Children are particularly prone to the late side effects of normal tissue irradiation. For this reason, pediatric solid tumors are a commonly cited indication for proton therapy worldwide. The aim of this survey was to assess pediatric patterns of care across proton centers in the United States.
A survey was developed and distributed annually to each clinical proton therapy facility in the United States in operation during the years of 2010, 2011, and 2012. Anonymized patient information including age range, tumor site, and diagnosis were collected annually for each patient 18 years old or younger treated between January 1, 2010 and December 31, 2012.
There was a 100% response rate from the United States proton therapy centers in operation for each year surveyed. All facilities treated at least 1 pediatric patient each year. A total of 694 pediatric patients were treated in 2012, an increase from 613 patients in 2011 and 465 patients in 2010. Fifty-seven percent of patients in 2012 were under 10 years of age, similar to 2011 and 2010. Anesthesia was required for 45% of patients. The six most common tumor diagnoses treated were ependymoma, medulloblastoma, low-grade glioma, rhabdomyosarcoma, Ewing sarcoma, and craniopharyngioma. In 2012, 19% of treated patients originated in countries outside the United States.
The total number of children treated at proton centers in the United States continues to increase, rising 33% since 2010, consistent with the international perception that pediatric patients derive a relative benefit from this rare technology. The average patient is a child <10 years old with a curable brain tumor or axial sarcoma.
Pediatric cancer treatments have dramatically improved survival over the past 3 decades. Accompanying the increased survival are the late effects associated with surgery, chemotherapy, and radiation therapy (RT) [1–3]. The long term effects of RT are well known and dependent on the dose administered and volume irradiated. RT effects are further magnified in children as developing organs/tissues are more susceptible to RT-related late effects such as abnormal bone growth, impaired muscle development, organ atrophy, and neurocognitive delay. Likewise, as childhood survivors of cancer have the potential for many decades of life after therapy, they risk survivorship issues from late effects of irradiation that can manifest decades after irradiation.
With central nervous system (CNS) tumors accounting for the highest percentage of pediatric solid tumors, neurocognitive deficiencies resulting from RT contribute to substantial psychological and social debilitating late effects. Notably, intelligence quotient is significantly impacted in pediatric patients undergoing cranial RT. Merchant et al  reported on quantification of the decrease in neurocognitive function in childhood cancer survivors as a function of dose intensity, dose volume, and age.
As survival rates improve in many childhood malignancies, current research strategies are aggressively focusing on decreasing the side effect profile while maintaining high cure rates. Innovations such as computer based three dimensional treatment planning have dramatically improved the delivery of x-ray based RT. Heavy charged particle RT presents a further advance in conformal RT delivery due to unique physical characteristics to allow for optimal dose distribution to tumors while minimizing normal tissue irradiation. The concept of using high-energy proton radiation as a mode of treatment for cancer patients was first proposed by Dr. Robert Wilson in 1946 . Proton RT (PRT) is the most commonly utilized heavy charged particle strategy due to the Bragg peak which minimizes unnecessary radiation exposure to surrounding tissues as compared to x-ray RT. Loma Linda University's proton therapy center treated its first patient in 1990 and became the first proton center designed primarily for patient care. Massachusetts General Hospital's clinical proton center began treatments in 2001 and had been previously treating patients at the Harvard Cyclotron in the physics department of Harvard University since 1974. Due to the few numbers of proton centers, as well as limits in beam availability, a relatively small proportion of pediatric patients in the United States (US) were initially treated with proton therapy. In the past decade, however, there has been a proliferation of proton therapy centers with 11 proton centers now in operation in the US, and several more in various stages of development and construction.
Given the particular benefit of proton therapy in the pediatric milieu, and as more proton centers are established in the US – and worldwide – it is important to understand patterns of care for pediatric patients. For this reason, the Pediatric Proton Foundation distributed voluntary surveys to all operational proton centers between the years 2010–2012. The results were compiled and interpreted in a semi-quantitative, longitudinal manner to form the content of this project.
Patients and Methods
The surveys collected anonymized patient information from each proton center in the US during the years 2010, 2011, and 2012. For the years 2010 and 2011, there were 9 proton therapy centers in operation in the US (3 of which began treatments mid-2010) and 10 centers in 2012. The first survey was sent out in the beginning of 2011 to aggregate the data on patients who began PRT between January 1, 2010 and December 31, 2010. The survey defined pediatric patients as those who were 18 years of age or younger at the start of PRT.
