTC17-0226: Preliminary Results of Locally Advanced Prostate Cancer Treated with Pelvic Proton Therapy Followed by Prostate Carbon Ion Boost at SPHIC

Q. Z Hang1, L. Ping1, Q. Weixiang1, C. Xin1, Z. Jingfang 2, S. Kambiz 2, F. Shen1

1Shanghai Proton and Heavy Ion Center, Radiation Oncology, Shanghai, China 2Shanghai Proton and Heavy Ion Center, Medical Physics, Shanghai, China

Purpose: To evaluate the safety and efficacy of pelvic proton therapy followed by prostate carbon ion boost for patients with locally advanced prostate cancer.

Material and Methods: From June 2014 to Nov 2017, a total of 5 patients with cT2b-4N1M0 prostate cancer underwent whole pelvic proton therapy followed by prostate carbon ion boost. Planning target volumes included the prostate, seminal vesical, whole pelvic lymphatic drainage and the positive lymph nodes. IMPT was used to treat the whole pelvic in 45-46GyE/23-25Fx, followed by carbon boost to the prostate in 24.66-27GyE/9Fx. Either simultaneous proton boost or carbon boost technique was used to the positive lymph nodes.

Results: PSA was used to evaluate the response of particle therapy. 5 of 5 (100%) patients had biochemical control (PSA < 0.2 ng/ml) right after one month of irradiation and PSA values were used for evaluate the efficacy of the treatment every 2 months after treatment. 3 of 5 (60%) patients underwent 99mTc- PSMA SPECT/CT before and after particle treatment, which showed that the abnormal PSMA uptake deceased dramatically and all reached 0 after irradiation. No one presented with G3 or higher grade particle therapy related acute toxicities. Based on CTCAE4.03 creteria, 2 patients presented with GU toxicity (G1, 1/5 ; G2 1/5), most of them were Grade 1-2 hematology toxicites.

Conclusion: Our primary data showed that pelvic proton therapy follwed by prostate Carbon ion boost were well tolerated by prostate cancer patients and immediate effect was encouraging. Long-term results need to be further investigated.

PTC17-0411: A Multi-Institutional Analysis of the Toxicities of Carbon Ion Radiotherapy for Prostate Cancer: The Japan Carbon-Ion Radiation Oncology Study Group

T. Nomiya1, H. Tsuji 2, K. Hidemasa3, O. Tatsuya3, T. Shingo4, S. Yoshiyuki4, N. Yuko1, N. Kenji5, T. Hirohiko 2, K. Tadashi 2

1Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan 2National Institute of Radiological Sciences, Department of Radiation Oncology, Chiba, Japan3Gunma University Heavy Ion Medical Center, Department of Radiation Oncology, Maebashi, Japan4Ion Beam Therapy Center- SAGA-HIMAT Foundation, Department of Radiation Oncology, Tosu, Japan5Yamagata University Hospital, Department of Radiation Oncology, Yamagata, Japan

Purpose: We performed a multi-institutional analysis of the outcomes of CIRT in patients with prostate cancer.

Materials and Methods: Data of patients enrolled in prospective clinical trials performed at 3 CIRT institutions were retrospectively analyzed. All patient risks were reclassified according to the D'Amico risk classification. A short-term neo-adjuvant hormonal therapy and a long-term neo/adjuvant hormonal therapy (≥24 months) were combined with CIRT for the intermediate-risk group and the high-risk group, respectively. The biochemical failure was censored according to the Phoenix consensus (a rise of >2.0 ng/mL above PSA nadir).

Results: Between December 2003 and December 2014, the total of 2157 patients were enrolled from the three institutions. The number of patients in low-risk, intermediate-risk, and high-risk groups were 263, 679, and 1215, respectively. The median follow-up periods of surviving patients was 29 months (range 0.5-133.7). The 5-year overall late G2 GU toxicity was 6.1%, and the 5-year incidence of G2 hematuria was 3.9%. The 5-year late G2 GI toxicity was 0.8%, and the 5-year late G3 GU toxicity and G3 GI toxicity were 0.0%. The ten-year biochemical relapse-free survivals (bRFS) in low-risk, intermediate-risk, and high-risk patients were 77%, 70%, and 79%, respectively.

Conclusion: No severe acute/late toxicities were observed in this study. Analysis of the first multi-institutional data on CIRT for prostate cancer suggested that CIRT showed favorable overall outcomes especially in high-risk patients with less toxicities.

PTC17-0460: Advantage of Hydrogel Spacer in IMPT with Anterior Fields and Prostate Cancer Patients: Reduction in Range Uncertainty

B. Davis1, T. Whitaker1, C. Beltran1, B. McMenomy 2, D. Routman1, B. Stish1, R. Choo1, T. Pisansky1

1Mayo Clinic, Department of Radiation Oncology, Rochester, USA 2Mayo Clinic, Department of Radiology, Rochester, USA

Purpose: To evaluate radiation dosimetry in prostate cancer (CaP) patients undergoing intensity modulated proton therapy (IMPT) using two anterior oblique and two lateral fields (Ant+OppLat) with hydrogel spacer placed between the rectum and prostate.

Materials and Methods: A total of 18 primary CaP patients received IMPT to the prostate +/- seminal vesicles, using two anterior oblique fields (30-50% weighting) from 30 to 55 degrees from the AP as part of Ant+OppLat. Eleven patients had hydrogel spacer (WS) placed, while 7 had no spacer (NS). Patients were treated with conventional (1.8 – 2.0 Gy) or moderate hypofractionation (2.5 – 3.0 Gy per fraction). Rectal and CTV dosimetry were normalized to prescription dose. All patients underwent CT-MR based treatment planning with a uniform 4-5 mm margin around the CTV.

Results: CTV volume was 52.4 cm3 (+/- 21). For WS patients, the mean spacer volume was 7.9 cm3 (+/- 2.1), and spacer thickness between the prostate and rectum was > 5mm. Dosimetry results are presented. Reduction in D2cc[%] and other rectal dose parameters, while accounting for range uncertainties, was observed in WS patients.

Conclusion: The use of hydrogel spacer for CaP patients undergoing IMPT with a 4-field (2 anterior oblique and 2 lateral) approach improved rectal dosimetry while assuming a +/-3% range uncertainty (RU) without compromise of target volume coverage.

PTC17-0466: Is Neoadjuvant ADT Necessary for Intermediate-Risk Prostate Cancer Treated with Proton Therapy?

M. Takagi1, Y. Demizu 2, K. Terashima 2, O. Fujii3, D. Jin 2, F. Nagano 2, Y. Niwa4, M. Mima5, N. Fuwa 2, T. Okimoto 2

1Sapporo Teishinkai Hospital, Proton Therapy Center, Sapporo, Japan 2Hyogo Ion Beam Medical Center, Radiology, Tatsuno, Japan3Hakodate Goryoukaku Hospital, Radiation Oncology, Hakodate, Japan4Uji-Tokushukai Medical Center, Radiology, Uji, Japan5Nishikobe Medical Center, Radiology, Kobe, Japan

Purpose: Presently no prospective, randomized trials have clearly defined the role of neoadjuvant androgen-deprivation therapy (N-ADT) for patients with intermediate-risk (IR) prostate cancer with proton therapy (PT). In this study, we assessed the impact of adding N-ADT to PT on biochemical relapse.

Materials and Methods: A total of 600 patients with IR prostate cancer (any of the 3 risk factors: PSA 10-20 ng/mL, Gleason Score (GS): 7, T2b-T2c) treated with PT at Hyogo Ion Beam Medical Center from 2003 to 2014 were included in this study. By the following methods, whether N-ADT improved biochemical relapse free rate (bRFR) of the IR patients was investigated. 1) Prognostic factor analysis including N-ADT for all IR patients. 2) Subgroup analysis with each risk factors (T classification, GS, PSA and % of positive core) and number or risk factors.

Results: The median follow-up was 74 months (range, 4 - 135 months). All patients received 74 Gy (RBE) in 37 fractions and 262 (43.7%) patients received N-ADT with the median duration of 7 months (range, 1 - 84 months). For all IR patients, 5- and 10- year bRFR were 90.8% and 82.4%, respectively. N-ADT did not improve bRFR (90.7% vs. 90.8%, P = 0.925) for all IR patients. On the subgroup analysis using each risk factors and number of risk factors, N-ADT did not improve bRFR with statistically significant differences.

Conclusion: PT provided excellent biochemical control for patients with IR prostate cancer. In this study, N-ADT did not improve bRFR of patients with IR prostate cancer.

PTC17-0498: Acute and Late Toxicity Report of Post-Prostatectomy Proton Therapy for Prostate Cancer Patients Undergoing Adjuvant or Salvage Radiotherapy

A. Jain1, N. Vapiwala1, K. Woodhouse1, S. Both 2, P. Gabriel1, Z. Tochner1, C. Deville3

1University of Pennsylvania, Radiation Oncology, Philadelphia, USA 2Memorial Sloan Kettering Cancer Center, Deptartment of Medical Physics, New York, USA3Johns Hopkins University, Radiation Oncology, Baltimore, USA

Purpose: To report the acute and late genitourinary (GU) and gastrointestinal (GI) toxicities associated with post-prostatectomy proton therapy (PT).

Materials and Methods: All patients (N=100) undergoing post-prostatectomy PT on an IRB-approved protocol from 2010-15 were assessed. Patients received 70.2 Gy relative biological effectiveness median dose in 39 fractions using passive scattering (13%), pencil beam scanning (86%), or both (1%) to the prostate bed (80%) or whole pelvis and prostate bed (20%). Thirty-one (30.7%) utilized combined IMRT to achieve predefined dose constraints, and 34% received androgen deprivation. Toxicity was scored by CTCAE v4.0. Median follow up was 25 months (range 3–47).

Results: Median age was 64 years. Median months from surgery were 25. Mean pre-radiation PSA was 0.67±1.48 (0.00–8.50), respectively. There were no grade ≥3 toxicities. Grade 2 GU toxicities consisted of hematuria and urinary incontinence, retention, urgency, and frequency. All grade 2 GI toxicities consisted of constipation. Median time to maximum GU and GI toxicity was 4 (3-39) and 3 (3-36) months, respectively. The mean baseline International Prostate Symptom Score, International Index of Erectile Function-5, and Expanded Prostate Cancer Index Composite bowel function, and bowel bother scores 6.6±6.1, 10.5±7.3, 90.9±10.8, 93.3±11.2, respectively, and after 2 years, the mean scores remained largely unchanged (6.3±3.6, 11.1±6.3, 92.8±5.8, and 90.9±10.3).

Conclusion: We report the clinical feasibility and favorable acute and late GU and GI toxicity profile of post-prostatectomy PT.

Head and Neck

PTC17-0035: Proton Therapy for Head and Neck Non-Squamous Cell Carcinoma: Planning Comparison and Initial Results of Early Toxicities

H. Iwata1, T. Toshito 2, K. Hayashi3, E. Nikawa3, M. Yamada4, C. Omachi 2, S. Hashimoto1, H. Ogino1, Y. Shibamoto5, J.E. Mizoe1

1Nagoya Proton Therapy Center- Nagoya City West Medical Center, Radiation Oncology, Nagoya, Japan 2Nagoya Proton Therapy Center, Proton Therapy Physics, Nagoya, Japan3Nagoya Proton Therapy Center, Proton Therapy Technology, Nagoya, Japan4Nagoya City West Medical Center, Radiation Therapy, Nagoya, Japan5Nagoya City University Graduate School of Medical Science, Radiology, Nagoya, Japan

Purpose: The purpose of this study was to investigate the optimal treatment planning using proton beams for head and neck non-squamous cell carcinoma (HNNSCC). We have implemented a patient-specific aperture system (PSAS) capable of attaching an energy absorber and reducing the lateral penumbra. Dose distributions of plans involving spot scanning proton therapy (SSPT) with or without a PSAS, passive-scattering proton therapy (PSPT), and X-ray IMRT were compared. Early toxicity results of SSPT with the PSAS were reported.

Materials and Methods: Twenty patients with HNNSCC treated using SSPT were selected at random, and the dose distributions of the four plans were compared. The dose data were transferred to MIM Maestro from each treatment planning system, and dosimetric parameters were compared. Neutron exposures by proton therapy were calculated using Monte Carlo simulation. Early toxicities were scored according to CTCAE ver.4.0.

Results: The conformity number 95% of SSPT with the PSAS plans was the best, and significant differences were detected among the four plans (P < 0.05, Bonferroni-tests). SSPT with the PSAS and PSPT plans protected the skin more efficiently than the other two plans, especially in low to intermediate dose ranges, but no difference was detected among the four plans. Neutron exposures by the PSAS were approximately 1.1 times higher and within an acceptable level. No grade 3 or higher dermatitis was observed.

Conclusion: The optimal treatment planning using proton beams for HNNSCC is considered to be SSPT with the PSAS. It is expected that intensity-modulated proton therapy with a PSAS may further improve the dose distribution.

PTC17-0057: Dosimetric and Radiobiological Benefits on Organs-At-Risk for Intensity Modulated Proton Therapy as Compared with Helical Tomotherapy for Nasopharyngeal Carcinoma

W.W. Lam1, H. Geng1, C.W. Kong1, Y.W. Ho1, W.K.R. Wong1, B. Yang1, T.L. Chiu1, T.Y. Lee1, K.Y. Cheung1, S.K. Yu1

1Hong Kong Sanatorium & Hospital, Medical Physics & Research Department, Hong Kong, China

Purpose: To evaluate the dosimetric benefits and impact on possible clinical toxicity of organs-at-risk (OARs) in terms of generalized equivalent uniform dose (gEUD) and normal tissue complication probability (NTCP) of intensity modulated proton therapy (IMPT) with multi-field optimization (MFO) as compared with helical tomotherapy (HT) for nasopharyngeal carcinoma (NPC).

Materials and Methods: Ten NPC patients treated with HT using 2.5cm field-width were retrospectively re-optimized using IMPT with MFO and simultaneous integrated boost technique using Varian Eclipse proton TPS. Nasopharynx gross tumour volume (GTVNP), nasopharynx, left and right nodal planning target volumes (PTVNP,PTVLtN,PTVRtN) were aimed to achieve 95% volume covered by the prescribed dose. From ICRU78, relative biological effectiveness-weighted dose [Gy(RBE)] was used for reporting, assuming RBE of 1.1 and 1 for proton and photon, respectively. GTV and PTV coverage, dose conformity and homogeneity were reported by dose received by 95% target volumes (D95), conformity index (CI) and homogeneity index (HI). Maximum dose (Dmax), near-maximum dose (D2) and mean dose (Dmean) of brainstem, optic chiasm and parotid glands were reported. The possible OAR toxicities were quantified by gEUD and NTCP using the Lyman-Kutcher-Burman (LKB) model and biological parameters suggested in Emami et al. Statistical analysis was performed using Wilcoxon-signed rank test. A two-tailed p<0.05 was considered statistically significant.

Conclusion: As compared with HT, IMPT for NPC could significantly lower doses to the evaluated OARs and reduce their probability of complication rate without compromising target dose coverage, conformity and homogeneity.

PTC17-0105: Investigation of Treatment Indication Restriction Related to Tumor Localization in Boron-Neutron Capture Therapy for Recurrent Head and Neck Cancer

K. Hirose1, T. Kato1, K. Arai1, T. Harada1, T. Motoyanagi1, R. Shimokomaki1, T. Nakamura 2, H. Wada 2, Y. Kikuchi 2, Y. Takai1

1Southern TOHOKU BNCT Research Center, Radiation Oncology, Koriyama- Fukushima, Japan 2Southern TOHOKU Proton Therapy Center, Radiation Oncology, Koriyama- Fukushima, Japan

Purpose: Boron-neutron capture therapy (BNCT) can achieve a cancer cell-selective particle therapy, and damage boron compound-penetrating cells with alpha and lithium particles. We are now conducting BNCT treatment with hospital-installed cyclotron-based BNCT system for recurrent head and neck cancer (HNC). In performing BNCT, there are some limitations of treatment application due to short tracts of neutron beam that restricts the beam settings and patient setup conditions. Therefore, we examined the relationship between tumor localization and treatment indication in BNCT for HNC.

Materials and Methods: For fourteen recurrent HNC patients with a history of radiotherapy, re-planning for BNCT treatment was performed with Simulation Environment for Radiotherapy Applications (SERA) system using the past planning CT. Optimizations of beam alignment and dose calculation were performed with considering patient setup after delineation. All treatment plans were normalized to deliver maximum dose of 12 Gy-Eq to mucosa in oral cavity and pharynx, and we evaluated whether each tumor localization can be indicative for BNCT with adequate dose distribution.

Results: The each size of effective neutron field for treatment was diverse and relied on tumor location. Parotid gland tumor was given extremely higher dose and thought to have good indication for BNCT. Oral cavity tumor had relatively lower dose than parotid gland tumor. Ethmoid sinus, nasopharyngeal, and mesopharyngeal tumors were considered to have no indication for BNCT in this setting.

Conclusion: Tumor localization might become an important clue to judge treatment indication in recurrent HNC.

PTC17-0134: Defining the Role of Range Shifter in Treating Head and Neck Cancer in the Era of Intensity Modulated Proton Therapy

X. Ding1, X. Li1, A. Qin1, J. Zhou1, D. Yan1, P. Chen1, C. Stevens1, R. Deraniyagala1, P. Kabolizadeh1

1Beaumont Health System, Radiation Oncology Proton Therapy Center, Royal Oak, USA

Purpose: To explore the possibility to eliminate the use of range shifter (RS) in treating bilateral head and neck cancer(HNC) patients.

Materials and Methods: Study1: A bilateral HNC case was planned via IMPT plans with RS (IMPT_RS) and without RS (IMPT_noRS) while increasing the number of beam angles from 3 to 10. Robust optimization parameters were set to ±3.5% range and ±3mm in x,y,z directions' setup uncertainties . The final plan objective values were plotted as a function of beam angles numbers. Study2: To find out whether the plan quality could be improved via IMPT_noRS , ten patients with bilateral HNC cases were planned with 4-field IMPT with RS (4F IMPT_RS) and without RS (4F IMPT_noRS) plans. The plan robustness was evaluated using the root-mean-square deviation doses (RMSDs).

Results: The first study showed a very interesting phenomenon that IMPT_noRS plan objective value was actually lower than IMPT_RS when the beam number was increased to four or more. Study 2: The result showed that the mean dose to ipsilateral parotid gland and skin was reduced from 34.1 Gy [RBE] to 32.10 Gy [RBE] (p=0.007) and 16.26 Gy [RBE] to 11.99Gy[RBE] (p=0.031) respectively. The RMSD comparison indicates that 4F IMPT_noRS is able to provide a comparable or even better robustness.

Conclusion: It is not necessary to use RS in treating bilateral HNC via IMPT if four or more beam angles are used and the plan quality could be further improved. However, our study confirm that RS is highly needed for treatment of bilateral HNC if less than 4 fields are used.

PTC17-0180: Custom-Made Mouthpiece Was Useful for Reducing the Oral Dose during Proton Beam Therapy for Pediatric Patient; A Case Presentation

M. Suzuka1, K. Ohmori 2, K. Minowa1, K. Tsuchiya3, T. Hashimoto4, K. Yasuda5, S. Takao6, H. Shirato5, A. Iguchi7, Y. Cho7

1Hokkaido University, Department of Oral Radiology- Division of Oral Pathobiological Science- Graduate School of Dental Medicine, Sapporo, Japan 2Hokkaido University, Department of Oral Radiology, Sapporo, Japan3Hokkaido University Hospital, Department of Radiation Oncology, Sapporo, Japan4Hokkaido University, Department of Radiation Medicine- Hokkaido University Graduate School of Medicine, Sapporo, Japan5Hokkaido University, Department of Radiation Medicine- Hokkaido University Graduate School of Medicine- Global Institution for Collaborative Research and Education GI-CoRE, Sapporo, Japan6Hokkaido University Hospital, Proton Beam Therapy Center, Sapporo, Japan7Hokkaido University Hospital, Department of Pediatrics, Sapporo, Japan

Reducing radiation dose to the normal tissue surrounding the target is important not only for reducing the acute adverse events but also for preventing the late adverse events especially for pediatric cancer patients. In our institution, we use a custom-made mouthpiece to reduce the irradiation dose to the oral cavity for proton beam therapy in head and neck region. We present a case of a 4-year-old girl with facial rhabdomyosarcoma, successfully decreasing the irradiation dose to the maxillary bone and mucosa using custom-made mouthpiece in proton beam therapy. She was introduced to our hospital for post-operative additional therapy. Proton beam therapy was delivered to the tumor bed around right nasal wing to a total dose of 50.4 GyE in 28 fraction, using vincristine, actinomycin D, cyclophosphamide concurrently. The mouthpiece was made with dental material and designed to insert into the upper lip sulcus to reduce the dose to the upper gum during proton beam therapy. Acute mucositis and small ulcer was observed in right upper lip, but no mucositis occurred at maxillary gingiva during proton beam therapy. Custom-made mouthpiece was useful to reduce the oral dose, and it may decrease the risk of late adverse events, such as facial bone asymmetry and tooth eruption failure.

PTC17-0378: Custom Made Mouthpiece to Reduce Radiation-Induced Side Effect during Carbon Ion Radiotherapy for Head and Neck Cancer

H. Ikawa1,2, M. Koto3, R. Takagi3, T. Nomura 2, T. Kamada3

1Hospital of the National Institute of Radiological Sciences, Dental Section, Chiba, Japan 2Tokyo Dental College, Department of Oral Medicine- Oral and Maxillofacial Surgery, Tokyo, Japan3Hospital of the National Institute of Radiological Sciences, Radiation oncology division, Chiba, Japan

Purpose: Carbon ion radiotherapy (CIRT) seems well indicated for non-squamous cell carcinoma of the oral, nasal cavity and paranasal sinuses. To prevent severe complications such as tongue mucositis, taste disorders, and osteoradionecrosis (ORN), the dose to normal tissue needs to be attenuated using intraoral devices, involving fixation and placement of tongue, intraoral mucosa and jaw. In our hospital, 3 patterns of intraoral devices to shield adjacent tissues were made by tumor location. In this presentation, we describe the efficacy of the custom-made mouthpiece in CIRT to reduce radiation-induced side effect.

Pattern 1: For palatine or nasal cavity and paranasal sinus tumors, use of a mouthpiece to open the mouth. Then, the tongue is pushed to the caudal side to remove the tongue from the irradiation field.

Pattern 2: In cases where the tumor is located in the submandibular gland, sublingual gland or floor of mouth, elevate the tongue cranially. This reduces the dose to the dorsum of tongue, reducing the risk of mucositis and taste disorders.

Pattern 3: In the case of tumors of the tongue base, a spacer is interposed between the tongue and the mandible to reduce the dose to the mandible.

Conclusion: Mouthpieces are made considering the localized region of the tumor, the incidence direction of the beam, and the irradiation field. It has been suggested that the risk of tongue mucositis and ORN can be reduced by shaping the mouthpiece. In addition, cooperation with radiation oncologists and radiation diagnosticians was considered important in mouthpiece design.

PTC17-0420: Sensitive Predictive Marker for Early Response of Olfactory Neuroblastoma Treated by Carbon-Ion Radiotherapy

S. Komatsu1, J.I. Saitoh1, K. Shirai1, A. Musha 2, T. Abe1, T. Ohno1, T. Nakano1, Y. Takayasu3, K. Takahashi3, K. Chikamatsu3

1Gunma University, Heavy Ion Medical Center, Maebashi, Japan 2Gunma University Heavy Ion Medical Center, Heavy Ion Medical Center, Maebashi, Japan3Gunma University Graduate School of Medicine, Department of Otolaryngology- Head and Neck Surgery, Maebashi, Japan

Purpose: Early changes in standardized uptake value (SUV) of 18F-fluorodeoxyglucose positron emission tomography and apparent diffusion coefficient (ADC) value of diffusion-weighed magnetic resonance imaging have been reported to predict response of head and neck squamous cell carcinoma to photon radiotherapy before significant changes in tumor size. The purpose of this study was to investigate the potentials of ADC value and SUV to monitor early response of olfactory neuroblastoma (ONB) to Carbon ion Radiotherapy (CIRT).

Materials and Methods: From May 2011 to April 2016, 6 patients with ONB underwent CIRT to a total dose of from 57.6 to 64.0 Gy(RBE). No patient had distant and lymph node metastasis.

Results: The median pre-treatment tumor volume was 7.2 cm3 [range, 0.4 - 34.3]. The tumors in four patients responded to treatment and decreased consistently, but in two patients' tumors transiently developed locally but they have almost disappeared after a year. All cases showed complete response. Tumor volume, mean-ADC value and max-SUV change was measured according to our CIRT protocol. The median pre-treatment max-SUV was 5.91 [4.4 - 6.85]. Max-SUV corresponded to that of tumor volume, which decreased consistently or followed by transient increase soon after treatment. The median pre-treatment mean-ADC value was 1.241 mm−2/sec [1.125 - 1.398], and the median mean-ADC value two months after CIRT was 1.713 mm−2/sec [1.652 - 1.978]. The mean-ADC value showed a rapid increase prior to tumor regression.

Conclusion: Our results demonstrate the potential usefulness of mean-ADC value for early post-treatment evaluation compared with tumor volume decrease or max-SUV.

PTC17-0444: Dosimetric Comparison of Intensity-Modulated Radiation Therapy (IMRT) Versus Carbon-Ion Beam Scanning Treatment for Locally Advanced Maxillary Sinus Carcinoma

N. Mizoguchi1, T. Nonaka1, T. Nomiya1, S. Shibata1, Y. Hagiwara1, M. Kutooka 2, T. Kusunoki 2, Y. Kusano 2, Y. Tokiya3, Y. Nakayama1

1Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan 2Kanagawa Cancer Center, Division of Medical Physics, Yokohama, Japan3Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan

Purpose: The purpose of this study was to compare the target dose distribution, conformity and normal tissue avoidance in intensity-modulated radiation therapy (IMRT) and carbon ion radiotherapy (CIRT) using pencil beam scanning method in locally advanced maxillary sinus carcinoma.

Materials and Methods: Five patients with locally advanced maxillary sinus carcinoma underwent IMRT in practice. The virtual plan of CIRT was simulated on the treatment planning computed tomography images of IMRT. Target coverage was evaluated with the D98, D50, D10 and D2. Target dose distribution and conformity were evaluated with the homogeneity index (HI) and conformity Index (CI). Normal tissue avoidance of organ at risk (OAR) was evaluated with the V10, V20, V30 and V40.

Results: In comparison with IMRT and CIRT plans, target coverage was similar to both plans. HI and CI were 1.17 ± 0.02 and 0.98 ± 0.06, respectively in IMRT, as compared to 1.10 ± 0.01 and 0.48 ± 0.09, respectively in CIRT. V10 of brain was 45.2 ± 4.18 in IMRT and 3.11 ± 1.33 in CIRT. V20 of brain was 22.6 ± 8.42 in IMRT and 1.97 ± 1.32 in CIRT. V10 and V20 of brain were significantly reduced in the CIRT plans (p=0.001 and 0.035, respectively).

Conclusion: CIRT plans could improve target conformity and could reduce the dose of OAR.

PTC17-0462: Customized Mouthpiece in Reducing Tongue Mucositis in Carbon-Ion Radiotherapy for Nasal Tumors; Technical Report

A. Musha1, J.I. Saitoh1, K. Shirai1, T. Abe1, Y. Kubota1, H. Shimada1, S. Yokoo 2, T. Ohno1, T. Nakano1

1Gunma University, Heavy Ion Medical Center, Maebashi, Japan 2Gunma University, Stomatology and Maxillofacial Surgery, Maebashi, Japan

Purpose: Acute radiation mucositis (ARM) is the most common acute adverse effect of radiotherapy to the head and neck. Customized mouthpieces have been used to minimize motion and allow a reproducible set-up for head and neck in Carbon-ion radiotherapy (C-ion RT). Theoretically, the incidence of ARM in the displaced tongue from irradiated area may be reduced by a customized mouthpiece. However, there is little published information about the methods that can be used to create customized mouthpieces and the planning techniques that can be used to reduce ARM.

Materials and Methods: Between 2011 and 2012, 18 patients with nasal cavity tumors were treated via C-ion RT at our institute. Patients with malignant melanoma were 12, adenoid cystic carcinoma were 4 and others were 2. Total dose to the tumor was determined according to the clinical protocol. There were 18 patients 64.0 Gy(RBE) in 16 fractions. ARM was graded by Radiation Therapy Oncology Group grading system. A maximum dose of palate and tongue was evaluated in each patient by using on the MIM software. In addition, location of the high dose area was compared to pictures of ARM taken weekly during and after C-ion RT.

Results: The mean grade of ARM was significantly lower in the tongue than in the palate, probably because the mucosal dose of the tongue was decreased by spacing due to the customized mouthpiece.

Conclusion: We conclude that the use of a customized mouthpiece can play an additional role in reducing ARM by displacing the tongue.

PTC17-0473: A Retrospective Multi-Institutional Study of Proton Beam Therapy (PBT) for Head and Neck Cancer with Non-Squamous Cell Histologies

N. Nakamura1, T. Akimoto1, S. Zenda1, Y. Demizu 2, T. Okimoto 2, S. Murayama3, H. Sakurai4, T. Nakamura5, K. Yamamoto6, H. Shirato7

1National Cancer Center Hospital East, Radiation Oncology and Particle Therapy, Kashiwa, Japan 2Hyogo Ion Beam Medical Center, Radiology, Tatsuno, Japan3Shizuoka Cancer Center Hospital, Proton Therapy, Syunto, Japan4University of Tsukuba, Radiation Oncology, Tsukuba, Japan5Southern Tohoku General Hospital, Radiation Oncology, Koriyama, Japan6Fukui Prefectural Hospital, Proton Therapy Center, Fukui, Japan7Hokkaido University Hospital, Radiation Oncology, Sapporo, Japan

Purpose: The purpose of this multi-institutional study is to evaluate the long-term clinical outcomes and toxicities of PBT for head and neck cancer with non-squamous cell histologies to clarify the efficacy of PBT.

Materials and Methods: We collected clinical data of patients who had biopsy-proven head and neck cancers with non-squamous histologies and were treated with PBT between 2003 and 2013 in 5 institutions, and analyzed clinical outcomes including overall survival (OS), local control rate (LCR) and incidence and grades of treatment-related toxicities.

Results: A total of 339 patients whose follow-up duration was 64 months ranging from 20-91 months were analyzed. Distribution of histological diagnosis was 141 (42%) in malignant melanoma, 83 in adenoid cystic carcinoma, 63 in olfactory neuroblastoma and 62 patients in others. OS, progression-free and LCR of all patients at 5 years was 61.2%, 36.8% and 71.2%, respectively. According to histologies, OS and LCR at 5 years was 40.2%, 64.2% in malignant melanoma, 72.9%, 71.5% in adenoid cystic carcinoma, 86.2%, 79.0% in olfactory neuroblastoma and 60.0% in others. Regarding the incidence and grades of late toxicities, 34 (10%) patients developed grade 3 or 4 late toxicities.

Conclusion: The results of multi-institutional study demonstrated excellent clinical outcomes and acceptable late toxicities, indicating that PBT would be effective treatment option for head and neck cancer patients with non-squamous cell histologies.

PTC17-0482: Initial Experience of Hypofractionated Re-Irradiation with Proton Beam in Patient of Recurrent Nasopharyngeal Carcinoma

H. Mammar1, B. Duan 2, L. De Marzi1, Y. Liang 2, K. Liao 2, J. Yu 2, T. Yuan 2, P. Bey1, R. Dendale1, X. Zhang 2

1Institut Curie, Radiotherapy, Paris, France 2Affiliated Cancer Hospital & Institute of Guangzhou Medical University, China

Purpose: To show our preliminary experience in using proton beam for re-irradiation of locally recurrent nasopharyngeal carcinoma. For the 10% of patient locally recurrent nasopharyngeal carcinoma after definitive IMRT, reirradiation therapy is the only potentially curative treatment option. Proton beam therapy (PBT) improve the therapeutic ratio by reducing doses to normal tissue.

Materials and Methods: We present here the cases of 4 consecutive patients with local recurrence of nasopharyngeal carcinoma confirmed by biopsy. There were 2 male and 2 female patients. All patients had a Karnofsky Performance Status greater than 80 at the time of treatment. All patients were treated using hypofractionated PBT. Mean prescribed dose was 40 Gy RBE (35 – 50 Gy RBE) with 5 Gy RBE per fraction. Patient and tumor characteristics are summarized.

Results: All patients concluded PBT and showed good tolerability. The median follow-up was 35,5 months (range : 82 –14). Only one patient (# 3) with the largest lesion showed a local recurrence of his lesion at one year from the end of treatment.

Conclusion: Proton beam for hypofractionated re-irradiation of nasopharyngeal carcinoma was safe and effective in 3 treated patients. Longer follow-up and a larger population of study is needed to confirm these promising results.

Lung

PTC17-0033: The Clinical Results of Proton Beam Therapy in Eighteen Patients with Idiopathic Pulmonary Fibrosis

T. Ono1, M. Hareyama 2, T. Nakamura1, K. Kimura1, Y. Azami1, K. Hirose1, M. Suzuki1, H. Wada1, Y. Kikuchi1, K. Nemoto3

1Southern Tohoku Proton Therapy Center, Radiation Oncology, Koriyama, Japan 2Sapporo Teishinkai Hospital, Radiation Oncology, Sapporo, Japan3Yamagata University Faculty of Medicine, Radiation Oncology, Yamagata, Japan

The purpose of this study is to evaluate the incidence of lung toxicities after proton beam therapy (PBT) in patients with idiopathic pulmonary fibrosis (IPF). IPF patients with lung tumors treated with PBT between January 2009 and December 2015 were recruited from our database retrospectively. Lung toxicities were evaluated using the Common Terminology Criteria for Adverse Events version 4.0, and the Fletcher-Hugh-Jones classification of respiratory status was used to evaluate pretreatment and posttreatment respiratory function. Eighteen IPF patients received PBT for lung tumors. The cohort was composed of 16 men and 2 women, with a median age of 76 years (range: 63–89 years). The median follow-up time was 17 months (range: 7–57 months), and all patients were followed at least 25 months or until death. The median dose of PBT was 80.0 Gy relative biological dose effectiveness (RBE) (range: 66.0–86.4 Gy [RBE]). The cumulative incidence of grade ≥2 pneumonitis was 22.6% (95% confidence interval [CI]: 3.0%–42.2%), including 2 case of grade 5 pneumonitis (both patients were received PBT for two lung tumors in different lesion). Reduced respiratory function was observed after PBT in 7 patients, including one patient with pleural dissemination; 5 of these patients required home oxygen therapy after PBT. However no patients experienced acute exacerbation within 30 days. This study suggests that PBT has become one of a treatment choice for lung tumors of patients with IPF, although the adverse events warrant serious attention.

PTC17-0041: Reirradiation of Thoracic Cancers with Intensity Modulated Proton Therapy

J. Ho1, Q.N. Nguyen1, H. Li 2, P. Allen1, X. Zhang 2, Z. Liao1, R. Zhu 2, D. Gomez1, S. Lin1, J. Chang1

1MD Anderson Cancer Center, Radiation Oncology, Houston, USA 2MD Anderson Cancer Center, Radiation Physics, Houston, USA

Purpose: Reirradiation of thoracic malignancies is a treatment challenge. Intensity modulated proton therapy (IMPT) may allow safe delivery of a higher dose of radiation while minimizing toxicities.

Materials and Methods: Between 2011-2016, 27 patients who received IMPT for reirradiation of thoracic malignancies with definitive intent were retrospectively analyzed. Doses were recalculated to an equivalent dose in 2-Gy fractions (EQD2). Patients received IMPT for recurrence of thoracic cancer (93%) or sequentially after thoracic stereotactic ablative radiotherapy (7%), to a median dose of 66 EQD2 Gy (range 43.2 – 84 Gy).

Results: The median time to reirradiation was 29.5 months. At a median follow-up for all patients of 11.2 months (25.9 surviving patients), the median overall survival (OS) was 18.0 months, with a 1-year OS of 54%. Four patients (15%) experienced local failure (LF), with a 1-year freedom from LF rates of 78%. The 1-year freedom from locoregional failure (LRF), and 1-year progression free survival (PFS) rates were 61% and 51%, respectively.

Patients who received ≥ 66 EQD2 Gy had improved 1-year freedom from LF (100% vs. 49%, p=0.013), 1-year freedom from LRF (84% vs. 23%, p=0.035), and 1-year PFS (76% vs. 14%, p=0.050). Reirradiation was well tolerated, with only 2 patients experiencing late grade 3 pulmonary toxicity, none with ≥ grade 3 esophagitis, and no grade 4-5 toxicities.

Conclusion: This represents the largest series of patients treated with IMPT for definitive reirradiation of thoracic cancers, demonstrating that IMPT provided durable local control with minimal toxicity, and suggests that higher doses may improve outcomes.

PTC17-0050: Study on Indoor Radon Measurement and Dose Assessment in Taiwanese Dwellings Effect Lung

K.C. Hsieh1, J.G. Jheng1

1E-DA Hospital, Radiation Oncology, Kaohsiung City, Taiwan- Province of China

This study planned to select more than 100 typical Taiwanese homes to measure average indoor Rn level in more than 2 rooms each by using an active instrument. We will get more than 200 effective radon level data. The population dose will be reassessed by using the newest announced lung dose model. The indoor Rn level and population dose in Taiwanese dwellings are expected to increase apparently.

According to the planned schedule, we have completed indoor radon sampling and measurement methods including instrument calibration until June 30. More than 46 residential indoor radon field measurements have been accomplished.

PTC17-0182: GTV CT Number, Volume and Mass Changes with IMRT vs. Passively Scattered Proton Therapy (PSPT) for Locally Advanced NSCLC Patients

A. Liu1, Y. Xu 2, Z. Liao3, R. Mohan1

1MD Anderson Cancer Center, Radiation Physics, Houston, USA 2Zhejiang Cancer Hospital, Radiation Oncology, Hangzhou, China3MD Anderson Cancer Center, Radiation Oncology, Houston, USA

Purpose: To investigate and compare changes in CT number (CTN), volume and mass of gross tumor volume (GTV) derived from the weekly CTs for NSCLC patients on an IMRT vs. PSPT randomized trial.

Materials and Methods: Twelve pairs of matched IMRT and PSPT patients with stage IIIA-IIIB NSCLC who had received 74 Gy RBE in 37 fractions were retrospectively analyzed. Each pair met the same inclusion pathological criteria with matching tumor size and location. Contours on each weekly CT were generated starting with the original planning contour using rigid image registration and manual editing if needed. Weekly changes in CTN, volume and mass of the GTV were extracted from CT images using an in-house software. Statistical analysis (paired t-test) was performed to assess the significance of differences between IMRT and PSPT.

Results: The mean CTN reduction within the GTV, averaged over IMRT patients, was found to be significantly larger than for PSPT patients (88 HU vs. 44 HU, p =0.02), with an average weekly CTN change rate of 38.9% for patient treated with IMRT and 28.8% with PSPT. The mean GTV and mass shrinkage were also larger for IMRT patient than for PSPT patients (p=0.029, p=0.04 respectively).

Conclusion: The GTV CTN, volume and mass decreased for NSCLC patients during the delivery of the IMRT and protons. The changes for IMRT were larger than for PSPT. A larger cohort of lung patients will be analyzed in the future.

PTC17-0286: Pencil Beam Scanning Particle Radiotherapy for Stage I Non-Small Cell Lung Cancer: A Mono-Institutional Retrospective Analysis

N.Y. MA1, J. Mao 2, J. Chen1, X. Cai1, J.F. Zhao3, X. Liu4, W.W. Wang4, G.L. Jiang 2

1Shanghai Proton and Heavy Ion Center, Radiation Oncology, Shanghai, China 2Shanghai Proton and Heavy Ion Center- Fudan University Shanghai Cancer Center, Radiation Oncology, Shanghai, China3Shanghai Proton and Heavy Ion Center- Fudan University Shanghai Cancer Center, Radiation physics, Shanghai, China4Shanghai Proton and Heavy Ion Center, Radiation Physics, Shanghai, China

Purpose: To evaluate the safety and efficacy of proton and carbon-ion radiotherapy for stage I non-small cell lung cancer (NSCLC) using pencil beam scanning technique.

Materials and Methods: From 2014.08 to 2015.12, 10 consecutive patients with stage I NSCLC who were inoperable or refused surgery were treated by proton +/- carbon-ion radiotherapy. Primary lesions were irradiated using 2-4 portals with 45-degree beams. A total dose of 50-72 GyE in 10-24 Fx were administrated to patients based on tumor location (peripheral, middle, or central).

Results: At the last follow-up in 2016.12 with the median follow-up of 18.1 months (11.9-28.1), local disease control was found in all patients per CT or PET/CT scanning, including 6 complete response; 3 partial response; 1 stable disease. However, 2 patients with partial response and stable disease experienced a distant failure at 9.8 and 25.6 months after radiotherapy, respectively. All patients tolerated the treatment well, while no radiotherapy-related Grade 3-5 toxicity happened. Grade 2 toxicities were found in 2 patients including acute skin reaction (1/10), and leucopenia (1/10). At 1, 3-5 months after finishing radiotherapy, the pulmonary function tests showed a slightly increase in FVC, FEV1 and DLCOsb compared with those before radiotherapy without statistical significance (P=0.94, 0.85, and 0.92).

Conclusion: The particle RT using pencil beam scanning technique was safe, and yielded encouraging outcome for patients with stage I NSCLC who were inoperable or refused surgery. Further follow-up and prospective clinical studies are warranted in the future.

PTC17-0358: Short Time Evaluation of Carbon Ion Radiotherapy for 2 Cases of Tracheal Adenoid Cystic Carcinoma

J. Chen1, J. Mao 2, N. Ma1, X. Liu3, W. Wang3, J. Zhao3, W.C. Hsi3, G. Jiang 2

1Shanghai Proton and Heavy Ion Center, Radiation Oncology, Shanghai, China 2Shanghai Proton and Heavy Ion Center- Fudan University Cancer Hospital, Radiation Oncology, Shanghai, China3Shanghai Proton and Heavy Ion Center, Physics, Shanghai, China

Purpose: Complete surgical resection for primary tracheal adenoid cystic carcinoma (ACC) is difficult, and photon irradiation is not sensitive enough. Here we report the short-time outcome of carbon ion therapy for two inoperable tracheal ACC patients.

Materials and Methods: The tumor of Patient A located in the upper third trachea with two lesions of 1.5 and 3.5cm length, respectively. Patient B had a tracheal stent implanted 21 month ago for her recurrent disease, which was treated with brachytherapy of 500cGy once and local laser treatment 5 years ago. Shortness of breath and dysphagia were observed before treatment. Both patients received 69 Gy E/23 fractions carbon ion radiotherapy using pencil beam scanning technique. Acute toxicities were scored per the Common Terminology Criteria of Adverse Events version (CTCAE) 4.0.

Results: The tumor of patient A shrank quickly within 1 month of finishing carbon ion therapy, achieved nearly complete response at 3 months follow-up, and remained the same at 6 months (picture). The short of breath of patient B relieved soon after radiation started. She got stable disease 1 month after radiation. No grade 3 or higher toxicities were found. Only observed grade 2 toxicity was hoarseness during treatment. Grade 1 toxicities included esophagitis, dermatitis, pneumonitis and hematologic toxicity.

Conclusion: To the best of our knowledge, this is the first report of using carbon ion radiotherapy for tracheal ACC, which was well tolerated and effective. A long-term follow-up with promising result is expected, and more patients should be accumulated for further study.

PTC17-0520: Comparison of Different Planning Concepts for NSCLC

M. Marti1, J. Hrbacek1, D. Oxley1, A. Mayor1, M. Peroni1, Y. Zhang1, A. Bolsi1, R. Poel1, T. Lomax1, D.C. Weber1

1Paul Scherrer Institut, Center for Proton Therapy, Villigen PSI, Switzerland

Purpose: To compare different target volume approaches for motion mitigation of PBS proton therapy

Materials and Methods: Using 4DCT data of ten thoracic patients, three target definition approaches were compared: mid-ventilation (MV), geometric (gITV) and range adapted (raITV) internal target volumes. Static SFUD plans with 2-3 fields were optimized on a reference CT (full exhale phase CT for gITV&raITV, mid-ventilation phase for MV) delivering V95%=100% to PTV and 1, 4, and 8 time volumetric scaled rescanning simulated. 4D doses for CTV (V95% and D5-D95%) and organs-at-risk were evaluated for each starting phase (64 combinations for 2-field plans and 8-bin 4DCT and 1000 combinations for 3-field plans and 10-bin 4DCT).

Results: For all 3 approaches, V95% and D5-95% to CTV improved, and it's variation as function of phase reduced, with the number of rescans. For one typical patient and 8 rescans, fluctuation of V95%, expressed by the 1st and 3rdquartile difference (Q1-Q3), were approx. 8%, 2% and 3% for gITV, raITV, and MV, respectively. Similarly, Q1-Q3 for D5-95% were 7%, 3%, and 3% respectively. Median values of V95% and D5-95% to CTV for 10 all patients are presented in a box plot and shows that even 8 rescans may not guarantee sufficient coverage in all cases. Differences in doses to OARs were statistically insignificant.

Conclusion: For this patient group, best results were achieved for the raITV approach, while gITV&MV exhibit unsatisfactory V95% for some patients. gITV shows strong dependence on the starting phase.

PTC17-0525: Anatomical Comparisons of Thorax Patients Imaged in Both the Seated and Lying Positions

M. Pankuch1, B. Kreydick1, D. Hecksel1, A. Panchal1, S. Boyer1, H. Ramirez1, S. Laub1, M. Gao1, J.C. Chang 2, W. Hartsell 2

1Northwestern Medicine Chicago Proton Center, Medical Physics, Warrenville, USA 2Northwestern Medicine Chicago Proton Center, Radiation Oncology, Warrenville, USA

A prior study on healthy individuals has suggested that positioning in the seated position may increase lung volume and decrease the magnitude of respiratory motion when compared to patients in the supine position. A CT scanner capable of obtaining phase binned, 4-D volumetric image sets of patients in the seated position has been installed and brought into clinical use. In the initial phases of implementation, 4-D treatment planning images of patients were obtained in both the seated and lying position and were evaluated for anatomical differences and potential planning benefits. Lying patients were imaged in either the supine, or prone position, depending on the location of the target regions with consideration of the treatment delivery system's capabilities. At the time of abstract submittal, active patient accrual is ongoing and the most updated analysis will be presented.

Evaluation of patient anatomical differences between the seated and lying positions include comparisons of lung volumes on average scans and at phases of full inspiration and full expiration. Additional comparisons include target volumes, target motion vector comparisons and net motion magnitude differences. The position of the center of mass of the heart will be determined and compared between the lying and seated studies.

PTC17-0537: Intensity Modulated Proton Therapy for Re-Treatment of Thoracic Malignancy: Scripps Experience

F. Giap1, R. Lepage 2, L. Dong3, H. Giap4

1University of Texas Southwestern, Medical School, Dallas, USA 2Scripps Proton Therapy Center, Medical Physicis, San Diego, USA3Scripps Proton Therapy Center, Medical Physics, San Diego, USA4Scripps Proton Therapy Center, Scripps Proton Therapy Center, San Diego, USA

Purpose: IMPT has been used exclusively at our center for re-irradiation of recurrent malignancies in the chest over the past 3 years. This study describes our technique and reports our early experience.

Materials and Methods: We identified 15 patients in this category. Group 1 (N = 5 patients with extra-thoracic disease) was treated with palliative intent (doses from 45-55 Gy). Group 2 (N = 10 patients with no extra-thoracic disease) was treated with curative intent (doses from 60-70 Gy). The primary endpoint was loco-regional failure. Secondary endpoints included acute toxicities and overall survival. Patients, immobilized with a Vac-Q-Fix cushion, were set up in the supine position with arms over their head. All patients underwent 4-D CT simulation for treatment planning. For tumor motion < 1 cm, the ITV technique was done. For motion > 1 cm, motion was managed with the Deep Inspiration Breath Hold SDX system. CT with contrast and PET were used. Weekly adaptive CT simulation was done. 1-3 beams using IMPT with SFO technique was used.

Results: With median follow-up of 18 months, most patients had grade 1-2 lung and skin toxicities. There were no grade 3 or greater acute or late toxicities. 2 patients of 5 in Group 1 and 7 patients of 10 in Group 2 patients are still alive, of which, 5 of 7 has local control.

Conclusion: IMPT is a feasible and safe modality for re-treatment of recurrent cancer in the thorax after previous radiation. Further validation with more patients and longer follow-up is needed.

Miscellaneous

PTC17-0382: Creating the Clinical Nursing System for CIRT

F. Schwarz1, R. Ohtaka1, M. Wakatsuki1, M. Moteki1, H. Kanazawa1, Y. Nakayama 2

1Kanagawa Cancer Center, Nursing, Yokohama, Japan 2Kanagawa Cancer Center, Medicine, Yokohama, Japan

Purpose: Kanagawa Cancer Center had stared CIRT since December 2015. For the best outcomes with limited resources, Nursing Department defined and organized roles of the related sections to create the nursing system for CIRT. The purpose is to create the clinical nursing system for CIRT

Materials and Methods: Retrospective study with documents at working gropes and meetings.

Results: The nursing depertement decided the clinical nurse assigned outpatient section and the consulting service nurse assigned Patient Support Center. The clinical nurse attended most of the preparations and therapies, such as: process of decision-making, becoming aware of expectations, worries, and difficulities to the therapy, attending to prepare immobilization devices and imaging simulation CT. Additionally create tools for empowering patient self-care abilities and supporting after-care of the therapy.

Conclusion: It was significantly important to define each role of the nurse to concentrate their specific tasks, such as: clinical care or consulting service. Creating the clinical nurse roles for CIRT were to support for patient decision-making continuously, to care about adjustment to the theraputic environment, to assist active self-care management, to support a patient as a person through the managemant of the side-effects or symptoms due to the therapy, and to appreciate all the efforts from patients undergoing CIRT.

PTC17-0402: A Prospective Study to Evaluate the Safety of the World-First Spot-Scanning Dedicated, Small 360-Degree Gantry, Synchrotron-Based Proton Beam Therapy System

K. Nishioka1, S. Shimizu1, K. Yasuda 2, K. Ono3, T. Hashimoto 2, N. Katoh4, T. Inoue4, K. Tsuchiya4, R. Onimaru 2, H. Shirato 2

1Hokkaido University, Radiation Oncology, Sapporo- Hokkaido, Japan 2Hokkaido University, Radiation Medicine, Sapporo- Hokkaido, Japan3Hokkaido University Hospital, Clinical Research and Medical Innovation Center, Sapporo- Hokkaido, Japan4Hokkaido University Hospital, Radiation Oncology, Sapporo- Hokkaido, Japan

Purpose: Since March 2014, we have prospectively investigated the safety of the treatment with a spot-scanning dedicated, small 360-degree gantry, synchrotron-based proton beam therapy (PROBEAT-RT) system. Here we report the incidence of the primary endpoint, early serious adverse events (SAE), Grade 4 or higher, related to proton beam therapy (PBT).

Materials and Methods: All patients were treated with the schedules which had been reported to be safely used in PBT literatures. The prescribed total dose ranged from 20 to 76 GyE (RBE=1.1) with the median dose of 65 GyE in 4 to 35 fractions. Fifty-six patients who followed 12 months or more were included in this study. Primary tumor sites were prostate (n=17), bone/soft tissue (n=10), liver (n=7), lung (n=6), central nervous system (n=5), colon (n=2), pancreas (n=2), kidney (n=2), and others (n=5). Chemotherapy was given concurrently for one patient and in adjuvant setting for 10 patients. Adverse events were recorded every week during the treatment period, then every month until 3 months and every 3 months thereafter. All SAE was reported and the relationship with PBT was then investigated.

Results: There were no SAE related to PBT within 3 months, nor 12 months, after the treatment. Two patients experienced Grade 4 adverse events which were attributed to chemotherapy; thrombocytopenia 1/56, neutropenia 2/56.

Conclusion: The PBT using PROBEAT-RT system was administered without early SAE for various sites. Its efficacy and long-term safety are under evaluation.

PTC17-0422: New Insights into Metallic Nanoparticles for Enhancement of Particle Therapy in Hypoxic Tumors

M. Bolsa Ferruz1, E. Porcel1, D. Salado Leza1, L. Stefancikova1, S. Lacombe1

1Institut des Sciences Moléculaires d'Orsay - CNRS, Biophysics and Biophotonic, Orsay, France

Low oxygen concentration in tumors results in lower cell death after exposure to radiation. The oxygen effect is expressed by the Oxygen Enhancement Ratio (OER) which depends, among others, on the oxygen level and the linear energy transfer (LET) of the radiation. Particle therapy benefits the treatment of hypoxic tumors compared to radiotherapy due to a decrease of OER when LET increases [Scifoni et al., 2013]. However, the irradiation of healthy tissues at the entrance channel remains a major limitation.

Nanotechnology brought new perspectives of using high-Z nanoparticles (NPs) to increase local radiation-effect. Previous studies performed by the group demonstrated that the radio-enhancement due to PtNPs is mostly related to the production of water radicals (OH) produced in the vicinity of the NPs [Porcel et al., 2010]. In parallel, it has been observed that the contribution of OH-mediated cell damage is strongly influenced by the presence of oxygen [Hirayama et al., 2013].

In the collaborative works performed at the Heavy Ion Medical Accelerator in Chiba (HIMAC, Japan) and at GSI-HIT (Germany), we investigated the effect of metallic NPs on human cancer cells incubated in oxic and hypoxic conditions and irradiated by carbon and helium ions. In particular, the addition of NPs improves the cell killing induced by carbon ions (LET=100keV/μm) in both oxic and hypoxic conditions. For the first time it is shown that the dose deposited in the normal tissue in front of the tumor can be decreased while maintaining the same tumor control even in the hypoxic cells.

PTC17-0510: Dynamic Operational Staffing Models Based on Patients under Beam (PUB) for Proton Therapy Centers

J. Dixon1, T. Wong1, A. Andrews1, R. Rengan1

1Seattle Cancer Care Alliance Proton Therapy Center, Proton Therapy Center, Seattle, USA

Purpose: Developing an efficient, dynamic staffing model that is responsive to changes in patient disease acuity, patient treatment complexity, per patient treatment time, and patient volume is critical to the success of a newly opened multi-room particle beam facility. Failure to account for all the relevant parameters can result in workflow inefficiencies and ultimately can negatively impact patient care. Here, we present a reliable method of creating a staffing model that is based upon Patients under Beam (PUB).

Materials and Methods: In our model, we considered the workflow of numerous proton centers when assigning clinical as well as administrative activities for staff. Patients under Beam (PUB) is the primary driver for staffing. PUB number is a product of other factors: leads, qualified patients, consults, financially cleared patients, CT simulations, and starts. These intermediate steps demonstrate the complexity involved in ancillary (non-treatment) staffing and their participation in the patient experience. Beyond clinical activities, we took into consideration operational and departmental meetings, education, and training requirements– both external and internal, which are volume-independent staffing requirements. As volumes continued to steadily grow, the center could layer PUB numbers over the baseline staffing needs that was responsive to patient demand.

Results: With all variables under consideration, we hire staff at appropriate intervals with high confidence.

Conclusion: A dynamic staffing model, based upon the above parameters, allows for safe and fiscally responsible center growth.

PTC17-0529: Implementation and International Use of a Proton Therapy User Training Program

C. Hill Kayser1, M. Kirk1, C. Misher1, S. Lowitz1, M. Hampshire1, R. Maughan1, N. Vapiwala1, J. Metz1

1University of Pennsylvania, Radiation Oncology, Philadelphia, USA

Purpose: New proton centers are being built and opening rapidly. Very limited proton-specific educational programs and certifications have historically existed, and an expansion of formal training programs for clinicians is needed to meet rising demand.

Materials and Methods: A training program for proton therapy training was developed, consisting of e-learning modules with lectures delivered by Penn Medicine faculty and staff, on-site visits to Penn Medicine in Philadelphia, and supervision of commissioning at the training center. Certificate of completion is provided to participants who complete an examination.

Result: A cooperative program was developed between Penn Medicine's Department of Radiation Oncology, OncoLink (www.oncolink.org) and IBA (Ion Beam Applications) to provide specific proton education. The program focuses on proton-specific knowledge including operations, treatment planning, quality assurance, beam measurement/ calibration, and image guidance. Over 2 years, the course has been attended by 324 trainees from 7 countries (US, Sweden, Poland, Korea, Belgium, China, Denmark). Most attendants have been physicists (30%) and clinicians (26%), followed by dosimetrists (20%) and therapists (10%). Trainee satisfaction with the session and instructor quality has been high: 97% report session quality to be good-excellent, and 98% instructor quality good-excellent.

Conclusion: A training program resulting in certificate of completion has been successfully developed used internationally. Trainee satisfaction with program quality is very high. As more proton therapy centers are planned to open, rigorous user training is essential to establish centers of clinical excellence. This may be accomplished with a model that includes e-learning modules as well as a train-the-trainers approach with on-site and hands on instruction.

PTC17-0535: Concurrent Pencil Beam Scanning Proton Therapy and Hyperthermia: Initial Clinical Experience

J. Snider1, J. Molitoris1, A. Chhabra1, T. Diwanji1, W. Regine 2, Z. Vujaskovic 2, C. Simone 2

1University of Maryland Medical Center, Maryland Proton Treatment Center/Department of Radiation Oncology, Baltimore, USA 2University of Maryland School of Medicine, Maryland Proton Treatment Center/Department of Radiation Oncology, Baltimore, USA

Purpose: Concurrent hyperthermia (HT) with radiotherapy is known to increase efficacy through multiple mechanisms including radiosensitization of hypoxic cells and inhibition of DNA-repair. HT reduces the oxygen enhancement ratio of “low”-linear energy transfer (LET) radiation (photon/proton) and increases radiobiologic effect (RBE), potentially mimicking high-LET particle therapy (12C ion) [Datta et al. 2014]. Naturally, both enthusiasm for improved outcomes and concerns regarding increased toxicity have arisen, yet limited data exists to date.

Materials and Methods: At the Maryland Proton Treatment Center, over 200 patients have been treated with pencil beam scanning proton therapy (PBSPT). Eighty-seven patients have been treated with HT in 4 years. All HT has been delivered on the BSD-500 platform with the 40-42°C target tumor temperature. Three patients have been treated with concurrent PBSPT and ETT: two patients, postoperatively for myxofibrosarcoma, and one, for inguinal recurrence from vulvar squamous cell carcinoma (SCC). The patients' treatment courses were: inguinal vulvar SCC (reirradiation-45 Gy(RBE),9xETT), high-grade chest wall myxofibrosarcoma (reirradiation-60 Gy(RBE),5xETT), intermediate-grade shoulder myxofibrosarcoma (66 Gy(RBE),4xETT).

Results: All patients completed their courses of proton and hyperthermia treatment without substantial acute complication. The following toxicities were encountered: G1-fatigue (3), G2-radiation dermatitis (3), G1-hyperpigmentation (3), G1-pain (2), G1-range-of-motion limitation (1), and G2-soft-tissue necrosis (1). The patients with myxofibrosarcoma have currently no evidence of disease; the patient with SCC has persistent disease outside of the reirradiation field.

Conclusion: Concurrent PBSPT and ETT appears safe, effective, and promising. Further investigation and expansion of clinical experience is warranted amongst institutions with technical capabilities.

Pediatric

PTC17-0155: The Place of Particle Therapy among Novel Technologies in Pediatric Tumors. An Update on Toxicity

J.L. Habrand1, D. Stefan1, S. Bolle 2, D. Lecomte1, J. Datchary1, S. Helfre3, C. Alapetite3

1Centre François Baclesse, Radiation Oncology, CAEN, France 2Gustave Roussy Cancer Campus, Radiation Oncology, Villejuif, France3Institut Curie, Radiation Oncology, Paris, France

Purpose: Assessing the impact of modern technologies including particle therapy [PT] on toxicity and sequelae in children and adolescents.

Materials and Methods: We performed an extensive review of published articles, from 01/2005 through 12/2015 through MEDLINE, according to the following key-words: Radiotherapy, English, Toxicity (acute & late sequelae), Novel technologies.

Results: 128 articles were selected, and classified in 3 groups: I] Pre-clinical studies: (dosimetrical, mathematical models): 42 (33%) II] Clinical studies: 71 (55%), ranging from small (S<20pts:19), to medium (M<50pts:22), large (L<100pts:13), very large (XL<500pts:15), and extra-large (XXL≥500 pts:2) cohorts, totalizing >7,000 pts. III] Other studies (economical, general): 15 (12%). Types of toxicity were: CNS/Quality of life [QOL]: 32%; Second cancers [K2]: 21%; (Neuro)endocrine: 15%; acute toxicity: 14%; cosmesis: 5%; others: 13%. Types of technologies: PT: (36%); advanced XR esp. IMXRT: 36%; 3D-XR: 18%; others: 9%. The place of PT increased with time (2015: PT:20, IMXRT:8, 3D-XR:5). As far as intercomparisons: dosimetrical: 87%, models:59%, economical:25%, clinical:18%, others:25%. Only 1 clinical study compared 3D-XR vs IMXRT and 13 XR vs PT (Brain & QOL:4, endocrine:3, acute:3, and others:3). Conclusions were: IMXRT>3D-XR:1 (p>.001), PT>XR:8 (p>.05: 6), XR>PT:3 (p>.05: 2), PT=XR:2.

Conclusion: Most pre-clinical evaluations pointed-out PT superiority (early/late toxicity). Few clinical comparative studies between PT & XR were performed, and none randomized. Most PT were related with protons. Through literature, the interest for preventing side-effects by PT has become considerable. Most pre-clinical studies display a clear advantage for PT. Clinical comparative studies remain relatively rare, but most show similar benefit esp. for brain/QOL, and endocrine toxicity.

PTC17-0380: Proton Beam Therapy for Pediatric Cancer with Endotracheal Intubation under the General Anesthesia: A Report of Two Cases

T. Hashimoto1, K. Tsuruga 2, H. Kobayashi3, A. Iguchi4, S. Honda5, N. Fujita 2, S. Shimizu6,7, S. Terasaka3, Y. Morimoto 2, H. Shirato1,7

1Hokkaido University Graduate School of Medicine, Department of Radiation Medicine, Sapporo, Japan 2Hokkaido University Graduate School of Medicine, Department of Anesthesiology and Critical Care Medicine, Sapporo, Japan3Hokkaido University Graduate School of Medicine, Department of Neurosurgery, Sapporo, Japan4Hokkaido University Graduate School of Medicine, Department of Pediatrics, Sapporo, Japan5Hokkaido University Graduate School of Medicine, Department of Gastroenterological Surgery I, Sapporo, Japan6Hokkaido University Graduate School of Medicine, Department of Radiation Oncology, Sapporo, Japan7Hokkaido University, Global Institution for Collaborative Research and Education GI-CoRE, Sapporo, Japan

Purpose: Proton beam therapy (PBT) provides dosimetric benefits in sparing normal tissue when treating pediatric cancers. We report two pediatric cases requiring general anesthesia with endotracheal intubation during PBT.

Case 1: An 1-year-old boy with posterior fossa tumor was referred to our hospital. Tumor removal was performed and gross total resection was achieved. Pathological findings indicated anaplastic ependymoma (WHO grade 3). He received PBT (CTV D99%=54.0 GyE in 30 fractions) under general anesthesia with endotracheal intubation. No severe late toxicity, including hoarseness, has been observed 1 year and 10 months after PBT with local control.

Case 2: A 3-year-old boy was diagnosed as adrenal or retroperitoneal neuroblastoma with skeletal and bone marrow metastases (stage 4). He underwent five courses of chemotherapy with vincristine, cyclophosphamide, tetrahydropyranyladriamycin, and cisplatin. Surgical resection of residual tumors in the area of primary tumors was performed after umbilical cord blood transplantation. Then he was referred to our hospital to receive PBT. The total dose delivered to 99% of the CTV was 19.8 GyE in 11 fractions. Although the observation period was limited, no severe toxicity was observed during and after PBT under general anesthesia with endotracheal intubation.

Conclusion: We experienced 2 pediatric cases of PBT with endotracheal intubation under the general anesthesia. Further studies of a large number of patients with long term follow-up are required to evaluate the efficacy and the safety of this treatment.

Sarcoma/Lymphoma

PTC17-0102: Proton Beam Therapy in Sarcomatous Tumors at West German Proton Therapy Center Essen (WPE)

D. Geismar1, T. Steinmeier 2, S. Peters1, S. Plaude 2, B. Timmermann1

1Clinic for Particle Therapy- University Hospital Essen, West German Proton Therapy Center Essen WPE, Essen, Germany 2University Hospital Essen, West German Proton Therapy Center Essen WPE, Essen, Germany

Purpose: Proton beam therapy (PT) is an attractive part of treatment in tumors with close proximity to critical structures.

Materials and Methods: Between May 2013 and November 2016, 197 patients (82 adults, 115 children, age 14.9 y (0.9-84.6 y) with sarcomatous tumors were treated at West German Proton Therapy Center Essen (WPE) and were prospectively enrolled in the in-house registry. Histology types were rhabdomyosarcoma (37.1%), chordoma (15.7%), EWING sarcoma (12.7%), chondrosarcoma (6.6%) and miscellaneous (27.9%). Radiation sites were brain/head and neck (58.9%), spinal/paraspinal (22.8%) and pelvis (18.3%). In 49.2% of cases, concomitant CTx was applied.The median PT dose was 55.8 Gy (36.0-74.0 Gy) applied in mean 31 fractions (18-41) by mostly using uniform scanning (51.3%) and pencil beam scanning (40.6%), respectively.

Results: The median follow-up after last fraction is 0.7 years (0.0-3.0 y). 157 patients (79.7%) showed disease control. Local and distant or both failure occurred in 16, 19 and 5 patients, respectively. 16 patients died so far (local progress (n=10), systemic progress (n=6)). PT was well-tolerated. New high-grade (CTCAE ≥°3) acute toxicities predominantly occurred in the fields of gastrointestinal (n=10), hematological (n=8) and general disorders (n=6). Long-term data for 12 and 24 months after PT show a low number of new high-grade (CTCAE °3) toxicities regarding the groups general (pain, lymphedema, musculosceletal) (n=3) and anemia (n=1). No new grade 4 or grade 5 effects occurred.

Conclusion: Current data support safety, tolerance and effectivity of PT in sarcomatous tumors. However, long-term follow-up data is still necessary to assess long-term outcomes.

PTC17-0284: Proton Radiotherapy for Mediastinal Hodgkin Lymphoma: Single Institution Experience

J. Kubes1, K. Dědečková1, P. Vítek1, S. Vinakurau1, B. Ondrova1, V. Vondracek1, M. Andrlik1, N. Radostová1

1PTC Prague, Proton therapy dept., Prague, Czech Republic

Purpose: Patients with Hodgkin lymphomas indicated for radiotherapy are good candidates for proton radiotherapy due to potential of proton beam to reduce doses to healthy structures. This sparing effect is most pronounced in mediastinal disease. Mediastinal proton radiotherapy can be safely provided with motion management.

Materials and Methods: Between May 2013 and June 2016, 39 patients (pts) with Hodgkin lymphoma (HL) underwent mediastinal proton radiotherapy (RT). Pencil beam scanning (PBS) technique was used in all pts. Overall 33 of 39 pts were evaluable for acute toxicity and early response. Proton RT was indicated in the first-line treatment in 30 pts, 3 pts were re-irradiated after photon RT. Median age at the time of RT was 32 years (range, 13-59 years). RT volume definition: involved field 9 pts, residual disease 10 pts, involved site 14 pts. 23 pts had PET negative disease and 10 pts had PET positive residual disease. Median total dose was 30 GyE (range, 19.8-40 GyE). Overall 17 pts underwent PBS RT in deep inspiration breath hold (DIBH), repainting strategy was used in rest of patients.

Results: Of evaluable pts, 31 are in complete remission. Progression outside of treatment volume was observed in two patients. Acute and subacute RT toxicity was mild (pharyngeal mucositis gr. 2 in 3 pts, leukopenia gr.3 in 1 pt, leukopenia gr 2 in 1 pt, radiodermatitis gr.2 in 1 pt). No case of radiation pneumonitis or Lhermitte syndrom was observed.

Conclusion: Proton RT offers promising and safe option for most pts indicated for mediastinal RT.

PTC17-0443: The Role of Gradient Matching Technique Using Intensity Modulated Proton Therapy in Treatment of Sarcoma

J. David1, X. Ding1, X. Li1, D. Moore 2, P. Kabolizadeh1

1William Beaumont Hospital, Radiation Oncology, Royal Oak, USA 2William Beaumont Hospital, Orthopedic Oncology, Royal Oak, USA

Purpose: Given the majority of spot scanning proton units have limited field sizes, treatment volume may exceed the maximum field dimensions. Herein we developed a novel spot scanning proton planning technique in management of patients with large sarcoma target volumes using gradient matching technique.

Materials and Methods: One preoperative and one post-operative male patient with large upper thigh sarcoma was chosen. IMPT plans were created via multi-field gradient matching technique with two parallel posterior oblique fields utilizing superior and inferior iso-centers respectively. In the overlapping region of multi-field plan, an effective 10 cm slope gradient matching was created to ensure an adequate robust planning. As a comparison, IMRT and 3DRT were also created using similar plan objectives.

Results: A linearly gradient matching with penumbra of 20%-80% over 10cm was created with homogeneity between the matching areas. All plans were evaluated based on RTOG and resulted in adequate target coverage. IMPT resulted in a superior sparing of adjacent normal structures including bone, femoral head, anal canal, genitals, testicles and longitudinal skin. Specifically the mean dose to the testicles were 2.6Gy, 1.8Gy, and 0Gy for IMRT, 3DCRT, and IMPT respectively in the preoperative case. Similar results were noted in the postoperative case with 4.16 Gy, 2.14 Gy, and 0.33 Gy mean dose to testicles respectively.

Conclusion: IMPT was superior in delivering the lowest total dose to normal tissues, relative to IMRT or 3DCRT. While IMRT and 3DCRT were acceptable in most constraints, IMPT is shown to prevent any radiation effect causing azospermia.

PTC17-0447: Definitive Proton Therapy for Pelvic Sarcoma Using a Retroperitoneal Tissue Expander for Bowel Displacement

J. Ashman1, S. Korte1, A. Anand1, T. Vern-Gross1, S. Keole1, M. Bues1, R. Gray 2

1Mayo Clinic, Radiation Oncology, Phoenix, USA 2Mayo Clinic, Surgical Oncology, Phoenix, USA

Purpose: Proton therapy provides a clinically significant advantage over photons for unresectable sarcoma. However, safe delivery of escalated, curative-intent doses may be limited by adjacent bowel. We describe a case of pelvic sarcoma where definitive proton therapy was facilitated by the placement of a retroperitoneal tissue expander.

Case Report: A 70-year-old woman presented with left hip pain, leg weakness, and foot drop. Imaging demonstrated a 14.5 × 12.5 × 11 cm destructive mass of the left ilium without metastasis. Biopsy confirmed high-grade pleomorphic sarcoma. Multidisciplinary tumor board recommended against radical surgery in favor of neoadjuvant doxorubicin, ifosfamide, mesna. A partial response with no metastatic progression was achieved after 3 cycles. Definitive proton therapy was recommended next, but bowel immediately abutting tumor was dose limiting. Surgical oncology placed a 10 × 12 cm tissue expander with 250 ml of saline in the retroperitoneal space. A dose of 74 CGE in 37 fractions was delivered using a 3-field plan with pencil-beam scanning and single-field optimization technique. D1cc to large and small bowel was 4522 CGE and 1170 CGE, respectively. Acute toxicity included only Grade 1 dermatitis. At 3-month follow-up, the patient had significant improvement in pain and mobility correlating with a marked decrease in tumor size to 11.5 × 7.5 × 9.4 cm. Follow-up will be further updated.

Conclusion: Close coordination with surgical oncology for placement of a tissue expander facilitated optimal delivery of high dose proton therapy for curative-intent treatment of pelvic sarcoma.

PTC17-0475: Dose Escalation with Pencil-Beam-Scanning (PBS) Proton Therapy and VMAT Photon Therapy for Recurrent Chordomas: A Feasibility Study

S. Yan1, G. Broussard1, T. Nguyen1, T. DeLaney1, K. De Amorim Bernstein1, Y. Hsiao-Ming1, Y. Wang1

1Massachusetts General Hospital, Radiation Oncology, Boston, USA

Purpose: Chordomas locally recurrent after surgery are difficult to control. Increasing the total radiation dose has the potential of increasing local control. We investigated the feasibility of dose escalation (84.6GyRBE, 1.8GyRBE/fraction) to the gross-tumor-volume (GTV) for recurrent chordomas while respecting dose constraints to organs-at-risk (OARs) using different PBS beam spot sizes and beam angles as well as VMAT planning.

Materials and Methods: Radiation plans were generated to provide dose escalation (84.6GyRBE) to the recurrent GTV with a 3-mm margin. For the PBS plans, two sets of beams were used: one with posterior-anterior beam only, and one with two beams of left- and right-posterior-oblique. Spot sizes used: 8-mm to 15-mm sigma and 2.3-mm to 4.4-mm sigma in air at isocenter with energies from 220-MeV to 90-MeV. VMAT plans used four partial arcs.

Results: PBS (spot size of 2.3-mm to 4.4-mm) provided the best mean and max dose to OARs and better GTV coverage. Two posterior-oblique beams reduced the mean dose to small bowel and bladder compared to a single PA beam. Intergluteal-fold dose was lower with two posterior-oblique beams. VMAT plans had similar max and mean dose to some OARs compared to spot size of 8-mm to 15-mm.

Conclusion: High dose radiation (84.6GyRBE) can be delivered to GTV while respecting dose constraints to OARs. PBS generally provides lower mean OARs dose compared to VMAT. Institutions with larger spot size should compare PBS and VMAT plans for certain OARs. Institutions with smaller spot size, two posterior-oblique beams may be favored compared to posterior-anterior beam.

PTC17-0532: Impact of High Dose Proton Based Radiation on Computed Tomography Bone Density of Healthy Bone in Sacral Chordoma Patients

Y.L. Chen1, O. van Wulfften Palthe 2, M. Bredella3, K.W. Jee1, J. Schwab4, T. DeLaney1

1Massachusetts General Hospital, Radiation Oncology, Boston, USA 2Massachusetts General Hospital, Orthopedic Oncology Service, Boston, USA3Massachusetts General Hospital, Radiology, Boston, USA4Massachusetts General Hospital, Orthopedic Oncology, Boston, USA

Purpose: High dose proton based radiotherapy is an effective adjuvant therapy for sacral chordomas. However, up to 2/3 of patients suffer insufficiency fractures after surgery and protons. The mechanism for the fractures is unknown.

Materials and Methods: Twenty-one patients who underwent pre-operative high dose radiotherapy (50.40 GyE) for sacral chordoma between 2009 and 2015 had both proton simulation CT and surgical planning CT (post-RT) available. We used Hounsfield unites (HU) as a surrogate for bone density. Volumetric HU were measured outside (L1 and L2) and within the proton field. To adjust for contrast-induced variations, we calculated a ratio of in field versus out of field HU. For control we compared L1 and L2 HU changes pre- and post-RT. Statistical analyses were preformed using the paired t-test.

Results: Among the 21 analyzed patients the ratio of HU in L2 relative to L1 did not change after radiotherapy (pre-RT mean, 0.976 ± 0.009 and post-RT mean, 0.979 ± 0.009; p = 0.799). However, the ratio of the HU within the sacral proton field relative to to L1 did change significantly after radiotherapy (pre-RT mean, 0.895 ± 0.050 and post-radiation mean, 0.658 ± 0.050; p < 0.001).

Conclusion: We were able reliably to compare pre- and post-proton changes in bone density in sacral chordoma patients using retrospectively collected CT scans. We observed a significant decrease in bone strength in the irradiated sacrum. This finding may impact surgical stabilization or pharmacologic management to minimize the risk of fracture as a complication of high dose proton therapy.

Acceptance, Commissioning, and Quality Assurance

PTC17-0017: Apollo Proton Therapy Center Planning & Installation - Indian Experience

J. Chandy1

1Apollo Proton Centre, Proton, Chennai, India

The Proton Treatment Centre at Apollo Hospitals, Chennai will be the first proton cancer centre in South East Asia to identify priority areas for paediatric cancers, Indian-specific research studies, and a 1-year proton fellowship program. Foundation will develop cancer control strategies which will include newer modalities of prevention, early diagnosis and treatment of cancer.

The Proton Treatment Centre will focus on organ specific cancer management and will have a dedicated Apollo oncology team. The Proton Beam Therapy provides an advanced radiation treatment option for the oncologists. The treatment provided is highly precise and focused radiation directly into the tumor, without causing any harm to the adjoining tissues. Tumours in difficult to access areas such as, in head, neck, brain, pancreas, and prostate will be targeted. Paediatric cases will also benefit to a greater extent from this technology.

The Proton Treatment Centre is an exalted project for Apollo Hospitals with the below mentioned technical specifications:

Model - Proteus Plus with

  1. Two gantry treatment rooms

  2. One fixed beam treatment room

  3. All Rooms with latest IMPT Technology

  4. Dedicated pencil beam scanning modes in all three rooms

  5. 3 D Cone beam CT in Two gantries for Online IGRT verification

PTC17-0067: Residual Radioactivity of Brass Aperture Induced by a Proton Beam of Wobbling System: Volume Source Model Investigation

H.Y. Tsai1, B.Y. Wang 2, H.H. Chen3, R.J. Sheu 2

1Chang Gung University, Medical Imaging and Radiological Sciences, Taoyuan City, Taiwan- Province of China 2National Tsing Hua University, Institute of Nuclear Engineering and Science, Hsinchu, Taiwan- Province of China3Chang Gung Memorial Hospital at Linkou, Department of Radiation Oncology, Taoyuan, Taiwan- Province of China

The purpose of this study is to determine the residual radioactivity of the patient specific brass aperture induced by a proton beam of single-ring wobbling system and the corresponding radiation exposures to patients and staff. A HPGe detector was used to determine the gamma ray spectrometry of the activated brass block. Then the radionuclides were analyzed. We used the efficiency transfer method to determine the residual activity. The volume source models were developed according the activity distribution within the brass block. The fitting equations of efficiency transfer factors were established considering the activity distribution along the proton beam direction. Although the ambient dose rate caused by the activated brass drops lower to the level of background radiation within around 3 hours, the residual activity from induced radionuclides with longer half-lives such as the cobalt serious should be concerned with respect to radiation safety.

PTC17-0071: Initial Experience in Calibration of CT for Proton Therapy

K.W. Ang1, W.Y.C. Koh 2, C.L.J. Lee1

1National Cancer Centre Singapore, Division of Radiation Oncology, Singapore, Singapore 2Nanyang Technological Institute-Singapore, School of Physical and Mathematical Sciences, Singapore, Singapore

Purpose: To evaluate CT calibration for proton therapy.

Materials and Methods: CIRS and Gammex_467 phantoms, with mixed lung, tissue, bone material plugs and individually placed plugs in an acrylic phantom, are scanned with Siemens Somatom Definition AS 64 CT-simulator, using head, thorax, abdomen and pelvis scan protocols as configured in our centre. The stoichiometric method of CT calibration by Schneider and Pedroni is followed to calculate 24 sets of coefficients due to photoelectric absorption, coherent and incoherent scattering. To evaluate the goodness of the coefficients, they are used to calculate the theoretical HU of the plugs based on their known chemical compositions, and compared against measured HU. 24 sets of human tissue HU are also calculated based on chemical compositions estimates from ICRU46. These are plotted against their respective relative stopping power ratios at 150MeV proton energy. They are used in Varian's Eclipse Treatment Planning System version 13.6. A nasopharyngeal carcinoma plan with 2 oblique posterior fields using multi-field optimization IMPT is planned & calculated with each calibration set. The dose distributions are analysed to gain an insight of the range of uncertainties associated with CT calibration.

Results: Choice of scan protocols does not significantly affect (<1%) theoretical HUs. Using mixed or individual plugs arrangements show at most 1.6% difference in theoretical HU. CIRS and Gammex_467 tissue, lung plugs give no significant differences in theoretical HU. CIRS bone plugs show 1.23% better results than Gammex_467. Treatment plan evaluations are still in progress.

Conclusion: We have established an initial calibration technique for our centre.

PTC17-0077: Monte Carlo Characterization of Relative Response of Liquid-Filled Ionization Chambers in Monoenergetic Proton Pencil Beams

M. Chan1, O. Blanck1, H. Shimada 2

1Universitätsklinikum Schleswig–Holstein, Klinik für Strahlentherapie, Kiel, Germany 2Gunma University, Heavy Ion Medical Center, Gunma, Japan

Purpose: To investigate the response of a liquid-filled ionization chamber (LIC) array for patient-specific plan quality assurance (QA) in monoenergetic scanned proton beam therapy.

Materials and Methods: PENELOPE Monte Carlo code extended to proton which models elastic electron coulomb scattering (MSC) and non-elastic nuclear interaction was used to simulate monoenergetic proton pencil beams (FWHM = 5.5 mm) of 100, 150, and 200 MeV. Each simulation used 1x106 particles. Detailed modelling of the LIC array (Octavius SRS 1000) was carried out according to the blueprint provided by the manufacturer (PTW, Freiburg, Germany), and its response relative to water at the water equivalent path length (WEPL) depth was estimated at the pristine Bragg peak (BP), and the build-up region at 2 cm for 100 MeV and 5 cm for 150 and 200 MeV proton.

Results: By estimating the shift of the BP, the estimated WEPL of the build-up material was within 1 mm of the physical thickness independent of the proton beam energies. The relative response varies with the proton energies from 1.21 to 1.18 at 2 cm and BP for 100 MeV, 1.13 to 1.11 for 150 MeV, and 1.07 for 200 MeV at 5cm and BP, respectively, with an estimated uncertainty of ∼ 2%.

Conclusion: Response correction may need to be applied to the LIC array for the use in patient plan QA measurement. Further investigations of the LIC's response with better uncertainty in the SOBP, and clinically relevant field sizes are warranted to understand its potential in proton plan QA.

PTC17-0078: Preclinical Trials of the Proton Therapeutic Facility “Prometheus"

S. Ulyanenko1, I. Gulidov 2, V. Balakin3, S. Koryakin1, A. Lychagin1, A. Solovev1, E. Beketov1, E. Koryakina1, V. Galkin4, A. Kaprin5

1A. Tsyb Medical Radiological Research Centre, Radiation biophysics, Obninsk, Russian Federation 2A. Tsyb Medical Radiological Research Centre, Radiation therapy, Obninsk, Russian Federation3Closed Joint Stock Company PROTOM, Administration, Protvino, Russian Federation4A. Tsyb Medical Radiological Research Centre, Administration, Obninsk, Russian Federation5National Medical Research Radiological Centre, Administration, Obninsk, Russian Federation

Purpose: Physico-dosimetric and radiobiological results of preclinical studies as start clinical research of head and neck tumors on the proton therapeutic facility “Ptometheus” with active scanning beam.

Materials and Methods: Compact synchrotron with 5 m diameter, manufactured by CJSC PROTOM, energy range 30-250 MeV, beam size is 1.4-7 mm depending on the extracted energy. Computational studies ware performed using software packages based on Monte Carlo: SRIM_2013; MCNPX 2.6, GEANT4. Absorbed dose measurements were conducted using ionization chambers: plane-parallel PPC40 (IBA), TM30010-1; Gafchromic EBT3 dosimetry films. Radiobiology testing was conducted both in vitro (B-16, V-79 cells) and in vivo (rats with sarcoma M-1).

Results: Calculations and dosimetry studies results have shown that the proton facility “Prometheus” parameters meet the quality assurance requirements for the radiation therapy. The radiobiological experiments verified the RBE (1.0-1.25) in case of single or multiple beams irradiation. The high precision of patient positioning is achieved due to built-in X-ray tomography system with a conical beam. Conformal proton radiation therapy was provided to more than 60 patients with different head and neck tumors. The results (as for December-2016) indicate a high efficiency of the radiation therapy using “Prometheus” facility.

PTC17-0080: Positron Emitter Predictions for Carbon Ion Therapy Based on a Filtering Method

A. Fochi1, T. Hofmann1, K. Parodi1, M. Pinto1

1LMU Munich, Medical Physics, Garching, Germany

The characteristic depth-dose curve of ions allows for a lower integral dose in the healthy tissue and a better dose conformation to the tumor volume with a reduced number of fields, in comparison to photon radiotherapy. The high ballistic precision and steep dose gradients demand also an accurate verification of the delivered dose. One approach relies on positron emission tomography to measure the positron emitter distribution (PED) produced by nuclear interactions of the therapeutic beam with the target nuclei. The measured PED is compared with a prediction, usually obtained by Monte Carlo simulations. In proton therapy, it was shown that a filter can be developed to predict the PED directly from the depth-dose profile [1]. In this work, a modified filtering formalism is developed for carbon ion therapy, which takes energy dependence into account, as well as different behavior of positron emitters produced by the projectile and the nuclei in the tissues. It provides submillimeter accuracy in the distal PED falloff, and adequate description of the fragmentation tail and the build-up region, yielding an overall good agreement when compared to simulated data. The approach was verified with simulations in different homogeneous and inhomogeneous phantoms and is currently being assessed with patient data. Future steps comprise the benchmark with actual measured PET data, and the implementation of the method in a research TPS.

[1] Parodi and Bortfeld 2006, Phys. Med. Biol. 51

PTC17-0087: Dosimetric Impact of Spine Hardware in IMPT Treatments

N. Depauw1, T. Ruggieri1, E. Batin1, J. Schwab 2, F. Hornicek 2, Y.L. Chen1, T. DeLaney1, H. Kooy1, K. De Amorim Bernstein1

1Massachusetts General Hospital, Radiation Oncology, Boston, USA 2Massachusetts General Hospital, Orthopaedics, Boston, USA

Spine sarcoma patients often undergo post-surgery radiotherapy. As a consequence of the surgery, bone-supporting metallic implants (e.g. spine cage, rods) are often located in the vicinity of the area requiring radiation. The implants create large imaging artifacts on CT scans. Although attempts are made to avoid and mitigate its effects at the time of treatment planning, this hardware remains the source of large dosimetric uncertainties during dose calculation and beam delivery. These uncertainties become far more relevant in IMPT for which individual spot would be affected, possibly resulting in large dose heterogeneities within and around the target volume.

Therefore, 2 homemade phantoms were built in order to accurately evaluate the dosimetric impact of spine hardware in IMPT treatment. Both phantoms, rectangular in shape, are made with Lucite walls and contain bony-equivalent vertebrae (1.36 RSP) with realistic shape, as well as a removable spinal-cord equivalent insert (0.96 RSP) which can be replaced with an equivalent insert including a micro ionization chamber (IC) indexed at various depths. The first phantom also contains an actual spine cage and titanium rods anchored in the vertebrae.

The second one contains an actual sacral plate with cobalt chrome rods. CT scans of these two phantoms will be reconstructed with and without metal artifact reduction. Various clinically realistic IMPT plans will then be generated and delivered. Planar measurements of the exit dose, as well as IC measurements around the target will be performed for each plan and compared to the expected dose.

PTC17-0095: Current Status of Carbon-Ion Therapy Facility Project of Yamagata University

T. Iwai1, K. Nemoto 2, H. Yamashita1, I. Kubota3, T. Kayama1

1Yamagata University, Faculty of Medicine, Yamagata, Japan 2Yamagata University, Yamagata University Hospital, Yamagata, Japan3Yamagata University, Administrative office, Yamagata, Japan

Carbon-ion therapy facility of Yamagata University is currently under construction. Main accelerator is a 430MeV/u synchrotron, similar to i-ROCK, located on the basement floor. The accelerated carbon ion beam is transported upward to a fixed beam treatment room and a 360° rotating gantry treatment room located on the 2nd floor. This vertically stacked configuration, first for the carbon ion therapy center, can limit the construction area to as small as 2000 m 2. Approximately two years long construction work of the facility building will begin in spring 2017, and the treatment is scheduled to begin in 2020. Updated information on the project overview, schedule, facility design and current status will be presented.

PTC17-0108: Survey of Practical Procedure of Respiratory Gating and Metal Artifact Management in Treatment Planning for Carbon-Ion Radiotherapy in Japan

H. Mizuno1, S. Minohara 2, N. Kanematsu3, A. Fukumura4, S. Yonai5, M. Tashiro6, K. Yusa6, T. Yanou7, M. Suga8, M. Mizota9

1National institute of Radiological Sciences- QST, Dept. of Radiation Measurement and Dose Assessment, Chiba, Japan 2Kanagawa Cancer Center, Section of Medical Physics and Engineering, Kanagawa, Japan3National institute of Radiological Sciences- QST, Medical Physics section, Chiba, Japan4National institute of Radiological Sciences- QST, Dept. of Management and Planning, Chiba, Japan5National institute of Radiological Sciences- QST, Dept. of Accelerator and Medical Physics, Chiba, Japan6Gunma University Heavy Ion Medical Center, Physics Division, Maebashi, Japan7Hyogo Ion Beam Medical Center, Dept. of Radiation Technology, Tatsuno, Japan8Hyogo Ion Beam Medical Center, Dept. of Radiation Physics, Tatsuno, Japan9Ion Beam Therapy Center- SAGA HIMAT Foundation, Dept. of Physics, Tosu, Japan

The QA team of Japan Carbon-ion Radiation Oncology Study Group (J-CROS), composed of the medical physicists of carbon-ion radiotherapy centers, has conducted several QA activities such as dose inter-comparison and on-site dose audit (end-to-end test) to assure and improve the quality of multi-center clinical trial of carbon-ion radiotherapy in Japan. Through these interactional activities among medical physicists of each center, it was recognized that there were some variations in practical procedures for treatment planning and beam delivery processes. We decided to make survey for the procedures of respiratory gating and metal artifact management in treatment planning process as those were considered to have big effect on the carbon-ion radiotherapy. For the respiratory gating procedure, 17 items were surveyed such as internal/external gating, type of sensor, delayed time commissioning, methodology of CT acquisition, how to decide the gate level, how to manage with respiratory level drift during treatment, QA and so on. For the metal artifact management for treatment planning process, 12 items were surveyed such as calculation methods for the beam penetrating metal including replacement of stopping power ratio value and evaluation of its volume, how to manage streak artifacts or contrast materials and so on. The results were shared with the QA team of J-CROS to standardize and improve the methodology of each center.

PTC17-0117: Personnel Disposition of Proton Therapy in Taiwan

H.H. Chang1, H.M. Ting1, Y.R. Chen1, C.C. Sung1, C.L. Kang1, Y.J. Huang1, Y.Y. Huang1

1Kaohsiung Chang Gung Memorial Hospital, Radiaiton Oncology, Kaohsiung, Taiwan- Province of China

Proton therapy is an advanced facility. In addition to hardware, the staff configuration is the most important thing. According to the specified order of ministry of health and welfare in Taiwan, a single-room proton is asking to allocate five full-time radiation oncology specialists, of which at least two of the specialists' years of experience should more than five years. Besides, more than three full-time medical physicists and more than three full-time radiologic technicians should also be assigned. For multi-room proton, in addition to the single-room conditions, each additional treatment room is asking to append two radiation oncology specialists, one medical physicist, and two radiologic technicians. Considering the delivering time and switching time of proton beam for multi-room facility, two-room proton may be the most effective for setting up costs and personnel expenses in Taiwan.

PTC17-0121: Monte Carlo Simulation of Proton Wobbling Nozzle at the Chang Gung Memorial Hospital (CGMH)

S.M. Su1, C.J. Wu1, M.J. Lin 2, T.C. Chao1, C.C. Lee1,2

1Chang Gung University, Department of Medical Imaging and Radiological Sciences, Taoyuan City, Taiwan- Province of China 2Chang Gung Memorial Hospital, Department of Radiation Oncology, Taoyuan City, Taiwan- Province of China

Purpose: The aims of this study are commissioning comprehensive Monte Carlo simulation systems for the wobbling nozzle at CGMH and finding optimized simulation parameters at different beam energies using the particle therapy simulation framework (PTSIM), a Geant4-based simulation framework.

Materials and Methods: Range modulation of the CGMH proton wobbling nozzles involved two different techniques: ridge filter and layer-stacking. Beam lateral spreading was commissioned into three wobbling options: small, middle and large. For Monte Carlo simulation, initial physics parameters will impact on depth dose distribution (DD) and lateral profile in water. These includes initial beam energy (E), energy spread (σE), and beam size (σr) at the nozzle entrance. PTSIM simulations were carried out for beam propagation through nozzle components of the three wobbling options at beam energies from 70 to 230 MeV and initial physics parameters were optimized to fit commission beam data. Bragg peak characteristic indices including Rpeak, R90, R80, R50, full width at half maximum (FWMH), and peak-to-entrance ratio were used to validate the simulated DDs. Similarly, simulation parameter optimizations were performed to fit the measured in-air profiles at three positions along the beam axis for both the inplane and the crossplane directions.

Results and Conclusion: Commissioning comprehensive Monte Carlo simulation systems for the wobbling nozzle at CGMH has been performed. Bragg peak characteristic indices comparison showed good agreement with discrepancy less than 1 mm between measurement and simulation. Also, it is possible to reproduce measured penumbra (80-20%) with 2 mm difference for the lateral profiles.

PTC17-0126: Beam-On Imaging of Short-Lived Positron Emitters During Proton Therapy

H.J.T. Buitenhuis1, F. Diblen1, K.W. Brzezinski1, S. Brandenburg1, P. Dendooven1

1University of Groningen, KVI - Center for Advanced Radiation Technology, Groningen, Netherlands

In-vivo dose delivery verification in proton therapy can be performed by positron emission tomography (PET) of the positron-emitting nuclei produced by the proton beam in the patient. A PET scanner installed in the treatment position of a proton therapy facility that takes data with the beam on will see very short-lived nuclides as well as longer-lived nuclides. The most important short-lived nuclide for proton therapy is 12N, which has a half-life of 11 ms. The results of a proof-of-principle experiment of beam-on PET imaging of short-lived 12N nuclei are presented. A 90 MeV proton beam from the cyclotron at KVI-CART was used with the PDPC ModuleTEK PET-system to investigate the energy and time spectra of PET coincidences during beam on. Events coinciding with proton bunches, such as prompt gamma rays, were removed from the data via an anti-coincidence filter with the cyclotron RF. The resulting energy spectrum allowed good identification of the 511 keV PET counts during beam-on. A method was developed to subtract the long-lived background from the 12N image by introducing a beam-off period into the cyclotron beam time structure. A range shift of 5 mm was measured as 6 ± 3 mm using the 12N PET profile. A larger, more efficient, PET system with a higher data throughput capability will allow beam-on 12N PET imaging of single spots in the distal layer of an irradiation with an increased signal-to-background ratio and thus better accuracy. This makes fast and accurate feedback on the dose delivery during treatment possible.

PTC17-0132: Using Monte Carlo Modelling for Evidence Based Tolerances in Quality Assurance

R. Mackay1, S. Pearce 2, A. Aitkenhead3, J. Richardson1

1The Christie, CMPE, Manchester, United Kingdom 2University of Manchester, Physics, Manchester, United Kingdom3The Christie, CMPE, Warrington, United Kingdom

Purpose: Quality Assurance (QA) should be efficient to enable proton centres to maximise the time for treating the patient. Often the quality assurance tolerances are governed by machine performance rather than informed by the clinical effect of deviations from established treatment parameters. In this project we seek to establish a framework to set QA tolerances using Monte Carlo methodology.

Materials and Method: The effect of deviations in key treatment parameters was simulated in a full clinical Monte Carlo model. The effect of changes in the nominal energy of treatment beams, the position and size of spots and the dose delivered per spot were all examined. The effects on clinical treatments were assessed using gamma analyses.

Results: The results of the Monte Carlo simulations show that the gamma analysis pass rate is sensitive to many parameters such as nominal energy and shifts in spot position. Changes in nominal energy of a treatment beam of 2% reduce the 2% 2mm gamma criteria pass rate to less than 95% of pixels. Other simulations show that clinical plans were less sensitive to changes in spot size.

Conclusion: A Monte Carlo system has been developed that can simulate the effect of changes in treatment beam parameters. Initial results highlight the effects of deviations of important parameters such as nominal energy and spot position. The system will also be developed to assess the effects of combinations of uncertainties in treatment positions and to simulate the effects of real variations found in proton QA.

PTC17-0143: Mevion S250i Hyperscan with Adaptive Aperture Treatment Workflow Results Using RayStation and ARIA

A. Langenegger1, B. Rakes 2, J. Conrad3

1Mevion Medical Systems, Product Manager, Littleton, USA 2Mevion Medical Systems, Software Engineering, Nashville, USA3Mevion Medical Systems, Product Management, Littleton, USA

The Mevion S250i with HYPERSCAN is the world's most compact proton pencil beam scanning system and together with the Adaptive Aperture is capable of delivering the smallest penumbra of treatment fields.

Using the IHE-RO Treatment Delivery Workflow profile, end to end testing was performed with a treatment plan generated in RayStation 6 and exported as RT Ion Plan to ARIA v13.7. The plan was scheduled for the Mevion S250i Treatment Delivery System using DICOM Unified Worklist and Procedure Step and subsequently delivered.

Full and partial deliveries were tested. The treatment records were then generated and sent to ARIA, which successfully generated a new Beams Delivery Instruction for the subsequent fraction. This round trip of end to end testing demonstrated the ability to deliver a full course of treatment using the Mevion S250i with adaptive aperture and pencil beam scanning.

PTC17-0176: Predictive Linear Modeling for Robust in Vivo Range Verification from Prompt Gamma Imaging

Y. Xing1, G. Janssens 2, J. Smeets 2, B. Macq1, L. Bondar1

1Université Catholique de Louvain, UCL, Louvain-La-Neuve, Belgium 2Ion Beam Applications S.A., IBA, Louvain-La-Neuve, Belgium

With the emergence of clinical prototypes and first patient acquisitions, the research on prompt gamma (PG) imaging is shifting from hardware development to data treatment, aiming at making most use of the PG data for in vivo estimation of any shift from expected Bragg peak (BP) position. The simple problem of matching the measured PG depth-detection profile of each pencil beam with a reference simulation from the treatment plan is actually made complex in heterogeneous anatomies due to the shape distortions induced by the potential errors. We will illustrate this challenge and demonstrate the robustness of a predictive linear model we proposed for BP shift prediction based on principal component analysis (PCA) method.

4115 error scenarios were simulated in noiseless conditions for the input and output of the learning model, considering the first clinical knife-edge slit camera design in use with anthropomorphic phantom. The linear predictive model was trained by 500 randomly chosen scenarios which were processed by PCA for eliminating data collinearities. This model improved the BP shift estimation by an average of 63±19% in a range between -2.5% and 86%, comparing to state-of-the-art profile shift (PS) matching method. We demonstrated the robustness by a comparative study by applying 1000 times Poisson noise to each profile. 70.26% cases obtained by the learning model had lower prediction errors than those obtained by the PS method. The estimation accuracy ranged between 0.31±0.22mm and 1.84±8.98mm for the learning model, while for the PS method it ranged between 0.33±0.25mm and 20.71±8.38mm.

PTC17-0179: Energy Dependence of Dose Response between Film and Optical Dosimeter for Scanning Particle Beams

Y. Li1, Z. Huang1, W. HSI1, K. Shahnazi1, S. Sun1, Q.A. Group1

1Shanghai Proton and Heavy Ion Center, Medical Physics, Shanghai, China

Purpose: The energy-dependence and dose-response of high spatial resolution EDR2 film and phosphor-based optical dosimeter was investigated for spot-scanning proton and carbon-ion beams.

Materials and Methods: To convert the reading of transmitted light passing through film in an IBA spot imager and an Epson flatbed scanner or light emitted from a phosphor plate viewed by a Canon camera, two-step procedure for calibrating the device was used. A Stouffer light step-wedge with calibrated optical density (OD) was applied to calibrate the film scanner. Fitting of linear relationship between normalized ODs and delivered doses resulted in different slopes for different energy ions.

Results: An exponential trend between the reading and OD for Epson scanner was seen. Due to the internal conversion in IBA spot imager, the reading was linear to OD. The camera suppresses desired light intensity resulting in a power-law trend between the reading and norm-dose. For film and optical dosimeter, the dose-response was linear as a function of dose at each energy. The extracted slope values showed small variation between 80 and 250 MeV for protons, but large variation between 80 and 450 MeV/n for carbon-ions. For optical dosimeter, small variation was found for proton and carbon-ion.

Conclusion: A smaller energy dependence for optical dosimeter compared to film allows smaller corrections to be applied when measuring spot sizes and shapes. Another advantage of optical dosimeter is that lateral profile measurements can be more convenient to analyze in real time during routinely daily and weekly QA than EDR2 film.

PTC17-0185: Feasibility Study of Iso-Centric Rotating Devices to Treat Patients at a Seating Position for Head/Neck Cancer

W. HSI1, Y. Li1, X. Zhang 2, F. Yang 2, Z. Wang 2, R. Zhou 2

1Shanghai Proton and Heavy Ion Center, Medical Physics, Shanghai, China 2College of Physical Science and Technology, Sichunan University, Chengdu, China

Purpose: Under only one fixed either horizontal or 45-degree incline beam in each treatment room, the feasibility of iso-centric (ISO) rotating device to deliver doses at patient's anterior or/and posterior direction for head/neck cancer was investigated.

Materials and Methods: A chair with 3-arm robot faced the interference between the feet-rest and the robot arms for 360 rotation without additional rotational mechanism in chair itself. Therefore, a six-pod 6D movement mechanism were utilized. A remote controlling software can correct both effects of weight-induced sagging and mechanical deformation. To rigorously perform the acceptance test, an optical tracking system using passive infer-red reflection was chosen. Measurement accuracy of optical system was studied on measuring movements of a calibrated 3-arm robotic couch.

Results: Because the translation at horizontal plan is significantly reduced for chair near its maximal vertical position, a coarse translation mechanism was attached at top of six-pod platform to achieve a 30cm x30cm x 30cm treatment volume of 6D movement. The accuracy of optical tracking system was found with 0.5 mm to measure a circular movement of robotic couch. To obtain >20,000 chair positions for various weights for correcting effects of sagging and deformation, a communication interface to interactively execute each movement and to record the position will be built.

Conclusion: An accurate ISO 6D rotating chair is achievable with careful design with measured correction for both effects of sagging and deformation. Additional 3-meter sliding rail allows exchanging the chair with a table with +/- 20-degree rolling to treat thoracic tumor.