The 2011 survey was sent in early 2012, and requested information for patients who started PRT between January 1, 2011 and December 31, 2011. The 2012 survey was sent in early 2013 and requested information for patients who started proton treatments between January 1, 2012 and December 31, 2012. Each year, the survey was slightly modified to clarify its content and expand its scope; the 2012 survey added two additional questions.
The surveys requested aggregated non-identifiable patient information including numbers of patients treated within specified age ranges, tumor sites, and histologic diagnoses. The 2012 survey also asked for the number of patients who lived in the US as well as the number of patients who also received x-ray therapy as part of the treatment course.
The distribution of the surveys, collection of the responses, and collation of data was conducted by the Pediatric Proton Foundation and the National Association for Proton Therapy.
For each of the 3 years, every center in operation participated in the surveys for a 100% response rate; the majority of centers were able to answer all the questions. In 2010, 465 pediatric patients were treated with proton radiation; 613 patients in 2011, and 694 patients in 2012. Each year, every center treated at least one pediatric patient, with a range of 1 to 111 patients in 2010, 4 to 124 patients in 2011, and 6 to 140 in 2012. When compared to the total number of patients treated with PRT in the US , pediatric patients consisted of 10% (range, 2% to 24%), 12% (range, 1% to 30%), and 13% (range, 2% to 40%) of the total patients treated in 2010, 2011, and 2012 respectively.
Slightly more than half of all the patients treated over 2010 to 2012 were 9 or under (Figure 1), making up 62.4% of the patients in 2010 (290 patients), 58.7% of the patients in 2011 (330 patients), and 56.8% (394) in 2012. Anesthesia was used for at least half of the fractions in 47% of the patients in 2011, and ranged from 0% to 53% of patients between the centers. In 2012, the range was from 11% to 65%, with a weighted average of 45%.
In both 2011 and 2012, rates of treatment intent for PRT defined as curative were 97.1% (range, 88.6% to 100%) and 98.6% (range, 93.2% to 100%) respectively. The majority of patients were treated with 21 to 30 fractions of proton radiation, with 31 to 40 fractions being the second most common (Figure 2). In 2012, 2 centers utilized x-ray therapy as a component of their treatment course for 29.3% and 5.7% of their patients. The stated reasons for usage of this mixed modality therapy included: (1) the relative skin sparing of photons; (2) the mitigation of dose uncertainty due to metal hardware implants; and (3) to avoid missed treatment days due to machine maintenance.
The histologic distribution of tumor types also remained very similar over the 3 years of the surveys. Table 1 lists every reported histology that had more than 1 patient treated in any single year. Figure 3 displays the most common histologies treated, with the six most common tumors remaining unchanged from year to year: ependymoma, medulloblastoma, low-grade glioma, rhabdomyosarcoma, Ewing sarcoma, and craniopharyngioma. These 6 histologies comprised 62.4%, 61.0%, and 62.2% of the tumors treated in 2010, 2011, and 2012, respectively.
With regard to treatment site, one center each year was unable to provide the information. Other centers had patients for whom multiple sites were treated; each of the sites listed as being treated were tallied (Figure 4). The CNS was the most common site treated with PRT, accounting for 71.5% of the patients in 2010, 64.5% in 2011, and 60.7% in 2012. Highly complex craniospinal treatments accounted for almost a third of the CNS treatments – comprising 22.7% and 20.1% of all patient treatments in 2011 and 2012, respectively.
In 2012, 132 patients (19%) were non-US patients; this ranged from 0% to 60.8% of patients at individual centers.
Our survey results are the first collaborative investigation involving every operational proton center in the US. With a 100% response rate for each of the three years the survey was conducted, we can comprehensively present patterns of care for the pediatric patient treated with proton therapy. The absolute number of pediatric patients treated annually in the US has steadily increased each year of the survey. Furthermore, the proportion of pediatric to adult patients treated with PRT in the US continues to increase as well. This trend reflects growing acceptance of the benefits of PRT with regard to late effects abatement for this population of patients at increased risk for radiation induced morbidity. Currently, the majority of the Children's Oncology Group trials permit proton therapy among the standard of care radiation treatment options.