PTC17-0209: Range Monitoring with the Inside PET Scanner at CNAO: Testing During a Clinical Treatment Delivery

P. Cerello1,2, M.G. Bisogni 3,4, N. Camarlinghi 3,4, M. Ciocca5, V. Ferrero1,2, E. Fiorina1, G. Giraudo1, M. Morrocchi 3,4, F. Pennazio1, R. Wheadon1

1INFN, Sezione di Torino, Torino, Italy 2University of Torino, Department of Physics, Torino, Italy3University of Pisa, Department of Physics, Pisa, Italy4INFN, Sezione di Pisa, Pisa, Italy5Fondazione CNAO, Unita' di Fisica Medica, Pavia, Italy

The INSIDE collaboration has built and installed at the CNAO synchrotron facility an in-beam PET scanner that features two 10x25 cm 2 planar heads, made by 10 modules each, at a default distance of 60 cm from each other, positioned above and below the patient bed.

After the successful testing of the PET scanner on PMMA phantoms, in terms of spatial resolution, response uniformity and reproducibility, the first data taking during a treatment session to a patient took place on Dec, 1st, 2016.

The treatment consisted in the delivery of 3.7E+10 protons in the 66.3 – 144.4 MeV energy range, in a 4 minutes time interval. After the end of the delivery, the PET acquisition continued for 30 seconds, the time needed to access the room and start removing the scanner. The reconstructed 3D activity map was analysed and compared to an accurate simulation. The difference between the measured and expected range values was evaluated in every 1.6x1.6 mm 2 region of the delivery surface. The agreement is very satisfactory in terms of spatial resolution, while the average value likely suffers from uncertainties in the manual positioning of the prototype.

After upgrading the mechanical system so as to take data in different bed positions and improve the scanner positioning accuracy, testing with patients will continue during 2017, also for carbon ion treatments.

PTC17-0215: Clinical Implementation of 2D Fluence Measurement Device and Software for Commissioning and Quality Assurance of Pencil Beam Scanning Delivery System

V. Letellier1, R. Dreindl1, J. Osorio1, L. Grevillot1, H. Palmans1,2, S. Vatnitsky1, M. Stock1

1MedAustron, Medical Physics, Wiener Neustadt, Austria 2National Physical Laboratory, Dosimetry, Teddington, United Kingdom

The implementation of 2D fluence measurement equipment is a pre-requisite to acceptance, commissioning and periodical QA of pencil beam scanning delivery system (PBS) in light-ion beam therapy center (LIBT). Emphasis is given to IBA Lynx PT, a 2D detector based on a scintillator and CCD camera with a 0.5mm resolution.

To improve the positioning repeatability, a Lynx holder was designed and assembled in-house to position the Lynx with a higher level of accuracy and reproducibly on the treatment couch. With its high resolution CCD and its homogeneity lower than 2% the Lynx, used in integration mode, is an interesting real time device for spots and fields characterization. In image sequence mode (maximum 7.5Hz), the Lynx can even measure intra-spill beam variations for slow-extraction synchrotron-based facility. Unfortunately the acquisition tool does not provide all this essential analysis so software named LynxQA based on a Python core was developed and validated in-house in order to extract easily and automatically the spots and 2D fields properties (size, position, skewness, ellipticity, penumbra, homogeneity, symmetry, intra-spill variation, etc.) from Lynx integrated images and image sequences.

LynxQA was verified and validated as medical device software, it is also able to open data from scanned film and output from PTW Octavius, a 2D detector based on a matrix of ionization chambers. The software allows to distinguish, measure and characterize spots or fields from a 2D fluence measurement device placed into the holder with a high efficiency and reproducibility level.

PTC17-0230: Application of the Discrete Range Modulation (DRM) Proton Range Verification Method to the CIRS 620M Electron Density Phantom

C.C. Lee1, Y.C. Tsai1, T.L. Tsai1, T.C. Chao1

1Chang Gung University, Medical Imaging and Radiological Sciences, Tao Yuan, Taiwan- Province of China

Purpose: A discrete range modulation (DRM) method was developed in our lab for proton range verification using proton radiography images. This report described test result of applying the DRM technique to the CIRS 620M electron density phantom.

Materials and Methods: This study adopted the MCNPX 2.7.0 package to simulate ideal parallel proton beams penetrating through the CIRS 062M phantom to form a radiographic image on a virtual detector. Density and material composition of the inserted rods were obtained from the vendor and ICRU material composition. For any scoring pixel on the detector, a plot of energy deposition verses beam energy (70-230 @ 2.5 MeV step) was obtained to locate a signature energy (E80) at 80% of the maximum energy deposition. The water-equivalent pathlength (WEPL) along the straight beam path to this detector pixel was then determined from a fitting function to the pre-simulated R80 values for beam energies from 70 to 230 MeV in a homogeneous water phantom. Theoretical WEPL values were calculated and compared to those from DRM simulations for rod materials ranging from low density lung and high density bone.

Results and Conclusions: For all the tested materials, range difference between theoretical and simulated values is less than 1 mm with minimum signal fluctuation (<0.5%) for the center 100 pixels. x- and y- profiles passing the detector center showed small dose perturbation near rod edges demonstrating insensitivity of the DRM technique to energy spectra broadening due to multiple Coulomb scattering.

PTC17-0239: Acceptance and Commissioning Tests of a New Treatment Room at SAGA HIMAT Foundation

T. Himukai1, Y. Tsunashima1, M. Mizota1, M. Kanazawa1, M. Endo1, T. Furukawa2, T. Inaniwa3, C. Tsukishima4, I. Takahashi4, Y. Shioyama5

1Ion Beam Therapy Center- SAGA HIMAT Foundation, Physics Section, Saga, Japan 2National Institute of Radiological Sciences- National Institutes for Quantum and Radiological Science and Technology, Department of Accelerator and Medical physics- Advanced Particle Therapy System Research Team, Chiba, Japan3National Institute of Radiological Sciences- National Institutes for Quantum and Radiological Science and Technology, Department of Accelerator and Medical physics- Treatment Beam Research Team, Chiba, Japan4Mitsubishi Electric Corporation, Advanced Electro-Magnet Applications and Medical Systems Department, Hyogo, Japan5Ion Beam Therapy Center- SAGA HIMAT Foundation, Department of Radiology, Saga, Japan

Ion Beam Therapy Center, SAGA HIMAT Foundation (SAGA HIMAT) has 3 treatment rooms (Room A, B and C). Rooms A and B started treatment of cancer patient with carbon-ion radiotherapy in August 2013. Room C equipped horizontal and vertical beamline with scanning irradiation method will start treatment in 2017. For starting in the new treatment room, we started acceptance tests in September 2016. After completed acceptance test, we will start clinical commissioning tests in 2017.

PTC17-0248: Reliability Monitoring of Medical Accelerator

S. Fukuda1, H. Inokuchi 2, K. Okumura 2

1QST/NIRS, Radiation Quality Control Section, Chiba-shi, Japan 2QST/NIRS, Accelerator and Medical Physics Devision, Chiba-shi, Japan

The reliability of the medical accelerators is required to be as high as possible. Because particle therapy accelerators (PTA) are more complex than Linac accelerators for conventional X-ray therapy, it is not easy to keep and maintain the stability. The standards of irradiation system have make a transition from the passive irradiation method to the scanning one. This means the connection between the accelerator system and the scanning irradiation system becomes stronger than that between the accelerator one and passive irradiation one. The displacement of the beam from the accelerator affects directly the position of the irradiation beam at the isocenter and may make the unexpected integrated dose distribution. So, the way to monitor the reliability of the PTA is required.

We proposed the way to monitor the reliability of PTA with parameters of Availability, Failure Rate and Repair time as a index for reliability monitoring. We introduced these monitoring parameter to the HIMAC (Heavy Ion Medical Accelerator in Chiba) including ion sources, linear accelerators, synchrotrons and irradiation system and has started to monitor its reliability since 2014.

All failures that occurred in a month have been listed with cause of failure, defective parts and downtime to calculate above parameters. If the trend of parameters shows low quality, the cause would be investigated and corrective action would be examined. This presentation will explain how we decided the parameters as reliability indexes and how we have used these, with the cycle of Monitor-Analysis-Correction- Evaluation that maintains and/or improves the accelerator reliability.

PTC17-0252: Commissioning of New Prototype Supraconducting Synchro-Cyclotron PBS Proton Therapy System

M. Vidal1, A. Gerard1, V. Floquet1, J. Herault1

1Centre Antoine Lacassagne, Institut Méditerranéen de Protonthérapie, NICE, France

Purpose: We report the commissioning process of new prototype Proteus® One proton therapy system with pulsed PBS high dose rate proton beam.

Materials and Methods: The supraconducting synchro-cyclotron has a frequency of 1kHz and delivers a pulsed proton beam of dose rate between 2.65 μGyE/pulse and 230 μGyE/pulse for the lowest (96 MeV) and highest (226 MeV) energies respectively. The facility is equipped with a compact gantry which rotates from -35° to 188° and delivers PBS proton beams of field sizes up to 20x24 cm 2. The nozzle monitor ionization chambers (IC) were designed to be able to measure high dose rate per pulse delivered charges. A prototype 14 cm plane parallel IC and a multi-layer IC were used to measure integrated depth dose curves. Spot profiles, field size and penumbra were acquired with a 2D-scintillator detector while 2D-fields uniformity and symmetry were measured with a prototype 2D-array IC.

Results: Proteus®One high dose rate per pulse beam causes mainly recombination effects in IC based detectors. Prototype detectors and nozzle IC were developed in order to take into account these beam characteristics. Measurements performed with these detectors show lateral and longitudinal dose uniformity better than 2.5 % for clinically significant field sizes, ranges, modulations and air-gaps. Absolute dose measurements were achieved in single-layer fields with plane parallel IC and with water calorimeter. Measured dosimetry data agreed within 2% for the lowest and the highest energies.

Conclusion: The new prototype Proteus® One PBS proton therapy system was successfully commissioned and released for clinical use.

PTC17-0261: Application of Building Information Modeling for Designing Proton Therapy Center at National Taiwan University Cancer Center

S.H. Kuo1,2, Y.C. Liu3, C.N. Chong3, C.W. Wang1, K.H. Yeh1, J.C.H. Yang1, A.L. Cheng1

1National Taiwan University Hospital and National Taiwan University Cancer Center- National Taiwan University College of Medicine, Department of Oncology, Taipei, Taiwan- Province of China 2National Taiwan University College of Medicine, Cancer Research Center- and Graduate Institute of Oncology, Taipei, Taiwan- Province of China3YongLin Healthcare Foundation, YongLin Healthcare Foundation, Taipei, Taiwan- Province of China

Purpose: To accelerate constructing the new proton therapy center at National Taiwan University Cancer Center, we applied for the Building Information Modeling (BIM) design projects. We further assessed whether the use of BIM provides faster construction program, fewer conflicts and the reduced rates of order changes when compared to “traditionally” designed projects.

Materials and Methods: Under the BIM-designed project, we utilized Revit MEP 2013 for drawing the BIM model of the new proton therapy center with 20,000 square meters. We exploited the Autodesk Navisworks Manage 2013 for further collision detection and 3D Design Review of the new proton therapy center.

Results: Till now December 2016, we have completed the structural, mechanical, electrical, plumbing, fire protection and telecom systems of the new proton therapy center through the BIM platform. Overall, more than 310 wall-penetrated conduits with a total length of approximate 5,000 meters were designed for considering radiation shielding capability for proton-treating and photon-treating rooms. A total length of 400 meters cable trays was implemented using the BIM platform. We found that BIM resolved 407 clashes from the proton bar, reduced the order changes, and avoided rework during design, pre-construction and construction stages for proton therapy center when compared with traditional design.

Conclusion: Through digitally building up an accurate virtual model for new proton facility, BIM helps architects, engineers, and constructors identify any potential issues in a simulated environment. Furthermore, BIM allows future users, including physicians and physicists, for visualizing what is to be built and for validation of their desired requirements.

PTC17-0282: The Manchester Proton Therapy Research Facility

M.J. Taylor1,2, M.J. Merchant1, H. Owen3, N.F. Kirkby1, A. Chadwick1, T. Mee1, A.H. Aitkenhead4, R.I. Mackay4, K.J. Kirkby1

1The University of Manchester, Division of Molecular and Clinical Cancer Sciences, Manchester, United Kingdom 2Christie NHS Foundation Trust, Clinical Oncology, Manchester, United Kingdom3The University of Manchester, School of Physics and Astronomy, Manchester, United Kingdom4Christie NHS Foundation Trust, Medical Physics and Engineering, Manchester, United Kingdom

The Department of Health has committed £250 million to developing high-energy proton beam therapy services in the UK. Two facilities are currently under construction located at the Christie NHS Foundation Trust, Manchester, and University College London Hospitals NHS Foundation Trust.

The Christie proton beam therapy centre will host three clinical treatment rooms featuring ProBeam gantries supplied by Varian Medical Systems. A joint investment by the Department of Health and the Christie Charity has allowed the development of a dedicated proton therapy research space located in a fourth room at the Christie proton therapy centre.

This research room will house two static horizontal beam lines, and have the capability for spot scanning beam delivery. The design philosophy for these two beamlines is to provide modular, adaptable end-stations for each beamline that will give the flexibility to allow an extensive range of experiments to be performed, encompassing radiobiology, physical, and technical experiments. The research room will be a UK national facility for proton therapy research.

In this talk, I will present the design and planned technical capabilities of the Christie Proton Therapy Research Facility and highlight how this facility will be used to address some of the key challenges in proton beam therapy.

PTC17-0287: Experience, Challenges and Results from Commissioning of the New Gantry 3 at the PSI Proscan Facility

A. Koschik1, J. Duppich1, A. Gerbershagen1, M. Grossmann 2, A. Lomax 2, D. Meer 2, S. Safai 2, J.M. Schippers1, D.C. Weber 2

1PSI Paul Scherrer Institut, Department Large Research Facilities GFA, Villigen, Switzerland 2PSI Paul Scherrer Institut, Center for Proton Therapy CPT, Villigen, Switzerland

With the new Gantry 3, PSI is extending its research and clinical capabilities in the field of proton therapy and pencil beam scanning technology at the Center of Proton Therapy (CPT). The additional treatment room at the PSI PROSCAN facility features a 360-degree downstream scanning gantry, which is built in research collaboration with Varian Medical Systems (VMS).

The main research goals at the PSI PROSCAN facility include further development of precise spot scanning and optimized beam delivery with low dead-time for treatment of moving targets. Consequently Gantry 3 is designed to feature advanced pencil beam scanning technology for the energy range 70 – 230 MeV with a large scan field size of 30x40cm, integrated cone beam CT functionality and will allow in the future fast energy layer switching times < 200 ms with beam accuracies at iso-center < ± 1 mm. The main challenge in realizing Gantry 3 is the integration of the VMS gantry and its subsystems into the existing PROSCAN beamline and control system environment, allowing seamless beam operation.

Gantry 3 has been installed and technically commissioned without interruption of patient treatments in the existing facility. We report about our experiences and successfully mastered challenges during this demanding technical commissioning phase and present results of the system performance that was achieved. The project status is presented and an outlook is provided.

PTC17-0291: A Direct Comparison of Helium and Proton Computed Tomography Using TOPAS Simulations and Experimental Data

P. Piersimoni1, C.A. Collins Fekete1, V.A. Bashkirov 2, B.A. Faddegon3, R.P. Johnson4, C.E. Ordoñez5, J. Ramos Méndez3, R. Schulte 2, L. Volz1, J. Seco1

1German Cancer Research Center - DKFZ, Medical Physics in Radiooncology, Heidelberg, Germany 2Loma Linda University, Basic Science, Loma Linda CA, USA3University of California San Francisco, Radiation Oncology, San Francisco CA, USA4University of California Santa Cruz, Institute for Particle Physics, Santa Cruz CA, USA5Northern Illinois University, Center for Research Computing and Data, DeKalb IL, USA

Purpose: To compare helium and proton CT using a prototype particle computed tomography (pCT) scanner.

Materials and Methods: A prototype pCT scanner that tracks individual particles and measures their water equivalent path length (WEPL) was installed on the beam line at the Heidelberg Ion-Beam Therapy (HIT) facility, Germany. A Monte Carlo simulation study was performed using the TOPAS tool to compare the accuracy of the pCT reconstructed relative stopping power (RSP) values obtained with a low fluence of protons and helium ions (200 MeV/u). For the simulation, in addition to the prototype scanner, an ideal totally absorbing energy-range detector was simulated in order to estimate the best theoretically achievable RSP accuracy. For both the simulation and the experiment, phantoms with inserts of different materials, sizes and spatial distribution, as well as an anthropomorphic phantom were reconstructed using an iterative reconstruction technique.

Results: In the ideal configuration, an average error of 0.2% was obtained for all materials with both protons and helium ions. For the real configuration, the average error was about 1%. Work is underway to improve these results by optimizing the energy to WEPL calibration procedure for helium. A comparison of the preliminary simulated and experimental results with the latest energy to WEPL calibration procedure will be presented.

Conclusion: The installation of the prototype pCT scanner at the HIT center will allow a direct comparison of proton and helium CT.

PTC17-0320: Research and Discovery at the Cincinnati Children's Hospital / UC Health Proton Therapy Center: Facility and Initial Experience

A. Mascia1, M. Sertorio 2, R. Vatner3, M. Lamba3, J. Breneman3, S.I. Wells 2, Y. Zheng 2, J. Perentesis 2

1Cincinnati Children's / UC Health Proton Therapy Center, Radiation Oncology, Cincinnati, USA 2Cincinnati Children's Hospital, Cancer and Blood Disease Institute, Cincinnati, USA3University of Cincinnati, Department of Radiation Oncology, Cincinnati, USA

The availability of advanced proton beam delivery systems, such as pencil beam scanning, is driving both clinical availability and expanded indications. In order to complement this surge, research and development in biology, physics, and engineering is necessary in order to optimize proton and particle therapy usage. In this context, the recently opened Cincinnati Proton Therapy Research Facility was designed specifically for authentic and rapid translation of research discovery to the clinic.

The proton therapy center is a 3-room facility, with a 250 MeV cyclotron, with two clinical gantries and one gantry dedicated exclusively to research. All have full 360 degree rotation, are dosimetrically equivalent, and equipped with a dedicated pencil beam scanning nozzle with a 4-6 mm in-air spot profile, an imaging system (dual kV planar or volumetric CBCT), and robotic treatment table positioner. The research facility includes a laboratory for molecular and cellular studies, an animal room, and expansion capabilities. Taking advantage of the treatment planning system and its connectivity to the research gantry console, a library of irradiations with varying energy and dose as well as a prototype animal positioning system greatly accelerate cell and mouse irradiation workflow. A unique immobilization and collimation system for small field, focal irradiations is under design.

A fully functional proton pencil beam-scanning gantry dedicated exclusively to research and development is, to our knowledge, unique. The goal is to facilitate rapid advances in physics, engineering and biology in proton therapy towards improved clinical treatment delivery and outcome.

PTC17-0327: Comparison of Accuracy of Parameterized Proton Range Models

H.E.S. Pettersen1, I. Meric 2, O. Odd Harald1, S. Jarle Rambo 2, R. Dieter3

1Haukeland University Hospital, Department of oncology and medical physics, Bergen, Norway 2Bergen University College, Faculty of Engineering, Bergen, Norway3University of Bergen, Institute of Physics and Technology, Bergen, Norway

Purpose: A correct and accurate calculation of proton ranges in homogenuous materials and phantoms is crucial for correct decision making in proton therapy and related activites such as proton imaging. The measurement of ranges in phantoms performed during commissioning and Quality Assessment, and serves as a ground truth for the calculation between range and energy in water in daily usage. Several parameterizations of the range-energy relationship exist, which can be of great value for interpolated calculations between arbitrary energies and ranges. In this study we compare the accuracy of some of the different parameterizations of the range-energy relationship.

Materials and Methods: In this study, different models for the relationship between range and energy are evaluated based on their ability to correctly reproduce data from the PSTAR database. 175 CSDA range values for protons in water, up to therapeutic energies, are split into two groups. One group (N=25) is used for finding the model parameters, the remaining control group (N=125) is used to evaluate the model calculations at small range intervals. Four models are evaluated: Two interpolation-based models using linear and spline-based interpolation, as well as two semi-empirical models using the Bragg-Kleeman equation and an exponential sum approximation of the Bethe equation.

Results: The deviations between the range values in the control group and the model calculated values are shown.

Conclusion: Overall, the spline interpolation model has the highest accuracy. A sub-percent range calculation accuracy is shown for all models above 100 MeV, and for the spline model above 10 MeV.

PTC17-0342: Characterization and Calibration of a Proton PBS Dose Delivery System Using Innovative Acceptance and Commissioning Procedures

R. Dreindl1, J. Osorio1, V. Letellier1, A. Carlino1,2, A. Elia1,3, L. Grevillot1, H. Palmans1,4, S. Vatnitsky1, M. Stock1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2University of Palermo, Department of Physics and Chemistry, Palermo, Italy3Université de Lyon, CREATIS, Lyon, France4National Physical Laboratory, Dosimetry, Teddington, United Kingdom

Purpose: Acceptance and commissioning of the proton PBS dose delivery system (DDS) for the first MedAustron beamline was finished in 2016. This work reviews the applied procedures and results, with particular focus on the commissioning and calibration of the beam intensity monitors.

Materials and Methods: Measurements were performed using scintillating screen detectors, linear PinPoint- and 2D-ionization chamber arrays, Roos- and Bragg-peak ionization chambers in water. An innovative method to separate DDS intensity monitor homogeneity from fluctuations of the overlaying scanned delivery contributions during 2D field-irradiation was developed. Two methods for beam intensity monitor calibration were applied (Dw of single-layer scanned fields as reference method, DAPw of static single-energy beamlets as redundant method). Comprehensive uncertainty analysis of the full beam intensity monitor calibration was done.

Results: All technical aspects (charge collection, background current, dose linearity, position accuracy, 2D homogeneity) of the DDS intensity monitors were found to meet acceptance criteria. The 2D homogeneity of the beam intensity monitors was within 1.3%. Scanned delivery was found to decrease the 2D homogeneity as the homogeneity index increases to 1.9%. Knowledge gathered during the acceptance phase allowed accurate calibration of the beam intensity monitors with a relative standard uncertainty of 2.5% and 2.6% for the reference and redundant methods, respectively.

Conclusion: A comprehensive investigation of the beam monitor homogeneity and two redundant reference dosimetry methods established a robust calibration of a proton PBS dose delivery system at MedAustron.

PTC17-0361: Commissioning Report of the Beam Delivery System in I-ROCK at Kanagawa Cancer Center

E. Takeshita1, T. Furukawa 2, K. Mizushima 2, N. Saotome 2, Y. Saraya 2, Y. Hara 2, R. Tansho 2, K. Noda 2, S. Minohara1, Y. Nakayama3

1Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan 2National Institute of Radiological Sciences, Dept. of Accelerator and Medical Physics, Chiba, Japan3Kanagawa Cancer Center, Radiation Oncology, Yokohama, Japan

Purpose: Our project of new facility, named as ion-beam Radiation Oncology Center in Kanagawa (i-ROCK), at Kanagawa Cancer Center has been constructed since 2010. From December 2015, the treatments were started as a clinical trial. The real treatments have been performed from February 2016 as advanced medical care.

Materials and Methods: A 3D pencil beam scanning system was installed for the beam delivery system in KCC. During the scanning irradiation, the stabilities of the position, size, and the intensity time structure (spill) for the pencil beam should be evaluated at the beginning because the performance of the pencil beam directly affects the dose distribution. To ensure the beam quality, we checked the performance of the pencil beam firstly. The stability and flatness of the irradiation field with scanning were also evaluated in any case.

Result: From the measurement results, it was checked that the position and the size of the beam were be able to supply within +-0.5mm independently of the extraction time. The beam size for every energy cloud be adjusted to a circular shape. The distributions of the uniform irradiation filed were measured by a screen monitor to check the position dependence of the dose monitor. A flatness of the field within 3% was deduced from the measurement data.

Conclusion: We success to perform the commissioning of the beam delivery system so that the quality both of the pencil beam and the irradiation field is suitable for the treatment.

PTC17-0362: Report of QA and Quickly Maintenance for the Beam Delivery System at Kanagawa Cancer Center

E. Takeshita1, Y. Kusano1, Y. Matsuzaki1, T. Shimoju 2, Y. Takahashi 2, N. Sasaki 2, T. Katori 2, H. Fujii 2, Y. Taki 2, Y. Mukaiyama 2

1Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan 2Accelerator Engineering Corporation, Kanagawa Office, Chiba, Japan

Purpose: From December 2015, the heavy-ion therapy was started as a clinical trial at Kanagawa Cancer Center. The real treatments have been performed from February 2016 as advanced medical care. To ensure the quality in the treatment, the quality assurance procedure must be performed in daily, weekly, monthly and annually.

Materials and Methods: The beam intensity and the transmission at the accelerator are checked at in every day. The beam positions of 11 energies are also verified within the tolerance. If the disagreement will be more than 2 mm, the position should be corrected using the steering magnets. The beam range of typical energy is measured to see the tolerable value.The absolute dose is calibrated using the daily-factor measured by a Daily QA phantom to reduce the daily difference in a little. Finally, we check any performance of the beam delivery system using a 3D irradiation field.

Results: Up to the present time, the beam delivery system has no problem for the treatment and the medical physicists checked and accepted them in every day. In our facility, the electrical data base for the QA and the maintenance were produced and they have been worked very well.

Conclusion: The QA and the maintenance of the beam delivery system have been succeeded to work safety and quickly since starting the treatment.

PTC17-0369: Optimization of a Dual Particle Facility for Protons: Acceptance and Commissioning Results of the Whole Treatment Workflow at MedAustron

M. Stock1, G. Kragl1, L. Grevillot1, A. Ableitinger1, H. Palmans1, M. Mumot1, J. Hopfgartner1, P. Steininger 2, S. Vatnitsky1, E. Hug3, V. Letellier1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2Paracelsus Medical University PMU, Institute for Research and Development on Advanced RadiationTechnologies radART, Salzburg, Austria3EBG MedAustron GmbH, Medical Department, Wiener Neustadt, Austria

With start of patient treatment with protons at MedAustron in 2016, the 6th dual-particle facility worldwide has become operational. From the design our synchrotron-based facility was specified to provide not only protons and carbon ions in general, but proton treatment quality competitive with dedicated proton facilities. Pencil beam scanning for carbon ions usually requires fixed beam nozzles and hence the same setup is used for protons. As a result proton beam quality is usually compromised due to a large air gap between nozzle and iso-centre. In order to preserve ballistic capabilities of protons in a dual-particle facility, the patient needs to be positioned as close as possible to the nozzle and we chose as design a non-isocentric treatment planning and patient positioning system to reduce the air gap. This workflow is supported by a collision avoidance system at different stages of the treatment. It starts with an automated collision detection integrated into the treatment planning phase. In the treatment room robotic table movements, imaging and beam delivery are supported by the record and verify system and checked upfront for collision. Finally, during robot movements the same checks are performed in real time for avoiding potential collision. The beam delivery and the patient alignment systems were optimized for non-isocentric treatments. All systems were successfully commissioned and are already in clinical routine operation beginning with the first patient.

The presentation highlights acceptance and commissioning aspects for the TPS, Record and Verify System, patient alignment system and beam delivery system operational at MedAustron.

PTC17-0374: Construction of the Full Paperless Information System for Carbon-Ion Radiotherapy

W. Maehana1, S. Minohara 2, Y. Kusano 2, E. Takeshita 2, K. Shioiri1, S. Yoshino1, Y. Tokiya1, S. Hirai1, S. Ide1, Y. Nakayama3

1Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan 2Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan3Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan

Purpose: In order to manage a safe and efficient treatment workflow of several steps at carbon-ion radiotherapy (CIRT) and to share the progress at each patient with medical staffs, we constructed the full paperless information system (FPIS) based on MOSAIQ-OIS (MQO; Ver.2.6.3, Elekta, Inc.).

Materials and Methods: Along with the treatment workflow from the forming of the immobilization device, the planning CT acquisition, the treatment planning, the patient beam QA and to the fractionated irradiation, the order and its record are transmitted while confirming at each step. The FPIS manage these workflows without printed materials. The template of the document was developed by combining Visual Basic for Application (VBA) with the eSCRIBE function in MQO. At first, the radiation oncologist input the outline of treatment plan using the graphical user interface designed originally. Then the VBA program automatically extracted the order and record sheet, and the radiation technologist add the work results at each step of workflow. This progress was checked and shared on MQO terminals. And these results are summarized to one-page report.

Results: The workflow of CIRT became efficient by the automatic content extraction program. The failure to confirmation of the patient information was prevented using the check sheet. We incorporated the FPIS in the clinical implementation. The FPIS enabled us to perform a safe and efficient workflow of CIRT.

PTC17-0377: Effect of CTSP Variation on Range Calculation in Particle Radiotherapy

Y. Kusano1, Y. Tokiya 2, S. Minohara1, K. Shioiri 2, W. Maehana 2, E. Takeshita1, Y. Matsuzaki1, S. Yoshino 2, S. Hirai 2, Y. Nakayama3

1Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan 2Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan3Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan

Purpose: The dose calculation algorithm for particle radiotherapy is based on the electron density along the beam pass in patient body. This relation is calibrated by the conversion factor from CT-number to relative stopping power ratio (CTSP). As noted by AAPM reports, CTSP depends on the scan conditions. As a part of commissioning in TPS, we check the CTSP at various conditions.

Materials and Method: According to the poly-binary CTSP calibration method, we prepared the tissue substitute samples and five ED-phantoms with different size. These CT images are collected under the various scan conditions such as slice thickness, field of view, X-ray tube current and voltage.

Results: The voltage dependence and object size dependence of CTSP was observed. The condition of 120 kV was about 50 HU smaller than that of 135 kV at the region of high CT-number. The distinct differences appeared around CT-number of bone region. In our case of prostate cancer, CTSP calibrated by the ED-phantom of 300 mm in diameter is usually used. However, the size of patient's body is different individually and is not true circle. So we estimated the impact the patient's body size on the range calculation in TPS. The variations of the beam range calculated by CTSPs of different ED-phantom size was approximately ±1.5 mm in worst case. Based on these commissioning results, we would like to evaluate the accuracy of calculated dose distribution in clinical case.

PTC17-0384: Using Radial Projection to Reduce the Statistical Uncertainty of Spot Lateral Profile Generated by Monte Carlo Simulation

X. Ding1, W. Liu1, J. Shen1, A. Anand1, J. Stoker1, Y. Hu1, M. Bues1

1Mayo Clinic Arizona, Radiation Oncology, Phoenix, USA

Monte Carlo (MC) simulation can been used to generate commissioning data for the beam modeling of treatment planning system. The purpose of this work is to present a method, called radial projection (RP), for post-processing MC-simulation-generated data. We used the RP method to reduce the statistical uncertainty of the lateral profile of proton pencil beams with axial symmetry. The RP method takes advantage of the axial symmetry of dose distribution, and uses the mean value of multiple independent scores as the representing score. Using the mean as the representing value instead of any individual score results in significant reduction of the statistical uncertainty as shown.

In this abstract, the concept of the RP method was presented, followed by step-by-step implementation, as well as comparison to demonstrate the advantage of the RP method. The statistical noise of MC data is reduced significantly by the RP method. Direct lateral profile comparisons were performed to demonstrate the uncertainty reduction qualitatively, while standard error comparisons were performed to demonstrate the reduction quantitatively. By using the RP method to post-process MC data, the corresponding MC simulation time was reduced by a factor of ten without quality reduction in the generated result from the MC data. We concluded that the RP method is an effective technique to increase the MC simulation efficiency for generating lateral profiles for axially symmetric pencil beams.

PTC17-0385: Patient Positioning and Operator's Workstation Solutions for Proton Therapy

I. Erokhin1, V. Konchikov1, I. Kancheli 2

1FSBI “SSC RF Institute for Theoretical and Exprerimental Physics” SRC “Kurchatov, Medical physics, Moscow, Russian Federation 2SSC RF “Institute for Theoretical and Experimental Physics” SRC “Kurchatov Institute”, Medical physics, Moscow, Russian Federation

Purpose: Prototyping a hardware-software solutions for patient positioning and operator's workstation in proton therapy (PT).

Materials and Methods: For the improvement of patient positioning device it was decided to develop dedicated controllers and a software for interaction with it. Controller are based on FPGA logic. Controller manipulates servomotors of the positioning device and is connected to main PT operator's workstation via RS-485 network. Software for operator's PC is developed with Microsoft.Net platform and WPF (MVVM template). It does not directly affect the position device operation for obtain the overall stability. Also, software has a structure, what allows scalability (such as adding an extra control terminals via Ethernet, adding an extra PT workflow hardware and so on).

Results: A working controller for positioner and a piece of PT workflow control software what included a visualization oriented for touch displays and extra terminals support was developed. Additionally, an emulator of patient positioning device was developed. It needs for debugging the software functionality.

Conclusion: This prototyping process gives us the future opportunity of creating more cost effective and more compact solutions for PT automation and operator's workstation including the main procedure devices (positioning, dose delivery, target centering and so on) and operator's data input devices (such as other control terminals and technological process data visualization systems).

PTC17-0387: Dosimetric Tests of a Multilayer Ionization Chamber (MLIC) in Proton Scanning Beam

N. Mojżeszek1, L. Stolarczyk1, W. Komenda1, R. Kopec1, P. Olko 2

1Institute of Nuclear Physics, Cyclotron Centre Bronowice, Kraków, Poland 2Institute of Nuclear Physics, Division of Applications of Physics, Kraków, Poland

Purpose: The MLIC (Giraffe, IBA Dosimetry) is widely used as a tool for a quick range determination of proton scanning beams. The purpose of this study was to verify how the range measurements depend on the chamber uniformity, linearity of dose response and small variation of beam energy.

Materials and Methods: Measurements were performed using pristine proton scanning beams aligned to the center of MLIC chambers. In case of uniformity tests the beam was also delivered on the cross and diagonal up to 50 mm from the chamber center (in total 69 beams). The highest range difference comparing the measurement in the middle and at distance of 40 mm from chambers center was 0.3 mm. The MLIC sensitivity to range changes was tested by changing energy values in the delivery system to get theoretical range differences from 0.1 mm to 0.5 mm and using a plastic plates of known WET. These were measured using 80 MeV, 150 MeV and 200 MeV proton beams. The difference of a theoretical range and those obtained experimentally was not exceeding 0.2 mm. The MLIC linearity was examined by changing monitor units MU in the range between 0.02 MU and 15 MU.

Results: The signal of MLIC chambers shows linear dose response with maximum range changes at the level of 0.1 mm.

Conclusion: The MLIC chamber is sensitive to range changes of 0.2 mm which indicates it is suitable to detect difference in range during quality assurance tests and is a useful tool in WET measurements.

PTC17-0393: A Geant4 Monte Carlo Model of Proton Pencil Beam Scanning Nozzle

X. Ding1, W. Liu1, J. Shen1, J. Stoker1, A. Anand1, Y. Hu1, M. Bues1

1Mayo Clinic Arizona, Radiation Oncology, Phoenix, USA

A Monte-Carlo (MC) code was developed to generate commissioning data for the spot-scanning proton therapy facility in Mayo Clinic Arizona. The MC code was built on the Geant4 platform, and ported to a computation cluster for fast calculation. It took three months to generate all commissioning data.

The MC generated commissioning data included in-air and in-water lateral profiles, as well as chamber size correction factors for 107 proton energies. The code was validated by measurements, including integrated depth dose (IDD) and measured in-air lateral profiles. The MC data served as the input data for the treatment planning system (TPS) Eclipse version 13.6. Since the facility started treating proton patient in March 2016, the TPS has been used to design treatment plans for 220 patients. All dose calculations by the TPS agree well with patient-specific-QA measurements.

PTC17-0397: Shielding Design at Ion-Beam Radiation Oncology Center in Kanagawa (i-ROCK)

Y. Kusano1, S. Minohara1, H. Tanaka 2, T. Katori3, T. Miyoshi3, E. Takeshita1, K. Shioiri4, S. Shinohara5, H. Nakayama5, Y. Nakayama6

1Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan 2K.ITO Architects & Engineers Inc., Section of Design, Tokyo, Japan3Accelerator Engineering Corporation, i-ROCK, Yokohama, Japan4Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan5Kanagawa Cancer Center, Office of Carbon-ion facility management, Yokohama, Japan6Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan

Purpose: The building design of i-ROCK started at 2010, and the construction of the building completed in October 2014. In this process, the simulation of radiation shielding was repeated, and that result have been reflected to the building design. Finally, i-ROCK facility has passed the examination of Nuclear Regulatory Agency (NRA) in Japan. We report a way of the facility design of shielding.

Materials and Methods: (1) We discussed the basic specifications such as beam energy, beam intensity and number of treatment rooms, etc. in considering with clinical requirements. (2) We have adjusted the arrangement of treatment rooms, the synchrotron-ring and beam transport-line under the limited space and the clinical usability. (3) The shielding conditions were calculated with the Monte Carlo Simulation (PHITS code). Referring the simulation and the Japanese regulations of radiation protection, we have repeated the correction of building design such as the concrete thickness, the iron-plate installation and the piping route. (4) After the document review by NRA, the shielding performance was measured under the condition of maximum beam energy.

Results: The boundary of controlled area was checked in detail. The shielding performance was measured at several points by the survey meter and neutron detector, and was confirmed within regulation.

Conclusion: We achieved the adequate design of radiation shielding that was not excessive. These processes would be the standard of the particle facility design in Japan.

PTC17-0401: Quality Control of the Proton Built Project

K.J. Juan1, E. Huang1

1Kaousiung Chang Gung Memorial Hospital, Radiation oncology, Kaousiung, Taiwan- Province of China

Purpose: The world medical trends, proton and heavy particle equipment is also vigorously developed around the world, Taiwan is the same, Linkou Chang Gung hospital built the first proton center and treatment of patients in 2015, according to plan, Kaohsiung protron center will be completed in 2018.

Materials and Methods: Step 1, implementation review and evaluation: market assessment, investment amount, benefit analysis and so on. Step 2, the establishment of preparatory organizations and the window of responsibility. Step 3, application and contrution set theaim of completion date. Step 4, the division of works and responsibilities are reviewed. Step 5, project quality checks regularly. Step 6, acceptance and payment.

Result: Under Integrity control of the contruction and quality testing, the progress of the project will completein line with the qualityfrom 2018/3/16 to 2017/12/15 in advance and start the installation.

Conclusion: The construction of the proton center is very complex and rigorous investment, so the complete organization and control mechanism is very important, this article provides this experiencefor your reference.

PTC17-0403: Verification of The “XiO” TPS Using Radiochromic Films

K. Shipulin1, C. Oancea1,2,3, D.M. Borowicz1,4, G. Mytsin1, J. Vilimovsky5, V. Vondracek5

1Joint Institute for Nuclear Research, Medico-Technical Complex, Dubna, Russian Federation 2Horia Hulubei National Institute for Nuclear Physics and Engineering, DRMR, Magurele, Romania3University of Bucharest, Faculty of Physics, Bucharest, Romania4Greater Poland Cancer Centre, Department of Medical Physics, Poznan, Poland5PTC Czech, Department of Medical Physics, Prague, Czech Republic

The objective of the present study was to verify all the stages of proton pencil beam scanning technique applied at PTC, Prague, Czech Republic. A proton beam irradiation simulation was carried out using “XiO” treatment planning system (TPS) developed by ELEKTA, Sweden, which is currently used on daily bases at PTC.

To evaluate the 2D dose distributions of different clinical treatment plans measured in an anthropomorphic phantom, from Alderson, and compare them to computer simulation predictions from a TPS we used Gafchromic EBT2 films, Ashland, USA. A simulated target has been drawn on CT images of the phantom in a high heterogeneity region. The target irradiation was first simulated from 3 irradiation angles (240°, 305°, and 110°) and in the second case only from one direction, angle 240°, dose 2.5 Gy.

Radiochromic films cut in the shape of phantom's slice were placed between the slices containing the target. The films were calibrated for a dose range from 0 to 3 Gy. The irradiated films have been scanned (Epson 11000XL Pro scanner) and transformed into a matrix of doses using measured calibration curve.

Quantitative comparisons between the TPS and the EBT2 film data were carried out using gamma index procedures. Our tolerance criterion for geometric accuracy was 3 mm and 3% for the values of dose. The first and second target coincides with the calculated one with at least 95.8% and 85.7% accuracy, respectively.

PTC17-0414: Monte Carlo Simulations for Second-Check Dose Validation

J. Hartman1, X. Zhang 2, X.R. Zhu 2, S.J. Frank3, J.J.W. Lagendijk1, B.W. Raaymakers1

1University Medical Center Utrecht, Department of Radiotherapy, Utrecht, Netherlands 2The University of Texas MD Anderson Cancer Center, Department of Radiation Physics, Houston, USA3The University of Texas MD Anderson Cancer Center, Department of Radiation Oncology, Houston, USA

Purpose: The purpose of this study is to use a Monte Carlo (MC) beam model of the MD Anderson Cancer Center clinical scanning proton beam, which was developed in TOPAS, as a second-check dose validation method in patient specific quality assurance (PSQA) in intensity modulated proton therapy.

Materials and Methods: In regular PSQA the dose distribution of the treatment planning system (TPS) was validated against planar dose measurements (PDMs) at several depths in a water phantom. Using the MC beam model, an MC PSQA method is developed. For seven pelvic and lung plans, a total of 52 PDMs were validated with MC PSQA. The outcome was compared to the results of regular PSQA using a 3%-3mm gamma analysis of the relative dose. This was in fact a comparison between the TPS calculated and MC simulated dose.

Results: For 57 of 58 planes (98%), the MC PSQA produces clinically acceptable pass rates (>95%), of which 47 (82%) have pass rates of >99%. These results are significantly better (p=0.03) than the results in regular PSQA. In 7 of 58 PDMs (12%) the gamma pass rates of regular PSQA were <90%, while the MC PSQA yielded pass rates of >95%.

Conclusion: MC PSQA is a feasible method for second-check dose validation besides regular PSQA. In case of dose disagreements in regular PSQA, MC PSQA can help to get insight in and solve such discrepancies. In the future, MC PSQA might partially replace regular PSQA, eliminating the need for some of the resource consuming PDMs.

PTC17-0423: A Simplified Empirical Model for Monitor Unit Calculation in Proton Broad Beam with Wobbling and Ridge Filter System

C.Y. Yeh1

1Chang Gung Memorial Hospital at Linkou, Radiation Oncology, Taoyuan, Taiwan- Province of China

Purpose: The wobbling technique combined with ridge filter providing a formation of 3D uniform dose distribution is introduced in two gantry rooms. In order to deliver the monitor unit (MU) with prescribed absolute dose at reference point, a patient specific dose measurement is always mandatory in QA procedure.For quality control purpose, we need to perform a double check in MU measurement. This report is to present a simplified empirical model for MU calculation with wobbling and ridge filter technique.

Materials and Methods: Three wobbling size, 10cmx10cm, 15cmx15cm and 25cmx25cm, and step of 1cm range modulation ridge filter providing 3cm to14cm SOBP were used in clinic. The relationship between the cGyE/MU and treatment range was studied from the MU measurement in ∼1000 fields. The curve fitting formula was used to establish a MU conversion curve for MU calculation.The predicted MU by using this model is substitute for the first MU measurement.

Results: The relationship between the cGyE/MU and energy in different combination of wobbling size and ridge filter is presented by a formula. A simplified empirical model for MU calculation shows the agreement within ±3% in the condition of frequently used treatment parameter. The patient aperture shape shows an insignificant effect in MU calculation. The small variation of MU calculation in different gantry room shows the compatibility of beam performance in two gantry rooms.

Conclusion: The more measurement data improves the higher accuracy in MU prediction by this model. The long-term evaluation by this model will be conducted in the future.

PTC17-0425: Magnetic Field Survey in a Proton Facility

E. Batin1,2, M. Bussière1,2, S.G. Bradley1

1Massachusetts General Hospital, Radiation Oncology, Boston, USA 2Harvard Medical School, Radiation Oncology, Boston, USA

An inquiry about potential access limitations for a therapy technology student prompted a review and verification of original magnetic field surveys in our proton treatment rooms.

While malfunctions of implantable cardiac devices in patients receiving proton therapy have been reported in the literature, there is no reported data about latent static (DC) and time varying (AC) magnetic fields in treatment rooms. No regulatory limits exist for occupational exposure to magnetic fields but recognized organizations provide maximum exposure guidelines of 1-2 Gauss (G) for pacemaker wearers to 50,000 G peak exposure for extremities.

Static and time varying fields were measured in passive scattering and scanning proton therapy gantry rooms and a room with two fixed beamlines. AC magnetic fields were below 0.5 Gauss throughout the treatment rooms - well within conservative limits. DC magnetic fields up to 50 G were measured in localized area (door magnets, phone speakers). More surprising results were found at the proximity of the permanent magnet array of the linear motor actuating the imaging panels with values greater than 100 G at 1 cm from magnet surfaces and decreasing to less than 5 G at 5 cm. The opposite side of the imaging panel arms had a maximum less than 1 G.

As a result of this survey, pacemaker dependent personnel are restricted access to the gantry imaging panel travel areas and patients with pacemaker are preferentially directed to the gantry room with magnetic rails guiding the imaging panels located on the opposite side of the patient.

PTC17-0431: End-To-End Test Procedures Using Alanine Dosimetry and Anthropomorphic Phantoms in Active Scanning Proton Beam Therapy

A. Carlino1,2, H. Palmans1,3, G. Kragl1, E. Traneus4, C. Gouldstone3, S. Vatnitsky1, M. Stock1, J. Osorio1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2University of Palermo, Department of Physics and Chemistry, Palermo, Italy3National Physical Laboratory, Radiation Dosimetry, Teddington, United Kingdom4Raysearch laboratories AB, Particle Therapy, Stockholm, Sweden

At MedAustron a quasi-discrete scanning beam delivery with protons has been established. The clinical implementation requires comprehensive end-to-end testing to ensure accurate patient treatments. We present dosimetric end-to-end test procedures for protons based on customized phantoms and different dosimetric techniques.

A homogeneous polystyrene phantom and two anthropomorphic phantoms (pelvis and head phantom) have been customized to allocate different detectors. All treatment planning steps were performed with RayStation TPS. The plans were delivered to the phantoms loaded either with alanine pellets (measurements of dose distribution) or Farmer chambers (measurement of the dose at reference point).

The alanine pellets and their read-out were provided by the National Physical Laboratory. One of the challenges of alanine for dosimetry in particle beams is the known response dependency (quenching) on the charge and the energy of the particles constituting the mixed radiation field. Corrections for this were derived by a Monte Carlo dose calculation platform implemented in a research version of RayStation.

The measured absolute dose to water obtained with the Farmer chamber in eight delivered plans was within 2% of the planned dose. Doses determined with the alanine pellets after correction for the quenching effect showed a mean deviation within 4% and a maximum deviation below 7% for all the phantoms. The end-to-end test procedures developed at MedAustron showed that the entire chain of radiation treatment works efficiently and with accurate dosimetric results. Our experience shows that alanine pellets are suitable detectors for dosimetry audits in active scanning proton beams.

PTC17-0442: Acoustic Imaging for Proton Therapy Range Verification; Clinical and Technological Limitations; Future Outlook

S. Yousefi1, S. Tzoumas1, L. Xing1

1Stanford University, Radiation Oncology, Palo Alto, USA

There have been multiple efforts to quantify proton-induced thermoacoustic for range in the past few years. However, clinical translation of this technique has not yet achieved due to current limitations.

We performed simulation studies to model proton-induced thermoacoustics. The proton pulse energy deposition in the tissue was modeled using Geant4. The results provided initial pressure for ultrasound wave propagation at the Bragg peak location that was modeled using Finite Element Model. The detection sensitivity, noise equivalent pressure and spectral density of the acoustic pulse at the receiver channel was measured based on the existing ultrasound transducer technology.

Assuming a current clinical scenario, a 20 us, 250 MeV proton beam delivering 100 mGy per pulse into the water tank can produce 20 mPa maximum pressure peak at the Bragg peak location. If the pulse width is reduced to 20 ns, a 0.1 Pa at Bragg peak is generated. The Noise Equivalent Pressure (NEP) of a single-element transducer (center frequency: 5 MHz and 2 cm diameter) is around 0.5 Pa. In order to increase the signal-to-noise ratio, an array of transducers and a novel reconstruction algorithm is proposed.

The proton pulse from clinical machines is typically on the order of microseconds or even milliseconds (larger that stress confinement). That will reduce the amplitude of the acoustic pulse and reduce the frequency content (resolution). To enhance the proton-induced thermoacoustics detection, array-based ultrasound acquisition and model-based reconstruction technique is required.

PTC17-0461: Demo Heavy Ion Cancer Therapy Facility in China

G. Xiao1, X. Cai1

1Institute of Modern Physics- Chinese Academy of Sciences, Heavy Ion Therapy Project, Lanzhou, China

A hospital-based tumor therapy facility HIMM (Heavy-Ion-Medical-Machine) was designed and developed at Institute of Modern Physics, Chinese Academy of Sciences, China. Two demo centers of HIMM, one in Wuwei, and another one in Lanzhou are under construction. The installation of HIMM Wuwei facility has been finished and the first beam has been obtained on Dec. 23, 2015. The registration detection by third part is now in progressing. The clinical trial will be followed after the detection. The HIMM Lanzhou is now in installation.

In HIMM, heavy ion beams of C5+ from ECR source are accelerated to 7 MeV/u by the cyclotron injector. C5+ ions are striped to C6+, and then injected to the synchrotron. The carbon ion beams will be accumulated and accelerated in the synchrotron with the maximum energy of 400 MeV/u, and extracted to the treatment terminal. The radiation fields of 200×200 mm 2 may be formed after the scanning by the magnet. There are 4 treatment terminals including one horizontal terminal, one vertical terminal, one combined terminal of horizontal and vertical direction, and one 45° terminal. All terminals are equipped with beam delivery system, dose monitoring and recording system, patient positioning system, and safety interlock system. With the circumference of 56m the synchrotron is a compact ring for heavy ion therapy. HIMM is the first heavy ion accelerator for tumor therapy in China with independent intellectual property rights.

The progress of the demo facility will be presented.

PTC17-0490: Characterization of the Low-Dose Halo of Proton PBS Spots with and Without a Range-Shifter

C.L. Teng1, L. Lin1, C. Ainsley1

1University of Pennsylvania, Radiation Oncology, Philadelphia, USA

Purpose: The modelling of the low-dose spot “halo” is an important factor in determining the accuracy of dose calculation for proton PBS. We characterize this halo as a function of air-gap and water depth for different proton beam energies, both with and without a range-shifter.

Materials and Methods: We reported previously on our observation that the variation in output with field size at the centre of uniformly-irradiated energy layers is sensitive to the underlying spot shapes. Here, we make measurements of this variation for field-sizes covering 40–250 mm at several depths in water spanning from the surface to near the end-of-range, for three different air-gaps, with and without a range-shifter, and for incident beam energies of 100 and 225 MeV (the two clinically-deliverable extremes). Via a novel technique to be presented, we extract the weight and width of the halo contribution to double-Gaussian functional descriptions of the spot profile at each point.

Results: With no range-shifter present, the shape of the output-versus-field-size curve (thus, the halo) for the low energy beam shows very little depth-dependence, but varies by room; by contrast, the curve for the high energy beam (thus, the halo) has significant depth-dependence, but exhibits little inter-room variation. The effects of including a range-shifter and of variable air-gap are currently being studied.

Conclusion: Without a range-shifter, the halo is dominated by the in-air contribution at 100 MeV and by the in-water contribution at 225 MeV. The consequences of including a range-shifter will be determined.

PTC17-0499: Testing a Mobile In-Room Tumor Blood Flow Rate Measurement System for Proton Therapy Patients

Z. Su1, P. Okunieff 2, Z. Li1

1University of Florida Health Proton Therapy Institute, Radiation Oncology, Jacksonville, USA 2University of Florida, Radiation Oncology, Gainesville, USA

Purpose: Positron emission nuclei of 15O, 13N and 11C are created inside tumors during patients' proton therapy treatment. A coincidence photon detection based system was developed and tested in measuring flow rates of proton irradiated water and the half-lives of water and carbon and their combination.

Materials and Methods: NaI2 based coincidence detection system was mounted on a modified x-ray C-arm platform. The system was tested using proton irradiated water, carbon and their combination. Proton irradiated water was drained at preset flow rates of 10, 15 and 25 mL/min using a peristaltic pump and the radioactivity was recorded by the system. The flow rates of 15O were obtained. Measurements of radioactivity of prostate patient were also performed.

Results: Fitted half lives from measurements of proton irradiate carbon and irradiated water were on average of 109.7 minutes, 19.8 minutes and 118 seconds, respectively. For proton irradiated combination of carbon and water, the fitted half lives are 20.37 minutes and 113.4 seconds, respectively. The measured average flow rates of the proton irradiated water were 8.1 mL/min for 10 mL/min preset, 13.45 mL/min for 15 mL/min preset and 23.45 mL/min for 25 mL/min preset. The measurements data of radioactivity from prostate patient are analyzed and preliminary results shows large signal variations.

Conclusion: The developed C-arm based coincidence photon detection system is capable of measuring various element radioactivity and water flow rates in a proton therapy setting. The patient radioactivity data indicates potential revision of the system to accommodate high counting rates.

PTC17-0503: Evaluation of Tissue Stopping Power Calculation in RayStation

S. Flampouri1, Z. Su1, E. Olguin 2, R. Slopsema1, Z. Li1

1University of Florida, UFHPTI, Jacksonville, USA 2University of Florida, Biomedical Engineering, Gainesville, USA

Commonly, treatment-planning systems (TPS) calculate proton travel distance in tissue from CT images through a calibration curve that translates Hounsfield numbers to relative proton stopping power (SP). As part of TPS commissioning users provide conversion data which directly assign a relative SP value at each CT voxel. RayStation (Raysearch, Sweden), however, requires a curve converting CT numbers to density and then internally calculates proton SP.

Proton SP calculated by the Bethe-Bloch formula, requires both physical density and elementary composition. Previously, the UFHPTI calibration curve was based on 67, clinically relevant, tissues (ICRU46). The corresponding HU were estimated by the stoichiometric method and verified with proton tissue equivalent materials. In RayStation, SP calculations are based on a set of 10 core materials and 50 discrete interpolations (density and composition). Due to this change in methodology the transition to RayStation required SP verification. To identify any differences in proton range, two density-to-CT calibration curves were inserted in the TPS. The first curve (C1) is formed from the ICRU46 tissues (density vs stoichiometric HU). C2 was created by implementing the RayStation SP calculation and forced density to values that result in same SP as the original UFHPTI curve.

To assess the effect on proton range, IMPT treatment plans for prostate, brain and lung were created based on C1 and then recalculated with C2.

The difference in proton range for all calculated fields was sub-mm (0.3-0.8mm). Dose differences up to 3% of the prescribed dose were observed.

PTC17-0511: Patient-Specific Log File Analysis and Dose Re-Computation at MedAustron

T. Bohlen1, J. Osorio1, A. Carlino1, G. Kragl1, M. Stock1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria

Purpose: The dose distribution administered by proton pencil beam scanning is controlled via a dose delivery system (DDS). In addition to patient-specific dose verification by measurements, patient-specific log file analysis and dose re-computation allow verifying the quality of the delivered dose distribution. Log file analysis combined with independent dose computation are planned to reduce time-consuming patient-specific dosimetric measurements in future.

Materials and Methods: As a first step towards establishing patient-specific QA via log file analysis, software tools for the re-computation of the delivered dose distribution in the TPS for a given beam set based on treatment log files were created. Furthermore, a standardized evaluation of DDS log files was established which allows monitoring the performance of the DDS in a condensed form for a given treatment fraction.

Results: All evaluated beam deliveries show dose differences to be always within +/-3% of the planned dose, even in high gradient regions. We display dose differences in % between the planned and delivered dose distributions for a meningioma patient. The average difference and standard deviation of the planned position compared to the delivered position as a function of number of particles (NP)/spot. For NP/spot>2×106, a feedback position correction loop improves positioning notably.

Conclusion: Log file-based dose re-computation and log file analysis are viable tools to help assuring correct dose delivery and may help to reduce patient-specific QA measurements at MedAustron in future. However, first an independent dose computation needs to be established and verified.

PTC17-0512: Automation of Patient Specific Quality Assurance

J. Stoker1, D. Hernandez1, W. Liu1, M. Fatyga1, J. Shan 2, K. Augustine1, C. Beltran3, D. Mundy3, M. Davis1

1Mayo Clinic Arizona, Radiation Oncology, Phoenix AZ, USA 2Arizona State University, Physics, Tempe AZ, USA3Mayo Clinic Rochester, Radiation Oncology, Rochester MN, USA

Purpose: As Intensity Modulated Proton Therapy (IMPT) has matured, and as the elements of proton Patient Specific Quality Assurance (PSQA) have become clearer in our clinic, we developed a system to automate routine portions of the workflow: second dose check, reconstruction of delivered dose from machine log files, and gamma analyses.

Materials and Methods: A web-based interface was developed to access in-house analytical and Monte Carlo dose engines for a secondary dose validation and evaluation of the proton Linear Energy Transfer. Treatment plans that fulfill criteria are further compared against measured dose planes and machine log files. Measured dose planes are acquired for individual fields at 2-3 depths with a 2D array operated over a local area network and immediately saved to a monitored network share drive, prompting an automated 2D/3D gamma analysis and report-generation script. Machine log data is compared to the planned spot position and MU and used for dose volume histogram investigation (DVH) after import into the treatment planning system.

Results: The developed web interface links several QA algorithms that reduce per-patient FTE required within the PSQA workflow. The automated 2D/3D gamma algorithm analyzes the data in less time than our previous method, as demonstrated. Analyzing the log files allows the reconstruction of a true dose volume for further analysis of its DVH.

Conclusion: The in-house developed algorithms allow an efficient and comprehensive analysis of treatment plans. By automating the processes, the time spent on PSQA procedures is substantially reduced without compromising quality.

PTC17-0534: Proton Beam Range Verification Based on Nuclear Activation Analysis Using Monte Carlo Radiation Transport Codes

A. Pourmoghaddas1, V. Moskvin 2, F. Pirlepesov3, Z. Li4, D. Indelicato4, T. Merchant3, W. Yao1

1St. Jude Children's Research Hospital, Radiation Oncology, Memphis, USA 2St Jude Children's Research Hospital, Radiation Oncology, Memphis, USA3St Jude Children's Research Hospital, Radiation Oncology, Memphis, USA4University of Florida Proton Therapy Institute, Radiation Oncology, Jacksonville, USA

Purpose: Proton range inaccuracy is a major source of uncertainty in proton therapy treatment planning. Studies have explored proton range verification using the annihilation signal of activated positron emitters created in the path of proton treatment beams. However, clinical feasibility of this technique has not be thoroughly assessed. In this study, a large scale (73 patient) comparison between PET images acquired after proton radiotherapy vs Monte Carlo (MC) generated PET distributions will be presented.

Materials and Methods: Seventy three SJCRH patients referred for proton radiotherapy were treated at UFPTI between 1/2012 and 3/2016. One field/day treatment plans were generated with the aim that the prescribed dose be covered by at least 95% of the CTV. Qualifying fields were approved and PET imaging followed the delivery of each field. Patient treatment plans and PET images were transferred to SJCRH for processing. The beam nozzle at UFPTI was modeled with TOPAS MC for each patient and phase-space particle data were generated to model beam delivery using FLUKA. The MC set-up was verified against measured dose in a water phantom. Also, ion-chamber measurements were made inside an anthropomorphic phantom to verify the delivered dose against MC (FLUKA).

Results: FLUKA-generated dose distributions matched those generated by our treatment planning system (Gamma test, 97.9% for 3%, 3mm), showing the suitability of FLUKA calculations for this purpose. MC modelling of nuclear activation and clinical validation are currently in progress.

Conclusion: Our preliminary results show that modelling the absorbed dose and nuclear activation using FLUKA is a feasible approach.

PTC17-0536: Development of the Beam Control Technology Using Multiple-Energy Operation at HIMAC

K. Mizushima1, T. Furukawa1, Y. Iwata1, Y. Hara1, N. Saotome1, Y. Saraya1, R. Tansho1, S. Sato1, T. Shirai1, K. Noda1

1National Institute of Radiological Sciences, Department of Accelerator and Medical Physics, Chiba, Japan

HIMAC has provided the carbon-ion beams with various energies for carbon-ion radiotherapy in NIRS. At the HIMAC synchrotron, the beam is slowly extracted by using third-order resonance with the RF-knockout method. Multiple-energy synchrotron operation was developed to realize a fast change of the output beam energy at HIMAC. To shorten the commissioning time, the systematic tuning depending on the beam energy was carried out for the operation parameter adjustment. In addition, we have installed the beam-intensity control system and the fast beam chopper system in the HIMAC beam control system. These systems function to keep the beam intensity high within the limit of the irradiation system and to prevent uncontrollable spilled beams caused by varying the beam energy. By the systems, the changes of the beam energy can be performed quickly and reliably, and the irradiation time can be saved. The irradiation tests were carried out to verify the performance of the beam control system using multiple-energy operation. In the tests, the system could perform the change of the output beam energy and could output the beams of many different energies in a short time. The test results proved that our system can greatly reduce the irradiation time.

Dose Calculation and Optimization

PTC17-0015: An Efficient Method to Determine Double Gaussian Fluence Parameters in the Eclipse Proton Pencil Beam Model

J. Shen1, W. Liu1, J. Stoker1, X. Ding1, A. Anand1, Y. Hu1, M. Herman 2, M. Bues1

1Mayo Clinic, Radiation Oncology, Phoenix, USA 2Mayo Clinic, Radiation Oncology, Rochester, USA

Purpose: To find an efficient method to configure the proton fluence for a commercial proton pencil beam scanning (PBS) treatment planning system (TPS).

Materials and Methods: An in-water dose kernel was developed to mimic the dose kernel of the pencil beam convolution superposition (PCS) algorithm, which is part of the commercial proton beam therapy planning software, EclipseTM(Varian Medical Systems, Palo Alto, California). The field size factor (FSF) was calculated based on the spot profile reconstructed by the in house dose kernel. The workflow of using FSFs to find the desirable proton fluence is presented. The in-house derived spot profile and FSF were validated by a direct comparison with those calculated by the Eclipse TPS. The validation included 420 comparisons of the FSFs from 14 proton energies, various field sizes from 2cm to 20cm and various depths from 20% to 80% of proton range.

Results: The relative in-water lateral profiles between the in house calculation and the Eclipse TPS agree very well even at the level of 10−4. The FSFs between the in house calculation and the Eclipse TPS also agree well. The maximum deviation is within 0.5%, and the standard deviation is less than 0.1%.

Conclusion: Our method significantly reduced the time to find the desirable proton fluences of the clinical energies. The method is extensively validated and can be applied to any proton centers using PBS and the Eclipse TPS.

PTC17-0028: Proton Minibeam Radiotherapy Reduces Side Effects - Characterization of Beam Shapes and Arrangements Based on Simulated Dose Distributions

M. Sammer1, C. Greubel1, S. Girst1, G. Dollinger1

1Institut für Angewandte Physik und Messtechnik, Universität der Bundeswehr München, Neubiberg, Germany

Proton radiotherapy using “minibeams” of sub-millimeter dimensions allows to enhance tissue sparing in the entrance channel by spatial fractionation additionally to advantageous proton depth dose distribution. Spatial fractionation leads to less cells suffering from direct radiation damage and thus to reduced side effects compared to conventional proton therapy, which has already been shown experimentally [1]. Tumor control is maintained via homogeneous irradiation of the tumor due to beam widening with increasing depth.

In this work beam-to-beam distances for quadratic, hexagonal and planar minibeam arrangements in dependence of tumor depth and thickness are calculated to achieve a homogenous dose in the treatment volume. Treatment plans are calculated for a tumor in 10-15 cm depth in a water phantom for the three minibeam scenarios with initial beam size of σ=0.2 mm. Mean cell survival in different depths is approximated with the linear quadratic model based on the resulting 3D-dose distributions.

In the entrance channel minibeam irradiation leads up to ∼85% less cell death, especially in the superficial tissues. The pencil minibeams result in higher cell survival compared to the planar minibeams while all proton minibeam irradiations show higher cell survival than conventional broadbeam irradiation.

These results indicate that proton minibeam radiotherapy reduces side effects but keeping tumor control as in conventional proton therapy. This makes minibeam, especially pencil-minibeam radiotherapy an attractive new approach for radiation therapy.

References: [1] Girst et al. “Proton Minibeam Radiation Therapy Reduces Side Effects in an In Vivo Mouse Ear Model.” International Journal of Radiation Oncology* Biology* Physics 95.1 (2016): 234-241.

PTC17-0075: Helium Ion Fragmentation in Radiotherapy Beams: Studies in Water and in PMMA Targets

G. Arico1,2,3,4, T. Gehrke1,2,3, A. Ferrari4, A. Mairani5,6, T. Tessonnier1,3,7, J. Jakubek8, O. Jäkel1,2,3,5, M. Martisikova1,2,3

1Heidelberg University Hospital, Radiation Oncology and Radiation Therapy, Heidelberg, Germany 2German Cancer Research Center, Medical Physics in Radiation Oncology, Heidelberg, Germany3Heidelberg Institute for Radiation Oncology, HIRO, Heidelberg, Germany4European Organization for Nuclear Research CERN, EN/STI, Geneva, Switzerland5Heidelberg Ion-Beam Therapy Center, HIT, Heidelberg, Germany6National Centre of Oncological Hadrontherapy, CNAO, Pavia, Italy7Ludwig Maximilians University, Experimental Physics - Medical Physics, Munich, Germany8Institute of Experimental and Applied Physics, Applied physics and technology, Prague, Czech Republic

Purpose: Helium ions offer a narrower penumbra in comparison to protons, and a reduced fragmentation tail and lower RBE with respect to carbon ions. Smaller RBE might be advantageous especially for pediatric patients. However, the current lack of experimental data regarding Helium ion fragmentation limits the precision achievable in the Helium beam modeling.

Materials and Methods: Water and PMMA targets were irradiated with 221 MeV/u Helium ion beams, at the HIT facility in Heidelberg. The semiconductor Timepix detectors [Llopart et al. NIM A 581, 2007] were used for single particle measurements. Particle species identification and particle tracking were performed behind the targets. The experimental results were compared with FLUKA Monte Carlo simulations.

Results: Beam attenuation, build up of projectile-fragments and lateral particle distributions in the forward direction were compared between water and PMMA targets with the same water equivalent thickness (WET), and between experiments and simulations. The amount of residual He ions behind equivalent water and PMMA targets agreed within 2%. However, more fragments (15-20%) were detected behind the PMMA than the water targets. A greater amount of residual Helium ions (3-8%) was measured in the simulations than in the experiments.

Conclusion: This research aims to provide some of the missing data, which can be used for benchmarking FLUKA and to enhance the currently used nuclear interaction models. The gained information can increase the accuracy in the treatment planning.

PTC17-0079: A Parameter Study of Proton Grid Therapy (PGT)

T. Henry1, A. Ureba1, A. Valdman 2, A. Siegbahn1

1Stockholm University, Medical Radiation Physics, Stockholm, Sweden 2Karolinska University Hospital, Department of Oncology and Pathology, Stockholm, Sweden

PGT with cross-fired and interlaced proton beams has recently been proposed by our research group. In this work, Monte-Carlo simulations of a simple PGT treatment were performed with varying beam sizes and center-to-center (c-t-c) distances between the beams. The aim was to determine which combinations of those two parameters would produce the most therapeutically desirable dose distributions (high target dose and low valley dose outside of the target).

The grids were aimed towards a cubic target at the center of a water tank. The target was cross-fired in an interlaced manner. Three beam widths (1/2/3 mm FWHM) and a wide range of c-t-c distances (3-12 mm) were studied. The planning objective was to obtain a nearly homogeneous target dose in combination with low peak doses in normal tissue as well as high peak-to-valley dose ratios (PVDRs) close to the target.

The most appropriate c-t-c distances, for 1/2/3 mm beams, were 7/8/10 mm respectively. With these c-t-c distances, a very high PVDR was obtained for the 3 beam sizes at entrance (>10000) and at 1 cm distance from the target (9/10/14 respectively). Inside the target, a high dose homogeneity could be obtained for these cases (σ=±4%).

We studied the possibility to use beam widths in the millimeter range for PGT combined with crossfiring. For each beam size studied, an optimal c-t-c distance was determined according to the selected planning objectives. With the optimal c-t-c distances, high target dose homogeneity could be combined with high PVDRs outside of the target.

PTC17-0093: Independent Monte Carlo-Based Dose Verification Program for Proton Pencil Beam Scanning

Z.Y. Yang1, S.C. Lee1, R.J. Sheu1, C.C. Chen 2

1National Tsing Hua University, Institute of Nuclear Engineering and Science, Hsinchu, Taiwan- Province of China 2ProCure, Proton Therapy Center, Somerset- NJ, USA

Purpose: An independent dose verification program with GUI using 4 different Monte Carlo (MC) codes for proton pencil beam scanning (PBS) was developed for practical applications, e.g. being a helpful tool in routine clinical use or comparing dose results predicted by various MC codes for academic interests.

Materials and Methods: A simplified MC simulation of dose calculation for proton PBS was validated and demonstrated in our previous works. The proton PBS source with measured spot characteristic was used for 4 widely used MC codes including FLUKA, GEANT4, MCNP6, and PHITS. The 3D dose distributions in water could be calculated in an hour using a 32-core computing node for a typical PBS treatment field with statistical uncertainties less than 3%.

Results: The program features intuitive user interfaces in MC execution, output display and result comparisons between MC, measurements, and TPS. The DICOM ION plan exported from TPS can be imported directly into the in-house program. Multiple 2D plane doses at designated depths can be calculated using the user-selected MC code. The 2D dose comparisons between the MC and TPS calculations, MC calculations and measurements as well as MC calculations using different codes can be presented with the gamma evaluations respectively.

Conclusion: The independent MC-based dose verification program can be used prior to the patient treatment without the occupancy of treatment room for patient-specific QA with measurements. The inter-comparison between various MC codes would be more relevant with a developing module of voxel CT phantom for dose calculation in heterogeneity.

PTC17-0122: Effective Particle Energies for Stopping Power Calculation in Radiotherapy Treatment Planning with Protons and Helium, Carbon, and Oxygen Ions

T. Inaniwa1, N. Kanematsu 2

1National Institute of Radiological Sciences, Department of Accelerator and Medical Physics, Chiba, Japan 2National Institute of Radiological Sciences, Medical Physics Section, Chiba, Japan

The stopping power ratio (SPR) of body tissues relative to water depends on the particle energy. For simplicity, however, most analytical dose planning systems do not account for SPR variation with particle energy along the beam's path, but rather assume a fixed energy for SPR estimation. The range error due to this simplification could be indispensable depending on the particle species and the assumed energy. This error can be minimized by assuming a suitable energy referred to as an “effective energy” in SPR estimation. To date, however, the effective energy has never been investigated for realistic patient geometries. We investigated the effective energies for proton, helium-, carbon-, and oxygen-ion radiotherapy using volumetric models of the reference male and female phantoms provided by the International Commission on Radiological Protection (ICRP). The range errors were estimated by comparing the particle ranges calculated when particle energy variations were and were not considered. The effective energies per nucleon for protons and helium, carbon, and oxygen ions were 70 MeV, 70 MeV, 131 MeV, and 156 MeV, respectively. Using the determined effective energies, the range errors were reduced to ≤ 0.3 mm for respective particle species. For SPR estimation of multiple particle species, an effective energy of 100 MeV is recommended, with which the range error is ≤ 0.5 mm for all particle species.

PTC17-0130: Validation of Proton Stopping Power Ratios for Tissue Surrogates in an Electron Density Phantom

M. Gustafsson1,2, E. Pettersson1,3, A. Thilander-Klang1,3

1Sahlgrenska University Hospital, Department of Medical Physics and Biomedical Engineering, Gothenburg, Sweden 2Skandionkliniken, Medical Physics, Uppsala, Sweden3University of Gothenburg, Department of Radiation Physics - Institute of Clinical Sciences, Gothenburg, Sweden

Purpose: To validate the calculated proton stopping power ratios (SPRs) for the inserts of the electron density phantom used by our institution in the conversion of Hounsfield Units to SPR (when using the stoichiometric method).

Materials and Methods: Eight tissue surrogate inserts of an electron density phantom (062A, CIRS) were evaluated. Their mass density, relative electron density and chemical composition were provided by the manufacturer. The mass densities of the inserts were carefully validated with a calliper and a scale. The SPRs were calculated for 226 MeV protons using expressions from Schneider [1]. The residual ranges for the tissue surrogates were measured with an ionization chamber (Bragg peak 34070, PTW) for a 226 MeV single spot proton beam using the cyclotron (Proteus PLUS, IBA) at the Skandionkliniken in Uppsala, Sweden. Five additional well-known materials were also measured and compared to calculated and tabulated data [2] for validation purposes.

Results: The mass densities given by the manufacturer were incorrect with up to 10.8% and the calculated SPRs were hence also incorrect. The measured SPRs agreed within 1.0% of SPRs calculated with the measured mass densities. The measured SPRs of the well-known materials agreed within 1.3 % of tabulated data.

Conclusion: The mass densities provided by the manufacturer were not correct for some materials. Using measured mass densities, the agreement between calculated and measured SPRs were within 1.0 %.

References: [1] Schneider U et al. Phys Med Biol. 1996;41:111-124. [2] PSTAR, National Institute of Standards and Technology (NIST), http://www.nist.gov.

PTC17-0140: Secondary Neutron Characterization in Proton Therapy Using an Organic Scintillator

K.W.S. Chung1

1University College London, Department of Medical Physics and Biomedical Engineering Building, London, United Kingdom

Purpose: Secondary neutrons with energy ranging from thermal level up to the incident proton energy are generated, which could potentially impose radiation hazard on patients and radiation workers. Its effects, including secondary malignant neoplasm and biological effective dose, on the remote non-target organs remain uncertain. Numerous detectors have been reported for neutron measurement.1 Prolonged measurement time, insensitivity of fast neutrons and lack of gamma/neutron discrimination are the known drawbacks. EJ299-33 has been reported its ability in detecting neutrons with energy range from 10 to 100 MeV 2, which reveals potential in neutron measurement in PT. Thepurpose of this study is to simulate and measure secondary neutrons generated in proton therapy (PT) with organic scintillator.

Materials and Methods: 1.)Secondary neutron characteristion was performed using Geant4 simulation. A tissue-equivalent plastic was irradiated with a 150MeV proton beam. Spatial and energy distribution were investigated. 2.) Calibration of EJ299-33 organic scintillator with Pulse Shape Discrimination(PSD) for neutron detection is in progress.

Results: For neutrons leaving at the surfaces parallel to the beam axis, they having median KE of (9.0±3.5)MeV leave the phantom at about 60° corresponding to the beam direction. For those leaving at the surfaces perpendicular to the beam axis, their median KE is about (27.8±0.8)MeV at the distal end and (0.8±0.5)MeV at the proximal surface. In the future, we will integrate the results from simulation and experiment to estimate the neutron biological effective neutron dose among non-target organs in PT.

Reference: [1] De Saint-Hubert M., et al, Radiat Prot Dosimetry. 170(2016) 336. [2]Buffler A., et al, IEEE. 62(2015) 1422.

PTC17-0146: Proton Stopping Power Ratio Determination with Basis Material Decomposition

E. Pettersson1, M. Gustafsson 2,3, A. Bäck 2,3, A. Thilander-Klang1,2

1Department of Radiation Physics, Institute of Clinical Sciences. Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden 2Department of Medical Physics and Biomedical Engineering, Sahlgrenska University Hospital, Gothenburg, Sweden3Skandionkliniken, Uppsala, Sweden

Purpose: Dual energy CT (DECT) scanners with projection-based material decomposition can produce material density (MD) images. Here, the photon attenuation of an arbitrary material is represented as a combination of two materials (e.g. water and iodine) [Alvarez & Macovski 1976]. This work investigates the accuracy of a quadratic parameterization from theoretical MDs of human reference tissues to their proton stopping-power-ratios (SPRs).

Materials and Methods: The water and iodine MD combination for 73 human reference tissues [White et al. 1987] were determined by linear fits of the mass attenuation coefficients of water and iodine to the attenuation coefficients of each reference tissue for 9 photon energies between 60 and 140 keV. The attenuation coefficients were calculated using their tabulated physical densities, elemental composition and elemental photon attenuation cross-sections. SPR values for 226 MeV protons in the tissue were calculated with the Bethe formula. Two quadratic fits between the MDs and the SPRs were performed for soft tissues and bone, respectively.

Results: The calculated combinations of water and iodine MDs are shown. The root mean square errors (RMSE) of the fitted SPRs are within 0.09% of the calculated SPRs, the maximal and minimal values are 0.22% and -0.34% for “skeleton red marrow” and “eye lens”, respectively.

Conclusion: Given perfect MD accuracy, the SPRs of soft tissues and bone can be determined to within ±0.4%.

References: Alvarez R.E. and Macovski A., Phys. Med. Biol., 1976;21(5):733-744. White D.R., Woodard H.Q. and Hammond S.M. Br J Radiol 1987;60(717):907-13.

PTC17-0154: The Use of Monte Carlo Simulations and Proton Radiography to Calibrate Hounsfield Unit to Relative Stopping Power Curves

F. Baker1, A. Gibson 2, P. Doolan3, V. Rompokos3

1The Christie NHS Foundation Trust, The Christie Medical Physics and Engineering, Manchester, United Kingdom 2University College London, Medical Physics and Bioengineering, London, United Kingdom3University College London Hospitals NHS Foundation Trust, Radiotherapy Physics, London, United Kingdom

Converting Hounsfield units (HU) to relative stopping powers (RSP) causes uncertainties in the range of a proton beam. We have shown that these uncertainties can be reduced using a HU to RSP calibration curve made patient-specific with proton radiography. Here, we extend this previous work, which assumed that protons travel in straight lines through tissue, by using a Monte Carlo simulation to more realistically calculate the proton path.

Time-resolved proton radiographs of anthropomorphic phantoms were acquired at Massachusetts General Hospital and simulated in Geant4 by scoring the proton flux on an imaging plane behind the phantom. Any differences from the real proton radiograph were assumed to be due to an incorrect definition of density, elemental composition or excitation energy from the CT, and corrected by varying the segmented material densities in the simulation, altering the HU-RSP curve. Simulations were iteratively improved until they agreed with the measurements.

The simulation agreed closely with existing measured depth-dose curves. The phantom CT was converted into different materials in Matlab and a HU-RSP curve produced. By optimising the density of materials in the simulated proton radiographs, the accuracy of the HU-RSP curve was significantly improved after relatively few iterations.

This work shows that Monte Carlo simulations could be used alongside proton radiography to generate patient specific HU-RSP curves, giving estimates of the patient anatomy. By iteratively altering the material density between simulations, a more accurate, patient-specific calibration curve was generated.

PTC17-0202: Development of Dose Verification System Using a Simplified Monte Carlo Method for Scanned Proton Therapy

S. Mizutani1, H. Kenji 2, H. Baba 2, T. Yamaguchi1, T. Mukawa1, Y. Aoki1, T. Miyashita3, T. Akimoto 2

1Sumitomo Heavy Industries- Ltd., Technology Research Center, Yokosuka-shi, Japan 2National Cancer Center, Research Center for Innovative Oncology, Kashiwa-shi, Japan3Sumitomo Heavy Industries- Ltd., Industrial Equipment Division, Niihama-shi, Japan

Purpose: The use of monitor log files to perform patient specific quality assurance for photon therapy has been established. Therefore, we have developed a dose verification tool using log files and the simplified Monte Carlo (SMC) method for scanned proton therapy to check the dose delivery accuracy. An SMC method [1] has been developed to obtain an accurate dose distribution estimate in heterogeneous regions within a reasonable calculation time.

Materials and Methods: Each delivered proton spot beam passes through dose and flatness monitors which provide information about the beam size, position and dose. We have made a program to reconstruct the patient delivered dose distribution based on logs of these parameters. We have calculated the reconstructed dose distribution for a prostate cancer patient that was treated at the National Cancer Center Hospital East (NCCHE).

Results: The reconstructed dose calculated by the SMC method in heterogeneous regions reproduces the irregular shape of the dose distribution accurately. The influence of bone structure on the dose distribution can be clearly observed.

Conclusion: We have developed dose verification software using monitor log files calculated with the SMC method for scanned proton therapy. This tool can calculate the dose distribution in a patient accurately. Therefore, this tool can enhance the treatment plan quality and improve safety of treatment. In future work, we will use gamma analysis to compare the dose verification tool results with the dose distribution from the treatment planning system.

References: [1] S. Mizutani et al, J. Appl. Clin. Med. Phys. 17 (2016), pp.315-327

PTC17-0224: Experimental Validation of a Commercial Proton Monte Carlo Treatment Planning System

C. Chen1, C. Zeng1, E. Traneus 2, M. Janson 2, E. Van Wie1, D. Mah1

1ProCure Proton Therapy Center, Medical Physics, Somerset, USA 2RaySearch Laboratories AB, Proton Planning, Stockholm, Sweden

Purpose: A pre-release commercial Monte Carlo treatment planning system (RayStation, v6, RaySearch Laboratories AB, Sweden) was validated by the 2D-dose measurements at depths downstream of (1) fresh animal tissue and (2) breast implant with a metallic tissue expander.

Materials and Methods: An animal tissue (fresh pig butt and bone), and a silicone-filled breast implant over a water-filled tissue expander (ALLERGAN, USA) were bound with Plastic Water (CIRS, USA) and Lucite respectively. The permanent magnet and titanium needle guard in the tissue expander were contoured based upon the manufacturer supplied physical dimensions and assigned with corresponding mass densities. The 2D-dose distributions at the depths close to the distal-dose fall off of a single spot scanning field were measured with the MatriXX-PT detectors (IBA-dosimetry, Germany) and compared to the calculated results in the RayStation TPS using both Monte Carlo (MC) and pencil beam convolution (PBC) methods.

Results: There was good overall agreement throughout the volume between the MC and PBC calculations. Notably, the MC calculation gave better agreement with measurement in the sharp gradient at the edges of high Z interfaces (e.g. magnet) than PBC. The depth for different calculated dose planes was changed until optimal agreement was found between measurement and calculation. The measured depth was <2.75% compared to calculated which is ascribed to a complex combination of range uncertainty and scattering.

Conclusion: PBC is sufficient for most situations, but MC method has better edge gradient modeling, which could be clinically relevant for treatment volumes close to bony or high-Z materials.

PTC17-0238: Determining the Impact of Tumor Hypoxia on the Clinical Dose in Carbon Ion Radiotherapy

A.E. Paz1, Y. Hirano 2, M. Tashiro 2, T. Kanai1

1Gunma University, Graduate School of Medicine, Maebashi, Japan 2Gunma University, Gunma University Heavy Ion Medical Center, Maebashi, Japan

The clinical dose system in Japan was established to achieve a homogenous biological response within the spread-out Bragg peak (SOBP). However, the relative biological effectiveness (RBE) considered in the optimization accounted only for the linear energy transfer (LET) dependence and neglected the oxygen effect. The enhancement in radioresistance of cells with severe oxygen deprivation is pronounced for therapeutic beams with low LET but is observed to diminish at very high dose-averaged LETs of about 300 keV/μm. Unfortunately, a typical 60-mm carbon ion SOBP has LET values ranging only from 30 to 180 keV/μm. Therefore, treatments with carbon ions are still not immune to the oxygen effect. This paper examines the effectiveness of the carbon ion SOBP on hypoxic tumors by using a three-dimensional tumor model with a spatially varying oxygen distribution. The model simulates oxygen diffusion and consumption by solving the reaction-diffusion equation with the Finite-difference method. The biological dose is determined by applying the physical dose from a 60-mm SOBP of a 290 MeV/n-carbon beam to the tumor model. The difference in the effective biological dose with and without hypoxia clearly highlights the impact of the oxygen effect on the clinical dose system.

PTC17-0242: Clinical Impact of Range Shifter Air Gap: Can Bolus Range Shifting Improve Plan Quality in IMPT of Head-And-Neck Cancer?

S. Michiels1, A. Barragán 2, K. Souris 2, K. Poels3, W. Crijns3, J. Lee 2, E. Sterpin1,2, S. Nuyts1,3, K. Haustermans1,3, T. Depuydt1

1University of Leuven, Department of Oncology- Laboratory of Experimental Radiotherapy, Leuven, Belgium 2Université catholique de Louvain, Institut de Recherche Experimentale et Clinique- Center of Molecular Imaging- Radiotherapy and Oncology, Woluwe-Saint-Lambert, Belgium3University Hospitals Leuven, Department of Radiation Oncology, Leuven, Belgium

Purpose: In intensity-modulated proton therapy of head-and-neck cancer, range shifter (RS) air gap compromises spot size and penumbra. The impact on plan quality, the potential of Monte Carlo (MC) based optimization to mitigate these effects, and the advantage of bolus compared to current RS solutions were investigated.

Materials and Methods: Plans for 5 oropharyngeal cancer patients were generated. Three RS configurations were compared: bolus, snout mounted and nozzle mounted. The air gap effect was quantified by recalculating the optimal bolus plan on the snout and nozzle configuration. Subsequently, the plans for the snout and nozzle were optimized using MC to re-establish clinical goals in the presence of the air gap. Finally, organ at risk (OAR) dose and late toxicity risks were calculated for all configuration-specific optimized plans.

Results: Disregarding the air gap during optimization yielded average loss of coverage (D98%) of the planning target volume (PTV) ranging from 6% (snout; low dose PTV) to 12% (nozzle; high dose PTV). Configuration-specific optimization re-established PTV coverage, but increased the mean dose to all OARs. Bolus reduced the average contralateral parotid gland complication risk with 3.0% and 3.8% compared to snout and nozzle RS, respectively. For dysphagia, bolus reduced the risk with 2.7% and 3.6% compared to snout/nozzle.

Conclusion: Including the air gap in plan optimization is essential, but cannot cancel out the impact of spot degradation on OAR dose. Bolus improves plan quality compared to snout/nozzle mounted RS, with significant risk reductions for parotid gland dysfunction and dysphagia.

PTC17-0272: Impact of Pencil Beam Modeling on Dosimetric Accuracy in Proton Therapy Treatment Planning: Experimental and MC Validation

S. Russo1, S. Molinelli1, G. Magro1, A. Mairani1, E. Mastella1, A. Mirandola1, A. Vai1, E. Ciurlia 2, A. Iannalfi 2, M. Ciocca1

1CNAO Foundation, Medical Physics Unit, Pavia, Italy 2CNAO Foundation, Clinical Department, Pavia, Italy

Aim of this work is to describe the validation procedure adopted for the implementation of the RayStation 5.02 TPS for pencil beam scanning proton therapy based on the inter-comparison with the Siemens Syngo VC13 TPS, currently in clinical use. Benchmarks to estimate involved uncertainties were experimental data and Monte Carlo (MC) simulations.

Plans were optimized with Syngo for cubic volumes (2 to 10 cm-sided) positioned at increasing depths in water, exported for re-calculation with RayStation and with the available independent Fluka-MC engine. Criteria for dosimetric accuracy evaluation were particle range, target DVH deviations and 3D-gamma evaluation (3%-2 mm). Plans were measured in a water phantom with pin-point chambers. This process was then repeated for 15 previously delivered patient plans.

Dose calculation on regular volumes in water showed a good level of agreement with MC simulations for both TPSs for cube sizes > 3 cm, requiring no range shifter. Patient DVH analysis showed a significant increase of the target dose, as estimated by RayStation in comparison with MC, which becomes more critical with the use of range shifters (target median dose variations up to around 6%). Doses outside target volumes were consistently in agreement with MC simulations for both TPSs, based on gamma evaluation results. Pin-point measurements confirmed data obtained from the MC simulation analysis.

The two TPSs under examination present different approaches to proton beam modeling that can lead to unexpected clinically significant differences in the calculated dose, in particular in presence of tissue heterogeneities and for superficial tumor volumes.

PTC17-0292: Monte Carlo Simulations of Magnetic Field Effects on Proton Dose Distributions in a 1T Measurement Setup

S. Schellhammer1,2, B. Oborn3,4, A. Lühr1,2,5, S. Gantz1,2, M. Bussmann6, A. Hoffmann1,2,7

1Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiooncology, Dresden, Germany 2OncoRay - National Center for Radiation Research and Oncology, Medical Radiation Physics, Dresden, Germany3Wollongong Hospital, Illawara Cancer Care Center, Wollongong, Australia4University of Wollongong, Centre for Medical Radiation Physics, Wollongong, Australia5German Cancer Consortium DKTK, Partner Site Dresden, Dresden, Germany6Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiation Physics, Dresden, Germany7Faculty of Medicine and University Hospital Carl Gustav Carus at the Technische Universität Dresden, Department of Radiation Oncology, Dresden, Germany

For an integration of proton therapy and magnetic resonance Imaging (MRI), mutual effects of these two modalities need to be assessed. We studied the magnetic field-induced proton beam deflection as well as the radiation-induced activation and demagnetization of the magnet material by simulating irradiation experiments with a realistic magnet.

Geant4 Monte Carlo simulations were performed for 80-180 MeV proton pencil beams traversing the 0.95 T transverse magnetic field of a dipole magnet. A PMMA slab phantom containing a radiochromic EBT3 film was placed between the magnetic poles, such that the incident beams were stopped in the film plane. The magnetic field was modelled using 3D finite-elements and validated by magnetometry. Beam trajectories were analyzed from the film's planar dose distributions. Upper bounds for radioactivation were deduced by analyzing the most common mother nuclides, and demagnetization was assessed by relating the simulated magnet dose to previously published data.

Considerable magnetic field-induced dose distortions can be observed from the planar dose distributions. Lateral displacement of the Bragg peak ranged from 1-11 mm for 80-180 MeV beams. Initial activation of the magnet material was less than 25 kBq, and the mean dose to the magnet poles was ∼20 μGy when delivering a dose of 2Gy at the Bragg peak to the film.

These results indicate that the Lorentz force-induced dose distortions are substantial, and measurable with the presented setup. Radiation-induced activation and demagnetization effects are small but should be monitored during the irradiation experiments.

PTC17-0294: Heterogeneity Correction for Proton Beams in Lung Tissue

Z. Belal1, D. Mirkovic1

1MD Anderson, Radiation Physics, Houston, USA

Purpose: The microscopic heterogeneities in lung tissue can cause degradation of distal edges of proton beams affecting the accuracy of dose calculations in volume averaged lung models that are currently used in clinical treatment planning systems. In this work, we investigated changes in proton beam characteristics using an accurate microscopic model of lung parenchyma with an aim to find if the dose accuracy can be improved by using an appropriate heterogeneity correction.

Materials and Methods: To determine the necessary heterogeneity corrections, we used a hierarchical model of lung tissue comprised of a standard macroscopic CT-based model of the whole lung with ∼2 mm3 resolution and a microscopic model of lung parenchyma approximated accurately using a grid of truncated octahedrons. TOPAS (Tool for Particle Simulation) was used to simulate the transport of mono-energetic protons with incident proton energies in the therapeutic range of 50 to 200 MeV through the heterogeneous and the corresponding homogeneous phantoms of equal average density. Phase spaces were collected downstream of each phantom and the properties of resulting particle distributions were compared. Additionally, we compared the dose and linear energy transfer (LET) in homogeneous and heterogeneous materials.

Results: We compared the scattering power of the homogeneous versus the heterogeneous material. Preliminary results indicate that the scattering power of the heterogeneous material is roughly 78% greater than that of the homogeneous material with no dependence on incident proton energy. Our results also show small to moderate differences in dose and significant differences in LET between the homogeneous and heterogeneous phantoms.

PTC17-0326: Validation of the CT Number to Mass Density Conversion Curve for Proton Dose Calculation: Proton Range Measurements in Animal Tissues

J. Gora1, G. Kragl1, S. Vatnitsky1, T. Bohlen1, M. Teichmeister1, M. Stock1, V. Letellier1

1EBG MedAustron, Medical Department, Wiener Neustadt, Austria

Proton dose calculation in the TPS is based on HU information taken from the CT scans and its relation to the relative stopping powers (RSP). However, tissue-equivalent-substitutes used for commissioning purposes may not reflect the properties of human tissues. Therefore, animal tissues (pig) were used for validation of the HU-to-mass density (MD) conversion implemented in the TPS (RayStation, v5.0.2).

10 tissue samples were placed in specially designed PMMA phantoms (head and pelvis) and CT scans were acquired. Phantoms design not only helped to reduce imaging artifacts but also allowed to apply a slight pressure on the tissues (for removal of unwanted air). Subsequently, the tissue phantom was attached to the front of the water phantom and range measurements were performed. HU acquired from the CT data were correlated with the measured RSP and validated against implemented HU-to-MD conversion curves. The measured data for all soft tissues were found to be within 1% agreement with the calculated data. Only for lung and bone the deviations were up to 3.5%.

The experimental validation of the conversion curve resulted in good agreement between measured and calculated data. There is a number of uncertainty sources related to these measurements, starting from HU to RSP model, real tissue heterogeneities or uncertainties related to acquisition of the CT data due to beam hardening. The last one, we tried to minimize by using especially dedicated phantom.

PTC17-0333: Monte Carlo-Based Biological Treatment Plan Optimization for Intensity Modulated Carbon Ion Therapy on GPU

N. Qin1, C. Shen1, M. Pinto 2, Z. Tian1, R. Taleei1, G. Dedes 2, A. Pompos1, S. Jiang1, K. Parodi 2, X. Jia1

1UT Southwestern Medical Center, Radiation Oncology, Dallas, USA 2Ludwig-Maximilians Universität, Medical Physics, Munich, Germany

Purpose: Biological effects must be considered in carbon-ion therapy treatment planning. For intensity-modulated carbon-ion therapy (IMCT), it is desirable to employ Monte Carlo (MC) methods to compute properties of each pencil-beam spot for treatment planning because of the accuracy. We have previously developed goCMC, a GPU-oriented fast MC engine for carbon-ion therapy. The purpose of this study is to build a biological treatment plan optimization system on top of goCMC.

Materials and Methods: Repair-Misrepair-Fixation model was integrated into goCMC to compute spatial distribution of linear-quadratic model parameters α and β for each spot. Based on prescribed biological dose distribution, we computed corresponding logarithm of cell survival log(S). Our treatment plan optimization problem then minimized difference between the prescribed and actual log(S). The choice of using log(S) simplified the optimization problem compared to the typical approach of minimizing difference between prescribed and actual biological dose. We tested our method in a water phantom case with two perpendicular beams and a prostate cancer case with two laterally opposed beams.

Results: In both cases, our system was able to generate spread-out Bragg peaks to cover the targeted regions with the prescribed biological dose, while sparing critical structures. For the prostate case with >9,000 spots, total computation time including spot simulation and optimization was within 2 hours using one NVidia Geforce GTX Titan-Black GPU.

Conclusion: We have developed a MC-based biological treatment plan optimization system for IMCT. With a single GPU, it can generate treatment plans in a clinically viable time frame.

PTC17-0349: Profiling of a CUDA and Monte Carlo Based Robust Optimization System for Proton Therapy on a GPU Cluster

A. Abdel-Rehim1, J. Ma1, H. Kamal Sayed1, H. Wan Chan Tseung1, M. Herman1, C. Beltran1

1The Mayo Clinic, Radiation Oncology, Rochester, USA

Methods: The locally developed Monte Carlo (MC) based and GPU accelerated robust optimization code was run on a cluster with 10 nodes, with 4 K80-GPUs each. The code was profiled for time and number of nodes. We monitored time for data pre-processing, MC dose influence maps calculation and storage, and optimization. Clinical cases were used to explore performance dependence on target size, number of ROIs, and complexity of constraints. 9 robust scenarios (position and range) were included.

Results: Both cases have 3 proton beams, 12 constraints, and 9 robust scenarios. Generating and storing the dose influence map is the largest component. The computational part of MC should have a perfect scaling, indicating that beyond 24 GPUs, data storage becomes the dominant factor. The scaling of the optimizer also shows a large data movement component with little improvement with the number of GPUs. We show the dependence of the problem size for 16 GPUs. bst-brain was 2.4x slower than p-brain. This is because the bst-brain case has 1.5x more spots.

Conclusion: Data movement is a bottleneck for scaling robust optimization with GPUs. For efficient clinical utilization of GPU clusters, this time must be minimized.

PTC17-0383: Dose Comparison between Pencil Beam and Monte Carlo Calculation Algorithm in Left Chest Wall Patients

A. Fung1, J. Saini1, D. Maes1, L.C. Fang 2,3, T. Wong1

1Seattle Cancer Care Alliance Proton Therapy, Medical Physics and Dosimetry, Seattle, USA 2Seattle Cancer Care Alliance, Department of Radiation Oncology, Seattle, USA3University of Washington, Department of Radiation Oncology, Seattle, USA

Purpose: Pencil beam dose calculation algorithm (PBA) for protons implemented in most treatment planning systems does not accurately account for lateral inhomogeneities for each of the ray traces and the dose scattered from range shifters. This may compromise the dose calculation accuracy for chest wall treatment in proton therapy. This study investigates the dose differences for chest wall patients between pencil beam and Monte Carlo dose calculation algorithm (MC).

Materials and Methods: 3 previously treated left chest wall patients to 50.4 Gy (RBE) calculated with pencil beam dose calculation algorithm were retrospectively re-calculated with Monte Carlo dose calculation algorithm. One field was used for treatment with a range shifter of 7.5 cm water equivalent thickness. The air gap used ranging between 10.0 cm and 17.9 cm. PTV coverage, overall hotspot, doses to left lung and heart and homogeneity index (HI) were analyzed.

Results: With Monte Carlo dose calculation algorithm, coverage to planning target volume (PTV) V95 was significantly decreased, ranging from 89.79% to 71.58%. Overall hotspot was between 114.6% and 121.7%. Higher doses to left lung among V20, V5 and mean were noticeable. There were minimal changes in doses to heart. Homogeneity index ranged from 1.16 to 1.26.

Conclusion: Pencil beam dose calculation algorithm does not accurately account for proton transport in the situation when treating a shallow target volume combined with the use of range shifter and larger air gap. This will compromise the dose coverage to PTV with less dose conformity and underestimate the dose to lung.

PTC17-0389: Relative Biological Effectiveness and Its Impact on Dose Calculation in Proton Therapy

A. Madkhali1,2, M. Partridge3, C. Hanquist Stokkevåg4

1University of Oxford, Department of Oncology, Oxford, United Kingdom 2King Saud University, Department of Medicine, Riyadh, Saudi Arabia3St. Ola, Orkney Islands, United Kingdom4Haukeland University Hospital, Department of Oncology and Medical Physics, Bergen, Norway

Purpose: In proton therapy (PT), a relative biological effectiveness (RBE) of 1.1 (RBE1.1) is often applied clinically. In reality, RBE depends on dose, linear energy transfer (LET), biological end point, and tissue type. Using RBE1.1 may lead to inaccurate dose calculation and hence, influence modelled outcome.

Materials and Method(s): We used in-house built software to calculate biological dose using voxel-by-voxel dose maps (Timlin et al, 2015). A published RBE model was implemented to calculate structure-specific RBE, recalculate organ dose and compare it with planned dose calculations using RBE1.1. Dose was analyzed for a number of structures from PT plans for six pediatric patients diagnosed with medulloblastoma. Variable RBE (RBEMinMax ) was calculated using the Carabe-Fernandez model (Carabe-Fernandez et al. 2007) which is a function of dose (d), α and β and RBEMin and RBEMax:

 
formula

Results: We show that the effect of using RBEMinMax was more evident in some patients and structures than others. In the selected organs, mean dose varied between 2.9% and 7.2% (average =5%) while the mean dose varied between 0.3% and 4.9% (average = 2.2%). Variations seen could be a result of age-related structural/volume differences.

Conclusion: Using RBE1.1 may contribute to making PT dose and dose-dependent predictions less accurate. Our results using an RBE calculation model show systematic differences in dose. This may have clinical implications and can influence predicted risk of secondary cancer as well as normal tissue complication probability.

PTC17-0392: A Semi Monte-Carlo Method to Calculate Proton Linear Energy Transfer in Voxelized Patient Geometries

X. Ding1, W. Liu1, M. Bues1

1Mayo Clinic Arizona, Radiation Oncology, Phoenix, USA

To integrate radiobiological modelling with clinical treatment planning (TPS) for proton radiotherapy, we developed a semi Monte-Carlo (MC) method that calculates proton linear energy transfer (LET) in voxelized patient geometries. The method was implemented in our in-house TPS for the spot scanning pencil beam. The method is much faster than the MC method while the calculated LET agrees well with the LET from MC method. The method can be implemented in the inverse treatment planning optimization, allowing us to create LET-based objectives in inverse planning.

First, we developed a Geant4 code to model the proton treatment nozzle installed in our hospital. The code was used to generate dose-averaged LET (LETd) kernels for all 97 proton energies. LETd includes the contribution of primary and secondary protons. Second, the LETd kernels were incorporated in the in-house TPS to calculate LET in voxelized patient geometries. Since the LETd kernels were pre-calculated, the LET distribution in patient geometries takes much less time comparing with a full-blown MC calculation. The LET distribution calculated by the in-house TPS agrees well with the MC calculation. The calculated LET distributions in patient geometries were used to evaluate potential clinical benefit and toxicity in our facility for various tumor sites, including lung, esophageal and brain. The dose and LET distribution can be further extended, using radiobiological models, to include radiobiological effectiveness (RBE) calculation in TPS.

PTC17-0457: Current Developments of a Monte Carlo-Based Treatment Planning System for Proton Therapy

W. Sauerwein1, C. Gomà 2, J. Sempau3, M. Rodriguez4, E. Sterpin5, C. Bäumer6, A. Lallena Rojo7, A. Carnicer8, A. Wittig9, L. Brualla1

1University Hospital Essen, NCTeam- Department of Radiation Oncology, Essen, Germany 2University Hospital Leuven, Department of Radiation Oncology, Leuven, Belgium3Universitat Politècnica de Catalunya, Institut de Tècniques Energètiques, Barcelona, Spain4Centro Médico Paitilla, Radiation Oncology, Panama City, Panama5Université Catholique de Louvain, Institut de Recherce Expérimentale et Clinique, Louvain-la-Neuve, Belgium6University Hospital Essen, Klinik für Partikeltherapie und Westdeutsches Protonentherapiezentrum Essen WPE, Essen, Germany7Universidad de Granada, Departamento de Física Atómica- Molecular y Nuclear, Granada, Spain8Centre Aontoine-Lacassagne, Centre de Protonthérapie, Nice, France9Philipps-Universität Marburg, Klinik für Strahlentherapie und Radioonkologie, Marburg, Germany

The accurate delivery of absorbed dose distributions is essential to assess the effectiveness of radiotherapy in terms of tumor control and normal tissue complication probabilities. The Monte Carlo method is considered the state-of-the-art for radiation transport problems. Although some research groups have developed Monte Carlo-based systems to compute proton dose distributions in patient geometries (e.g., GATE, TOPAS, MCTP), their integration into treatment planning systems is limited.

In 2013 the PRIMO system for the Monte Carlo simulation of external beam radiotherapy with medical linacs was released. PRIMO uses PENELOPE as the Monte Carlo engine for its computations, which in the context of electron-gamma showers is one of the most accurate general-purpose radiation transport Monte Carlo codes.This software system is wrapped under a user-friendly graphical interface conceived for the setup, computation and analysis of linacs and the subsequent absorbed dose distribution tallied in phantoms or computerized tomography images of patients. The software system is freely distributed through the website http://www.primoproject.net. Recently a set of subroutines has been developed by the author of the code to simulate electromagnetic interactions of protons with matter.

Our ongoing project aims at the integration of this new proton transport subroutines, as well as the peculiarities of proton therapy, into a modified version of PRIMO for proton beams. This new software will be initially commissioned for the proton nozzles in Leuven, Essen and Nice. Owing to the state-of-the-art multiple scattering algorithm implemented in PENELOPE, the system can be used for dosimetric studies of very small fields and pencil beam scanning.

PTC17-0506: Dosimetric Validation of the Monte Carlo Dose Engine in the Treatment Planning System RayStation for Scanned Proton Fields Including Apertures

M.S. Janson1, E. Engwall1, M. Gao 2, J. Lundberg1, T. Nordström1, R. Somers1, F. Tamm1, E. Traneus1

1RaySearch Laboratories, Development, Stockholm, Sweden 2Northwestern Medicine Chicago Proton Center, Physics, Warrenville- Illinois, USA

92 scanned proton fields including a block aperture has been delivered to a solid water phantom at the Northwestern Medicine Chicago Proton Center. The plans were either single energy layers or multi energy layers with spot patterns optimized to yield uniform doses in box shaped targets of a water phantom. Rectangular shaped apertures of various sizes were used. Plans were created iterating combinations of: proton distal range, target width, height, and extension in depth, aperture to phantom surface air gap, and inclusion of a range shifter. The measurements included 468 lateral profiles in X and Y at depths ranging from the near surface region to end of range, 56 central axis depth dose curves, and 36 absolute doses at the center of the targets. The doses of the plans were calculated using the Monte Carlo dose engine of the treatment planning system RayStation version 6.

The agreement between calculated measured doses was quantified in terms of gamma(3%, 3mm) passing rates (profiles and depth doses), penumbra width (20 to 80%), full width at half maximum (profiles), 90% proton range (depth doses), and relative dose error (absolute dose points). The agreements of all measurements were found to pass the defined acceptance criteria, demonstrating, among other things, the correct handling of the so-called edge scatter effect.

PTC17-0513: Dose Comparison between Proton Pencil Beam and Monte Carlo Dose Calculation Algorithm for Lung Lesions

D. Maes1, S. Bowen 2, A. Fung3, J. Saini1, C. Block 2, A. Eagen1, J. Zeng4, R. Rengan4, T. Wong1

1SCCA Proton Therapy Center, Medical Physics, Seattle, USA 2University of Washington, Medical Physics, Seattle, USA3SCCA Proton Therapy Center, Dosimetry, Seattle, USA4University of Washington, Radiation Oncology, Seattle, USA

Purpose: Pencil beam dose calculation algorithms (PB) for protons lack the capability to accurately calculate dose in heterogeneous environments leading to potential underdosing of the target volume for lung treatments. Monte Carlo dose calculations (MC) yield a more accurate estimation of dose in heterogeneous environments. This study investigates dose differences between PB and MC for lung lesions using proton pencil beam scanning (PBS).

Materials and Methods: Five previously treated lung patients calculated with PB were recalculated using MC. Each patient was treated with PBS using varying gantry angles with dose prescriptions ranging from 60 Gy (RBE) to 66.6 Gy (RBE). Dose differences between PB and MC were investigated through analysis of the following quantities: CTV V95, CTV homogeneity index, total lung V20 and global max dose.

Results: When recalculated using MC, reduction in coverage to the CTV V95 ranged from 4.6% to 13.3% compared to treatment plans calculated with PB. The average homogeneity index for the CTV was 1.03 and 1.10 for PB and MC respectively. Changes to the total lung V20 were minimal with all measured quantities falling within 1%. All measured global max doses were within 8%.

Conclusion: A retrospective analysis of five lung patients showed a reduction in coverage to the CTV when previously treated patients were recalculated using MC. This indicates that the use of PB for proton beam lung lesion treatments potentially leads to an underdosing of the target volume necessitating the need of MC for these types of treatments.

PTC17-0519: Robustness Comparison of Double Scattering and Pencil Beam Scanning Proton Therapy Treatment Plans for a Head-And-Neck Patient

Z. Perko1, J. Daartz1, T.M. Madden1, N. Depauw1, D. Lathouwers 2, T.R. Bortfeld1

1Massachusetts General Hospital / Harvard Medical School, Radiation Oncology, Boston, USA 2Delft University of Technology, Department of Radiation Science and Technology, Delft, Netherlands

Proton therapy is advancing from double scattering (DS) to pencil beam scanning (PBS). PBS plans are more conformal, but contain more inhomogeneous beams and can be more sensitive to uncertainties. Since clinical experience is based on DS, quantifying robustness for the modalities is highly relevant for ensuring consistency with historical data and judging PBS plan quality.

We used Polynomial Chaos Expansion (PCE) to perform a comprehensive robustness comparison for a head-and-neck patient. We built appropriate PCEs for clinical PBS and DS plans, with the same assumed uncertainties: systematic and random positioning errors with 2 mm standard deviation, and systematic range errors with 2% standard deviation. To assess robustness a thousand realizations of the 28 fraction treatments were modeled. Systematic errors were first chosen randomly, then, with these fixed, 28 different random errors were sampled for all thousand simulations.

The maximum brainstem dose is lower with PBS, but has higher uncertainty than DS (range corresponding to 95% confidence level is 2Gy vs. 1Gy). For the left submandibular gland mean dose, PBS leads to higher dose and higher uncertainty (2.5 Gy vs. 2.1 Gy). For target coverage PBS outperforms DS: D95 is higher and uncertainties are similar (1.2Gy range for both). This example shows lower robustness of PBS plans, indicating potentially higher sensitivity of organ doses (without using robust optimization). Hence, for PBS plans with organ doses close to their tolerance care is advised, as uncertainties might be higher than historical experience.

Image Guidance and Adaptive Therapy

PTC17-0040: Clinical Use of Low-Tesla MR Simulation to Implement Surveillance MRI of Pediatric Cancer Patients Who Receive Proton Therapy at UFHPTI

S. Huh1, D. Indelicato1, M. Hall1, R. Rotondo1, M. Rutenberg1, Z. Li1, M. Hunter1, E. Kryck1

1UFHPTI, Radiation Oncology, Jacksonville, USA

Long scan times and patient motion are common barriers to effective MR simulation in pediatric radiation oncology. In an effort to overcome these obstacles, anesthesia or conscious sedation are often employed, but necessitate additional interventions and non-zero risk in this population. Our purpose is to reduce data acquisition time via target-specific scanning protocols and limiting patient motion with custom immobilization devices. Customized Magnetic Resonance (MR) imaging techniques using a 0.23T MR Simulator (Panorama, Phillips) with various immobilization devices were used to scan pediatric cancer patients with craniopharyngioma and low-grade glioma (LGG) who received proton therapy at our institution. Immobilization devices, including a universal head rest, an inflatable immobilization device, and an MR-safe headphones were fabricated to acquire clinically useful images for tumor monitoring in brain tumor patients between 2-20 years old. No anesthesia, conscious sedation, or contrast have been used. Two scanning protocols have been used: (a) regular scan with 1.5mm slice thickness, and (b) faster scans with 3mm slice thickness in both axial and sagittal planes for small Gross Tumor Volume (GTV). MR images are acquired weekly in craniopharygioma patients or at least once mid-treatment in LGG patients using the techniques described. The images are co-registered with the CT simulation and diagnostic MR images and reviewed by the physician for tumor progression and adjustment of target volumes in all patients.

PTC17-0046: Development of an Iterative Reconstruction Method for Low Dose CBCT in Proton Therapy Patient Positioning

T. Yamaguchi1

1Sumitomo Heavy Industries- Ltd., Technology Research Center, Yokosuka, Japan

Cone Beam Computed Tomography (CBCT) is used to determine patient positioning before each irradiation, which results in increased radiation exposure for the patient. Then, even if the X-ray tube current was reduced, we have developed an image reconstruction method employing an iterative algorithm that can achieve high image contrast. In a gantry-mounted CBCT, the X-ray tube and the flat panel detector (FPD) deviate some from their ideal position, resulting in a different geometry for each measurement angle. The proposed method calculates the system matrix for an ideal geometry in advance and performs positional deviation correction on the measured data, allowing us to obtain a reconstructed image with positional deviations taken into account.

We have investigated the relationship between the X-ray exposure dose and image quality (low contrast resolution). We treated the present measurement conditions as 100%, and calculated the projection data into which the dose was changed from 100% by the computation simulation. We reconstructed the calculated projection data by the proposed method and evaluated the images. The indices used for evaluation are the signal to noise ratio, standard deviation, and the detectability index.

PTC17-0048: Interfractional Variations of Prostate and Seminal Vesicle for the In-Room CT-Image Guided Proton Therapy with Lateral Beams for Prostate Cancer

Y. Maeda1, Y. Sato1, S. Shibata1, S. Bou1, H. Tamamura1, K. Yamamoto1, N. Fuwa 2, M. Sasaki1, S. Takamatsu3, K. Kume4

1Fukui Prefectural Hospital, Proton Therapy Center, Fukui, Japan 2Ise Red Cross Hospital, Radiotherapy, Mie, Japan3Kanazawa University Hospital, Radiotherapy, Ishikawa, Japan4The Wakasa Wan Energy Research Center, Research & Development, Fukui, Japan

Purpose: To quantify interfractional movement of prostate, seminal vesicle (SV) and rectum for prostate cancer treatment in the image-guided proton therapy with opposed lateral beams.

Materials and Methods: We have analyzed 375 sets of daily CT-images acquired throughout the treatment of ten patients undergoing proton therapy. The contours of prostate, SV and rectum were created and the daily movements of each anatomy were analyzed under three kinds of the image matching strategies, ie, bone matching (BM), prostate center matching (PCM) and prostate rectum side matching (PRSM). In BM, daily CT images were matched to the reference CT using bony structure. In PCM, the translational movements of the prostate center were corrected after BM. In PRSM, additionally to PCM, the correction along AP direction was performed matching the posterior wall of prostate in the rectum side. In each strategy, systematic and random errors were evaluated, and the margin values were estimated based on Van Herk formula.

Results: The estimated movement margins of prostate and tip portion of SV are displayed. Daily movements of SV were almost twice larger than that of prostate, and showed large variations over head-foot directions similar to the posterior wall of rectum. PRSM can reduce the overall margins compared with the value in BM or PCM.

Conclusion: The estimated margins for prostate and SV were restrained by PRSM compared with BM and PCM. PRSM by the CT-image guidance may be effective to reduce rectal complications for high risk prostate cancer treatment.

PTC17-0049: Study of Dosimetric Impact of Scanning Beam Delivery Parameters and Interplay Mitigation Strategies for Proton Therapy for Lung Cancer

H.L. Man1, R. Amos1,2, J. McClelland1,3

1University College London, Department of Medical Physics and Biomedical Engineering, London, United Kingdom 2University College London Hospitals NHS Foundation Trust, Department of Radiotherapy Physics, London, United Kingdom3University College London, Centre for Medical Imaging Computing, London, United Kingdom

Purpose: To evaluate the dosimetric impact of beam delivery parameters on proton pencil beam scanning (PBS) therapy for lung cancer, and to investigate strategies to mitigate interplay.

Materials and Methods: An ‘artefact free' 4DCT dataset was generated using a motion modelling approach developed at UCL, and a nominal plan based on this dataset was calculated using Eclipse (Varian Medical Systems, Palo Alto, CA). Dynamic doses were generated for a variety of delivery scenarios, including different dose rates, spot switching times, and breathing speeds, with spots being distributed across phases according to the temporal interplay between target position and beam delivery. Different rescanning strategies were evaluated as interplay mitigation techniques. Cumulative doses were determined for each scenario through deformation and summation, and benchmarked to the nominal plan.

Results: Compared to the nominal plan, D5-D95 increased from 8.28% to 11.20% and D1-D99 increased from 11.83% to 15.11% for the dynamic dose distribution. Volumetric rescanning decreased D5-D95 to 8.64% for two rescans, and 4.48% for three rescans. Considering factors such as scan speed and breathing cycle, the maximum dose within the body reached 120.44%.

Conclusion: Motion interplay decreases dose homogeneity for PBS-based proton therapy for lung cancer, however fast scanning delivery and multiple beams can mitigate most of these effects. Volumetric rescanning strategies without gantry angle repetition improve plan robustness further. The interplay effect may not be a significant concern for PBS-based proton therapy of lung tumours with appropriate delivery strategies.

PTC17-0064: Development of Metal Filtration to Improve Properties of CBCT

N. Kamiguchi1, T. Yamaguchi 2, T. Kaneko3, H. Miyanaga3, M. Yamada3, T. Kato3

1Sumitomo Heavy Industries- Ltd., Technical Reaearch Center, Yokosuka, Japan 2Sumitomo Heavy Industries- Ltd., Technical Research Center, Yokosuka, Japan3Sumitomo Heavy Industries- Ltd., Industrial Equipment Division, Niihama, Japan

Purpose: To maximize the effectiveness of proton therapy, it is important to perform patient positioning accurately because proton beam has fixed range. However, there are issues with frequent use of CBCT, particularly non-negligible exposure to X-rays. CBCT also suffers from beam hardening and scattering. Low energy photons below 30keV do not contribute to producing high quality x-ray images in our system.

Materials and Methods: Metal films are suitable for cutting low energy photons, but high energy photons are also attenuated. Additionally, inclusion of a filter requires increased mAs to compensate absorption, which may result in problematic. We conducted a survey of potential materials and their thickness at each tube voltage by transporting photons through metal filter and homogeneous cylinder phantom using Monte-Carlo simulations. Reconstruction of projected images was performed using the FDK method. Evaluated parameters were total amount of photon and improvement of the beam hardening effect.

Results: Appropriate materials have the following characteristics: (1) low Energy K-edge, (2) high density, and (3) workability. The relationship between a material's thickness and the amount of penetrated photon[HE6] was obtained. Cu is a good candidate for filtration film from a comprehensive viewpoint. A 1-2 mm copper sheet is enough to reduce beam hardening effect at 120 keV.

Conclusion: Metal filtration can be employed in CBCT systems to reduce low energy photon and to improve the beam hardening effect at each tube voltage. We were able to determine the optimal configuration of filter material and thickness.

PTC17-0073: Adequate Margins for Image-Guided Proton Therapy of Liver Tumors Using a Fiducial Marker

R. Muramatsu1, S. Hashimoto 2, K. Hayashi1, K. Tanaka1, K. Asai1, A. Shimomura1, T. Toshito3, H. Ogino 2, Y. Shibamoto4, J.E. Mizoe 2

1Nagoya Proton Therapy Center, Department of Proton Therapy Technology, Nagoya City, Japan 2Nagoya Proton Therapy Center, Department of Radiation Oncology, Nagoya City, Japan3Nagoya Proton Therapy Center, Department of Proton Therapy Physics, Nagoya City, Japan4Nagoya City University Graduate School of Medical Sciences, Department of Radiology, Nagoya, Japan

Purpose: Understanding characteristics of motion of the liver including internal motion variability is important for image-guided proton therapy (IGPT). The purpose of this study was to investigate characteristics of liver motion based on the movement of fiducial markers and the dosimetric impact for proton therapy.

Materials and Methods: This study analyzed characteristics of liver motion in 166 patients with liver tumors undergoing IGPT. Each patient had a fiducial marker percutaneously implanted around the gross tumor. The X-ray projection data of orthogonal views taken by Positioning Image Analysis System (Hitachi, Ltd, Tokyo, Japan) for each fraction and each patient were analyzed, and 3D motion trajectory of the fiducial marker was extracted. We analyzed the inter-fraction and intra-fraction liver motion variability and estimated adequate internal target volume margins with or without the fiducial marker. The dosimetric parameters of IGPT with the fiducial marker were compared with those of proton therapy without the marker.

Results: The adequate margins defined by Van Herk, et al. for left–right, anterior–posterior, and craniocaudal directions were 6.15, 5.58, and 11.3 mm, respectively, for the inter-fractional motion, and 1.69, 1.37, and 3.65 mm, respectively, for the intra-fractional motion. IGPT could reduce the irradiated volume of the organs at risk compared with proton therapy without the fiducial marker, while maintaining excellent PTV coverage.

Conclusion: To minimize the internal target volume margin and considering the dosimetric impact for proton therapy, fiducial markers should be placed for proton therapy with liver tumors.

PTC17-0083: Proton Radiation Therapy Accounting for Tumour Hypoxia - Development of a Treatment Planning Module Allowing for Treatment Individualization

A. Ureba1, J. Öden1,2, E. Lindblom1, J. Uhrdin 2, I. Toma-Dasu1,3

1Stockholm University, Medical Radiation Physics, Stockholm, Sweden 2RaySearch Laboratories AB, Research, Stockholm, Sweden3Karolinska Institutet, Department of Oncology and Pathology, Stockholm, Sweden

This study presents a Python module for the treatment planning system RayStation (RaySearch Laboratories, Stockholm, Sweden), developed to incorporate the information from PET hypoxia images into the proton treatment planning process, based on a previous work developed for photon planning [1].

The module converts voxel standard uptake values in the PET images taken using a dedicated hypoxia tracer (FMISO) into oxygen partial pressures (pO2) using either a linear or a nonlinear model. The prescribed dose required for achieving a predefined level of tumour control probability (TCP), taking the voxel pO2 into account, is obtained using a modification factor including a set of user- or default-parameters and the radio-sensitivity parameters for the linear quadratic model. The model for dose prescription also accounts for the dynamic character of tumour hypoxia. Thus, the dose prescription approach considers the heterogeneity and the dynamics of the radiation resistant regions, by means of the correlation between functional imaging data and hypoxia distribution. The module also generates and allows visualization of the pO2 maps in DICOM format.

The module is not only limited to hypoxia targeting, but could account for volume-based cell density distribution in the calculation of TCP values, thus allowing further degrees of freedom for the early identification of the clinical cases which might benefit from dose escalation.

From this perspective, the treatment planning module is a versatile tool that allows personalized treatments accounting for negative predictive factors.

[1] Toma-Dasu, I., et al., 2012. Dose prescription and treatment planning based on FMISO-PET hypoxia. Acta Oncologica, 51, 222-230.

PTC17-0118: Proton Range Verification Using the Discrete Range Modulation Method

T.L. Tsai1, Y.C. Tsai 2, C.C. Lee1,2,3, T.C. Chao1,2

1Chang Gung University, Medical Imaging and Radiological Sciences, Taoyuan, Taiwan- Province of China 2Chang Gung University/Chang Gung Memorial Hospital, Dose Assessment Core Laboratory of Institute for Radiological Research, Taoyuan, Taiwan- Province of China3Chang Gung Memorial Hospital, Department of Radiation Oncology, Taoyuan, Taiwan- Province of China

Purpose: This work has investigated the properties of proton radiography and water-equivalent path length (WEPL) to quantitatively assess the image quality and range uncertainties. We use the Discrete Range Modulation (DRM) method which was developed using layer stacking SOBP techniques. We will report advantages of DRM method over the Continuous Range Modulation (CRM) method.

Materials and Methods: In this study, we used MCNPX 2.7.0 to simulate depth dose distribution by pencil beam geometry (PBG) in a water phantom of 40 × 40 x 40 cm 2. Proton beam energies from 70 to 230 MeV at 2.5MeV/step were used. In DRM method, first step, we established the correlation between R80 (80% dose range of a beam at distal fall-off region) and proton energy. Second step, DRM method to determine the WEPL is by using the relationship between the proton energy and energy deposition at 80% of the proximal fall-off (E80) of the dose stacking distribution in water. For CRM method, a dose gradient plan was constructed to estimate relationship between WEPL and energy deposition. DRM and CRM were verified by different thickness of water phantoms.

Results: In DRM method, the difference of WEPL are < 1 mm, noise in the images was found to be σ <0.00075 mm (standard deviation). However, in CRM, difference of WEPL are >1mm, σ> 0.8 mm. SNR of DRM is much better than that of CRM.

Conclusion: Proton radiography obtained using DRM method is more accurate and precise than those using CRM method.

PTC17-0168: Dosimetric Characterization of Carbon-Ion Beam-Grid Irradiations: A Monte Carlo Study

T. Tsubouchi1, T. Henry 2, A. Ureba 2, A. Siegbahn 2, N. Bassler 2

1Osaka University, Department of Radiation Oncology, Suita-city, Japan 2Stockholm University, Medical Radiation Physics- Department of Physics, Stockholm, Sweden

Purpose: Research on grid therapy using arrays of milli- and micro-meter wide beams have been carried out in pre-clinical trials. The beam width and the center-to-center (c-t-c) separation of the beams are basic parameters which need to be determined for the grid irradiations. We investigated the relationship between these grid parameters and the peak-dose to valley-dose ratio (PVDR) variable for grids containing narrow carbon-ion beams and evaluated the potential therapeutic usefulness of crossfiring such grids over a target volume.

Materials and Methods: The dose distribution in water, produced by a single carbon-ion beam, was simulated with the PHITS Monte Carlo code. The incident carbon-ion beams were assumed to be of rectangular shape, unidirectional, of height 2.0 cm, and of widths varying from 0.1 to 3.0 mm. The spatial distribution within the incident beam had a Gaussian shape and the spread in kinetic energy was 1.0%.

Results: The simulated carbon-ion beams remained narrow down to the depths of the Bragg-peaks. The 0.5- and 3.0-mm-wide beams were selected for further studies based on the physical characteristics of the produced dose distributions, expected biological effects and the relative technical ease of treatment delivery. The optimal separation of the beams for each grid setup was determined with an objective function, using simulation results as variables.

Conclusion: A homogeneous dose distribution inside the target could be combined with high peak-to-valley dose ratios in the normal-tissue, when crossfiring beam grids, if the c-t-c distance was chosen optimally.

PTC17-0188: Using CBCT Projections to Triage Head and Neck Patients for Adaptive Proton Therapy

B. Winey1, J. Kim1, G. Sharp1

1MGH, Radiation Oncology, Boston, USA

Purpose: To develop a workflow that utilizes CBCT projections to triage head and neck (H&N) patients for adaptive proton therapy planning.

Materials and Methods: CBCT images require processing to reduce scatter effects to obtain an image useful for proton dose calculations but the anatomic variations observed in H&N patient populations might be visible in the projections. Raw CBCT projections of H&N patients, sampled on average every week, were acquired during patient positioning scans. The first CBCT was assumed to be the baseline for future comparisons. The images were cropped and renormalized to focus on the neck where the anatomic variation of H&N patients is often visualized. The differences between the projections of the first scan and subsequent scans were analyzed using histograms, raw intensities, localized standard deviation, local range variation, and entropy.

Results: The sum of the raw intensities in the difference images was found to be most sensitive to the anatomic variations at the edge of the patient volume. We display the histograms, the cropped, normalized projection images, and the metric evaluations. We display the sum of the intensity differences versus projection angle for two fractions, one near the beginning of treatment and one near the conclusion of treatment. The effects of the missing tissue can be visualized in the systematic decrease of the signal from fraction 24.

Conclusion: CBCT projections potentially capture anatomic variations of head and neck patients and can be used to quickly triage patients for adaptive proton therapy.

PTC17-0199: Study of Polarization-Based PET Coincidence Discrimination for Range Verification in Proton Therapy

A. Makarova1, H. Shirato1, K. Umegaki1,2, L. Xing3, H. Peng3

1Hokkaido University, Global Station for Quantum Medical Science and Engineering GSQ- Global Institution for Collaborative Research and Education GI-CoRE, Sapporo, Japan 2Hokkaido University, Division of Quantum Science and Engineering- Faculty of Engineering, Sapporo, Japan3Stanford University, Department of Radiation Oncology- Stanford University School of Medicine, Stanford, USA

One challenge of PET imaging is its limited photon detection sensitivity. The dependence of Compton scattering angle on the orthogonal polarization of two annihilation gammas can be utilized to enhance the sensitivity and thus increase image signal-to-noise ratio (SNR) up to 22% [1].

In this study we investigated potential use of this approach for the range verification in proton therapy with Geant4 modelling. The G4PolarizedComptonModel was selected as the physics model which uses Stokes vector for polarization. We present the correlation of azimuthal Compton scattering angles of annihilation gammas with polarization. The concentration of positron emitters was recorded in each voxel for a pencil beam of protons hitting the water phantom. The distribution was converted into activities for each voxel and each type of isotope, which were then used as the “voxel source” for PET procedure simulation.

Our preliminary results indicate that the polar scattering angle range should be centered around 80° and the difference between azimuthal angles centered around 90° with full width at half maximum (FWHM) equal 20% to maximize the correlation, which is in good agreement with [1]. The two 3D reconstruction images with and without polarization information will be reconstructed and compared.

Reference: [1] Toghyani, M., et al. “Polarisation-based coincidence event discrimination: an in silico study towards a feasible scheme for Compton-PET.” Physics in Medicine and Biology 61(15), p. 5803 — 5817.

PTC17-0218: On-Line PET Imaging Studies with Radioactive Ion Beams for Treatment Verification in Hadrontherapy Using FLUKA

R. Augusto1,2, J. Bauer3, A. Ferrari1, C. Gianoli 2, P.G. Ortega1,4, K. Parodi 2,5, W. Kozlowska1,6, T. Tessonnier 2,3, V. Vlachoudis1

1CERN, EN-STI, Geneva, Switzerland 2Ludwig-Maximilians-Universitat Munchen, Department of Experimental Physics – Medical Physics, Munich, Germany3Uniklinikum Heidelberg, Department of Radiation Oncology, Heidelberg, Germany4Instituto de Fisica Corpuscular- Centro Mixto CSIC-Universidad de Valencia, Department of Experimental Physics, Valencia, Spain5Heidelberg University Hospital, Heidelberger Ionenstrahl-Therapiezentrum HIT, Heidelberg, Germany6Medical University of Vienna, Department of Medical Physics, Vienna, Austria

In the context of hadrontherapy, Positron-Emission-Tomography (PET) imaging is used to check the accuracy of the delivered treatment by comparing the activity distribution induced by irradiation with a Monte Carlo simulation of the treatment delivery. In clinical practice, information on the in-vivo beam range can be of paramount importance, for it provides confirmation that the tumor was targeted effectively. Compared to in-room PET or off-line combined PET-CT (Computed Tomography) imaging, the possibility of an on-line PET monitoring is therefore beneficial, for it entails not only ideally immediate knowledge of the irradiated zone, but it suppresses the need of additional acquisition time afterwards as well. In this study, we assessed how radioactive ion beams of C-11 and O-15 compare to C-12 and O-16 beams with respect to the on-line imaging signal output, for an equivalent dose delivery plan. In particular, the Monte Carlo particle transport code FLUKA was used to simulate both mono-energetic and spread-out-Bragg-peaks in water and an anthropomorphic head phantom, including a realistic modeling of a PET scanner response. The signal to noise ratio (SNR) of the reconstructed PET image was also assessed. All the results support the hypothesis that radioactive beams outperform the corresponding stable counterparts in terms of imaging signal output[pic1], with equivalent[pic2] dose delivery. Future work will seek for experimental possibilities to confirm the simulation results in collaboration with colleagues from the National Institute of Radiological Sciences, which is one of the few places in the world able to accelerate radioactive beams up to therapeutic energies.

PTC17-0221: Contrast Optimization in 4D-MRI as an Essential Step Towards MR-Guided Particle Therapy of Pancreatic Cancer

K. Dolde1, F. Maier 2, M.T. Freitag3, P. Naumann4, N. Saito1, A. Pfaffenberger1

1German Cancer Research Center DKFZ, Medical Physics in Radiation Oncology, Heidelberg, Germany 2German Cancer Research Center DKFZ, Medical Physics in Radiology, Heidelberg, Germany3German Cancer Research Center DKFZ, Department of Radiology, Heidelberg, Germany4University Clinic Heidelberg, Department of Radiation Oncology, Heidelberg, Germany

Time-resolved volumetric Magnetic Resonance Imaging (4D-MRI) shows promising potential for application in MR-guided Particle Therapy (MRgPT). To fully employ the information on organ motion and deformation, high image quality and sufficient contrast between organs are required. The purpose of this study was to improve the generally low contrast between pancreas and surrounding organs to fully enable employment of 4D-MRI information in 4D treatment planning for MRgPT of pancreatic cancer.

Images of 4D-MRI recorded with a 3D-encoded gradient echo radial vibe sequence were acquired from volunteers to optimize repetition time (3.3-4.3 ms) and flip angle (6°-12°) for several echo times with respect to the Ernst angle to achieve high signal intensity. 4D-MRI images were reconstructed with a motion-compensated algorithm and image quality was scored by contrast calculations and visual inspection by a radiologist.

The highest signal intensity of the pancreas was obtained when matching the flip angle to the Ernst angle. However, this configuration did not yield the best results in terms of contrast to neighboring organs, which was independently confirmed by visual inspection. We were able to enhance the average relative contrast of pancreas by 50% from 0.175 to 0.262, in particular, the contrast to bowel (+135%) and kidney (+61%) were improved by increasing the flip angle.

From both visual inspection and contrast calculations a 4D-MRI parameter configuration close to the Ernst-angle enables a superior differentiation of the pancreas compared with imaging at the exact Ernst angle, allowing a more precise target contouring which is essential in particle treatment.

PTC17-0236: Evaluation of Prompt Gamma Energy and Angular Relationship to Optimize Compton Camera for Proton Therapy

P.Y. Liao1, M.L. Jan 2, I.T. Hsiao3

1Chang Gung University, Medical Imaging and Radiological Sciences, Taoyuan, Taiwan- Province of China 2lnstitute for Radiological Research- Chang Gung Memorial Hospital, Medical Imaging, Taoyuan, Taiwan- Province of China3Chang Gung Memorial Hospital, Nuclear Medicine, Taoyuan, Taiwan- Province of China

Hadron therapy has advantage for releasing most dosage within a finite range called “Bragg peak”. Due to its higher energy, range verification is important for accurate dose delivery in proton therapy. Prompt gamma can be detected for in-vivo range verification in proton therapy using the Compton camera.

We have previously found energy of a prompt gamma depends on its leaving angle out of the object. We would like to investigate the relationship between the energy and the leaving angle of prompt gamma from the object in a multi-stage Compton camera. The detector arrangement for optimal geometrical detection efficiency was also explored from the use of LYSO detector material. Three-stage Compton camera for detecting high energy prompt gamma with an irradiation of 160MeV proton pencil beam was modeled on a cylindrical PMMA phantom with a radius of 7.5cm and height of 20 cm. The GATE Monte Carlo simulation software based on Geant4 toolkit was applied.

The preliminary results showed the energy of detected prompt gamma is obviously depended on the detector position where the energy deviation can be up to approximately 20% for detectors in the same distance from phantom. Therefore, with a limited detector resources, detector position plays an important role in the optimal design of a Compton camera. To reduce detector size and increase detection sensitivity, we will further explore an optimal design of a two or three stage Compton camera for the range verification in the proton therapy in the future work based on the current study results.

PTC17-0251: Study of Compton Camera Designs Using Figures of Merit to Optimize Detector Configurations for Maximum Prompt-Gamma Detection Sensitivity

G.T. Liu1, M.L. Jan 2,3, S.H. Wang4, H.H. Lin1, Y.C. Ni4, K.S. Chuang1, J.H. Hong3

1National Tsing Hua University, Department of Biomedical Engineering and Environmental Sciences, Hsinchu, Taiwan- Province of China 2Chang Gung University/Chang Gung Memorial Hospital, Institute for Radiological Research, Kwei-Shan- Taoyuan, Taiwan- Province of China3Chang Gung Memorial Hospital, Department of Radiation Oncology, Taoyuan, Taiwan- Province of China4Institute of Nuclear Energy Research, Division of Health Physics, Taoyuan, Taiwan- Province of China

Compton camera has been proposed as a method for range verification by measuring prompt gammas induced by the proton beam in a patient. The feasibility of prompt-gamma Compton camera (PGCC) used in clinically realistic scenarios depends greatly on the design of highly efficient detectors. A new PGCC to perform high counting statistics is currently under development by our group. The scintillation-based PGCC consists of a scatterer with novel 3D side-on design, and an absorber with depth-of-interaction configuration. The aim of this study is to optimize the PGCC detectors by using figures of merit (FoMs) for maximum MeV-prompt-gamma detection sensitivity. In this work, all the dataset from the production of prompt-gamma spectra emitted from a water phantom to the event classification, was simulated using GATE/GEANT4. Four materials of LYSO, LaBr3, GAGG and CZT (for reference) in various thickness were evaluated. The SFoM (FoM of scatterer) as a function of crystal thickness was calculated based on single Compton fraction, single Compton detection efficiency, energy resolution and energy-band weightings for each material. The AFoM (FoM of absorber) as a function of crystal thickness based on energy-absorption ratio, photon-penetration depth and position mislocation were analyzed. The results of SFoM show that the GAGG-based scatterer with new designed configurations presents the highest true-event detectability. The AFoMs indicate that the depth-of-interaction absorber to offer high precision of Compton-cone-axis identification is important. More details will be presented. It is concluded that the FoMs we proposed can provide as metrics for PGCC design and potential performances evaluation.

PTC17-0264: Toward Heavy Ion Computed Tomography with Carbon Ions: A Monte Carlo Study

D. Shrestha1, N. Qin1, Y. Zhang1, F. Kalantari1, A. Pompos1, S. Jiang1, X. Jia1, J. Wang1

1UT Southwestern Medical Center, Radiation Oncology, Dallas, USA

As radiation treatment with heavy ions holds great promise in improving treatment outcomes for various cancers, enhancing the accuracy of range prediction of these ions inside the patient's body has become very important. An accurate localization of the Bragg peak provides greater conformity of the dose to tumors while sparing healthy tissues. Among several methods to create a stopping power map of a patient body, there exists a unique method to use the same heavy ions that are used for treatment, for relative stopping power calculation as well as patient alignment. We present a Monte Carlo simulation study using the Geant4 simulation toolkit and a modified reconstruction algorithm based upon the Algebraic Reconstruction Technique with Total Variation minimization (ART-TV) to obtain carbon Computed Tomographic (cCT) images using carbon ions (C-12) of energy 430 MeV/u. Our work includes a detailed analysis of propagation of the ions through a water phantom and reconstructing a 3D relative electron density map of a human head size phantom containing structures of different resolutions and contrasts. We have been able to obtain cCT images with spatial resolution of 6 line pairs/cm with a dose of 3mGy per scan to the center of the phantom under a straight line approximation for the path traveled by the carbon ions inside the phantom.

PTC17-0280: Plug'n'play Proton Radiography with Commercial QA Equipment - a First Hand Practical How-To

N. Krah1, L. De Marzi 2, A. Patriarca 2, G. Pittà3, I. Rinaldi1

1CNRS/IN2P3 and Lyon 1 University- UMR 5822, IPNL, Villeurbanne, France 2Institut Curie - Proton Therapy Center, IC-CPO, Orsay, France3DE.TEC.TOR. Devices & Technologies Torino S.r.l., DE.TEC.TOR., Torino, Italy

Proton imaging has long been proposed as alternative or complementary imaging modality in ion beam therapy offering a direct probe of the relative stopping power. However, it is yet to see an integration into the clinical workflow. Several groups around the world are working on the technical implementation of proton radiography/tomography set-ups based on more or less complex hardware designs. Many of these set-ups come along with prerequisites for the treatment facility: coupling with trigger output from the accelerator, unusually low particle fluence, additional integration of dedicated hardware. Furthermore, many set-ups rely on non-certified prototype hardware components. These factors limit the immediate accessibility of proton imaging in ion beam therapy facilities. In this contribution, we will report about a recent series of experiments conducted at the Institut Curie - Proton Therapy Center in Orsay, France. We acquired proton radiographies of various phantoms, using a commercial range telescope available in the facility. The detector was operated in free running mode and standard beam parameters were used, rendering the set-up plug'n'play. We will further elucidate the dedicated data and image processing involved to produce proton radiographies from the raw data and illustrate the good image quality which can be reached even with such a comparatively simple set-up. The purpose of this contribution is to share a first hand practical how-to and to encourage the idea that proton radiographic imaging can be realized with reasonable effort employing hardware already available in ion beam therapy facilities for quality assurance and/or positioning applications.

PTC17-0301: Recotom©: A New CBCT Reconstruction Algorithm Aiming to Mitigate Typical CBCT Artifacts, Such as Jittering, Patient Movement & Metal Artifacts

G. Sakas1, T. Braun 2, C. Meffert1

1MedCom GmbH, CEO, Darmstadt, Germany 2MedCom GmbH, Product Manager, Darmstadt, Germany

We present a new reconstruction algorithm for cone beam computed tomography (CBCT) called Recotom©. It offers several correction algorithms aiming to mitigate typical CBCT artifacts, such as jittering, patient movement & metal artifacts:

  • Misalignment artifacts are caused by inaccurate geometrical projection parameters due to mechanical inaccuracies of the scanner. Unlike conventional methods, Recotom© can correct these artifacts on-the-fly based only on the input X-ray images, i.e. without requiring any additional hardware or fiducial markers

  • Patient motion during image acquisition can also be compensated by the reconstruction algorithm. This correction results in reduced blurring and streaking artifacts.

  • The presence of high-density objects induces metal artifacts, which degrade reconstruction quality or even may obscure relevant anatomy. Such artifacts could be de-facto eliminated with Recotom©, i.e. details lost in the corrupted regions are recovered.

Recotom© has been successfully integrated in VeriSuite© 2.2 and is clinically used in Samsung Medical Center in Korea and Sapporo Teishinkai Hospital in Japan. Together with the dynamic gain functionality of the 4030d panel, soft tissue contrast could be highly improved. This enables VeriSuite© perform do a high accurate 3D/3D registration with the planning CT volume. The reconstruction result of a volunteer can be seen in the picture below.

PTC17-0307: Proton CT Using a Fluoroscopy Flat Panel Designed for X-Ray Imaging

R. Zhang1, K.W. Jee1, G. Sharp1, E. Cascio1, C. Finley1, J. Flanz1, H.M. Lu1

1Massachusetts General Hospital, Radiation Oncology, Boston, USA

Proton CT (pCT) is a promising approach to reducing range uncertainties induced by the X-ray CT HU to RSP conversion. This study aimed to realize pCT by proton radiography (pRG) with a flat panel and optimize the RSP for individual voxels. An amorphous silicon flat panel was placed behind the phantom to measure the dose delivered in time, termed as dose rate function (DRF), from a proton beam modulated by the modulator wheel. By rotating the phantom on the rotary stage, data were acquired at projection angles from 0 to 360 degrees with an increment of 2 degrees. For each projection angle, the WEPL image was retrieved from calibration models based on quasi-unique statistical features of the DRF for different WEPLs. X-ray CT scan of the phantom (xCT) was acquired and co-registered with the proton CT acquisition coordinates. RSPs were optimized iteratively by minimizing the difference between measured WEPL image and calculated WEPL image from ray tracing in xCT with RSPs converted from the HU. Pixels in measured WEPL images with severe proton range mixings were eliminated for the optimization. Initial results show that RSPs were iteratively optimized to give better agreement between the WEPL by ray tracing and by measurement. For the Gammex phantom inserts, good agreement was achieved between optimized RSPs and reference values. This study proved the concept of pCT by a single flat panel and an interactive method to optimize RSPs in individual voxels.

PTC17-0308: A Robust Reconstruction Algorithm for Low-Fluence Proton Radiography and CT

T. Plautz1, V. Bashkirov 2, C. Ordonez3, R. Johnson1, H. Sadrozinski1, R. Schulte 2

1University of California Santa Cruz, Santa Cruz Institute of Particle Physics, Santa Cruz, USA 2Loma Linda University, Basic Sciences, Loma Linda, USA3Northern Illinois University, Physics, DeKalb, USA

Purpose: To describe a novel algorithm for particle radiography and CT and evaluate its performance under conditions of very low fluences and dose.

Materials and Methods: An algorithm based on the most-likely path (MLP) concept was developed. It estimates the column voxel sums of water equivalent thickness (WET) of a discretized object using tracking and water equivalent path length (WEPL) data of individual particles. The algorithm was optimized and tested by using proton data provided by a prototype particle computed tomography (pCT) scanner. Proton data (200 MeV) of a homogeneous polystyrene step phantom and a pediatric head phantom were acquired, MLPs were reconstructed and WEPL voxel histograms of all protons intersecting a given voxel were summed along individual columns, renormalized, and their mode was calculated using a robust mode estimation method. The individual modes of voxel columns formed a pixelated radiograph for a given projection. For CT reconstruction, projection images were combined for 3D CT-reconstruction using a standard filtered back projection algorithm.

Results: The algorithm was found to produce high-quality radiographs even for very low fluences of 50 protons/ mm 2-pixel. Additional testing of fast FBP-based reconstruction using projections obtained with this algorithm is underway using quality assurance phantoms of the Catphan 600 series and its performance will be compared to previously established proton CT reconstruction algorithms.

Conclusion: A new algorithm has been developed using the mode of WEPL column histograms that promises improved performance and efficiency of particle radiography and CT at very low particle fluences.

PTC17-0334: Geometric Accuracy of a Couch Mounted Patient Position Verification Device

A. Ableitinger1, A. Utz1, A. Zechner1, M. Stock1, H. Deutschmann 2, P. Steininger 2, V. Letellier1

1EBG MedAustron, Medical Physics, Wiener Neustadt, Austria 2Institute for Research and Development on Advanced Radiation Technologies radART, Paracelsus Medical University, Salzburg, Austria

Purpose: The aim for patient alignment systems is to provide a very high geometrical accuracy and precision throughout the whole treatment volume. Compared to room based X-Ray systems this is even more crucial for couch mounted devices since additional alignment of the couch top with the beamline adds to the overall positioning accuracy. In this study the geometrical accuracy of a couch mounted imaging device is evaluated with respect to the couch coordinate system (CCS).

Materials and Methods: The positions of reflector probes were determined in CBCT and planar datasets acquired with the imaging system and compared to their coordinates measured with a Lasertracker in the CCS. Three different reflector setups and 16 different imaging positions/modalities led to an evaluation of 1672 reflector points. The 3D Euclidian distance is reported for CBCTs. For planar acquisitions the closest distance of the reflector to the projection line -probe in the image plane towards the x-ray source- is evaluated.

Results: The minimum, maximum, mean deviation and the percentile 0.50, 0.95 and 0.99 are listed. For the CBCT imageset the mean deviation was about 0.3mm and 0.2 mm for planar images. This difference can be explained by evaluation based on back-projection of the planar dataset. The overall accuracy for 99 percent of all measured reflector positions was better than 0.44 mm.

Conclusion: The study showed no obvious differences between the 2D and 3D imaging modality. This couch mounted imaging device provides an absolute position accuracy of better than 0.5 mm.

PTC17-0337: CT-On-Rails Commissioning for Image Guidance in a Half-Gantry Treatment Room

J. Johnson1, A. Deisher1, K. Furutani1, J. Ma1, M. Zakhary1, D. Mundy1, M. Herman1, J. Kruse1

1Mayo Clinic, Radiation Oncology, Rochester, USA

Purpose: A CT-on-rails system has been installed in a treatment room of our proton center.

Materials and Methods: The proton treatment room consists of a half-gantry-mounted nozzle and a six-degree-of-freedom robotic couch. The couch is capable of translating the patient approximately 1.5 meters from treatment isocenter to the imaging isocenter of a sliding CT gantry for volumetric image guidance. The system was commissioned both for volumetric treatment localization as well as for radiotherapy simulation according to the applicable AAPM TG-66 recommendations.

Results: Localization accuracy is sensitive to small differences in calibrated couch angles between the CT position and treatment isocenter. These differences can be due to either inconsistent couch corrections under applied weight or slight angular deviations between the couch and CT rails. Phantom localization accuracy is better than 1 mm over a full range of isocenter locations and weight loads on the couch. Rapid helical image acquisition allows for image guidance of breath held or respiratory guided treatments.

Conclusion: Our CT-on-rails system has been commissioned to provide accurate volumetric localization across a wide range of treatment scenarios. High quality helical CT image sets acquired during treatment sessions may be used for adaptive planning.

PTC17-0350: A Feasibility Study on Patient Positioning with Proton Range Image and Proton Range Contour

W. Yao1, T. Merchant1, M. Krasin1

1St. Jude Children's Research Hospital, Radiation Oncology, Memphis, USA

Purpose: Registration of image data between the planning computed tomography (CT) and cone beam CT (CBCT) of the patient on the couch by using the CT numbers is widely used in patient positioning. The registration goal in photon therapy is to match the tumor targets and surrounding tissues, but this goal is insufficient for proton therapy, where the target coverage depends on the proton range along the proton beam path, not just the matching of tumor targets and surrounding tissues. Therefore, we propose the proton range-based registration, in addition to CT number–based registration, to improve patient positioning in proton therapy.

Materials and Methods: The range of a mono-energetic proton beam in the patient body forms a two-dimensional (2D) image on the beam of view (BOV). This 2D image was calculated from the daily CBCT and was registered to the corresponding 2D range image calculated from the CBCT acquired on the first day of treatment. The water equivalent thickness along specific contours in each beam direction was calculated to estimate the range deviation.

Results: The registration results from the (volumetric) CBCT number based registration and the range image based registration were within 1.0±3.0 mm and 0.5±0.7 degree for the pelvis case, and within 0.6±2.2 mm and 0.1±0.4 degrees for the brain case. The root-mean-square-difference of the range deviation along CTV was 2.8±0.6 mm and 2.4±0.4 mm for the investigated pelvis and brain cases respectively.

Conclusion: Range image based registration introduces new, important dimensions to improve patient positioning and range verification before beam is delivered.

PTC17-0351: MRI-Guided Proton Spot Scanning with Simultaneous Integrated Boost for High-Risk Prostate Cancer

M. Moteabbed1, M. Harisinghani 2, A.V. Trofimov1, G.C. Sharp1, J.A. Efstathiou1, H.M. Lu1

1Massachusetts General Hospital, Radiation Oncology, Boston, USA 2Massachusetts General Hospital, Radiology, Boston, USA

Purpose: We explore the dosimetric feasibility and the potential for advancing tumor control in MRI-guided proton therapy with simultaneous integrated boost (SIB) to dominant intraprostatic lesions for high-risk prostate cancer.

Materials and Methods: Diagnostic multiparametric-MRI scans were retrospectively used to delineate the intraprostatic lesions for six high-risk prostate cancer patients who had received conventional radiotherapy. Delineation was based on the hypointense appearance of lesions on T2-weighted-MRI and apparent-diffusion-coefficient (ADC) maps. Contours were fused onto the planning CT, and spot-scanning proton therapy plans without and with lesion SIB were created. Doses prescribed to seminal vesicles, prostate and lesion were 51.2, 70.4 and 83.2 Gy(RBE) in 32 fractions, respectively. Dose-volume-histograms were analyzed in terms of 2Gy-fraction-equivalent dose (EQD2), assuming α/β=1.5 (prostate/lesion), =3 (rectum/bladder).

Results: The average prostate/lesion mean EQD2 increased by 18.5/34.2% for SIB compared to non-boosted plans. For these volumes the average D98 increased by 13.7/29.3%. Although in most cases lesions were in close proximity to organs at risk (OARs), the average increase in OAR mean dose was within 2 Gy(EQD2). The average increase in bladder/rectum D2 was more prominent at 6.7 Gy(EQD2), but remained below the clinical tolerance of 80 Gy(EQD2) in all cases. Tumor control probability (TCP) of the lesion was significantly increased (∼60%) for SIB plans. Normal tissue complication probability (NTCP) slightly increased but remained below 0.7% for both OARs.

Conclusion: Proton therapy with SIB to intraprostatic lesions is feasible up to mean dose of ∼100 Gy(EQD2), which could significantly enhance the local control for high-risk prostate cancer.

PTC17-0354: Integrating a Portable CT to a Single Room Proton Machine

B. Sun1, T. Zhang1, M. Goddu1, L. Santanam1, J. Bradley1, S. Mutic1, T. Zhao1

1Washington Univ in St. Louis, Department of Radiation Oncology, Saint Louis, USA

Purpose: To integrate a portable CT scanner into a compact proton unit for 3D volumetric imaging guided proton therapy and adaptive planning with superior CT imaging quality.

Materials and Method: The portable CT scanner was placed parallel to the setup position of the 6D robotic couch, which required a 90 degree rotation of the couch between the setup and imagining position. A rigid arm with 4 radio-opaque fiducials monitored by a ceiling-mounted camera was attached to the couch to maintain the rigid transformation between the couch and the room coordinates within 0.5mm. Images acquired with the CT scanner were used for patient setup and retrospective verification of the original plan with true fractional dose calculated on volumetric setup image. Physician reviewed the verification plan and ordered new plan if necessary.

Results: The portable CT passed successfully our test on image quality and geometric accuracy within 1mm in all three directions, including the direction of CT scanner movement. It was capable of localizing target with accuracy better than 1mm/1 degree including all uncertainties from the couch, CT scanner and machine isocentricity. The whole imaging and localization process took about 20 minutes. A stoichiometric CT calibration curve was commissioned in our treatment planning system for the calculation of true fractional dose on images from the CT scanner.

Conclusion: The use of portable CT scanner in proton therapy enhances our capability to localize tumors in soft tissue and provides additional benefits for off-line plan adaption with superior CT imaging quality.

PTC17-0365: Improvement of the Positional Reproducibility of Prostate Cancer by a Newly Setup Procedure

K. Shioiri1, W. Maehana1, S. Hirai1, S. Yoshino1, S. Minohara 2, Y. Kusano 2, Y. Tokiya1, S. Akita1, M. Ogura1, S. Ide1

1Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan 2Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan

Purpose: By our previous observations of inter-fractional target motion using in-room CT images of prostate cancer, the target in some patients has tended to shift to vertical direction although the location of the bones were in a good reproducibility. This systematic error seems to arise from the patient tension of muscles around buttock. So we tried to improve the procedure of immobilization and planning CT acquisition. We report on the effect of these improvements.

Materials and Methods: We compared the target position in two groups. Group A is eleven patients that are immobilized by our conventional method. Group B is twelve patients that are immobilized by our new procedure with taking several minutes to relax after the patient setup. Each patient has taken the CT images before the irradiation at each fractionation, and these CT images were analyzed for the systematic error and standard-deviation (SD) in organ motion.

Results: The means of systematic errors of group A and B were -0.9 and 0.2 mm in the vertical direction. Similarly, the means of the SDs of the groups were 1.4 and 1.2 mm, and the root-mean-square of the individual SDs of the groups were 1.3 and 1.0 mm, respectively. These results showed that there was statistically significant difference between two groups in terms of systematic errors (P<0.05).

Conclusion: The mean of systematic error of vertical direction was significantly improved, and random error became clearly small. These results suggest the relaxation of muscle would improve the accuracy and reproducibility of the patient positioning.

PTC17-0367: H&N Patients Undergoing Proton Therapy at PSI: A Preliminary Investigation Establishing a Control CT Scanning Protocol Identifying Changes during Treatment

L. Placidi1, M. Kountouri1, A. Bolsi1, M. Walser1, R. Schneider1, A.J. Lomax1,2, D.C. Weber1,3

1Paul Scherrer Institute, Centre for Proton therapy, Villigen PSI, Switzerland 2ETH Zurich, Department of Physics, Zurich, Switzerland3University Hospital of Zurich, Radiation Oncology Department, Zurich, Switzerland

Purpose: Evaluation of a protocol for control CT acquisition frequency during proton PBS therapy that detects potential anatomical changes and dosimetric effects for optimal clinical management of H&N patients.

Materials and Methods: Since March 2016, five randomly selected adult patients were entered in a pilot control CT protocol. All patients were treated in the H&N area and/or for a tumour involving the paranasal sinuses or had a beam arrangement that passed through these structures. They underwent control CTs every second week and, in case of sinus(es) involvement, also weekly control slices. Patients' weight was recorded weekly. For each repeated CT (or slices) the nominal plan was recalculated on the newly detected anatomy. Dose metrics (targets V95% and organs at risk maximum/ mean dose) and clinical parameters (such as patient weight variation, tissue and nasal cavity filling variation (cc)) were evaluated. These were compared with the results of five control patients.

Results: Of 43 recalculated plans, 55% show at least one dose metric with difference >5% if compared to the nominal plan, except for the target coverage (V95%). Typically, in presence of nasal cavity variations the target V95% is reduced if the cavity is filled, while there is an increase in maximum dose when the cavity is empty. All these differences were however within 5% and, due to the small sample size, it is not feasible to correlate them with clinical parameters.

Conclusion: Clinical parameters, such as nasal cavity variations and weight, can potentially be predictive factors of changes in the dose metrics.

PTC17-0370: Intrafraction Prostate Motion Comparison Between Endorectal Balloon and Rectal Spacer Hydrogel

M. Fagundes1, S. Hedrick 2, B. Robison 2, M. Blakey 2, J. Renegar 2, M. Artz 2, B. Wilkinson 2, N. Schreuder 2

1Miami Cancer Institute, Radiation Oncology, Miami, USA 2Provision Center for Proton Therapy, Radiation Oncology, Knoxville, USA

Purpose: Endorectal balloons (ERB) are often used to mitigate prostate motion during proton therapy. Rectal spacer hydrogel (GEL) was recently introduced to lower rectal dose, without ERB. We evaluated the effect of GEL on intrafraction prostate motion compared to ERB.

Materials and Methods: 10 ERB and 16 GEL patients (1502 fields) were analyzed. Intraprostatic fiducial markers were used for image-guidance, employing orthogonal x-rays. Fiducials had a 0.2 cm expansion for alignment tolerance. Pre- and post-field x-rays were obtained; X, Y, and Z shifts necessary to return the fiducials to within the expansions were recorded and a 3D vector was calculated. Time between pre- and post-field images was also recorded.

Results: For shifts <0.3 cm, the mean vector shift of 0.06 cm for ERB was significantly smaller than 0.09 cm for GEL. For shifts >0.3 cm, there was no statistically significant difference. There was a significant positive correlation between vector shifts and treatment time for ERB (r=0.2) and GEL (r=0.07), (p<0.05). The most common treatment interval was 5-9 minutes during which a majority of shifts were <0.2 cm, ERB (85.9%) and GEL (73.2%). No statistical difference was noted between ERB and GEL for treatment times >9 minutes.

Conclusion: While there was a statistically significant difference between ERB and GEL shifts, the submillimeter differences are not clinically significant. Planning margins to account for intrafraction motion in prostate therapy are commonly not less than 0.2-0.3 cm and for shifts larger than this margin, ERB and GEL and comparable. GEL had similar time dependency as ERB.

PTC17-0371: Workability of Patient Handling in the Treatment Room at I-ROCK

S. Hirai1, W. Maehana1, M. Ogura1, K. Shioiri1, S. Yoshino1, S. Minohara 2, Y. Kusano 2, S. Ide1, T. Nomiya3, Y. Nakayama3

1Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan 2Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan3Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan

Purpose: In December 2015, i-ROCK started the clinical operation for carbon-ion radiotherapy using the pencil-beam scanning. We report our workflow and its performance in the treatment room.

Materials and Methods: Two radiological technicians handle the patient at each room. Typical workflow in the treatment room is as follows:

  1. The patient is set on the table of robotic couch, and skin-marks is aligned to laser lines. We take care not to twist the patient's body.

  2. The patient body is fixed by the shell-type device. The initial couch position is automatically set to the calculated position at the treatment planning or the previous treatment position.

  3. Orthogonal X-ray FPD images are taken, and are compared to volumetric planning CT images. This is 2D-3D automatic image registration. It takes several seconds in calculation, and the couch position is adjusted semi-automatically.

  4. If necessary, we take CT image by the in-room CT and verify the target and OAR by overlay planning contours.

  5. The technician presses the beam-on button. The beam-on time is about one minute.

  6. These processes are managed and recorded by the oncology information system.

Results: Results of the patient staying time at each fraction in treatment room were shown. In prostate case, the treatment time was extended by about 8 minutes using in-room CT.

Conclusion: Over the first year after treatment started, we have established the treatment workflow for safety and smoothly clinical operation.

PTC17-0395: Verification of Target and Critical Organ Positions Using In-Room CT Images; First Step to the On-Demand Adaptive Carbon-Ion Radiotherapy

S. Minohara1, Y. Kusano1, K. Shioiri 2, W. Maehana 2, Y. Tokiya 2, S. Hirai 2, M. Ogura 2, S. Yoshino 2, Y. Matsuzaki1, Y. Nakayama3

1Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan 2Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan3Kanagawa Cancer Center, Department of Radiation Oncolog, Yokohama, Japan

The ion-beam Radiation Oncology Center in Kanagawa (i-ROCK) has four treatment rooms with fixed beam port for carbon-ion pencil beam scanning. Each treatment room has an on-rail in-room CT in addition of a set of orthogonal x-ray FPD imaging system for patient positioning.

Our standard method of patient positioning is 3D-2D automatic image registration based on the bone structure. 3D is a volumetric CT images for planning and 2D is a set of FPD images. The application of in-room CT is option to verify the target position. Contours of planed CTV, PTV and OAR are superimposed on the current in-room CT images, and we check their relations.

The dose distribution of particle beam is strongly affected by the density along beam path. So our first priority of patient positioning is to match the bone arrangement to the planed one. After that, we check reproducibility of the target and OAR positions against the bone structure using in-room CT images. If the target would not be within the PTV, we would re-adjust patient condition.

Our current situation is the first step to the on-demand adaptive carbon-ion radiotherapy in future. Now we are accumulating the CT data at many clinical situations and are re-calculating the dose distribution to discuss the validity of treatment planning. Decreasing the number of fraction in carbon-ion radiotherapy, the positioning error of one fraction would effect on clinical outcomes. In next step, we would like to establish the clinical criteria to whether or not to perform the re-planning.

PTC17-0408: Effectiveness of Dynamic Proton Range Adaptive with Computed Tomography Image Guidance in the Lung

S. Moriya1, H. Tachibana 2, K. Hotta 2, N. Nakamura 2, T. Sakae1, T. Akimoto 2

1University of Tsukuba, Graduate School of Comprehensive Human Sciences, Tsukuba, Japan 2National Cancer Center, Exploratory Oncology Research & Clinical Trial Center, Kashiwa, Japan

Purpose: To evaluate a dynamic adaptive system using CT-based three-dimensional image guidance in lung cancer patients.

Materials and Methods: We performed a retrospective analysis using CT images from lung cancer patients. The GTV and CTV were contoured by a radiation oncologist. Dose distribution was optimized using 180-deg. and 270-deg. two fields in passive scattering proton therapy. Original plan was generated using clinical protocol for lung, in which a distal margin and a smearing margin were added 4 mm to CTV and 7.5 mm in radius. Three dose distributions were generated using each daily respiratory-gated CT image dataset based on bone registration, tumor registration, and tumor registration with range adaptation. To assess the effectiveness of the system, the three dose distributions were compared to the original plan using the dosimetric parameters of D95% and D5%-D95% for the CTV dose coverage and homogeneity.

Results: For the bone registration, the D95% and D5%-D95% were decreased by -6.3±7.6 % and 3.7±4.2 Gy (RBE) respectively. The tumor registration achieved better dose coverage and homogeneity than the bone registration, however the D95% and D5%-D95% were decreased by -3.2±3.7 % and 1.9±2.1 Gy (RBE), respectively. The range adaptive plans provided similar dose coverage (D95%: -1.8±1.8 %) and homogeneity (D5%-D95%: 1.0±0.8 Gy [RBE]).

Conclusion: Our range adaptation approach enables us to achieve the planned target dose coverage and homogeneity.

PTC17-0413: The Impact of Positioning and Anatomic Deviations on Proton Range Change for Lung SBRT Treatment

Y. Lin1, M. Zhu 2, F.F. Yin3, C. Smith1, A. Waller3, F.F. Yin1

1Duke University Medical Center, Department of Radiation Ocology, Durham, USA 2Maryland Proton Treatment Center, University of Maryland School of Medicine, Department of Radiation Oncology, Baltimore, USA3Varian Medical Systems, Palo Alto, CA

To understand how the daily setup and anatomy change affects the proton beam range for lung SBRT treatment.

Lung SBRT patients previously treated using photon beams were analyzed as if they were treated using protons. The daily anatomy and setup variations were captured using daily CBCT which were used to calculate the daily change of patient water equivalent path length (WEPL). Deformable image registration using VelocityTM software was performed to map planning CT to daily CBCT. The change of WEPL for the distal of planning target volume (PTV) was calculated for each treatment fraction using EclipseTM proton planning software 13.7. The WEPL was calculated for beam directions spreading over 180 degrees with 5-degree interval.

We show an example how the change of WEPL to the distal end of PTV can be more than 1 cm for certain beam angles. The plots of gantry angle versus the change of WEPL to the distal end of PTV for three patients shows that some beam angles are less affected by the daily anatomy and setup variation. For instance, for patient case 2, beam angle ranges from 125-150 shows very little range changes throughout the treatment.

This study explores the possibility of performing proton beam range check before daily treatment. It also indicates the possibility of choosing beam angles that are less affected by daily setup and anatomy change. We are in the process of analyzing the dosimetric consequence of range change and the results shall be presented at the time of conference.

PTC17-0451: Novel 4D-CBCT Reconstruction Technique for Moving Target Using Fiducial-Marker Position

T. Fujii1, S. Takao 2, T. Matsuura3, N. Miyamoto 2, S. Hirayama1, S. Shimizu1, K. Umegaki3, R. Baba4, T. Umekawa5, H. Shirato1

1Hokkaido University, Graduate School of Medicine, Sapporo, Japan 2Hokkaido University Hospital, Department of Medical Physics, Sapporo, Japan3Hokkaido University, Graduate School of Engineering, Sapporo, Japan4Hitachi- Ltd., Department of Research and Development, Kokubunji, Japan5Hitachi- Ltd, Department of Research and Development, Sapporo, Japan

To acquire accurate information for anatomical structure which include not only bone but also soft tissue like moving target at patient positioning of real-time-image gated spot scanning proton beam therapy (RGPT), novel reconstruction technique of 4-dimensional Cone beam CT (4D-CBCT) using position of fiducial marker at exhale phase (= planning phase) in respiration was proposed. We developed in-house software named “Image Analysis Platform (IAP)” and evaluated the effect of the proposed technique by using IAP.

IAP was designed so that it can 1) Track the position of fiducial marker retrospectively on projection-images during CBCT, 2) Segment the projection-images into spatial phases from inhale to exhale according to the marker's position. 3) Reconstruct images from the segmented projection-images. As preliminary study for the novel technique, performance verification of IAP “effect of phase-segmentation” was implemented by using dynamic thorax phantom (CIRS, Ltd.) include a tumor-model (ϕ10 mm) and gold-marker (ϕ2.0 mm) as fiducial-marker. The phantom was activated with sin4 function. CBCT system with dual-orthogonal X-ray sources at Hokkaido University Proton Therapy Center was used for this verification.

The result of the performance verification showed that the reconstruction images segmented for exhale phase improve the position identification of tumor-model compared to that of non-segmented images. It seems possible that noise on the segmented image can be reduced by using reconstruction algorithm like Ordered Subset Exception Maximization (OS-EM) to reduce lack-effect of projection angle. We will improve this technique for the accurate positioning.

PTC17-0454: Dosimetric Effects of Patient Positioning Method on the Inter-Fractional Error in Carbon Ion Radiotherapy for Stage I Lung Cancer

M. Sakai1, Y. Kubota1, J.I. Saitoh1, D. Irie1, K. Shirai1, R. Okada1, M. Torikoshi1, T. Ohno1, T. Nakano1

1Gunma University, Gunma University Heavy Ion Medical Center, Maebashi, Japan

To evaluate the robustness for inter-fractional error in carbon ion radiotherapy for stage I non-small cell lung cancer (NSCLC), two positioning method (tumor matching (TM) and conventional bony-structural matching (BM)) were compared.

Sixty irradiation fields from 30 patients were analyzed. CT images acquired before treatment initiation for confirmation (Conf-CT) under the same settings as the treatment planning CT images were used. The dose distributions were recalculated for Conf-CT using both BM and TM, and the dose volume parameters and the acceptance rate (V95% > 90%) were evaluated. Additionally, the relationships between V95% and clinical factors such as age, position, and tumor displacement were evaluated. Furthermore, the optimal isotropic margin, which in all cases achieved the acceptable condition, was examined.

The means (ranges) of the V95% were 98.93% (70.05%–100%) and 100% (80.42%–100%) (p<0.001), and those of V5Gy(RBE) were 135.9 (62.2–272.0) and 125.8 (55.7–284.6) (p=0.694) with BM and TM, respectively. Particularly, TM offered benefits when Effective Displacement > 0. The optimal isotropic margins with BM and TM were 8 and 4 mm, respectively, while they increased by 61.1% and 6.3% of V5Gy(RBE) compared with planning.

TM was superior to BM because it ensured better dose distribution compared with BM in our treatment policy. Particularly when the tumor is located in the lower lung, the benefit If the divergence between bony structure and tumor position is large, emphasis should be placed on tumor position. Hence, it is important to obtain a CT scan instead of relying on only 2D X-rays.

PTC17-0463: Fiducial Marker Matching Versus Vertebral Body Matching: Dosimetric Impact of Patient Positioning in Carbon Ion Radiotherapy for Primary Hepatic Cancer

S. Abe1, Y. Kubota 2, K. Shibuya3, T. Abe3, T. Sutou1, T. Ohno 2, T. Nakano3

1Gunma University Hospital, Department of Radiology, Maebashi, Japan 2Gunma University, Heavy Ion Medical Center, Maebashi, Japan3Gunma University Graduate School of Medicine, Department of Radiation Oncology, Maebashi, Japan

Purpose: The aim of this study was to compare the dose-volume parameters of fiducial marker matching (MM) with vertebral body matching (VM) in patient positioning for carbon ion radiotherapy for primary hepatic cancer.

Materials and Methods: Twenty patients with primary hepatic cancer were retrospectively studied to assess changes in reproducibility of tumor position and dose distribution on two CT scans. One was for treatment planning and another was for dose confirmation, acquired the day before the first treatment day. The coverage of the clinical target volume (CTV) (D98) and normal liver volume excluding the CTV which received 20 Gy relative biological effectiveness (RBE) (V20) were used as evaluation parameters. Additionally, the correlation of tumor movement and D98 was calculated in VM and MM. The prescription dose was 60.0 Gy (RBE) delivered in four fractions (15 Gy/fx).

Results: The median (range) D98 for VM and MM was 57.9 (20.8–59.9) and 59.9 (57.2–60.3) Gy (RBE), respectively. The median (range) V20 for VM and MM was 17.9 (4.8–44.4) and 16.2 (4.7–44.9) Gy (RBE), respectively. The D98 for MM was significantly larger than that for VM (p = 0.001), although V20 showed no significant difference (p > 0.05). Twelve patients were clinically acceptable (D98 > 57 Gy (RBE)) with VM, while all patients were clinically acceptable with MM. Marker movement correlated with a decrease of D98 for VM (R = −0.814).

Conclusion: Compared with VM, MM was clinically acceptable in all patients. This suggests that MM is more robust than VM.

PTC17-0472: Image Reconstruction with a Fast Monolithic Proton Radiography System

F. DeJongh1, E. DeJongh1, V. Rykalin1, J. Welsh 2, M. Pankuch3, N. Karonis4, C. Ordonez4, K. Duffin4, J. Winans4, G. Coutrakon5

1ProtonVDA Inc, Physics, Batavia, USA 2Stritch School of Medicine Loyola University - Chicago, Radiation Oncology, Maywood, USA3Northwestern Medicine Chicago Proton Center, Physics, Warrenville, USA4Northern Illinois University, Computer Science, DeKalb, USA5Northern Illinois University, Physics, DeKalb, USA

Purpose: Proton radiography offers advantages for therapeutic dose accuracy over x-ray radiography. The requirements are that a clinical system be simple, lightweight, easily scaled to large field sizes, operate at high speed to maximize patient throughput, and expose the patient to the minimum possible radiation dose for a given resolution. We are developing a system to produce images of proton stopping power, by tracking individual protons before and after the patient and measuring the proton residual range after traversing the patient. To achieve optimal spatial resolution, an image reconstruction algorithm must fully exploit the individual three-dimensional proton position information.

Materials and Methods: We have developed an iterative algorithm for radiography fully exploiting proton path information. Simulations of our detector design, with and without multiple scattering effects included, determine the expected accuracy of our proton path reconstruction, and the impact on the spatial resolution of the reconstructed image.

Results: Protons typically scatter by 4 mm after 20 cm of water. Our simulations, demonstrate path reconstruction of individual protons to better than 1 mm. Our iterative algorithm successfully produces images with 1 mm sharpness. The iterative process necessarily increases pixel noise compared to estimates neglecting multiple scattering and additional protons will be needed to achieve a given contrast.

Conclusion: A proton radiography system optimizing image sharpness and dose to the patient will individually track protons before and after the patient. An iterative algorithm produces images with spatial resolution given by the tracking accuracy, at the price of increased pixel noise.

PTC17-0502: Implementation of a Weight Independent Patient Alignment System: Characteristics and Implications on Treatment Planning

A. Utz1, A. Ableitinger1, M. Mumot1, A. Zechner1, M. Stock1, P. Steininger 2, H. Deutschmann3, J. Osorio1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2MedPhoton GMbH, Informatics, Salzburg, Austria3MedPhoton GmbH, Medical Physics, Salzburg, Austria

Purpose: For particle therapy (especially for fixed beamlines), highly flexible patient alignment systems (PAS) are needed. At MedAustron a weight independent 7DOF robot with an integrated CBCT capable imaging system is in clinical use. Due to weight, the CT couch is bent and causes a rotation and shift of the CT dataset. These conditions have to be mapped into the treatment room. This study shows the impact of patient weight on the whole treatment chain.

Materials and Methods: Planning CTs of an indexed phantom were acquired with 0kg, 78kg and 104kg distributed over the CT table. The planning CT is linked to the PAS by an image device transformation (IDT) which was generated without load. The same load setups were used on the treatment couch. Planar images of the phantom were acquired with a table mounted imaging device and registered to the planning CT. The resulting correction vector was applied by the robotic system and the position verified with in-room-lasers. Afterwards, the weight on the PAS was removed and the procedure repeated.

Results: Changes of the weight on the CT results in a different correction vector. Weight changes on the PAS have no impact on the correction vector (Tab1). The accuracy for aligning the phantom is better than 1mm for each axes.

Conclusion: The bending of the CT table is mapped with the static IDT to the treatment table and affects the correction vector. Applying the correction vector, takes the CT table bending into account and executes the resulting shift and pitch.

PTC17-0514: Design Considerations, Installation and Clinical Implementation of a CT Scanner for In-Room Volumetric Imaging of Patients in the Seated Position

B. Kreydick1, M. Pankuch1, D. Hecksel1, W. Hartsell 2, N. Mohammed 2, B. Cramlett3, H. Deeke4, M. Stauffer5, A. Pruneau4, M. Marash6

1Northwestern Medicine Chicago Proton Center, Medical Physics, Warrenville, USA 2Northwestern Medicine Chicago Proton Center, Radiation Oncology, Warrenville, USA3Northwestern Medicine Chicago Proton Center, Machine Shop, Warrenville, USA4Northwestern Medicine Chicago Proton Center, Radiation Therapy, Warrenville, USA5Northwestern Medicine Chicago Proton Center, Dosimetry, Warrenville, USA6P-Cure, Administration, LOD, Israel

Gantry treatment rooms have been developed to rotate a particle beam around patients that are positioned on a horizontal treatment table. However, the flexibly of unlimited angles of incidence for the particle beam comes at the high cost of the gantry structure. An alternative option that maintains rotational flexibility is to rotate the patient in the seated position around a fixed particle beamline, greatly decreasing the complexity and cost of the delivery system. If a patient is to be treated in a seated position, a seated treatment planning image set must also be obtained for proper planning.

In collaboration with P-Cure, a commercial, large bore, 16 slice CT scanner has been mounted inside of an operational proton treatment room. The treatment room is configured with two, fixed beam treatment angles (horizontal and 30° from vertical). A treatment chair has been developed to fit on the room's existing patient positioning robot. The chair has been designed to allow 360° rotational access of the patient to either fixed beam position. The treatment isocenter and the imaging isocenter are offset due to spatial limitation in the room. In this configuration, the patient images can be used for treatment planning, verification of anatomical consistency or volumetric alignment. The scanner is fully capable of 4-D image acquisitions in the seated position making it ideal for treatments within the thorax. The system was installed in late 2016 and now is in early clinical use. Initial experiences will be discussed.

PTC17-0522: Radiation Dose and Risks of Radiation Carcinogenesis from Proton Radiography vs X-Ray Based Methods of Range Determination and Image Guidance

J. Welsh1, M. Pankuch 2, G. Coutrakon3, V. Rykalin4, E. DeJongh4, N. Karonis5, C. Ordonez5, J. Winans5, I. Polnyi4, F. DeJongh4

1Loyola University Stritch School of Medicine, Radiation Oncology, Maywood, USA 2Northwestern Medicine Chicago Proton Center, Department of Proton Therapy, Warrenville, USA3Northern Illinois University, Physics, DeKalb, USA4Proton VDA, Proton VDA, Batavia, USA5Northern Illinois University, Physics and Computer Science, DeKalb, USA

Proton beam therapy (PBT) has been promoted as a superior method of administering radiation therapy for pediatric patients since it better confines the excess, unwanted radiation dose. PBT has led to a reduction of acute and chronic adverse effects in children. To maximize its benefits, PBT demands highly accurate and precise proton beam range determination. Additionally, the highly conformal dose distributions demand precise patient positioning which is nowadays achieved through x-ray image guidance.

Radiation doses from 3 different modalities for improving set up and range determination were compared: dual-energy x-ray imaging, cone beam CT, and proton radiography. Cone beam CT imaging may deliver around 20-30 mGy per image depending on the anatomical region. Dual energy x-rays for determining proton beam range may give similar doses. Assuming daily application, when multiplied over a typical course of 35 fractions, this amounts to 660 to 1,050 mGy. Using the standard estimate of 5% per Sv (Gy) for radiation carcinogenesis, the calculated risk of fatal radiation-induced cancer is roughly 3-5%. Such rates of secondary cancers could overshadow the benefits of PBT, especially for children. The dose from proton radiography is approximately 0.1 mGy per study. When multiplied over a 35 fraction course, this amounts to an absorbed dose of 3.5 mGy.

Doses from proton radiography as a means of beam range determination and image guidance are perhaps two orders of magnitude lower than with x-ray based methods. Calculated risks for fatal cancer by proton radiography are well under 1%.

Motion Management

PTC17-0070: Exploiting Spot-Wise Dose Rate Modulation of Varian ProBeam® Systems for Motion Mitigation

Y. Zhang1, I. Huth 2, M. Wegner 2, D.C. Weber1, A.J. Lomax1

1Paul Scherrer Institut, Center for Proton Therapy, Villigen-PSI, Switzerland 2Varian Medical Systems, Particle Therapy GmbH, Troisdorf, Germany

For PBS, treatment time is, inter alia, influenced by dose rate, layer/spot switching time and system dependent minimum spot duration. For cyclotron-based system, maximum possible dose rate is limited by energy-dependent beamline transmission. This study investigated the potential and advantage of spot-wise dose rate modulation for breath-hold and re-gated tumour treatments, in comparison to layer-wise modulation.

Using 6 4DCT(MRI) datasets of liver tumours with irregular motions >10mm (CTV: 100-400cc; period: 5.3/6.3s), 4D treatments using breath-hold and re-gating (5mm-gating-window +5x layered-rescanning), were respectively simulated for the above-mentioned dose rate modulation models, assuming 4 prescription dose schemes (0.67/2/4/12Gy). All 4D plans were compared using dose-volume metrics (D5-D95, V95) for the CTV and treatment time.

For re-gating, comparable plan quality (interplay effect was mitigated within 5% of the static case) was obtained independently of the chosen dose rate model, but treatment times of spot-wise modulated dose rate plans were only 93/61/47/42% (0.67/2/4/12Gy) of those using layer-wise approach. For breath-hold treatments, treatment times for spot-wise dose rate were only 62/50/61/94% of the corresponding plans using layer-wise.

Our results show the significant reduction of plan delivery time for all simulated dose schemes and cases if beam current could be adjusted spot-by-spot. Such an approach can substantially enhance feasibility and efficiency of both breath-hold and re-gating delivery tumour treatments. In particular, for fraction doses up to 4Gy, spot-wise dose rate modulation could reduce treatment time per field for breath-hold treatments to 20-30s, thus allowing for a whole field to be delivered in one breath-hold.

PTC17-0091: Respiratory Guidance Method Used for Synchrotron-Based Scanned Heavy-Ion Beam Delivery

P. He1, Q. Li1, T. Zhao1, X. Liu1, Z. Dai1, Y. Ma1

1Institute of Modern Physics- Chinese Academy of Sciences, Department of Medical Physics, Lanzhou, China

Synchrotron-based heavy-ion accelerator operates in pulse mode at a low repetition rate that is comparable to the patient's breathing rate. To overcome inefficiencies and interplay effects between the target residual motion and the scanned heavy-ion beam delivery process, a novel respiratory guidance method was developed to help patients synchronize their breathing patterns with the synchrotron excitation patterns by performing short breath holds (BH) with the aid of personalized audio-visual biofeedback (BFB) system. Using 96 breathing traces from eight healthy volunteers who were asked to breathe freely and guided to perform short BHs with the aid of BFB, a series of dedicated 4D dose calculations (4DDC) were performed. Our results showed that with the respiratory guidance method, the treatment efficiency increased by factors of 2.23∼3.94 as compared to FB gating, depending on the duty cycle (DC) settings. The magnitude of dose inhomogeneity for the respiratory guidance methods was 7.5 times less than that of the non-gated irradiation and good reproducibility of breathing guidance among different fractions was achieved. Thus, our study indicates that the respiratory guidance method not only improved the overall treatment efficiency of respiratory-gated scanned heavy-ion beam delivery, but also had the advantages of less dose uncertainty and better reproducibility among fractions.

PTC17-0125: Re-Scanning Strategies, Methods and Measurements

M. Braeuer1

1Siemens Healthcare GmbH, AT RO PT MC CAP, Erlangen, Germany

The use of carbon ion-ions for irradiation of tumors in the abdomen is growing. These tumor sites require a motion management, especially if the raster scanning method is used. One possible option for motion management is the application of the re-scanning method.

For the case of layer scanning, new strategies for re-scanning are described. The methods are a validated and benchmarked by simulation models on the basis of a commercial planning system. In addition, results of validation measurements and the timing considerations for a commercial carbon and proton system are presented.

PTC17-0208: Sufficient Margins for Prostate Interfractional Motion at the Gunma University Heavy Ion Medical Center

D. Bridges1, K. Fukata1, H. Kawamura1, T. Kanai1

1Gunma University, Medical Physics and Biology for Ion Therapy, Maebashi-shi, Japan

Purpose: Our prostate carbon ion radiation therapy (CIRT) has excellent results. Because we only match bony anatomy for patient positioning, further clarification of clinical outcome is desired. Lacking CBCT for our carbon therapy patients, we investigate the robustness of our treatment margins by considering the interfractional motion of prostate patients treated with Elekta Intensity-Modulated X-ray Therapy (IMXT) at Gunma University Hospital.

Materials and Methods: 1513 daily table shifts (87 IMXT prostate cancer patients) were used to derive a probability function and probability-weighted dose distribution in MATLAB R2016b: Six carbon patients' RTDOSE files from XiO-N 4.47 were used to sum an ensemble-averaged dose distribution (i.e. weighted by shift frequency), assuming patient's average is population average. Results were analyzed with Velocity 3.2.0 and MATLAB.

Results: Our IMXT prostate interfractional distance empirical cumulative distribution function is shown. We assume this data represents CIRT prostate motion as well, implying a PTV margin of 6.2 mm will ensure coverage for 95% of treatment fractions. Mean prostate displacement: (0.0,0.34,-0.43) millimeters with standard deviation (1.0,2.1,1.8) millimeters from simulated position in patient's left, posterior, and superior directions, respectively. The mean distance shifted was 1.9 mm with standard deviation 2.3 mm. Our planned and estimated delivery for six patients is shown. This patient ensemble probabilistic method neglects stopping power changes and organ deformation.

Conclusion: This study supports the conclusion that with only AP-and-lateral bony-matching DRRs our margins ensure dose delivery to the CTV of prostate CIRT patients while establishing rectal sparing.

PTC17-0213: Reproducibility of the Lung Anatomy Using Active Breathing Control: Dosimetric Consequences for Scanned Proton Treatments

L.A. den Otter1, R.G.J. Kierkels1, E. Kaza 2, A. Meijers1, M.O. Leach 2, D. Collins 2, J.A. Langendijk1, A.C. Knopf1

1University of Groningen - University Medical Center Groningen, Radiation Oncology, Groningen, Netherlands 2CR-UK Cancer Imaging Centre, The Institute of Cancer Research and The Royal Marsden Hospital, London, United Kingdom

Purpose: The treatment of moving targets with scanning proton beams is challenging. By controlling lung volumes, Active Breathing Control (ABC) assists breath-holding for motion mitigation. The delivery of proton treatment fractions often exceeds feasible breath-hold durations, requiring high breath-hold reproducibility. Therefore, we investigated dosimetric consequences of anatomical reproducibility uncertainties in the lung under ABC, evaluating robustness of scanned proton treatments during breath-hold.

Materials and Methods: T1-weighted MRIs of five volunteers were acquired during ABC, simulating image acquisition during four subsequent breath-holds within one treatment fraction. Deformation vector fields obtained from these MRIs were used to deform 95% inspiration phase CTs of 3 randomly selected non-small-cell lung cancer patients. Per patient, an intensity-modulated proton plan was recalculated on the 3 deformed CTs, to assess the dosimetric influence of anatomical breath-hold inconsistencies.

Results: Dosimetric consequences were negligible for patient 1 and 2. Patient 3 showed a decreased volume (95.2%) receiving 95% of the prescribed dose for one deformed CT. The volume receiving 105% of the prescribed dose increased from 0.0% to 9.9%. Furthermore, the heart volume receiving 5 Gy varied by 2.3%.

Conclusion: Based on the studied patients, our findings suggest that variations in breath-hold have limited effect on the dose distribution for most lung patients. However, for one patient, a significant decrease in target coverage was found for one of the deformed CTs. Therefore, further investigation of dosimetric consequences from intra-fractional breath-hold uncertainties in the lung under ABC is needed.

PTC17-0219: Irradiation Time of Audio-Coached Respiratory-Gated Proton Therapy Using a Synchrotron-Type Accelerator

N. Kondo1, S. Niihara1, T. Arimura1, T. Ogino1, Y. Hishikawa1

1Medipolis Proton Therapy and Research Center, Medical Department, Ibusuki, Japan

Purpose: Synchrotron-type accelerators in clinical use generally operate at a constant frequency with a period of several seconds, which is comparable to one respiratory cycle. In practice, we have often experienced the phenomenon of synchronization between the acceleration cycle and the patient's respiration, while the relationship between the audio coaching and the time of irradiation was not revealed.

Materials and Methods: For a better understanding of the irradiation time of audio-coached respiratory-gated proton therapy, we statistically analyzed the treatment records in the past patients. Additionally, the numerical simulation of respiratory-gated irradiation using the simplified models of accelerator operation patterns and respiratory waveforms was performed.

Results: The computational results suggest that the irradiation time of respiratory-gated proton therapy is independent of the coaching rate of 10-20 breaths per minute (BPM) in principle, meanwhile the irradiation time can be shortened in the particular condition. For example, in the case of the coaching rate of 15BPM (=4[s] period) or 20BPM (=3[s] period) with the accelerator operation period of 2[s], the amount of extracted proton per unit time increases in the specific phase difference between the patient's respiration and the acceleration cycle. The tendency as mentioned above was verified through the actual therapeutic radiation.

Conclusion: The mechanism of the irradiation time of audio-coached respiratory-gated proton therapy using a synchrotron-type accelerator was clarified. The new knowledge obtained here will be applicable to the development of novel coaching systems for high-efficiency particle therapy.

PTC17-0243: Assessment of Dosimetric Uncertainties Induced by Deformable Image Registration Methods in 4D Proton Treatment Planning for Liver Tumors

C. O. Ribeiro1, A. Knopf1, A.J. Lomax 2, J. Langendijk1, D.C. Weber 2, Y. Zhang 2

1University Medical Center Groningen UMCG, Department of Radiation Oncology, Groningen, Netherlands 2Paul Scherrer Institut PSI, Center for Proton Therapy, Villigen PSI, Switzerland

Purpose: Different deformable image registration (DIR) algorithms can result in diverse motion estimations, directly influencing 4D dose distributions and clinical decision-making. The aim of this study is to evaluate DIR-induced dosimetric uncertainties for 4D dose calculations of scanned proton plans using 4DCT-MRI data sets (3DCTs generated with motion from 4DMRI). Such data sets utilize the unique advantage of having a prior known dense motion field instead of conventional sparse landmarks.

Materials and Methods: Four different DIR methods (1) ANACONDA (DIR1), (2) Morfeus (DIR2), (3) B-splines (DIR3), and (4) Demon's (DIR4) were respectively applied to three 4DCT-MRI data sets of liver cancer patients (8 mm motion amplitude). The extracted motion fields were used as input for a 4D dose calculation (4DDC). DIR-induced dosimetric uncertainty was assessed by individually comparing the resultant 4D dose distributions to those obtained with the originating 4DMRI motion. Differences for both single-and three-field plans were investigated.

Results: Pronounced differences in 4D dose distributions were observed among different DIR scenarios, as well as in comparison to the original 4DMRI calculations. DIR-induced dose differences in V95 to the CTV reached 7.91±3.46% for single-field and 4.40±2.17% for three-field plans. Of the tested methods, Morfeus (DIR2) provided the lowest dosimetric uncertainties with the best prediction of interplay effects.

Conclusion: Motions estimated from 4D imaging using DIRs can vary substantially. The results of this work indicate the necessity to therefore interpret individual 4D dose distributions for scanned proton plans with caution and ideally with an error bar.

PTC17-0258: Experimental Comparison of Spot, Raster and Line Scanning and Their Effectiveness in Mitigating Tumor Motion Using Rescanning

G. Klimpki1, Y. Zhang1, G. Fattori1, S. Psoroulas1, D.C. Weber1, A.J. Lomax1, D. Meer1

1Paul Scherrer Institute, Center for Proton Therapy, Villigen PSI, Switzerland

Purpose: We compare the capability of three different beam delivery techniques to mitigate organ motion using rescanning: discrete spot scanning (SS), raster scanning (RS) and continuous line scanning (LS).

Materials and Method: Based on a 4DCT of a liver carcinoma, we defined a geometric ITV encompassing target motion. The SS plan (0.6 Gy, single field) was optimized at end-exhale and converted to equivalent RS and LS plans. All plans were delivered to a water phantom on our Gantry 2 beginning at the same breathing phase. We reproduced patient-specific target motion with a sliding table (10.7 mm peak-to-peak and 3.7 sec. period). Based on 4D simulations, we determined 6 to be an appropriate number of rescans in layered and volumetric sequences. Absolute dose distributions were measured with a 2D array of ionization chambers (PTW seven29); relative ones were recorded with a scintillation screen coupled to a CCD camera.

Result: Converted RS and LS plans yield equal dose distributions in 11.1 cm water depth (center SOBP). Furthermore, RS and LS are substantially faster than SS saving roughly 40 and 50 sec., respectively, when rescanning volumetrically. In terms of motion mitigation, volumetric rescanning decreases inhomogeneity inside the water-equivalent CTV contour further than layered, especially for LS. At 6 volumetric rescans, SS and LS show comparable inhomogeneity scores (∼6%). Residual inhomogeneity for RS remains higher at ∼8%.

Conclusion: We conclude that LS could represent a fast and effective delivery technique to treat moving targets using rescanning.

PTC17-0276: Robustness of Conformal 4D-Optimization against Motion Detection Errors

K. Schuetze1, A. Eichhorn1, S. Hild 2, C. Graeff1, M. Wolf1

1GSI, Biophysics, Darmstadt, Germany 2University of Trento, TIFPA, Trento, Italy

Purpose: 4D-optimization offers the possibility of conformal treatment also of strongly moving tumors, but requires motion-synchronized delivery. Robustness versus measurement inaccuracies is therefore essential.

Materials and Methods: For 5 NSCLC patients (motion range 2 – 22 mm), 4D-plans were optimized delivering a homogeneous dose to the CTV in each motion phase. Motion-synchronized delivery was simulated on the HIT accelerator. Detection delays of 0.1 – 0.6 s were introduced, causing spots to be irradiated in erroneous motion phases. Dose coverage (V95) was assessed. The effectiveness of isotropic margins to recover dose errors was investigated.

Results: Dose errors scaled both with motion amplitude and measurement error. For realistic errors of 0.2 s, V95 of the strongest moving patient dropped to 93.4%, and was above 95% for all others. For 0.6 s, V95 ranged from 64.3% to 97.5%. Margins of 3mm restored dose coverage at 0.2 s delay for all patients to > 99%. For 5mm margins and 0.6 s error, V95 increased to a range of 89.6% to 100%.

Conclusion: Conformal 4D-optimization with uniform dose to motion phases is robust against motion detection errors. Underdoses occur at the borders of the target volume and can be recovered by isotropic margins. More realistic experiments are needed to verify these results.

PTC17-0278: Time and Memory Reduction for 4D IMPT Optimization

K. Anderle1, M. Prall1, S. Hild1, M. Wolf1, C. Graeff1

1GSI Helmholtz Centre for Heavy Ion Research, Biophysics, Darmstadt, Germany

Purpose: Range changes have to be considered when treating moving targets with IMPT. A full 4D optimization can solve this issue, but significantly increases the problem size. When several targets and OARs are included in 4D optimization, the calculation can take days, if at all possible. Therefore a novel approach is proposed to reduce memory requirements and speed up large 4D IMPT optimization problems.

Materials and Methods: An adaptive voxel grid was employed instead of original or downsampled grid. Our adaptive voxel grid focuses on relevant voxels in each 4DCT motion state, such as OAR and target borders or high dose gradient regions. A further increase in speed-up was achieved by splitting iterations into major and minor contributions from beamlets. The number of major contributions is significantly lower, therefore a speedup was achieved by optimizing major contributions alone and updating minor ones only at regular intervals. The new algorithm was tested on two complex lung cancer patient cases (both with two lesions, 6 and 4 fields, 149 and 440 cc total volume).

Results: The ordinary 4D optimization was possible only with downsampled grid. With our new algorithm the memory consumption was reduced by a factor of 3, while total time was reduced by a factor of 20 – from 80 to 4 hours, without any loss of treatment plan quality.

Conclusion: We have developed a novel optimization algorithm to significantly reduce memory and time consumption, which is essential for successful 4D IMPT optimization.

PTC17-0355: A New Algorithm for Motion Adapted Cone Beam CT Reconstruction Using an Advanced Motion Model

Y. Censor1, S. Rowland 2, T. Dou3, D. Low3, R. Schulte4

1University of Haifa, Mathematics, Haifa, Israel 2City University New York, Discrete Imaging and Graphics Group, New York, USA3University of California Los Angeles, Radiation Oncology, Los Angeles, USA4Loma Linda University, Basic Sciences, Loma Linda, USA

Purpose: To describe a novel algorithm for motion adapted cone beam CT (CBCT) reconstruction useful for 4D radiation therapy of lung tumors.

Materials and Methods: The presence of internal motion during CBCT acquisition is a challenge to CBCT reconstruction. We report on a novel algorithm that adapts to the presence of lung motion by using a free-breathing acquisition of CBCT ray sums, a surrogate breathing signal, and an advanced motion model1. The new algorithm solves the reconstruction problem as a feasibility problem with an iterative projection method. We are testing and optimizing the algorithm with a modification of the 4D FORBILD thorax phantom 2 and the simulation package jSNARK3.

Results: We simulated 25 helical scans of the 4D phantom and a motion described by 5 cosine terms. For helical-CT reconstruction, we used the Katsevich algorithm4. The deformable image registration toolbox Elastix was used to register the image from the first helical scan to the remaining 24 images establishing the motion model parameters1. We are currently simulating a CBCT acquisition of the 4D phantom and will use it to reconstruct the reference-breathing image of the phantom with the new algorithm.

Conclusion: A novel CBCT motion-adapted reconstruction algorithm in combination with advanced breathing motion model has been developed for applications in particle therapy.

References: [1] Thomas D, et al. Int J Radiat Oncol Biol Phys. 89:191-8, 2014. [2] http://www.imp.uni-erlangen.de/phantoms/thorax/4D_Thorax_Description.pdf. [3] http://jsnark.sourceforge.net/. [4] A. Katsevich. Advances in Applied Mathematics, 32:681–697, 2004.

PTC17-0363: Newly Patient Immobilization Technique to Reduce the Respiratory Induced Body Motion in the Prone Position

K. Shioiri1, Y. Kusano 2, W. Maehana1, S. Hirai1, Y. Tokiya1, S. Yoshino1, S. Minohara 2, S. Ide1, T. Nomiya3, Y. Nakayama3

1Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan 2Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan3Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan

Purpose: As carbon-ion radiation therapy facility with a fixed beam nozzle, the prone position is used for the irradiation from the back of the patient. In the prone position, there is a problem that the motion of spine and pelvis around abdomen caused by respiration tends to occur in the anterior-posterior (AP) direction. In this report, we introduce the patient immobilization technique to reduce the spine and pelvis motion in prone position.

Materials and Methods: In order to release the abdominal motion caused by breathing, the empty space was made between the thorax and the pelvis. The patient immobilization device consisted of the vacuum bags, the plastic foam, and the thermoplastics sheet. To confirm the effectiveness of this method, serial X-ray images with respiration waveform were acquired from nine patients with chordoma of the sacrum and spine, and the maximal displacement of bone were analyzed during the respiration.

Results: Maximal displacements of the bone around abdomen along with the respiration were 1.4 mm, 0.8 mm, and 1.4 mm in the lateral, vertical, and longitudinal directions, respectively. Especially in the sacral chordoma, the displacement in the AP direction was very small as 0.5 ± 0.5 mm (mean ± SD), and the influence of respiratory motion was negligible.

Conclusion: Our results suggest that the newly patient immobilization technique sufficiently suppressed the influence of respiratory motion in the prone position.

PTC17-0364: Development of an Easy-To-Use Respiration Sensor with Fast Response for Respiratory Gated Radiation Therapy

K. Shioiri1, S. Minohara 2, W. Maehana1, Y. Kusano 2, E. Takeshita 2, S. Hirai1, Y. Tokiya1, S. Yoshino1, S. Akita1, S. Ide1

1Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan 2Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan

Purpose: We developed the novel hybrid-respiratory sensor (IRP sensor) that consists of position sensitive detector (PSD), infrared camera, infrared light source and corner cube reflector. This IRP sensor is controlled on the Anzai respiratory gating management system (AZ-733VI). The purpose of this study is to determine the delay time in the response of IRP sensor system.

Materials and Methods: The response of IRP sensor corresponding to the motion of the dynamic phantom (CIRS, Dynamic Thorax Phantom) was measured with reference to the analog output of the laser displacement sensor (Keyence, LB-02). In addition, the respiratory gate signal from AZ-733VI was recorded on a digital oscilloscope to evaluate the delay time of gate signal.

Results: The respiration waveform of IRP sensor corresponded with the laser displacement sensor well, and there was the delay time of both waveform within 5 msec. Overall, the observed delay time of gate signal in the response of IRP sensor system was less than 20 msec. These results showed that delay time of gate signal was dramatically reduced in comparison with other commercial respiratory gating system.

Conclusion: The IRP sensor system is the useful device for respiratory gating that has high-speed response and is easily and simple to setup on the patient.

PTC17-0373: Respiratory Gated Irradiation Using Carbon-Ion Pencil-Beam Fast Scanning at I-ROCK

S. Minohara1, Y. Kusano1, E. Takeshita1, K. Shioiri 2, W. Maehana 2, Y. Tokiya 2, Y. Matsuzaki1, S. Hirai 2, S. Yoshino 2, Y. Nakayama3

1Kanagawa Cancer Center, Section of Medical Physics and Engineering, Yokohama, Japan 2Kanagawa Cancer Center, Division of Radiological Technology, Yokohama, Japan3Kanagawa Cancer Center, Department of Radiation Oncology, Yokohama, Japan

Our approach of the irradiation to the respiratory moving target is a combination of the pencil-beam fast rescanning and the respiratory gating with respiratory sensor on the body surface. In the phase controlled rescanning that was developed by NIRS, each slice-layer is re-painted by the dose at several times within a gate-on period. Our clinical criteria are to keep the respiratory motion less than 5 mm during the gate-on and to rescan each layer at six times.

In the treatment planning, 4D CT and respiratory gated CT images are used. And the CT simulation room has a set of orthogonal x-ray FPD that is available to record the serial images with respiration waveform. Using these dynamic image data, we delineate GTV, CTV, and PTV, and decide the gate level.

In the treatment room, the patient is positioned based on bone structures of x-ray FPD images along with our standard setup procedure. And just before the irradiation, the motion of implanted fiducials are checked by the oblique x-ray fluoroscopic images. The acceptable range of the motion during gate-on is compared to the template that is generated by the treatment planning system. If the respiration waveform are disturbed during irradiation, the gate signal is turn off manually and we re-check the relation of respiration waveform and motion of fiducials. If necessary, we can verify the detail using in-room CT images.

After the commissioning of respiratory gated irradiation system, we have started the clinical operation to lung and liver tumors.

PTC17-0418: Tumor Tracking System at CNAO: One More Step Toward Clinical Use

H. Ziri1,2, G. Baroni1,3

1Centro Nazionale di Adroterapia Oncologica CNAO, Clinical Bioengineering Unit, Pavia, Italy 2The University of Pavia, Department of Industrial and Information Engineering, Pavia, Italy3Politecnico Milano, Department of Electronics- Information and Bioengineering, Milan, Italy

Scanned ion beams have been shown to be successful in providing highly conformal treatment of different tumour sites. However, the treatment of moving organs situated in the thorax and abdomen is still a challenge. Interference of target and beam motion as well as the path length variation are still challenges to exact dose delivery in active scanning system.

The National Centre of Oncological Hadrontherapy (CNAO), provided with proton and carbon ion active beam scanning system, has a dedicated optical motion detection device. The aim was to provide real-time target localisation and correction of radiation beam geometry to compensate for respiratory motion. Preliminary experiences were conducted at CNAO and Gesellschaft für Schwerionenforschung (GSI) based on a dedicated interfacing between beam delivery system and motion monitoring system using internal-external correlation models. The feasibility of such complex system was proven. The average error in target localization accuracy was within 1.5 mm. Additional activities are planned for further validation before clinical use to fill the gap between phantom studies and clinical application of tumour tracking in particle therapy. First step, system evaluation and dosimetric experiments on more realistic respiratory motion phantom is a mandatory. An overview of tumour tracking experiment as well as potential adequate motion monitoring and quality assurance standards, will be given aiming for tumour tracking validation for clinical use at CNAO.

This project has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 675265.

PTC17-0421: Optimized Beam Scanning Direction: A Technique to Mitigate for Interplay Effects?

G. Fattori1, J. Hrbacek1, S. Safai1, G.M. Klimpki1, S. Psoroulas1, M. Marti1, L. Placidi1, D.C. Weber1, A.J. Lomax1

1Paul Scherrer Institut, Center for Proton Therapy, Villigen, Switzerland

The role of scanning direction for the mitigation of interplay effects in PBS proton therapy is well reported in the literature, with 12 out of 14 papers (published 2001-2015) mentioning a direction-dependence. This scanning effect is however mostly supported by simulations, or by experiments using simple geometries which may not represent a realistic clinical scenario. To clarify this controversy, we have analysed multiple 4D measurements at PSI to seek experimental evidence for the mitigating effect of scanning direction.

All experiments (acquired from multiple currently running 4D projects) were collected using a sliding-platform mounted dosimeter, using discrete PBS delivery with or without rescanning. Dose distributions were measured at mid-SOBP position with a CCD system and ionization chamber array, and the 4D dose distributions compared to the static reference irradiations using the gamma criterion (3%/3 mm). Delivered fields ranged from geometrical targets (2D dose layers or spheres) to lung cancer fields. Optical tracking was applied for motion-synchronized beam delivery, providing consistent measurement conditions for a variety of motion models, including patient-specific breathing curves and sinusoidal trajectories with amplitudes of up to 20 mm.

Results from single energy (2D) measurements tended to confirm that if the primary (fastest) scanning is parallel to the motion, a large mitigation of the interplay effect can be achieved. This was however not confirmed for realistic clinical settings using multiple energy levels (3D), where the interplay among energy layers and the use of highly modulated patient fields generally annihilate the scanning direction effect.

PTC17-0459: Evaluation of the Frequency and Length of Patient Couch Positions During Real-Time-Image Gated Spot-Scanning Proton Therapy (RGPT) for Prostate Cancer

H. Tamura1, S. Shimizu 2,3, K. Nishioka 2, T. Hashimoto4, T. Yoshimura1, Y. Matsuo1, T. Matsuura1,5, S. Takao1, K. Umegaki1,5, H. Shirato3,4

1Hokkaido University Hospital, Proton Beam Thrapy Center, Sapporo, Japan 2Hokkaido University, Department of Radiation Oncology, Sapporo, Japan3Hokkaido University, Global Station for Quantum Medical Science and Engineering- Global Institution for Collaborative Research and Education GI-CoRE, Sapporo, Japan4Hokkaido University, Department of Radiation Medicine, Sapporo, Japan5Hokkaido University, Division of Quantum Science and Engineering, Sapporo, Japan

Purpose: We treat prostate cancer with RGPT system detecting the location of the gold marker inserted into the prostate gland during proton beam delivery. The patient couch is adjusted so that the marker stays within a pre-determined 2mm displacement from the planned position. We evaluated the frequency, direction and length of the patient couch movement during each treatment session.

Materials and Methods: Three hundred treatment sessions with 10 patients from March 2016 to October 2016 were analyzed retrospectively. Right-left (RL), cranial-caudal (CC), and anterior-posterior (AP) coordinates of the gold marker were recorded from the system log files.

Results: Couch corrections occurred 104 times (34.7%), and were necessary at least once in 30 treatment sessions with 9 out of the 10 patients. The patient who needed corrections most often required 27 corrections in 30 sessions. The mean and SD of the couch shift distances were 0.56+/-0.65mm, 1.92+/-1.71mm, and 2.51+/-1.94mm in the RL, CC, AP directions respectively. The maximum distances in one treatment were 3.7mm, 9.7mm, and 13.7mm and the frequency of the couch shifts were 2%, 32%, and 61% in the RL, CC, and AP directions respectively. The most frequent time for the first couch shift was 4-5 minutes from the start of each session.

Conclusion: The frequency of the couch shift in the AP direction was higher than in the other directions. Checking the real time target position during beam delivery might be useful to evaluate doses irradiating normal tissue especially for the rectum wall.

PTC17-0493: Clinical Commissioning of Gated Proton Pencil Beam Scanning

M. Enmark1,2, N. Lundkvist 2, M. Fager 2, M. Kügele1,3, H. Nyström 2, S. Ceberg1

1Department of Hematology- Oncology and Radiation Physics- Skåne University Hospital, Lund, Sweden 2Skandionkliniken, Uppsala, Sweden3Medical Radiation Physics- Department of Clinical Sciences Lund- Lund University, Lund, Sweden

Purpose: To mitigate respiratory motion in pencil beam scanning (PBS), we intend to treat in deep inspiration breath hold (DIBH) with the use of an optical surface scanning system as an external gating signal. The purpose of this study was to evaluate beam accuracy in gated PBS delivery.

Materials and Methods: Two categories of spot-maps were measured with a 2D ion-chamber array for gated and non-gated PBS delivery, a typical “patient” spot-map (MAPpat) and an extended test-pattern (MAPtest) with small target volumes distributed over a field of 28x38 cm 2. The system was stressed by using a breathing frequency that is three times higher than for typical DIBH. The gated PBS delivery was evaluated by comparing to 1) TPS calculation and 2) non-gated PBS delivery. In total, 40 gated and non-gated PBS deliveries were carried out to address reproducibility for each delivery technique. The variation of the delivered dose in each volume was statistically evaluated using Wilcoxon rank sum test.

Results: Acquired data met the clinical gamma criteria of 95% pass rate at 3%/2mm in comparison with TPS- calculations, had reliable reproducibility in both MAPpat and MAPtest cases. The difference between non-gated and gated PBS was found to be not statistically significant (P>0.05).

Conclusion: This study concludes that gated PBS is delivered with the same reliability as non-gated PBS. Thus PBS DIBH treatment using optical surface scanning was shown to be feasible.

PTC17-0524: Evaluation of Intra and Inter-Fraction Motion during Accelerated Partial Breast Irradiation in Proton Therapy Using a Surface Imaging System

H. Singh1, Y. Zheng1, T. Twyford1, M. Chacko1, L. Harrell 2

1ProCure Proton Therapy Center, Medical Physics, Oklahoma City, USA 2ProCure Proton Therapy Center, Radiation Oncology, Oklahoma City, USA

Purpose: Proton therapy may help to reduce the late cardiac complications for patients receiving left breast accelerated partial breast irradiation (APBI). Setup accuracy and reproducibility are essential for proton therapy, especially when it is hypo-fractionated. In this study, we report the application of a surface imaging system on site set-up and motion monitoring during treatment for APBI patients.

Materials and Methods: We analyzed six left breast cancer patients treated with APBI for a total dose of 40 CGE in 10 fractions. Each patient was treated with two or three fields at two different couch angles. Realtime AlignRT monitoring was used during the site set-up for each couch angle to ensure arm position reproducibility. The setup was then verified with an orthogonal X-ray imaging system. During treatment, patient motion was monitored and recorded with AlignRT. The recorded data (a data entry every 0.3 second) were analyzed to evaluate the intra and inter-fraction motion and plan robustness.

Result: The average motion for a sample patient as a function of fraction number is shown. The motion magnitude ranged from 0.9 to 3.3 mm for field 1 and 0.8 to 2.7 mm for field 2. Patient motion was the least in lateral direction and most in vertical direction. For all six patients, the average motion was up to 3.5 mm during treatment.

Conclusion: Surface imaging is a useful tool to improve setup reproducibility and monitor motion for APBI patients. Further application in gated treatment is under investigation.

Nozzle Design, Beam Delivery, and Dosimetry

PTC17-0034: The Initial Study of 3D Gel Dosimeter in Carbon Ion Radiotherapy

Y. Sheng1, W. Wang1, S. Kambiz1, D. Wang 2, Z. Tao 2

1Shanghai Proton and Heavy ion Center, Medical Physics, Shanghai, China 2Shanghai Proton and Heavy ion Center, diagnostic radiology division, Shanghai, China

Purpose: Gel dosimetry can be used to measure dose distributions in three dimensions. In this study, the application of MAGAT (Methaacrylic Gelatin And Tetrakis) for carbon ion radiotherapy was investigated.

Materials and Methods: MAGAT gel was made from gelatin, methacrylic acid (MAA), tetrakis (hydroxymethyl) phosphonium chloride solution (THPC), and deionized water. The gel was poured into three types of vials for irradiation with carbon ions. Type A vials were used for dose calibration, were cylindrical in shape, had a diameter of 3 cm, and a length of 5 cm. Type B vials were used to measure depth dose distributions, were cylindrical in shape, had a diameter of 6 cm, and a length of 20 cm. Type C vials were used for measurement of dose distributions, were spherical in shape, and had a diameter of 10 cm. All vials were scanned with a clinical MRI unit after irradiation.

Results: Calibration curves relating the MRI R2 values to absorbed doses were generated. The depth dose distributions of carbon ions was well represented; the entrance region, Bragg peak, and fragment tail could be seen by both MRI and eye. For type C vials, dose distributions determined by the MRI scans showed good agreement with treatment planning calculations.

Conclusions: Gel dosimeters can be used to measure 3D dose distributions including plan verification for carbon ion radiotherapy. More research is needed to make the dosimeter an accurate method for proton and carbon ion therapy, such as the energy dependence, dose dependence, dose rate dependence, and stability.

PTC17-0042: Direct Experimental Determination of kQ for Precision Dosimetry of a Clinical Carbon Ion Beam

J.M. Osinga-Blättermann1,2, U. Ankerhold1, S. Brons3, S. Greilich 2, O. Jäkel 2,3, A. Krauss1

1Physikalisch-Technische Bundesanstalt PTB, Department of Dosimetry for Radiation Therapy and Diagnostic Radiology, Braunschweig, Germany 2German Cancer Research Center DKFZ, Division of Medical Physics in Radiation Oncology, Heidelberg, Germany3Heidelberg Ion-Beam Therapy Center HIT, HIT, Heidelberg, Germany

Until now, ionization-chamber based dosimetry of carbon ions has not reached the same level of accuracy as dosimetry of high-energy photons. The main reason for the three times larger standard uncertainty is the limited knowledge of the so-called kQ factor, which corrects for the response of the ionization chamber to the actual user beam quality Q (here: 12C) compared to the reference beam quality Q0 (here: 60Co) [TRS-398, IAEA, 2000]. In order to significantly improve the dosimetric accuracy, high precision water calorimetry using the transportable water calorimeter of the German National Metrology Institute (PTB) has been performed in the entrance channel of a scanned 6 cm x 6 cm radiation field of 429 MeV/u carbon ions at the Heidelberg Ion-Beam Therapy Center. This allowed for a direct calibration of ionization chambers and thus for the experimental determination of kQ. To substantially reduce measurement uncertainties, the irradiation parameters and the radiation field were characterized in great detail. In total, three separate series of measurements were performed to determine kQ for the two Farmer-type ionization chambers FC65-G (IBA) and TM30013 (PTW). We achieved an unprecedented standard measurement uncertainty of 0.8% for the experimental kQ values, which correspond to about a threefold reduction of the uncertainty compared to calculated values. This result shows the potential of high precision water calorimetry to significantly reduce the overall uncertainty related to ionization-based dosimetry of clinical carbon ion beams, which will be extended in the near future.

PTC17-0092: Experimental Investigation of the Linear Energy Transfer Dependent Response of Radiophotoluminescence Glass Dosimeter to Heavy Charged Particles

W. Chang1, S. Hidetoshi1, S. Kiyomitsu1, K. Yusuke 2

1Tokyo Metropolitan University, Radiological Sciences, Tokyo, Japan 2National Institute of Radiological Sciences- QST, Center for Radiation Protection Knowledge, Chiba, Japan

Radiophotoluminescence glass dosimeters (RGDs) are well-established dose verification tool in conventional X-ray radiotherapy. However, the linear energy transfer (LET) dependent response make the application in heavy charged particle (HCP) therapy difficult. According to the track structure theory, the detector response varies with not only the energy but also the type of irradiated particle. As a prerequisite for the application of RGD in heavy charged particle therapy, this work aim to investigate the RGD response to HCPs.

RGDs (GD-302M, Asahi techno.) were calibrated using 60Co γ-ray. To clarify the dependence of RGD's response in particle species and in energies, RGDs were irradiated at different water equivalent depths, i.e. different LET, in a 70 MeV u−1 proton and a 290 MeV u−1 carbon beam.

We show the LET dependence of RGD's response which was normalized at 0.2 keV μm−1 correspond to 60Co γ-ray. And the black circle shows the value obtained from past report. The difference of RGD's response between proton and helium is about 4 % when LET is around 2 keV μm−1, and it was about 10 % between proton and carbon when LET is around 10 keV μm−1.

In this work, response of RGD to HCP beams were reported. The response of RGD in helieum beam will be investigated in the future. Based on these results, the future work aims at developing a model for the RGD response to HCP as a function of particle type and energy.

PTC17-0109: Beam Tuning for Scanning Irradiation System at SAGA-HIMAT

M. Kanazawa1

1SAGA-HIMAT, Ion Beam Therapy Center, Tosu, Japan

In SAGA-HIMAT, 620 patients have been treated by use of two irradiation rooms in 2015 financial year, where passive irradiation method is adopted. To increase treatment capacity of our facility, we have started the construction of the third treatment room at the beginning of 2014 with a scanning irradiation system. In the new treatment room (room C), there are horizontal and vertical irradiation courses. This construction was required to carry out without interruptions on the treatments in room A and room B. At the end of 2016 financial year, the system tests are almost scheduled to be ready for treatment. In this presentation, we will give beam test results of our scanning system.

PTC17-0123: Secondary Doses from Proton Spot Scanning: TEPC Measurements and Monte Carlo Simulations

O. Ardenfors1,2, A. Dasu3,4, J. Lillhök5, L. Persson5, I. Gudowska1,2

1Medical Radiation Physics, Department of Physics- Stockholm University, Stockholm, Sweden 2Medical Radiation Physics, Department of Oncology and Pathology- Karolinska Institutet, Stockholm, Sweden3The Skandion Clinic, Uppsala, Sweden4Department of Medical and Health Sciences- Linköping University, Linköping, Sweden5Swedish Radiation Safety Authority, Stockholm, Sweden

Purpose: The rapid proliferation of proton therapy facilities has increased the interest in secondary dose assessments. This study presents theoretical and experimental evaluations of secondary doses produced in proton spot scanning therapy using tissue-equivalent proportional counters (TEPC) and Monte Carlo simulations.

Materials and Methods: The detectors were calibrated with a Cs-137 source and used to determine absorbed doses from secondary radiation produced in a cubic and a whole-body phantom. A W/e of 28 eV was used for the mixed neutron and photon field. Measurements were carried out at 50 and 100 cm distance from isocenter at 0°, 45°, 90° and 135° relative to the incident protons. Measurements were reproduced with MCNP6 and ambient neutron organ dose equivalents were determined for a brain tumor treatment calculated on a whole-body phantom computed tomography.

Results: The TEPC calibration determined from simulations and measurements agreed within 0.4%. Absorbed doses were consistently higher for higher energy irradiations with a maximum of 127 μGy.Gy−1 at 0° and 50 cm distance from isocenter. The lowest dose of 3 μGy.Gy−1 was detected at 135° and 100 cm. Simulations of the high energy irradiation agreed within 25% whereas lower energy irradiations agreed within 30% at 0° and 135° but differed up to 60% at 45°. Organ doses ranged between 1 μSv.Gy−1 (bladder) to 1 mSv.Gy−1 (eyes).

Conclusion: The results indicated that proton spot scanning leads to low secondary organ doses, largely dependent on proton energy and distance from isocenter. Additional analyses and calculations will further improve the results and conclusions.

PTC17-0127: Water Calorimetry in a Pulsed PBS Proton Beam

S. Rossomme1, R. Trimaud 2, V. Floquet 2, M. Vidal 2, A. Gerard 2, J. Herault 2, H. Palmans3,4, J.M. Denis5, D. Rodriguez Garcia5, S. Vynckier5

1Universite Catholique de Louvain, Molecular Imaging- Radiotherapy and Oncology MIRO, Brussels, Belgium 2Centre Antoine Lacassagne, Nice, France3EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria4National Physical Laboratory, Acoustics and Ionising Radiation Division, Teddington, United Kingdom5Cliniques Universitaire St-Luc, Radiotherapy and Oncology Dep., Brussels, Belgium

Purpose: To present a direct comparison between a water calorimeter and plane-parallel ionisation chambers in a pulsed PBS proton beam.

Materials and Methods: Measurements were performed using mono-layer maps of proton beams (10x10 cm 2), with incident beam energies of 96.17 MeV and 226.08 MeV, at a water-equivalent depth of 3.1 cm. The response of the calorimeter is corrected for heat transfer and non-water material inside the calorimeter. Using hydrogen-saturated high-purity water in the calorimeter, the chemical heat defect is assumed to be zero. Classical correction factors are applied to the response of ionisation chambers.

Results: We show preliminary relative differences of dose-to-water measured with the calorimeter and ionisation chambers, during two independent experimental campaigns. A small positioning uncertainty could explain that the differences obtained during campaign B are larger for the low energy beam. For campaign A, differences are larger for the high-energy beam, which cannot be explained by a positioning uncertainty.

Conclusions: The absolute relative differences between dose-to-water derived from calorimetry and ionometry are smaller than 2%, which is within the uncertainties of the TRS-398 protocol. Due to the depth-dose distribution, a depth inferior to 3.1 cm (e.g. 2cm where the gradient is lower) would be more suitable to minimise the positioning uncertainty. Further investigations are planned to confirm correction factors and improve the overall uncertainty on absorbed dose-to-water obtained using each system. The next experimental step is to perform the same comparison for a real clinical situation: a dose cube of 10x10x10 cm3, created by a superposition of mono-energetic layers.

PTC17-0128: Investigation of Clinical Implications When Using the Ultra-Fast Scanning Possibilities with a SC Gantry

A. Gerbershagen1, A.J. Lomax 2, D. Meer 2, J.M. Schippers1, M. Seidel1, D.C. Weber 2

1Paul Scherrer Institute, Large Research Facilities, Villigen PSI, Switzerland 2Paul Scherrer Institute, Center for Proton Therapy, Villigen PSI, Switzerland

In gantries employing scanning, energy modulation necessitates fast magnetic field changes. High speed field changes are well-known limitations of superconducting magnets, which are of interest for reducing gantry weight and size. We are investigating a gantry beam optics using superconducting magnets with very large beam energy acceptance of -20% to +26%, allowing treatments over large depth without change of the magnetic field.

We have investigated the necessary number of magnet settings for treatments of the indications that have been treated at PSI Gantry1. All 8275 fields delivered between 2007 and 2016 have been characterized by their maximum range and their maximum range variation (thickness). With the expected energy acceptance we found that 36.0 % of the fields can be treated with one magnet setting, 56.3 % would need two and 7.7 % would need three magnet settings. When also using a range shifter in the nozzle, we found that 98.9 % of the targets can be treated with one or two magnet settings per gantry field. Our results with Gantry1 patients confirm earlier results for other indications [K. Suzuki et al., Med Phys 38 (7) 2011].

In summary, we have demonstrated that by using a gantry with superconducting magnets with a very large beam energy acceptance and a range shifter almost all patients in proton therapy can be treated with only one or two settings of the gantry's magnetic field. Combination with a very fast degrader (a few ms per range change) enables the application of new scanning strategies.

PTC17-0147: A Pathway to Ultra-Compact, Truly Cost-Effective, Proton Therapy Solutions

S. Brink-Danan1, A. Zigler 2, S. Eisenmann1, C. Johnstone3, C. Spencer4, B. Weinfeld1, E. Papeer1

1HIL Applied Medical, Physics, Jerusalem, Israel 2Hebrew University in Jerusalem, Physics, Jerusalem, Israel3Fermi National Accelerator Laboratory FermiLab, External Beams, Batavia, USA4Stanford University, SLAC National Accelerator Laboratory, Menlo Park, USA

For much of the past two decades - since proton acceleration by laser was first demonstrated – its application to proton therapy has been dismissed as “being 20 years away from clinical practice”. Concerns abound – often justifiably – about wide spatial and energy distributions and “impossible” or “impractical” beamline designs, for example requiring 40 Tesla magnets.

HIL Applied Medical, together with scientists from the Hebrew University, FermiLab, SLAC, and others, have been systematically revisiting this concept.

We present experimental results of a novel acceleration scheme, yielding considerably improved energy and spatial distributions, while requiring significantly lower laser intensities (and hence smaller-size and lower-cost laser system). Enabled by advances in proton-rich, nano-structured targets that enhance the interaction with the incident laser beam, and by a unique laser-target geometry, this scheme has been demonstrated to accelerate protons to clinically-relevant energies with near-clinical dose rates. Proton energies have been recorded that are consistently 5-20 times above those achieved with other methods for laser-based acceleration, per given laser power.

We further present preliminary design of a clinical delivery system housed almost entirely in an ultra-compact gantry, 2.5m high x 4.5m long. The design allows for 360-degree isocentric rotation, PBS capabilities, and a footprint smaller than a standard LINAC vault – while utilizing magnets with B field of 4.8T at most, which are readily available today.

Combining the unique characteristics of laser-nanotarget-based acceleration, with innovative delivery system design, promises potentially groundbreaking advances in the overall compactness and capital expense of the entire proton therapy system.

PTC17-0153: Response of PTW-60019 MicroDiamond Detectors in Proton, Carbon, and Oxygen Ion Beams

S. Rossomme1, M. Marinelli 2, G. Verona-Rinati 2, P. Cirrone3, F. Romano3, A. Kacperek4, S. Vynckier1,5, H. Palmans6,7

1Universite Catholique de Louvain, Molecular Imaging- Radiotherapy and Oncology, Brussels, Belgium 2Università di Roma “Tor Vergata”, Dipartimento di Ingegneria Meccanica, Roma, Italy3Istituto Nazionale di Fisica Nucleare, Laboratori Nazionali del Sud, Catania, Italy4National Eye Proton Therapy Centre, Clatterbridge Cancer Centre, Wirral, United Kingdom5Cliniques Universitaires Saint-Luc, Radiotherapy and Oncology Department, Brussels, Belgium6National Physical Laboratory, Acoustics and Ionising Radiation Division, Teddington, United Kingdom7EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria

Purpose: To investigate the LET-dependence of PTW-60019 microDiamond detectors in particle beams.

Materials and Method: Measurements were performed in three non-modulated particle beams (proton, carbon and oxygen). The response of five microDiamond detectors was compared to the response of a Markus or an Advanced Markus ionisation chamber. The microDiamond detectors were used with their axis parallel to the beam axis and without any bias voltage. A high bias voltage was applied to the chambers, to minimise the ion recombination mechanism, for which no correction is applied.

Results: We show the ratio between the response of the microDiamond detectors and the response of the ionisation chamber in a 60 MeV proton beam and a 62 MeV/n oxygen beam, as a function of the LET. A similar figure is obtained in a 62 MeV/n carbon beam. Uncertainty bars of the measurements take into account the reproducibility of the measurement, the interpolation of the ionisation chamber data and the range determination. A LET dependence of about 0.03% per keV/μm can be estimated in the case of oxygen. A negligible dependence is thus to be expected with protons due to the lower LET-values, which is confirmed by the experimental results obtained.

Conclusion: A LET dependence was observed in the case of carbon and oxygen beams. Such an effect was found to negligible in proton beams. The under-response of the microDiamond detector could result from the recombination in the thin synthetic diamond layer, due to the high LET-values. More investigations are required to confirm this assumption.

PTC17-0158: EPID's Performance as 2D Detector for PBS Proton Beam

S. Deloule1, M. Togno1, O. Sevela1, D. Menichelli1, A. Giuliacci1, J.C. Celi1

1IBA DOSIMETRY, Physics and Innovation, Schwarzenbruck, Germany

The decrease in proton pencil beam sizes (1σ ≤ 3 mm) has raised recently a renewed interest for high-resolution 2D devices. Among the possible technologies, Si-based sensors known as EPIDs (or flats panels) offer a very good resolution (down to 100 μm), even in large sizes (up to 43x43 cm 2), and a large range of measurement. However, those devices, optimized for X-ray imaging, saturate if facing a proton beam. They also suffer for various image distortions (lag and ghosting effects), usually reduced by integrating only a few ms. Last but not least, EPIDs are known to be radiation sensitive, with a large increase in offset values under irradiation, leading to a reduced Signal to Noise Ratio and to an ineffective offset correction.

This work focused on optimizing the performance of such a device for proton beams and evaluating its radiation hardness, both in Cobalt-60 (Terabalt, at IBA dosimetry) and proton beams (PTC Prague). The stability of the panel response over 4 shifts was studied using intense irradiation sequences (in PBS mode) separated by 2h pauses, for a total of more than 1.13×106 MU delivered. The response was evaluated in terms of single spot shape at two energies (226 and 100 MeV). A follow-up of the EPID offset evolution (annealing effect) was then carried out over 3 weeks. The results were promising, with a limited background increase and a good consistency of the measured spot / field parameters.

PTC17-0166: Dependence of Proton Diode Response on Linear Energy Transfer

S. Anderson1, M. Howard1, K. Furutani1, M. Herman1, C. Beltran1

1Mayo Clinic, Radiation Oncology, Rochester, USA

Purpose: Proton diodes are an attractive tool for high spatial resolution characterization of clinical proton beams. Due to changing LET of the proton beam, however, any LET dependence of the diode response must first be understood. In this work, we measured the LET dependence of a PTW proton diode by comparing the diode response with an ion chamber as a function of LET in a clinical proton beam. A novel solid-state microdosimeter facilitated measurements of dose-weighted lineal energy, yD, a quantity related to LET.

Materials and Methods: We placed a PTW proton diode, Markus chamber, and MicroPlus Probe microdosimeter at identical depths within a water-equivalent phantom. The detectors were irradiated simultaneously with a spot-scanning proton beam of field size 20x20 cm 2. We took measurements in depth increments of 0.25 mm along pristine Bragg peaks of energy 71.3 MeV or 159.9 MeV. For one measurement, we used a variant of a ridge filter that broadened the Bragg peak. The diode response was divided by the Markus chamber response and plotted versus yD measured with the microdosimeter.

Results: We find that the ratio of diode to ion chamber response is nearly constant as a function of yD. If there were an LET dependence of the proton diode, we would expect an over-response for high yD-values. We observe no such systematic effect.

Conclusion: We find minimal LET dependence of the PTW proton diode. This detector is therefore well suited for characterization of proton radiotherapy fields.

PTC17-0174: Characterization of the Exradin W1 Scintillator for Use with Very Small Fields in Proton Therapy

C. Duzenli1, C. Penner 2, C. Hoehr3

1British Columbia Cancer Agency, Medical Physics, Vancouver, Canada 2National University of Ireland, Medical Physics, Galway, Ireland3TRIUMF, Life Science, Vancouver, Canada

Scintillator detectors are gaining importance for dosimetry measurements in radiotherapy [1], particularly for small fields. The Exradin W1 plastic scintillator is a commercially available, near-water equivalent detector, with a diameter of 1 mm and a length of 3 mm [2]. The scintillator has proven to be highly effective for small photon fields [3] and may be a good choice to perform dosimetry measurements for small proton fields typically used in the treatment of ocular cancer. W1 performance in passively scattered 74 MeV protons at TRIUMF (Vancouver, Canada) was evaluated for fields down to 5mm in diameter. Data was compared with a selection of other detectors including a BPW34 PIN photo diode (Vishay), a synthetic diamond detector (PTW) and Gafchromic film (Ashland). Depth doses, profiles, output factors, the Cherenkov light ratio, and fibre activation have been investigated. Excellent agreement between W1 and Gafchromic film for small field output factors down to 5 mm is demonstrated.

References: [1] P. Carrasco et al., Med Phys, vol 42, p. 297, 2015. [2] http://www.standardimaging.com/exradin/exradin-scintillator/. [3] T. Underwood et al., Phys Med Biol, vol 60, p. 6669, 2015

PTC17-0177: A Film Response Correction Method for EBT3 Film Based on Carbon Ion Linear Energy Transfer (LET)

W. Wang1, Z. Huang1, J. Sun1

1Shanghai Proton and Heavy Ion Center, Department of Medical Physics, Shanghai, China

Purpose: To build a carbon ion LET- dose responses table for EBT3 film, and use this table to correct the film responses.

Materials and Method: EBT3 films were sandwiched into a solid phantom and positioned parallel to the beam line. Three dose fields using the same energy of E120 (257.6 MeV/u) carbon ion beam were irradiated to the same piece of film. The particle numbers were set to create three dose levels. The irradiated doses were firstly measured by a farmer chamber located at 2 cm depth perpendicular to the beam line. Then, based on that, the doses along the beam paths were obtained using the percentage depth dose (PDD) from a treatment planning system (TPS). Three net Optical Densities (netODs) from three fields and their respective doses at the same depth were used to derive a dose calibration curve correlated with the LET there. To validate the effectiveness of LET-dose responses table, the other dose fields using E90 (221.72 MeV/u) and E150 (290.71 MeV/u) were irradiated to another piece of film using the same setup. This table was used to correct the netODs from film delivered by E90 and E150.

Results: by using this table, the corrected film PDDs for E90 and E150 showed great consistency not only in the entrance and also in the Bragg-peak region with the PDD from TPS.

Conclusion: this LET-dose responses table could effectively correct the film response in the beam entrance even in the Bragg-peak.

PTC17-0181: 2D Dosage Reconstruction of Proton Pencil Beam Scan by Using Strip Ionization Chamber

C.H. Lin1, C.H. Li 2, C.W. Hsieh 2, P.R. Tsai1, P.K. Teng1, A.E. Chen3, F.X. Chang4, C.C. Lee5, H.C. Huang6, C.Y. Yeh6

1Academia Sinica, Institute of Physics, Taipei, Taiwan- Province of China 2National Chiayi University, Department of Electrical Engineering, Chiayi, Taiwan- Province of China3Department of Physics, National Central University, Jhong-li, Taiwan- Province of China4Liverage Technology Inc., -, Hsinchu, Taiwan- Province of China5Chang Gung University, Department of Medical Imaging and Radiological Sciences, Linkou, Taiwan- Province of China6Linkou Chang Gung Memorial Hospital, Proton and Radiation Therapy Center, Linkou, Taiwan- Province of China

The fast pencil beam scanning (PBS) system has demonstrated its advance in dose delivery. The size of pencil beam is very small with few mm sigma only. Existing 2D dosage detectors with 7 to 8 mm channel spacing could not provide sufficient measurement resolution. In this paper, we report a novel method to reconstruct 2D relative dosage distribution by a XY strip detector, CROSS, with 20kHz sampling rate. With such high sampling rate, the shape of proton pencil beam can be measured precisely even during the high speed scanning up to 20mm/msec. It allows a good construction of the dynamic beam profile model and then rebuild the relative 2D dosage map. This method has been verified in the system of continuous proton beam scanning and was compared with the film measurement. The good agreement is achieved.

PTC17-0191: Dosimetric Evaluation of an Endorectal Balloon with EBT3 Film for In-Vivo Rectal Dosimetry in Proton Therapy for Prostate Cancer

Y. Goh1, S. Min1, E. Jeang1, K.H. Cho1, S.H. Choi 2, K. Jo3, S.B. Lee1, D. Shin1, U.J. Hwang4, Y.K. Lim1

1National Cancer Center, Proton Therapy Center, Goyang-si, Republic of Korea 2Korea Institute of Radiological and Medical Sciences, Departments of Radiation Oncology, Seoul, Republic of Korea3Samsung Medical Center, Proton Therapy Center, Seoul, Republic of Korea4National Medical Center, Departments of Radiation Oncology, Seoul, Republic of Korea

Purpose: We developed a prototype of two-dimensional dosimetric endorectal balloon (2DD-ERB) for prostate cancer radiotherapy. The purpose of this study is to dosimetrically validate 2DD-ERB for two-dimensional in vivorectal dosimetry during proton therapy for prostate cancer.

Materials and Methods: The 2DD-ERB was equipped with an unfoldable radiochromic film, which consisted of seven pieces of EBT3 film. This unfoldable film installed on the balloon was designed to be unrolled or rolled as the balloon inflated or deflated by filling water. We selected 10 prostate cancer patients who were treated by proton therapy using the double scattering (DS) technique in our hospital, and performed patient-specific quality assurances (pQAs) using a rectal phantom and 2DD-ERBs for all patients. 10 dose distributions corresponding to 10 patients were obtained from pQAs. The distorted dose distribution due to film gaps was recovered by a dose interpolation method. The absolute dose profile, the gamma passing rate and the dose-volume histogram (DVH) after the interpolation were evaluated for every patient.

Results: The delivered dose profiles agreed well with planned dose profiles for all patients. The gamma passing rates of the measured dose distributions were above than 90% for DS treatment plans under the criteria of 5%/3 mm. The DVH of anterior rectal wall obtained from the 2DD-ERB also agreed well with the plans.

Conclusion: The feasibility of 2DD-ERB was investigated for two-dimensional in vivo rectal dosimetry during proton therapy for prostate cancer. 2DD-ERB is expected to be a clinically useful tool for this purpose.

PTC17-0192: The Development of New Dosimetry System Using Cherenkov Radiation Sensors with an Optical Disk for Pencil Beam Scanning Mode

S. Cho1, J. Son1, Y. Lim1, S.B. Lee1, M. Yoon 2, S.Y. Park3, K. Hyung Jun1, D. Shin1

1National Cancer Center, Proton Therapy Center, Koyang-si, Republic of Korea 2Korea University, Bio-Convergence Engineering, Seoul, Republic of Korea3Great Lakes Cancer Institute, McLaren Regional Medical Center, Flint, USA

In this study, we characterized the Cerenkov radiation generated from an optical disk as fundamental research for the development and evaluation of a new dosimetric system for proton therapy using an array of optic disks Cerenkov radiation sensors. The aim of this system is to replace the depth profile measurement of Bragg peak or SOBP (spread-out Bragg peak) which is presently implemented by MLIC (multi-layer ion chamber) system.

The Cerenkov radiation due to the proton beam was measured using a homemade phantom, consisted of Blue wavelength shifting (B-WLS) optical disk surrounding with Green Wavelength shifting (G-WLS) optical fiber. It was connected to each channel of a multi-anode photomultiplier tube (MAPMT). Data were recorded using a National Instruments data acquisition system (NI-DAQ) and notebook computer. The currents from MAPMT pass through a current-to voltage converter circuit, with voltage signals from these converters acquired by the NI-9205 modules.

The characteristics of Cerenkov radiation generated from B-WLS optical disk were investigated. The relationship between Cerenkov light yield of B-WLS optical disk and dose rate measured by proton current was linear. The effectiveness of the system in proton dosimetry was assessed by comparing the results obtained using this system with those obtained using the conventional ion chamber system. PDDs measured using B-WLS optic disk show good agreement with the PDDs measured with ion chamber system for proton beams of 227.1 MeV. And B-WLS optic disk was superior to ion chamber in real-time data acquisition and cost effectiveness.

PTC17-0194: Methodology for Beam Range Determination for Gamma Electron Vertex Imaging System

S.H. Kim1, J.H. Park1, H.R. Lee 2, Y. Ku1, C.H. Kim1

1Hanyang University, Department of Nuclear Engineering, Seoul, Republic of Korea 2Korea Atomic Energy Research Institute, Neutron Utilization Technology Division, Daejeon, Republic of Korea

We have developed a prototype of gamma electron vertex imaging (GEVI) to monitor proton beam range by acquiring prompt gamma distribution. It was experimentally confirmed that the GEVI system can monitor proton dose distribution, but the falloff position on a GEVI image does not directly match with the beam range. In the present study, therefore, we developed a new methodology for beam range measurement, which is composed of three steps: (1) a sensitivity map, to flatten the distribution of imaging efficiency in an extended area, is produced and applied to the PG distribution, (2) a falloff position in the corrected PG distribution is determined with the sigmoidal curve fitting method, and (3) a correlation function between the PG falloff position and the real beam range is obtained. The correlation function can be then used to estimate the proton beam range. The methodology was tested by Geant4 simulations for proton beams of different energies (108-192 MeV) and proton numbers (1.0×108-8.0×108). The application of the sensitivity map flattened the imaging efficiency distribution in the area of 200×100 mm 2, which is two times larger than the previous one. Based on the falloff positions using 8.0×108 protons, the correlation function was established with adjusted R 2 of 0.9982. Our simulation results show that the proton beam range can be determined within ∼1% of error (= 1 σ) for 5.0×108 or more protons. This simulation study will be extended to experimental study using a therapeutic proton beam, and the results will be presented in the conference.

PTC17-0200: Development of an Online Dose Distribution Monitoring System by Using a Fluorescent Screen

F. Ito1, T. Hasegawa 2, M. Maeda1, K. Kume1

1The Wakasa Wan Energy Research Center, Research & Development, Tsuruga, Japan 2Hasetech LLC, Tsuruga Div., Tsuruga, Japan

An online verification of the radiation dose distribution is one of the noticeable technique for the evolution in the charged hadron therapy. We have been developing a novel dose measurement method using a fluorescent screen to realize an economical, simple, and easy real-time online dose distribution monitoring system.

A fluorescent screen was irradiated by a broad irradiation field of 200 MeV proton beam, and a resulting fluorescent emission was observed by a video camera with 30 images per a second. Captured images were accumulated and analyzed using Image-J (1.51i 17) software. In addition, we simultaneously measured the dose distribution with a 2-D ionization chamber array.

The distribution of the emission intensity of an accumulated image was compared to the dose distribution obtained by the 2-D ionization chamber array. Both of the distributions show a similar trend with some differences. The emission intensity bears a proportionate relationship to the dose at the irradiation field. The emission intensity has large margins of error with the reproducibility. This result suggests the following that the system with the fluorescent emission can give us a real-time, efficient and economical online dose distribution verification system with a dose distribution calibration equipment.

PTC17-0201: LET Dependence Correction of EBT3 Films in Proton Dosimetry Using Modified Bimolecular Chemical Reaction Model

M. Lee1, S. Ahn 2, Y. Han1,2, W. Cheon1

1Sungkyunkwan University, Department of Health Science and Technology- SAIHSTSamsung Advanced Institute for Health Sciences & Technology, Seoul, Republic of Korea 2Samsung Medical Center, Department of Radiation Oncology, Seoul, Republic of Korea

Purpose: The use of GafChromic films in proton dosimetry has been tried many years, but dependency on linear energy transfer (LET) and thus under-estimation of dose at high LET (Bragg peak region) has hindered the application to proton dosimetry. Therefore, we have investigated a method to correct the under-dose effect with energies from 70 to 190 MeV.

Materials and Method: For depth dose measurement, EBT3 film was placed parallel to the beam axis in water. A set of films was exposed to scanned pristine 112 MeV proton beam with 6 different dose intensities ranging from 0.373 to 4.865 Gy. This set of data was used to determine empirical parameters and calculate LET correction factor based on bimolecular reaction model. Another set was irradiated with energies ranging from 70 to 190 MeV. This set was applied with the correction method using the same parameters determined above. Finally, the corrected data was compared with the depth-dose measured with Zebra (IBA-Dosimetry, Schwarzenbruck, Germany).

Results: Bragg peak position was compared with Zebra data. The discrepancy of range was within submillimeter accuracy for all measured energies. In case of depth-dose, the dose discrepancy was within 4% in the entrance region when normalized at the peak.

Conclusion: Depth-dose of scanned pristine proton beam was evaluated by using EBT3 film. Furthermore, EBT3 film has potential as a modality for evaluating depth-dose distribution of superposed proton beams such as SOBP.

Acknowledgment: This research was supported by the NRF funded by the Ministry of Science, ICT&Future Planning(2012M3A9B6055201 ,2013M2A2A7043507), Samsung Medical Center grant[GFO1130081]

PTC17-0212: A Comprehensive Study of Lateral Fall-Off for Proton Pencil Beam Scanning (PBS)

C. Winterhalter1, S. Safai1, D. Oxley1, D.C. Weber1, T. Lomax1

1Paul Scherrer Institute, Center for Proton Therapy, Villigen PSI, Switzerland

Purpose: Different combinations of collimation, edge-enhancement and pre-absorption scenarios have been investigated in order to improve penumbra in PBS proton therapy.

Materials and Methods: Monoenergetic PBS fields with ranges between 4.1cm and 30.7cm have been simulated using the TOPAS 3.0.p1 Monte Carlo tool. Lateral penumbras (80-20%) at the Bragg peak have been evaluated for the following: Collimation alone (field-edge collimation of uniformly weighted beams), edge-enhancement alone (optimization of field-edge pencil beams) and edge-enhanced collimation (optimization of field-edge-collimated pencil beams). In order to deliver ranges <4.1cm, the additional effects of fixed (for the whole field), automatic (inserted only for the delivery of low energy Bragg peaks within the field) or variable (8x5mm mini-range shifter plates) pre-absorbers have also been investigated.

Results: For a 10cm airgap (collimator/pre-absorber to patient), collimation alone improves penumbra in comparison to edge-enhancement only for ranges between 4-10cm, whereas edge-enhanced collimation reduces penumbra by up to 2.6mm compared to edge-enhancement alone. For ranges <4cm, penumbra could be reduced by up to 2.7mm for the automatic compared to fixed pre-absorber approaches, whilst it could be reduced by an additional 2.0mm assuming a variable pre-absorber. For all combinations, penumbras are sharpest if the pre-absorber is downstream of the collimator and become progressively sharper as the airgap is reduced.

Conclusion: For clinically relevant airgaps, collimation alone improves penumbra only for depths of 4-10cm. Sharpest penumbras are achieved with a combination of collimation, edge-enhancement and a variable pre-absorber downstream of the collimator.

PTC17-0222: Non-Isocentric Scanned Proton Beam Delivery at MedAustron: Performance and Commissioning Specificities of the Beam Delivery System

L. Grevillot1, G. Kragl1, H. Palmans1,2, S. Vatnitskiy1, J. Osorio1, V. Letellier1, R. Dreindl1, A. Carlino1,3, M. Stock1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2National Physical Laboratory, Dosimetry, Teddington, United Kingdom3University of Palermo, Physics and Chemistry, Palermo, Italy

The increased spot size at patient entrance, due to beam scattering in the nozzle and beam divergence propagation in the air gap between the nozzle exit and the patient entrance is one of the main drawbacks of proton pencil beam scanning (PBS). We report on the commissioning specificities of our proton fixed beam line used for non-isocentric treatment delivery, when patients are moved towards the nozzle to reduce the air gap.

A non-isocentric reference point (NIRP) 50cm closer to the nozzle exit than the isocenter has been defined (15 cm air gap). In addition to the baseline data acquired at isocenter, baseline data have been acquired at the NIRP for a reduced set of energies representative of the 255 energies available for treatment and comprises: depth-dose profiles in water, spot maps in air with and without a 3cm range shifter, square fields in air and RW3, output factors (frame and field size factors), 3D cubes in water (absolute dose, SOBP) and in RW3 (transverse profiles) for shallow, intermediate and deep seated targets.

In open beam configuration, non-isocentric beam delivery allows reducing the spot sizes at patient entrance by a factor of about 2 at the lowest energies, corresponding to lateral penumbra reduction of about 30% for shallow cubic targets (6×6×6 cm centered at 6 cm). With range shifter, non-isocentric beam delivery allows reducing the spot sizes at patient entrance by a factor of about 3. Non-isocentric set-up improves the performance of scanned proton beam delivery at our fixed beam line.

PTC17-0234: Next Generation of Fast Cycling Hadron Driver with Continuous Energy Sweep Scanning

K. Takayama1, T. Adachi1, K. Okamura1, T. Kawakubo1, T. Sakae 2, K. Takada 2

1High Energy Accelerator Research Organization, Accelerator Laboratory, Tsukuba, Japan 2University of Tsukuba, Proton Therapy Center, Tsukuba, Japan

Now days, 3D spot scanning of hadron beams applied to cancer tissues in human organs is of great concern in society. Several ways to realize such spot scanning in depth are known. Beside the well-known classical technique of using a range shifter, a technique of changing the extraction beam energy itself has been demonstrated in the slow-cycling proton synchrotron, where the excitation ramping pattern of guiding magnets is varied by using a different acceleration cycle and the beam is continually and slowly extracted on the flat top by a combination of third-integer resonance and RF deflection. Respiration gating operation is realized by real-time feedback of human body information. The change in the ramping pattern is not continuous but discrete, changing pulse by pulse. The smoothness of spot scanning in depth is intrinsically limited by this characteristic. Continuous irradiation of moving organs, which requires energy sweeps of a few tens of milliseconds, has not been realized yet.

Recently, the idea of energy sweep extraction in a single acceleration cycle from a fast-cycling synchrotron has been proposed, where a fast cycling induction synchrotron is employed as a driver, a beam is continuously extracted by a novel beam handling method to allow continuous beam spill loss from the barrier bucket. For this purpose, the ring lattice with a large flat dispersion function and the electrostatic septum driven by a varying output voltage DC power supply are indispensable. Status of recent R&D works will be presented at the conference.

PTC17-0255: Preliminary Characterization of the Neutron Field in Top-Implart Proton Therapy Facility

M. Vadrucci1, P. Ferrari 2, L. Campani 2, F. Mariotti 2, L. Picardi1, C. Ronsivalle1

1ENEA, Development of Particle Accelerators and Medical Applications, Frascati, Italy 2ENEA, Radiation Protection Institute, Bologna, Italy

The proton therapy facility TOP-IMPLART is under construction at the ENEA Frascati research center. The phase I of the project, funded by the local Government, concerns the realization of a linear accelerator of 150 MeV protons upgradable to 230 MeV for deep solid tumors treatment in a second phase.

The accelerator is based on a sequence of modular resonance cavities accelerating protons to the final desired energy. The low-energy part is a 7 MeV proton injector operating at 425 MHz. Additional modules operating at 3gHz increase the proton energy to higher levels .

Currently the four first modules have been installed supplying a proton exit energy of 35 MeV. A series of numerical calculations have been carried on to characterize the main field and the secondary neutron radiation field due to the interactions of protons with the treated volume.

Secondary neutron radiation has been simulated in a water tank and evaluated along the beam axis and out of the primary field and repeated at different angles with respect beam axis, out of tank, in order to map the scattered field around the target volume. A set of measurements have been performed through CR-39 track detectors and rem-counters. In that case the limited energy of the current source set-up allows measurements with PADC, that are known to under-respond for energy higher than 15 MeV.

These preliminary results are aimed to test the feasibility of the measurements and validate the simulations. Further simulations and measurements are foreseen following the subsequent construction steps of the accelerator.

PTC17-0262: Spectral and Spatial Shaping of a Laser-Accelerated Proton Beam for Radiobiology Experiments

E. Bayart1,2, L. Pommarel1, B. Vauzour1,3, O. Delmas1, F. Megnin-chanet 2,4, F. Goudjil5, C. Nauraye5, E. Deutsch 2,6, A. Flacco1, V. Malka1

1ENSTA ParisTech- CNRS- Ecole polytechnique- Université Paris-Saclay, Laboratoire d'Optique Appliquée, PALAISEAU, France 2Gustave Roussy- INSERM- Université Paris Saclay, U1030, VILLEJUIF, France3CEA, DAM- DIF, ARPAJON, France4Institut Curie- INSERM- CNRS- Université Paris Saclay, INSERM U1196 / CNRS UMR9187, ORSAY, France5Institut Curie, Centre de protonthérapie, ORSAY, France6Gustave Roussy, Radiation therapy departement, VILLEJUIF, France

Radiation therapy is a cornerstone of cancer management. Because of the favorable ballistic of proton beams, proton therapy has proven its clinical effectiveness, particularly in pediatrics optimizing antitumor efficacy while decreasing treatment-associated side effects. Moreover, ultra-high dose rate irradiations (>40 Gy/s) have been shown to increase the differential response between normal and tumor tissue. Recent developments of laser-plasma based particle accelerators made possible the production of pulsed proton beams with exceptional parameters in terms of peak dose rate (108 Gy/s) and duration (ns), with great potential therapeutic applications. As laser-accelerated beams display a broad energy spectrum and a wide angular large angular divergence produced by the target normal sheath acceleration (TNSA) regime, a proper beamline needed to be set up in order to produce clean and controllable irradiation conditions compatible with radiobiological studies.

We will present the realization of a beam transport system to shape and control the proton beam from the multi-terawatt laser SAPHIR. To provide a wide irradiation surface and uniform dose distribution, a set of four permanent magnet quadrupoles was used to transport and focus the beam, and efficiently clean the spectrum. Shot-to-shot dosimetry was also implemented and calibrated for absolute dose deposition evaluation in the irradiation volume. Thanks to these improvements, protocols for dose delivery to biological samples were developed and validated. Much is to be explored on the biological impact of protons delivered in such short pulses and ultra-high dose rates on living tissues.

PTC17-0269: Response of Bragg Peak Ionization Chambers in Photon, Electron and Proton Beams

J. Osorio1, R. Dreindl1, L. Grevillot1, N. Zagler 2, S. Markus1, S. Vatnitsky1, H. Palmans1,3

1MedAustron, Medical Physics, Wiener Neustadt, Austria 2Landesklinikum Wiener Neustadt, radiotherapy department, Wiener Neustadt, Austria3National Physical Laboratory, NPL, Hampton Road- TW11 0LW Teddington, United Kingdom

Bragg Peak ionization chambers (BPIC) are plane-parallel ionization chambers (PPIC) that have been used mainly in relative particle beam dosimetry and recently for absolute particle beam dosimetry and small photon field dosimetry.

The lack of primary standards for protons and carbon ion beams to calibrate this type of PPIC, necessitates their cross-calibration in other beam qualities, ideally in a scattered proton field or alternatively in high-energy photon or electron beams. This work presents an experimental study on the response in terms of Dose Area Product (DAP) of BPIC chambers in three different beam types; a 6 MV photon beam, a 22 MeV electron beam and a 179.2 MeV scanned proton beam. Special attention was paid to the accurate establishment of all correction factors (ion recombination, polarity effect, lateral dose correction factors, etc.) and their dependencies with the used beam quality during the BPIC cross calibration. The experimental variation of the response was compared with theoretical beam quality correction factors calculated according to the Appendices of TRS-398.

The relative differences between the response reported by the manufacturer (in Cobalt-60 free in air) and the measured BPIC responses are: 9.7% for the electron beam, 0.1% for the photon beam and -1.3% for the active scanned proton beam. These values are within the uncertainties consistent with the relative variation of the theoretical beam quality correction factors. This absolute dosimetry characterization of BPIC provides an initial verification step towards the implementation of absolute dosimetry protocols based on BPIC.

PTC17-0271: Absolute Dosimetry in a Prototype Scanned Pulsed High Dose Rate Proton Beam

M. Vidal1, A. Gerard1, V. Floquet1, S. Rossomme 2, S. Vynckier 2,3, J. Herault1

1Centre Antoine Lacassagne, Institut Méditerranéen de Protonthérapie, NICE, France 2Institute for Experimental and Clinical Research- Université catholique de Louvain, Molecular Imaging- Radiotherapy and Oncology, Brussels, Belgium3Clinique Universitaire St-Luc, Radiotherapy and Oncology, Brussels, Belgium

Purpose: To present absolute dose/monitor unit calibration protocol for the new prototype Proteus®One proton therapy system.

Materials and Methods: The supraconducting synchro-cyclotron has a frequency of 1kHz and delivers a pulsed proton beam of dose rate between 2.65 μGyE/pulse and 230 μGyE/pulse for the lowest (96 MeV) and highest (226 MeV) energies respectively. Absolute dose/monitor unit calibration was based on dose measurements in single-layer fields of size 10x10 cm 2. Plane parallel ionization chamber was used with -500V voltage to limit the effect of recombination. Following the TRS398 protocol, temperature and pressure (kT,P), polarity effect (kpol), beam quality factor (kQ,Q0) and recombination effect (ks) correction factors were computed and applied to the considered ionization chamber. The TRS398 pulsed beams model and three additional Boag's models were investigated to compute the fraction of free electron for the recombination effect correction, while the kQ,Q0 factor evaluation was based on Monte-Carlo Goma's results for PBS proton beams.

Results: We show that the standard deviations between the ks values obtained with the four methods are not larger than 0.02% and 0.3% for the lowest and the highest energies respectively. ks value of 1.009 was selected for all the energies. We sum up the correction factors applied to the ionization chamber charge response to convert it into absolute dose (Gy). Water calorimetry measurements were also achieved for single-layer fields of size 10x10 cm 2 with the lowest and highest ranges, and differences between ionometry and calorimetry measurements were below 2%.

PTC17-0314: Correcting for Linear Energy Transfer Dependent Quenching in Radiochromic Three Dimensional Dosimetry of Proton Therapy

E.M. Høye1, M. Sadel 2, L.P. Muren1, J.B.B. Petersen1, P.S. Skyt1, L.P. Kaplan 2, J. Swakon3, L. Malinowski3, G. Mierzwińska3, P. Balling 2

1Aarhus University Hospital, Department of Medical Physics, Aarhus, Denmark 2Aarhus University, Department of Physics and Astronomy, Aarhus, Denmark3Polish Academy of Sciences, Institute of Nuclear Physics, Krakow, Poland

3D dosimetry allows detailed measurements of dose distributions in photon-based radiotherapy, and has potential to become a useful tool for verification of proton therapy. Linear-energy-transfer (LET) dependent quenching of the signal in the Bragg peak results in an under-response in most dosimeters. In this study we investigate whether LET-dependent quenching in a 3D dosimeter can be corrected by adjusting the measured optical density (OD) distribution using a calibration model generated from a Monte Carlo (MC)-simulation and a calibration dosimeter.

Two radiochromic (leucomalachite green) silicone-based 3D dosimeters were irradiated with unmodulated 60 MeV proton beams with entrance doses from 5 to 20 Gy. Read-out was performed before and after irradiation with optical CT-scanning. Dose and dose-averaged LET (dLET) distributions were calculated using MC-simulations. A calibration model was generated from two spatially separated beams in dosimeter A: Linear fits to OD as a function of dose was computed for all voxels within each limited dLET-range. The third beam in dosimeter A and dosimeter B were calibrated using these variables.

Quenching resulted in lower dose response in the Bragg peak (0.012 cm−1 Gy−1) compared to low-LET regions (0.028 cm−1 Gy−1 at 3.1 keV/μm). MC-simulated dose and calibrated measurement agreed well. Errors > 5% of maximum dose was found in 12% and 9% of all calibrated voxels in dosimeter A and B respectively. Discrepancies were mainly located in the lateral beam-edges.

We have presented the first LET-corrected dose measurements in a 3D dosimeter for proton therapy, showing that dose verification for single fields is possible.

Acknowledgment: PL-Grid

PTC17-0317: Monte Carlo Modeling of a Proton Fixed Beam Line Featuring Non-Isocentric PBS Treatment Capabilities

A. Elia1,2, L. Grevillot1, A. Carlino1,3, T. Böhlen1, H. Fuchs1,4,5, M. Stock1, D. Sarrut 2, J. Osorio1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2CREATIS, Université de Lyon- CNRS UMR5220- Inserm U1044- INSA-Lyon- Université Lyon 1- Centre Léon Bérard- 69007 Lyon- France, Lyon, France3University of Palermo, Department of Physics and Chemistry, Palermo, Italy4Medical University of Vienna / AKH, Department of Radiation Oncology, Vienna, Austria5Christian Doppler Laboratory for Medical Radiation Research for Radiation Oncology, Medical University of Vienna, Vienna, Austria

In non-isocentric proton scanned pencil beam treatments, it is of paramount importance to precisely describe beam optics for different air-gaps. Indeed, nozzle components and air-gaps drastically affect the lateral dose profile due to the multiple coulomb scattering. This work shows an extended Monte Carlo proton beam modeling based on final medical commissioning data, with special emphasis at non-isocentric conditions.

A full description of the nozzle design was implemented. Physics-builder QBBC_EMZ was selected in Gate 7.2/Geant4 10.02. Beam optics parameters (beam size, divergence and emittance) were adjusted to match the measured spot sizes in air at 7 different air gaps for 20 representative energies. The energy parameters (mean energy and energy spread) were optimized in order to match the measured beam range (R80) and Bragg peak width evaluated in water.

Absolute mean deviation from measured data used for beam modeling was 0.2 mm for FWHM at ISO and non-ISO positions. Higher deviations were observed at non-relevant clinical conditions: 62.4 MeV at position ISD20cm (6.6% relative agreement). Maximum deviation up to 10% was observed for 97.4 MeV at isocenter using range shifter. Residuals in terms of physical range were always within 0.1 mm. Discrepancies up to 2.5% at worst were found in depth dose profile while Bragg peak width mean deviation at 80% of maximum dose was 0.1 mm, corresponding to a relative deviation of 3%.

Presented results in 1D are very promising and further validation in 2D, 3D and absolute dose is ongoing.

PTC17-0329: Assessment of Accelerator Upgrade Scenarios and Associated Improvement of Treatment Delivery Times at a Synchrotron-Based Particle Therapy Centre

T. Böhlen1, G. Kragl1, G. Kowarik 2, P. Urschütz 2, M. Stock1, J. Osorio1

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2EBG MedAustron GmbH, Therapy Accelerator, Wiener Neustadt, Austria

Purpose: MedAustron is a synchrotron-based dual-particle quasi-discrete PBS therapy facility which started patient treatment in the end of 2016. The MedAustron Particle Therapy Accelerator (MAPTA) is developed and maintained in-house. With the advent of CE-marking and clinical operation, basic accelerator performance and design parameters have been achieved. Several MAPTA upgrade scenarios exist and prioritization should be driven by clinical benefits and needs. In this study, the impact of different upgrade scenarios on treatment delivery times was investigated.

Materials and Methods: In a first step, possible MAPTA upgrade scenarios were identified and current MAPTA performances and constraints, were evaluated, gathered and reviewed. Performance parameters included intensity, number of particles (NP)/spill, degradation and cycling times, while constraints included: min NP/spot, min time/spot, max permissible intensity and degrader settings. Then, an automated intensity selection and a parametric modelling of treatment delivery time was developed and validated for current MAPTA performances. Finally, predictions for different upgrade scenarios have been obtained which respect delivery constraints.

Results: Comparisons of simulated and measured delivery times showed generally an agreement within 10%. Identified MAPTA upgrade scenarios included: improved intensity selection, increased admissible max intensity, variable spill length, reduced inter-spill dead time, NP/spill increase, cycle-shortening for unused beam, multi-energy extraction and the usage of a ripple filter. We show reductions of delivery times obtained for different upgrade scenarios for exemplary plans.

Conclusion: The evaluation of merits of possible MAPTA upgrades in terms of treatment velocity provided quantitative information for decision making and allows prioritizing future developments.

PTC17-0330: A Leaf Positioning Algorithm for Mimicking Patient-Specific Apertures with the Mevion Adaptive Aperture

J. Cooley1, T. Smith 2, D. Catanzano1, K. Huang1, A. Ikica3

1Mevion Medical Systems, Advanced Development, Littleton, USA 2Dartmouth College, Physics, Hanover- NH, USA3Cosylab, Ljubljana, Slovenia

The Mevion Adaptive Aperture is a mechanically actuated, travelling multileaf collimator designed to improve the lateral penumbra of pencil-beam-scanned proton therapy treatments. In order to minimize size and complexity, the device was designed to shape only a small number of spots at one time, but to move with the scanned beam in order to collimate at any location in a treatment field. In this way, the Adaptive Aperture can dosimetrically replicate the effect of machined static apertures and blocks. This is an important feature as many currently available treatment planning systems have the capability of generating treatment plans that prescribe patient-specific static apertures.

We will present the details of a leaf-positioning algorithm that converts the static aperture curve and spot maps prescribed in a spot-scanned clinical treatment plan into a sequence of Adaptive Aperture leaf configurations. The algorithm is designed to optimize conformality to the aperture curve while minimizing the number of leaf positions in order to minimize the impact on treatment time. For several example collimated treatment plans, we will present leaf position sequences as well as Monte Carlo simulation results comparing the dose delivered with the Adaptive Aperture with that from the equivalent static aperture. An example of a floating block inside a field will be included to demonstrate the potential for the Adaptive Aperture to improve spot size limitations anywhere in a treatment plan.

PTC17-0331: Initial Quantification of the Beam Characteristics of the First Mevion S250i Hyperscan System

D. Pang1, H. Chen1, T. Zwart 2, S. Rosenthal3

1Georgetown University Hospital, Radiation Medicine, Washington- DC, USA 2Mevion Medical Systems, Product Development, Littleton- MA, USA3Mevion Medical Systems, Applications, Littleton- MA, USA

Purpose: To present the initial measurement results of the beam characteristics of the first Mevion S250i HYPERSCAN system.

Materials and Methods: The pencil beam spot size, position, profile and penumbra produced by the Mevion S250i HYPERSCAN system was measured using Matrixx and Octavius ion chamber arrays, and EBT3 films. The integrated depth dose (IDD) was measured using a plane parallel chamber. The energy switching speed was measured for a 14x14 cm 2 field size using machine log files.

Results: The spot sigma in air is 5±0.2 mm; the spot position accuracy is 0.5 mm; the flatness and symmetry are 3 and 4 %, respectively; the penumbra is 2.05±0.3 mm with the adaptive aperture. The time to scan a 1-litter volume is 6 S, and to deliver 2Gy dose to a 10cm3 water volume is 60 seconds. The energy switching time is 50 ms.

Conclusion: The Mevion S250i HYPERSCAN integrates the cyclotron, gantry and beam scanning module into a single compact system, eliminating the typical beam transport system required for other proton systems. Such a compact configuration has effectively reduced the footprint to permit installation in hospitals with tight operating space. The unique design has achieved a proton utilization efficiency of more than 99%, instead of the typical 1%. Furthermore, the rapid energy switching and penumbra trimming technology has achieved a scanning speed 50 times faster than that of the typical proton systems and a uniform spot size with a penumbra significantly sharper than what have been reported for the other proton systems.

PTC17-0336: LET Spectra Behind High-Density Dental and Hip Implants Irradiated with a Scanned Carbon Ion Pencil Beam

C. Oancea1,2,3,4, K. Brabcova3, I. Ambrozova3, G. Mytsin 2, S. Greilich5,6, M. Davidkova3

1University of Bucharest, Faculty of Physics, Bucharest - Magurele, Romania 2JINR, DLNP, Dubna, Russian Federation3Nuclear Physics Institute, Department of Radiation Dosimetry, Prague, Czech Republic4IFIN-HH, DRMR, Bucharest - Magurele, Romania5German Cancer Research Center DKFZ, Division of Medical Physics in Radiation Oncology, Heidelberg, Germany6Heidelberg Institute for Radiation Oncology HIRO, National Center for Radiation Research in Oncology, Heidelberg, Germany

We investigated the impact on absorbed dose and linear energy transfer (LET) spectra of a scanned carbon ion pencil beam caused by orthopedic or dental implants.

Tissue equivalent phantoms of 65 and 80 mm were used. The phantoms contained stair-like high-Z metallic components with known chemical composition corresponding to materials usually used in orthopedic and dental surgery. Those implants were made of titanium alloy (Ti) or stainless steel (AK) to represent a hip implant, and grade 2 or 5 titanium to represent a dental implant. Behind the phantoms, track etched detectors (TEDs) and fluorescent nuclear track detectors (FNTDs) were placed. The stacks of the phantom and the detectors were irradiated with carbon beam using the following parameters:

  • A nominal energy of about 400 MeV/u, the range of carbon ions coincided with the plateau region of the Bragg curve;

  • The ranges of the carbon beam at 80% of maximum dose coincided with the front side of the detector behind 15 mm hip implants, and 10 mm dental implants.

Our study did not show significant modifications of absorbed dose and LET spectra when the implants were localised in the plateau region of the Bragg curve.

However, for implants close to the Bragg curve maximum of the carbon beam, particles with very high LET between 100 and 1000 keV/μm were detected.

In conclusion, we present a complete study on the effects of metallic implants in carbon ion therapy.

PTC17-0340: Measurement and Modelling of the Time Response of the Beam Monitors Used for Fast Raster Scanning

U. Weber1, B. Voss 2, C. Schuy1, K. Zink3, C. Graeff1, K. Anderle1

1GSI Helmholtzzentrum für Schwerionenforschung, Biopyhsics, Darmstadt, Germany 2GSI Helmholtzzentrum für Schwerionenforschung, Detector laboratory, Darmstadt, Germany3University of Applied Sciences Mittelhessen THM, Institute of Medical Physics and Radiation Protection, Gießen, Germany

Purpose: The improvement of the speed for raster scanning is one of the most important prerequisites for the treatment of moving tumours such as lung or liver. One bottle neck in the existing raster scanning systems was turned out to be the gas filled beam monitors that were used for control of the scanning. Even when using fast amplifiers and fast data processing of the detectors, there remains the relatively low drifting time of the ions (from the ionized electron ion pairs) that delays the signal. In this work we modelled and measured the time response of the integral parallel plate ionization chambers (PPIC) and the positions sensitive MWPC detectors.

Materials and Methods: The PPICs and MPWCs from the GSI pilot project were tested. Four different types of gas fillings (Ar/CO2, air, Helium, Nitrogen) were compared. The output signals from the PPICs and from some wires of the MWPCs were amplified using fast current amplifiers (Stanford-SR570m) and were recorded via a fast storage oscilloscope. In order to measure the time response from a short beam pulse we used fast beam pulses (<5μs) from a 6 MeV Photon Linac (Varian) or from of the GSI synchrotron (C12, 300 MeV/u) using the fast extraction mode. The voltage of the detectors was varied from 100 to 2000V.

Results: The time response for Helium is much faster and that the drifting time is strongly dependent from the voltage.

Conclusion: Diligent selection of the detector gas and operation voltage dramatically improves the time behaviour.

PTC17-0341: A Dosimetric Analysis of Dose to Active Bone Marrow in Anal Cancer Patients Treated with Proton Therapy versus Photon Therapy

C. Freese1, M. Sudhoff1, T. Meier1, L. Lewis1, A. Mascia1, E. Wolf1, J. Kharofa1

1University of Cincinnati, Department of Radiation Oncology, Cincinnati, USA

Purpose: Patients undergoing chemoradiation for anal cancer are at risk of hematologic toxicity (HT) with rates of Grade 3-4 HT as high as 60%. Studies have demonstrated correlation between HT and the dose to total bone (TB) as well as the PET-defined active pelvic bone marrow (APBM) spared. We hypothesize that the use of intensity modulated proton therapy (IMPT) can reduce the dose to the TB and APBM compared to volumetric-modulated arc therapy (VMAT).

Materials and Methods: Comparative VMAT and IMPT plans were generated from 8 patients originally treated with chemoradiation for anal cancer. The TB volume was contoured as per Mell et al (IJROBP 2008). The APBM volume was defined as all regions within the TB with an SUV value >total body mean SUV. IMPT plans used multi-field optimization and split-target technique. Data was analyzed using paired sample t-tests.

Results: The volume of TB and APBM spared was significantly greater with IMPT plans compared to VMAT plans. IMPT plans increased the mean volume of APBM spared 10 Gy (485 cc vs 91 cc, p=0.0005), 20 Gy (569 cc vs 117 cc, p=0.0005), and 40 Gy (815 cc vs 759 cc, p= 0.04).

Conclusion: The volume of TB and APBM spared have been shown to be strong predictors of acute hematologic toxicity. IMPT planning results in significantly greater TB and APBM sparing compared to VMAT. The use of IMPT for anal cancer may result in lower HT rates in patients undergoing definitive chemoradiation through improved bone marrow sparing.

PTC17-0352: Clinical Data of Biological Washout Effect of Positron Emitter after Carbon Ion Treatment

T. Ishii1, Y. Hirano 2, K. Yusa 2, M. Torikoshi 2, K. Tsuda1, Y. Itabashi1, S. Takayuki1, T. Kanai 2, T. Ohno 2, T. Nakano 2

1Gunma Univ. Hospital, Dep. of Radiology, Maebashi, Japan 2Gunma University, Heavy Ion Medical Center, Maebashi, Japan

Purpose: Positron Emission Tomography (PET) is the useful modality for taking dose distribution related image of particle therapy. Carbon-11(half-life time:20.4min) is major positron emitter in various kinds of positron emitters, cause of the longer half-life time. We try to estimate washout effect of C-11 in a patient's body after carbon ion treatment. We acquire Autoactivation-PET(AA-PET) data in dynamic mode, then we can reconstruct a multiphase dynamic image and analyze as passed time. The biological washout effect in a patient's body is estimated by deference between decay of ROI data with theoretical decay of C-11. ROI(region of interest) is defined on CT image of Radiation treatment planning system (RTPs). We estimate the biological washout effect in a patient's body by deference between decay of ROI data with theoretical decay of C-11.

Materials and Methods: 1. Reconstruct dynamic AA-PET image (30sec. *10phase). 2. Transport AA-PET ,CT image and ROI data to 3D-WS of medical Image analysis. 3. Fusion a ROI data on RTPs to CT images of AA-PET study. 4. Analyze decay time of ROI.

Results: We can measure the washout effect of C-11 in a patient's body after carbon irradiation using off-line AA-PET. It is about 5minutes.

Conclusion: We will estimate another decay factor of positron nuclei, and create more sophisticated verification system of carbon ion irradiated region. We are able to define decay correction factor from Dynamic AA-PET study for human body.

PTC17-0353: Exploring Noise vs Pile-Up Effects for Different Sized MicroPlus Solid State Detectors as a Function of Proton Flux

B. Nelson1, S. Anderson 2, A. Rosenfeld3, M. Herman4, C. Beltran4

1Mayo Clinic Graduate School of Biomedical Sciences, Biomedical Engineering, Rochester, USA 2Mayo Clinic, Mayo School of Graduate Medical Education, Rochester, USA3University of Wollongong, Centre for Medical Radiation Physics, Wollongong, Australia4Mayo Clinic, Radiation Oncology, Rochester, USA

Purpose: Biological effects of radiation depend on microscopic patterns of energy deposition. In microdosimetry, these patterns are measured in volumes that mimic cell sizes, allowing correlation to biological effects. Solid-state microdosimeters offer advantages in microdosimetric measurements, but effects such as noise and pulse pile-up must first be understood to yield optimal measurement performance. In this work, we investigate a novel, silicon-on-insulator microdosimeter with several individual sensitive volumes (SV) of varying sizes. We characterize noise and pulse pile-up performance for each SV in a clinical proton beam.

Materials and Methods: The MicroPlus probe had several single 3D SVs, with active areas of 250, 200, 150, 100, and 50 μm 2 and a thickness of 10 μm. We made measurements in a water-equivalent phantom at depths along a pristine Bragg peak of energy 159.9 MeV. For each depth, we measured lineal energy distributions and calculated dose-mean lineal energy, yD. Proton flux was varied by placing range shifters (RS) in the beam to scatter protons and decrease flux at the detector: RS0 (No range shifter), RS25 (small scatter), RS45 (large scatter), and ERS45 (small scatter).

Results: For all cases, decreasing detector sensitive volume increased low lineal energy noise, thus lowering yD. We observe that the large sensitive volumes with high flux increase pileup, giving higher yD.

Conclusion: We examined noise and pulse pile-up properties for a solid-state microdosimeter with varying sizes of SV. We found decreasing SV increased low energy noise while higher proton flux yields pulse pileup.

PTC17-0356: Assessment of Thermal Modulation and the Quenching Effect in PRESAGE Dosimetry of Proton Beams

C.S. Wuu1, C.C. Chen 2, Y. Na1, P. Black1, J. Admovics3

1Columbia University, Radiation Oncology, New York, USA 2ProCure Treatment Centers- Inc., Radiation Oncology, Somerset, USA3Rider University, Chemistry, Lawrenceville, USA

Detailed 3-D dosimetric information of a proton beam is crucial for accurate treatment planning design and treatment delivery due to the sharp dose falloff at the end of the beam range and laterally at the field edges. Both radiochromic plastic PRESAGE and gel dosimeters, in conjunction with an optical CT scanner, have shown promising results in providing 3-D dose verification for complex photon radiotherapy such as IMRT and SRS/SBRT. However, radiation dosimeters such as EBT film, scintillators, gel dosimeters, and PRESAGE dosimeters usually show under-response at the Bragg peak (quenching effect). Our recent study has shown that the quenching effect in the PRESAGE dosimeters can be significantly reduced to less than 3% when the PRESAGE is heated to 50°C. However, the reduction of quenching effect in PRESAGE depends not only on the temperature, but also on the heating mechanism. In this study, PRESAGE dosimeters composed of a specific chemical formulation, with its temperature in the range of room temperature to 50°C, were irradiated with mono-energetic proton pencil beams of 128, 136 and 148 MeV. To assess the quenching effect in relation to the heating mechanism, both PRESAGE cylindrical phantoms of 10 cm diameter and PRESAGE films were used in the study due to their different thermal properties. Our results demonstrated that there was a 20-30% under-response in dose at the Bragg peak for dosimeters at room temperature. As the temperature of PRESAGE dosimeters increases under the specific heating mechanism, the quenching effect can be reduced to less than 3%.

PTC17-0359: Luminescence of Water during Radiation Irradiation Lower Energy Than Cerenkov-Light Threshold

S. Yamamoto1

1Nagoya University, Graduate School of Medicine, Nagoya, Japan

It is commonly thought that, no visible light is thought to be produced in water during irradiation of protons below Cerenkov-light threshold. Contrary to this consensus, we found that luminescence was emitted from water with proton irradiation below the threshold. We placed water phantoms set on a table, and luminescence images of water phantom were measured with a high-sensitivity cooled charge coupled device (CCD) camera during 100MeV proton-beam irradiation. The luminescence images of water phantoms showed clear Bragg peak, and the measured proton ranges from the images were almost the same as those obtained with an ionization chamber. We also tried the imaging of water for carbon-ion, alpha particles and low energy X-ray photons lower energy than Cerenkov-light threshold and the luminescence of water was observed and imaging was possible for all these radiations. We will also show the light output and spectra for the luminescence of water in the meeting.

PTC17-0366: Application and Consistency of PBS Beam Monitor Calibration Methods

R. Dreindl1, J. Osorio1, G. Loic1, B. Till Tobias1, C. Antonio1,2, K. Peter3,4, K. Gabriele1, V. Stanislav1, S. Markus1, P. Hugo1,5

1EBG MedAustron GmbH, Medical Physics, Wiener Neustadt, Austria 2University of Palermo, Department of Physics and Chemistry, Palermo, Italy3Department of Radiation Oncology - Division Medical Physics, Medical University of Vienna, Vienna, Austria4Christian Doppler Laboratory for Medical Radiation Research for Radiation Oncology, Medical University of Vienna, Vienna, Austria5National Physical Laboratory, Dosimetry, Teddington, United Kingdom

Purpose: This work presents the implementation of two machine-specific reference field methods and one plan-class specific reference (pcsr) field method for dosimetry of scanned proton beams.

Materials and Methods: As preparatory step of beam intensity calibration, all ionization chambers used (Farmer, Roos, Bragg Peak and PinPoint) were checked and corrected for ion recombination- and polarity-effects. Roos, Bragg-Peak and PinPoints were cross-calibrated against a calibrated Farmer ionization chamber in 179.2 MeV protons. Single-layer scanned fields (SLSFs) with identical number of protons in each beamlet and equidistant lateral spacing were used as reference method for the beam intensity monitor calibration in the clinically relevant proton energy range. As redundant method, dose-area-product (DAP) of static beamlets was measured with Bragg Peak chambers. The preliminary beam model for our TPS (RayStation 5.0.2) was afterwards tuned using pcsr-boxes of different modulation sizes centred at shallow, intermediate and large depths.

Results: SLSF and DAP-methods agreed on average within 1.0% and 1.5% for beam monitors 1 and 2. These differences are within the estimated uncertainties of both methods. Obtained baseline data led to a preliminary beam model that was subsequently adjusted by 2% in absolute dose after investigating the results from Farmer and PinPoint measurements in water using the pcsr-field approach.

Conclusion: The agreement between both dosimetry methods increases confidence in the beam monitor calibration and the estimated uncertainty. The 2% systematic correction resulting from the pcsr-field dosimetry emphasises the importance of accounting for the influence of the dose calculation model on absolute dose.

PTC17-0386: An Irradiation Control System for Spot-Scanning Mode in Shanghai Proton Therapy Facility

M. Liu1, C. Yin1, B. Zhao1, C. Miao1, Z. Chen1

1Shanghai Institute of Applied Physics- Chinese Academy of Sciences, Department of Beam Instrumentation and Control Technology, Shanghai, China

Shanghai Proton Therapy Facility is under construction. The irradiation control system was developed to achieve the spot-scanning mode of proton delivery. In order to irradiate the accuracy dose to the specified target and avoid unnecessary dose to the healthy tissues, the sensors and the controllers with various functions were designed to form the system. The ionization chambers and the related electronics detect and verify the dose and position information of the spot. The hall sensors inside the scanning magnets help to confirm the beam position and compensate the hysteresis. The irradiation controller and monitor execute and verify the whole delivery sequence. They are also linked with the accelerator timing system to control the slow extraction process and trigger the beam abort system. The simulation process, the details of the development and the preliminary results are described in the contribution.

PTC17-0405: Development of Ultra-Compact Scanning System

T. Furukawa1

1National Institute of Radiological Sciences, Medical Physics Research Program, Chiba, Japan

For our particle therapy society, this 10-20 years was transition phase form the passive delivery to the scanning delivery. In this transition from passive delivery, scanning delivery needs to keep specifications of 1) capability to treat moving tumor, 2) same or better dose rate, and 3) better dose conformation. Thus, the development of scanning system has been focused on faster scanning and smaller beam size. As a result of development, our scanning system has been treating the patients including moving target in both NIRS and KCC. Treating moving tumor with fast rescanning was one of important milestone in our society. However, next step of the development is strongly required, which means compact system with lower cost while keeping system performance. For this purpose, we have developed ultra-compact scanning magnet, which is really revolutionary device. This brand-new magnet makes the scanning system amazingly compact from 8 m to 3.5 m. Further, this new scanning system can deliver the field size of 34 cm x 34 cm. Not only a conceptual design, a prototype of this magnet is already manufactured and tested with carbon ion beam. Results of the beam test were successful and suitable for clinical use. Since smaller beam size is also important issue, we newly designed the irradiation port having a movable nozzle for further reduction of beam size, cooperating with the range shifter-less system. In this presentation, we will report this brand-new scanning system, which will be installed upcoming facilities.

PTC17-0407: An Eight-Year Review of More Than 15,000 Treatment Beam Results at Southern Tohoku Proton Therapy Center

T. Kato1, T. Nakamura 2, T. Matsumoto1, S. Oyama1, T. Ono 2, Y. Azami 2, K. Takayama 2, M. Suzuki 2, H. Wada 2, Y. Kikuchi 2

1Southern Tohoku Proton Therapy Center, Department of Radiation Physics & Technology, Koriyama, Japan 2Southern Tohoku Proton Therapy Center, Department of Radiation Oncology, Koriyama, Japan

Purpose: Proton beam therapy (PBT) has led to marked advances in treatment, mainly because of its excellent dose distribution. However, the huge capital requirement in PBT is a major obstacle for its wide use. In this study, we reviewed and analyzed the clinic operational data from 2008 to 2016 based on the institutional record at the Southern Tohoku Proton Therapy Center (STPTC).

Materials and Methods: At STPTC, two gantries and a fixed beam line are used. The gantries were fully commissioned in 2008, but a fixed beam line in 2009. From October 2008 to September 2016, more than 3,600 new patients were treated. Treatment condition records included more than 15,000 PBT fields. Information on the disease sites, field size, range, the number of fields per patient, and the beam angle utilization were analyzed.

Results: The disease sites and percentage of total include head and neck, lung, prostate, esophagus with 31%, 14%, 13%, and 11%, respectively. If a potential new beam line design resulted in a maximum range of 250 mm of water and a maximum field square of 130 mm, about 98% of all fields could be treated. Of all fields, 28% used only lateral beams, 42% used just the four normal angles (0°, 90°, 180°, 270°), and 58% used other gantry angles.

Conclusion: Knowledge of such distributions and trends in the clinical usage is important for planning a new proton center with reduced cost and also is expected to contribute useful information for the persons concerned.

PTC17-0410: 3D Printed Passive Modifiers for Particle Radiotherapy

A. Solovev1, A. Chernukha1, U. Stepanova1, M. Troshina1, A. Lychagin1, V. Kharlov 2, S. Ulyanenko1, V. Galkin3

1A. Tsyb MRRC, Radiation Biophysics, Obninsk, Russian Federation 2Fablab “Modelspectr”, CEO, Obninsk, Russian Federation3A. Tsyb MRRC, CEO, Obninsk, Russian Federation

Thus, the particle therapy is a modern way for cancer treatment nowadays, there are still lots of challenges to face during irradiation with these beams. It is not always possible to form the optimal dose distribution inside tumor region within the patient body only on its own. For degraded beams, there is need to use boluses, collimators, ripple and ridge filters of energy wheel etc. Even for active scanning beams, one is to use some filters because of some low energy constraints or to reduce the irradiation time. The 3D printing techniques suggests a balance between accuracy and fast prototyping and manufacturing.

In this study, we tested two main approaches using different modalities: the ripple filters, which are commonly used to obtain a slightly spreaded Bragg peak curve both in active and passive beam delivery systems, and the patient-specific bolus, designed to follow the patients head surface. The Geant4, ROOT (CERN) and in-house software NPLibrary used for Monte-Carlo simulations. The carbon ion monoenergetic beam of U-70 synchrotron, IHEP, Protvino and the compact proton synchrotron of the “Prometheus” therapeutic facility used to verify the depth-dose distribution, 2D and 3D dose field. The 3D printed samples obtained using ABS and PLA on Stratasys® Dimension Elite and MakerBot® Replicator printers. The dosimetry study performed in the water phantom using varoius ionization chambers and radiochromic film.

The preliminary study results shown that the 3D printers can be potentially used to form the semi-optimal dose field shape both for passive degraded beams and for active scanning techniques.

PTC17-0412: Degrader Design of SC200 Superconducting Proton Cyclotron

F. Jiang1, Y. Song1,2, J. Zheng1,2, M. Li1, X. Zeng1, J. Zhang1, W. Zhang1, B. Zhi 2, M. Han 2, O. Kamalou3

1Institute of Plasma Physics, Chinese Academy of Sciences, Hefei, China 2Hefei CAS Ion Medical and Technical Devices Co., Ltd, Hefei, China3Large Heavy Ion National Accelerator, GANIL, Caen, France

The proton beam energy decides the range of particles and thereby where the dose is imparted. According to the depth of tumors, an energy degrader is needed to modulate the proton beam energy in cyclotron based proton therapy. SC200 superconducting proton cyclotron is a key project supported by the Hefei local government of China and Chinese Academy of Sciences. SC200 superconducting proton cyclotron will provide a 200MeV fixed energy proton beam, and an energy degrader should be designed for degrading the beam energy from 200 to 70 MeV continuously. Through compiling programs, the parameters of beam were calculated when particles pass energy degrader, such as energy, momentum spread, angular spread, emittance, transmission fraction. These programs could be running by theoretical formulas method or Monte Carlo method. According to the result of calculation, the material, thickness and shape of degrader were confirmed as graphite, 150mm and triple-wedge, respectively. The collimators and the momentum slits which are placed after energy degrader were also designed and calculated for transmission fraction. The emittance and maximum momentum of beam was 16πmmmrad and 1.0%, respectively, at the exit of energy select system. Enventually, all these results were compared with program LISE, a Monte Carlo based particle transport software, from GANIL.

PTC17-0415: Beam Line Delivery System of SC200 Superconducting Proton Cyclotron

J. Zheng1,2, Y. Song1,2, F. Jiang1, M. Li1, X. Zeng1, J. Zhang1, M. Han 2, O. Kamalou3, M. Grossmann4, A. Gerbershagen4

1Institute of Plasma Physics, Chinese Academy of Sciences, Hefei, China 2Hefei CAS Ion Medical and Technical Devices Co., Ltd, Hefei, China3Large Heavy Ion National Accelerator, GANIL, Caen, France4Paul Scherrer Institute, PSI, Villigen-PSI, Switzerland

SC200 superconducting proton cyclotron is a key project supported by the Hefei local government of China and Chinese Academy of Sciences. The accelerator will provide 200 MeV proton beam with maximum current of 1μA in 2017-2018. The cyclotron is very compact and light, the estimate total weight is about 35 tons and extraction radius is 60 cm. We have performed simulations of all systems of the SC200 cyclotron and specified the main parameters of the accelerator. Average magnetic field of the cyclotron is up to 3.5 T and the particle revolution frequency is about 45 M. The proton beam should be delivered from the extracted point of accelerator to the entrance of nozzle by beamline in proton therapy. A new beam line delivery system has been designed to bring the 200 MeV proton beam from SC200 superconducting proton cyclotron to two proton therapy treatment rooms. To enhance compact form of beam line, the main trunk line will be achromatic by 45 degree dipole magnets for minimum area. The beam line delivery system also employs an energy selection system, including an energy degrader which can modulate the proton beam energy from 200 to 70 MeV continuously, two collimators which can adjust beam emittance to 16π mmmrad, and a pair of momentum slits. There will be a treatment room with a fixed horizontal beamline for eye treatments. In addition, there will be a room with iso-centric gantry for more complex treatments, and the FWHM of beam should be 4-10mm at iso-center.

PTC17-0435: Prompt Gamma Imaging with a Nuclear Emulsion for in Vivo Dose Verification in Proton Therapy

T. Toshito1,2, M. Kimura1, M. Nakamura3, H. Ogino4, Y. Shibamoto 2

1Nagoya Proton Therapy Center, Department of Proton Therapy Physics, Nagoya, Japan 2Nagoya City University, Graduate School of Medical Sciences, Nagoya, Japan3Nagoya University, Graduate School of Science, Nagoya, Japan4Nagoya Proton Therapy Center, Department of Proton Therapy, Nagoya, Japan

Purpose: We propose a new concept of prompt gamma camera using a nuclear emulsion technique for in vivo three-dimensional dose verification in proton therapy. It is dead-time-free at clinical dose rates, neutron-background-free, and of high three -dimensional resolution.

Materials and Methods: The system uses a nuclear emulsion. Prompt gamma-rays above 1 MeV are converted to e+e- pairs, and electrons and positrons are three-dimensionally tracked in nuclear emulsion. Tracks are measured by using a high-speed emulsion read-out system. Direction and energy of primary gamma-rays can be reconstructed from initial directions of secondary tracks, ranges and angular deflections caused by multiple Coulomb scattering. A first prototype emulsion gamma camera was tested at Nagoya Proton Therapy Center. The gamma camera which measures 50 × 50 x 3.4 mm (0.28 radiation length) was put on the sidewall of the phantom. A water phantom was irradiated with 200 MeV proton beams using a spot scanning nozzle at clinical dose rates. Emulsion films were developed and read out at Nagoya University.

Results: In the emulsion gamma camera, 540 thousand proton tracks and 180 thousand minimum ionizing particles including electrons and positrons from gamma-rays conversion were detected. Software to detect gamma to e+e- conversion is underdevelopment.

Conclusion: We examined a first prototype nuclear emulsion gamma-ray camera system for proof-of-principle. The detector system operated within expectations and will be used in further studies with enlarged scale devices capable for clinical application. Future experimental upgrading will also be discussed.

PTC17-0436: Feasibility of a LaBr3(Ce) Detector for Boron-10 Concentration Imaging in Boron Neutron Capture Therapy: A Simulation Study

K. Akabori1, K. Taki1, Y. Aoki1, T. Mitsumoto 2, S. Yajima 2, H. Tanaka3

1Sumitomo Heavy Industries- Ltd., Technology Research Center, Yokosuka-shi, Japan 2Sumitomo Heavy Industries- Ltd., Industrial Equipment Division, Niihama-shi, Japan3Kyoto University, Research Reactor Institute, Sennan-gun, Japan

In boron neutron capture therapy (BNCT), the absorbed dose in tumors and normal tissues strongly depends on the 10B concentration distribution in a patient. Determination of this distribution during the administration of clinical BNCT will lead to more refined control of BNCT treatment. To this end, we are developing a measurement system based on the 478 keV prompt gamma-single photon emission computed tomography (PG-SPECT) technique, which allows one to map out a three-dimensional distribution of the dose component due to the 10B(n,α)7Li reactions through detection of 478 keV gamma rays.

Our system will consist of a high energy resolution detector, efficient shielding against gamma rays and neutrons, and a collimator of moderate spatial resolution. As a detector, we chose a LaBr3(Ce) scintillator coupled to a photomultiplier tube. The gamma-ray and neutron shielding materials will be lead and polyethylene with lithium fluoride, respectively. The collimation component will be a parallel-hole collimator made of lead designed to have 1 cm spatial resolution. To test the feasibility of our system design, we simulated the response of the system with PHITS2.82 Monte Carlo simulation code, and estimated the system performance for detecting 478 keV prompt gammas originating from boron neutron capture reactions in a water phantom. The results imply that although reasonable uncertainties might be achieved in a half hour of BNCT administration with our current design, higher detection efficiency would be needed to measure the 10B concentration in real time.

PTC17-0440: Improvement of Penumbra by Optimizing Irradiation Parameter

T. Miyashita1, Y. Murakami1, T. Yamaguchi 2, T. Mukawa 2, S. Mizutani 2, K. Hotta3, H. Baba3

1Sumitomo Heavy Industries, ltd, Industrial Equipment Devision, Niihama, Japan 2Sumitomo Heavy Industries, ltd, Technology Research Center, Yokosuka, Japan3National Cancer Center Hospital East, Particle Therapy Division, Kashiwa, Japan

Intensity modulated particle therapy (IMPT) is a method to irradiate a tumor by combining inhomogeneous fields from multiple directions, and make a better dose concentration compered to single field uniform dose (SFUD) method which combines homogenous fields.

Although IMPT realize better dose concentration, there are several issues to make it practical. One of them is to reduce the penumbra size of the fields.

When IMPT is applied to fine structure such as head and neck, it is critical to reduce the penumbra to fit the structure, especially at the shallower region or lower energy, because the beam size becomes larger at the lower energy due to scattering in the thicker energy degrader.

In order to reduce the penumbra size, there are several methods; 1) reduce the size of the beam, 2) use a collimator, 3) enhance the edges of the filed by modulating the dose rate.

In this study, we combine and optimize these methods and make the penumbra size minimum and evaluate the effect to the final dose distribution.

PTC17-0445: Range Verification of 15O Beam Using OpenPET

A. Mohammadi1, E. Yoshida1, H. Tashima1, F. Nishikido1, A. Kitagawa 2, T. Yamaya1

1National Institute of Radiological Sciences, Department of Radiation Measurement and Dose Assessment, Chiba, Japan 2National Institute of Radiological Sciences, Heavy-Ion Radiotherapy Promotion Unit, Chiba, Japan

In advanced ion therapy, visualization of the range of incident ions in a patient body are important to exploit advantages of the ion therapy. PET has been applied to image positron emitters produced through fragmentation reactions, but a patient-dependent difference between the PET peak position and the dose peak position is preventing straight forward understanding of PET images. Therefore, radioactive beam irradiation, which enables direct visualization of beam stopping positions by PET, is recognized as an ideal method. We have shown feasibility of this idea through in-beam PET study by using a 11C ion beam and the OpenPET system. We have confirmed that the difference between the beam stopping position and the dose peak is shorter than the PET resolution, but the potential difference between them has not been analyzed in detail yet. Therefore, in this work, we observed the difference between the beam stopping positon and the dose peak position with a finer turned radioactive beam of 15O ions. Two min half-life of 15O, which is shorter than 20 min of 11C, also enables more accurate PET measurement. A PMMA phantom was irradiated by the 15O beam and in-beam PET measurement was done by the OpenPET. The dose distribution was measured in different depth of water using a cross ion chamber. We show the reconstructed image for the 15O beam. The difference between the PET peak and the Bragg peak positions was observed as 0.7 mm.

PTC17-0449: Quantitative Evaluation of Range Degradation According to the Gradient of the Range Compensator in Passive Scattering Proton Therapy

W.G. Shin1, C.K. Kim 2, H.S. Kim 2, J.H. Jeong 2, S.B. Lee 2, C.H. Min1

1Yonsei University, Radiation Convergence Engineering, Wonju, Republic of Korea 2National Cancer Center, Proton Therapy Center, Ilsan, Republic of Korea

It is reported that the sharp gradient of the compensator causes the range degradation due to the scattering effect. The purpose of this study is to quantitatively evaluate the range degradation due to the slope of the range compensator using Geant4 Monte Carlo (MC) tool. The thicknesses of the compensator in the isocentric line was set as 5 cm and the various slopes of 0-5 (dy/dx) were employed. Downstream of the compensator, the depth dose profile (DD) in a water phantom was assessed according to the compensator slope with pristine beam of 145 MeV and therapeutic beam of 15 cm range, 5 cm modulation width. To quantitatively analyze the range degradation, the distal fall-off width that the distance between the D20 and D80 was defined.

With the compensator of 0-5 slope, the range was decreased from 91.1 to 79.8 mm in pristine beam, 85.5 to 65.5 mm in therapeutic beam. The distal fall-off widths were from 2.1 to 24.8 mm in pristine beam, 5.3 to 32.0 mm in therapeutic beam. The DD was significantly distorted, interestingly, the distal fall-off width is linearly increased with compensator slope dy/dx. Our results show that not only patient geometry but also range compensator significantly contributes to the dose degradation.

PTC17-0465: Performance of Prompt Gamma Fall-Off Detection in Clinical Simulations

B. Huisman1, D. Sarrut1, E. Testa 2, N. Krah 2

1CREATIS- Université de Lyon, Team 4, Lyon, France 2IPNL, PRISM, Lyon, France

As the first Prompt Gammas (PG) cameras are deployed in clinical setting, we studied PG fall-off positions (FOP) estimation on a complete clinical simulations. The number of protons (spot weight) required for a consistent FOP estimate was investigated for two PG cameras, a multislit and knife edge design, for a single spot of a clinical fully clinical simulation of an patient treatment. A new spot-grouping method is proposed that combines better measurement statistics with fall-off preservation.

We considered a clinical head and neck treatment plan containing both a CT and re-planning (RP)CT. Monte-Carlo simulations were performed on both CTs. During the irradiation, two PG cameras implemented as published, recorded the PG profiles, spot by spot. We study shifted distributions as function of the spot weights, for each camera, from $10^6-10^9$. A FOP estimation was applied on 20 CT and 20 RPCT realizations to obtain FOP distributions for both images. A natural way to improve statistics is to integrate PG profiles over multiple spots. We investigate if and how spot-grouping methods improve FOP estimation.

By studying recent treatment plans from various proton clinics, we observe very few spots with weights over $10^8$. We did not manage to detect the morphological change present, an approximately 10mm shift, between the (RP)CT with either PG camera, with such statistics. Only for $10^9$ primaries, with one of the cameras, the change may be expected to be detected. Preliminary results of grouping spots in geometric layers seem to make an improved FOP estimation possible.

PTC17-0469: Beam Delivery Strategies in Proton Pencil Beam Scanning (PBS): The Continuous Approach

A. Patriarca1,2, R. Van Roermund3, L. De Marzi1, A. Mazal1, S. Meyroneinc1, R. Dendale1, V. Patera 2

1Institut Curie - CPO, Radiotherapy, Orsay, France 2Sapienza Università di Roma, Dipartimento di Scienze di Base e Applicate per Ingegneria, Roma, Italy3IBA, Protontherapy, Louvan La Neuve, Belgium

Different approaches have been studied and applied to deliver a given dose distribution to a target with protons pencil beams: the beam can be scanned perpendicular to the beam axis while the energy is kept constant until a new layer is started, or the beam can be scanned parallel to the beam axis changing the depth continuously. In each technique the delivery of the spot can be done either spot by spot (the beam is switched off between two locations), by raster scanning where the beam stays on all the time even between spot positions (typically with synchrotrons), or continuous scanning where the concept of spot is discarded, as the beam is moving all the time through iso-energy layers.

Currently, in proton therapy centers using a cyclotron as accelerator, the spot scanning is the only available technique. In the attempts to improve the treatment process in PBS, continuous scanning could represent an alternative to spot scanning in specific cases. Possible advantages of this delivery technique will be the reduction of the irradiation time allowing the treatment of moving target with repainting strategies. The study here proposed wants to show results on the comparison between models and measurements of several irradiations patterns in spot and continuous scanning for an IBA proton therapy system (C235 cyclotron and universal nozzle gantry). In particular, we will focus on how deliver equivalent maps in terms of dose, thus comparing the results of the irradiations in terms of beam delivery time, dose homogeneity and lateral penumbra.

PTC17-0477: An International Reference Dosimetry Intercomparison for Ocular Proton Radiotherapy Facilities

J. Swakoń1, J. Heufelder 2, T. Horwacik3, J. Hrbacek4, A. Kacperek5, B. Koska6, C. Nauraye7, P. Olko1, R. Slopsema8, P. Trnkova9

1Institute of Nuclear Physics PAS, Proton Radiotherapy Group, Krakow, Poland 2Charité - Universitätsmedizin Berlin, BerlinProtonen, Berlin, Germany3Institute of Nuclear Physics PAS, Cyclotron Centre Bronowice, Krakow, Poland4Paul Scherrer Institut, Center for Proton Therapy, Villigen, Switzerland5The National Eye Proton Therapy Centre, The Clatterbridge Cancer Centre NHS FT, Clatterbridge, United Kingdom6Westdeutsches Protonentherapiezentrum Essen Gemeinnützige GmbH, Klinik für Partikeltherapie am Universitätsklinikum, Essen, Germany7Institut Curie, Centre de Protonthérapie d'Orsay, Orsay, France8University of Florida Health Proton Therapy Institute, Radiation Oncology Department, Jacksonville, USA9HollandPTC, PTC, Delft, Netherlands

Within the OPTIC PTCOG Subcommittee, an international proton dosimetry intercomparison between seven proton ocular therapy facilities was organised at the Institute of Nuclear Physics in Krakow, Poland. Its aim was to evaluate the consistency in delivering the therapeutic dose to patients by comparing results of reference dosimetry using dosimetry systems with primary or secondary traceability to national reference standards.

The intercomparison was carried out at the INP PAS eye therapy facility, the proton beam being supplied by the IBA Proteus C-235 cyclotron. Two small-field proton beams, of 25 mm diameter and of 31.5 mm range in water, modulated over 15 mm or 31.5 mm, were used. The participants used four different types of semiflex ionization chambers, three types of parallel-plane Markus-type ionization chambers and three types of reference-class electrometers. The measurements carried out in reference conditions according to the TRS-398 IAEA protocol. The same pre-selected number of monitor units for each setup was measured, followed by appropriate temperature and pressure corrections. The administrated absorbed dose in water at the reference point was about 5 Gy.

We observed good agreement between the participant-measured and delivered reference dose. Some differences were observed, dependent on chamber types and their calibration. For the seven primarily referenced setups the standard deviation was 0.2% (1σ) and the maximum difference did not exceed 1%. For all of the 19 setups tested over two days the standard deviation (1σ) did not exceed 0.3%, while the largest differences between measured and delivered dose were within 2%, for some Markus-type ionization chambers.

PTC17-0492: Can a Pencil-Beam-Scanning Nozzle Deliver Eye Treatments with the Same Quality as a Dedicated Eyeline?

R. Slopsema1, M. Mamalui1, Z. Li1

1University of Florida Health, University of Florida Health Proton Therapy Institute, Jacksonville, USA

Purpose: Ocular proton therapy is typically delivered by a dedicated nozzle (eyeline). The eyeline is designed to deliver small dose distributions with extremely sharp dose fall-offs, at very high dose-rate. Addition of an eyeline to a proton-therapy facility is often cost-prohibitive because of the low incidence of ocular cancer. Most proton-therapy rooms installed today are general-purpose pencil-beam-scanning systems not suited for eye treatments. The purpose is to determine the feasibility of an eye treatment option in a pencil-beam-scanning nozzle with dosimetric properties comparable to a dedicated eyeline.

Materials and Methods: The PBS-based eye treatment option has a range shifter (RS) downstream of the ionization chambers and an aperture at 7 cm from isocenter. The energy is varied to adjust the range and to create the spread-out-Bragg-peak (energy stacking). Either magnetic defocusing of a single spot or spot scanning spreads the beam. RS thickness and position and beam optics are optimized to minimize penumbra while maximizing dose rate. Different configurations are simulated by an analytical model that models beam phase space propagation in presence of collimators and scatterers. Measurements of the lateral penumbra, distal fall-off, and dose rate are made in several test setups with the ‘IBA Universal Nozzle'.

Results: We show the penumbra and dose rate for several spreading methods and RS configurations.

Conclusion: An eye-treatment option can be implemented on a pencil-beam-scanning nozzle with dose fall-offs comparable to a dedicated eyeline, but with lower maximum dose rate.

PTC17-0505: Uncertainties in Normal Tissue Dose Delivery in the Pelvis: A Case for Concern?

P. Boimel1, M. Kang1, P. Kimstrand1, A. Carabe-Fernandez1, L. Lin 2

1University of Pennsylvania, Radiation Oncology, Philadelphia, USA 2MD Anderson Cancer Center, Department of Radiation Oncology, Houston, USA

Purpose: To investigate the impact of dose uncertainty to bowel in a patient receiving pencil beam scanning (PBS) proton therapy who developed a small bowel obstruction two months after treatment.

Materials and Methods: A single patient with history of abdominal surgeries and a pelvic nodal recurrence of endometrial cancer received PBS proton therapy to the whole pelvis 45GyE (1.8GyE) and a 20GyE boost (2GyE/fx) to nodal recurrence. CT simulation for planning was performed. To generate a PBS planning target volume (PBSTV), a geometric expansion of 7 mm was done with 3.5%+1mm in the direction of posterior oblique beams. Bowel gas was overridden. The LET effect was calculated using in-house biological modeling. kV-kV matching was performed prior to daily treatment. The inter-fraction dose accumulation was evaluated on biweekly CTs compared to the planning CT to evaluate maximum dose differential for target volume and organs at risk.

Results: kV-kV imaging demonstrated setup errors of 2mm. Bowel gas changes contributed small dose uncertainties. Using a CT Hounsfield unit to stopping power ratio of +/-3% and setup error of 2mm to generate 12 uncertainty scenarios, the 10Gy isodose line shifted by 1.5cm anteriorly. Bowel V45 was 154cc on initial planning CT, 173cc with LET correction, and 130cc accumulated dose on subsequent CT evaluation. However, due to uncertainty bowel V45 could be as high as 214.5cc.

Conclusion: Distal range proton dose uncertainties should be considered in patients receiving high dose radiotherapy to pelvic disease as volume of bowel in the high dose region could increase.

PTC17-0521: Dosimetry of Pencil Scanning Beam at Cyclotron Proteus-235 and Synchrocyclotron Proteus One Therapy Facilities

M. Liszka1, L. Stolarczyk1, M. Vidal 2, R. Kopec1, J. Kunst1, B. Michalec1, G. Mierzwinska1, W. Slusarczyk-Kacprzyk3, P. Kukolowicz3, P. Olko4

1Institute of Nuclear Physics Polish Academy of Sciences in Krakow - IFJ PAS, The Bronowice Cyclotron Centre - CCB, Krakow, Poland 2Centre Antoine-Lacassagne - CAL, Institute of Protontherapy and CyberKnife, Nice, France3The Maria Skłodowska Curie Memorial Cancer Centre and Institute of Oncology - MCMCC, Medical Physics Department, Warsaw, Poland4Institute of Nuclear Physics Polish Academy of Sciences in Krakow - IFJ PAS, Division of Applications of Physics, Krakow, Poland

Purpose: In synchrocyclotron based PBS systems dose rate can reach up to 200 Gy/s, what influences the response of IC due to charge recombination. In this work response of plane-parallel IC was investigated at therapy system based on SC2 synchrocyclotron (Proteus One, IBA) CAL-Nice and on C230 cyclotron (Proteus C-235, IBA) at CCB-Krakow. The measurements were verified with thermoluminescence (TL) and alanine detectors not dependent on the dose rate.

Materials and Methods: Markus-type IC TM 23343 with electrometer UnidosWebline (PTW) calibrated at SSDL-Warsaw traceable to IAEA and PPC05 IC with Dose1 electrometer (IBA-Dosimetry) calibrated at BIPM-Paris were used to measure Dw at the middle of uniformly covered 10 × 10 x 10 cm3 cubes. Treatment plans with ranges of 20, 25 and 30 cm and 10 cm modulation were prepared using RayStation TPS. At the same conditions MTS-7 TL and alanine detectors were irradiated with a known Dw. The ks correction factors were measured for each cube and determined by means of two-voltage method.

Results: ks measured for TM 23343 depends on the dose rate and varied between 1.024 and 1.030 at Proteus One whereas at Proteus C-235 varied between 1.001 and 1.003. Agreement between Dw measured by TM 23343 and PPC05 at Proteus One facility, after applying ks factors, was at the level of 1%. The results from passive detectors confirm this difference within the uncertainty range.

Conclusion: Accounting for recombination effect we got a very good agreement between measurements performed with ionization chamber and the solid state detectors.

PTC17-0527: A Custom Snout Extension for Treatment of Ocular Targets with Uniformly Scanned Proton Beams

A. Egan1, J. Saini1, D. Maes1, A. Fung1, B. Ridgeway1, R. Rengan 2, T. Wong1, C. Bloch 2

1SCCA Proton Therapy Center, Physics, Seattle, USA 2University of Washington, Radiation Oncology, Seattle, USA

Purpose: Proton beam ocular treatments are typically administered with dedicated eye lines. At the SCCA Proton Therapy Center (SPTC) a novel snout-ex