In the US, it is estimated that 6,000 pediatric patients receive radiation treatments annually, with PRT accounting for approximately 10% of cases. In comparison, less than 1% (∼0.85%) of all patients receiving RT in the US were treated with PRT . The utilization of PRT in pediatric patients is expected to continue to increase not only as a result of the operational centers treating more patients, but also as a consequence of the expanding numbers of proton centers. Currently, there are plans for several proton centers in development in the US, and in the next 3 years alone, at least 7 more centers that will begin patient treatments; in 2013, ProCure Seattle began operations, and it is expected that the Scripps Proton Center (San Diego, CA, USA), the Provision Center for Proton Therapy (Knoxville, TN, USA), McLaren Proton Therapy Center (Flint, MI, USA), and Washington University's proton center (St. Louis, MO, USA) will begin treatments in 2014. St. Jude Children's Research Hospital (Memphis, TN, USA) will open a proton therapy center in 2015. The Mayo Clinic will also open a center in Rochester, MN, USA in 2015, followed by a second facility in Phoenix, AZ, USA in 2016. While one of these will be dedicated solely to the treatment of the pediatric patient, 3 others will be affiliated with children's hospitals. There are also many centers in development outside the US, and in particular in Asia and Europe that are facing many of the same questions and challenges in the treatment of pediatric patients as faced in the US. The results of the survey will be invaluable in anticipating the needs of the pediatric patient as these proton centers move forward.
While the median age of the patients treated with PRT has risen over the past 3 years, younger children still predominate. Compared to pediatric patients treated with photons, the increased time of treatment with PRT, along with a lower median age of patients, contributes to a fairly high usage of anesthesia. This is an important consideration for future proton therapy centers in development, especially for free-standing facilities where anesthesia availability may be limited. By 2012, all proton centers in operation utilized anesthesia for a select proportion of their pediatric patients.
The six most common pediatric tumors treated each year have been consistent since 2010. Of these six diagnoses, three of them include concurrent chemotherapy as part of the treatment course; two of these regimens require inpatient chemotherapy administration and the possibility of hospitalization for the third remains high due to the risk of neutropenic fever. This complexity of care frequently involving anesthesia and/or chemotherapy underscores the importance of support across pediatric sub-specialties and the proximity of a tertiary pediatric medical center. In developing new proton therapy facilities, it is important to recognize that the overall care environment—not just the proton beam—is critical to pediatric cancer care.
Of the diagnoses treated, CNS tumors represent the majority of the tumor types treated with PRT. This mimics the distribution of pediatric oncology patients who undergo RT as a component of their therapy. Nevertheless, over the past 2 years, there appears to be a greater increase in the treatment of non-CNS malignancies. In particular, the axial soft tissue malignancies have increased as a percentage of the patients treated. This is due in part to increased awareness of late radiation side effects in long term Ewing sarcoma, rhabdomyosarcoma, and Hodgkin lymphoma survivors that may be mitigated by PRT .
Craniospinal irradiation (CSI) represents one of the highest value targets for the use of PRT in pediatric and adult populations, given the large volume of irradiated tissues using x-ray techniques that can be avoided using PRT. A substantial minority of the CNS treatments were for CSI. This reflects the benefits PRT can have in lowering morbidity as compared with x-ray CSI; x-ray CSI late effects are well established and include cardiac disease, hearing loss, endocrine abnormalities, decreased fertility, and secondary malignancies [9–11]. Early proton CSI dosimetric work by the Loma Linda University and Massachusetts General Hospital proton therapy groups have shown a significant decreased in the radiation dose delivered to normal structures [12, 13]. More recent work by Howell et al  have shown that proton CSI can substantially lower doses to normal tissues even in older patients with a larger range of heights and weights. Computer modeling data have predicted a decrease of secondary malignancies with proton CSI by as much as 7 to 12 fold [15, 16].
Aside from the improved therapeutic ratio, multiple studies from the US and Europe have documented the cost effectiveness of utilizing proton therapy in pediatric cancer patients [17, 18]. In countries such as the United Kingdom where the state is responsible for the burden of medical costs, this represents a public healthcare finance issue. Not surprisingly, multiple countries in Europe are pursuing the development of their own proton centers. In the meantime, as the 2012 data illustrates, many children are being referred to US centers for treatment.
However, there are several challenges to the treatment of the pediatric patient. One of these is accessibility – with the limited availability of proton centers, and the logistics of coordinating transference of care for a child who has recently undergone a neurosurgical procedure, many children who could otherwise benefit from proton CSI are unable to receive it in a timely manner. This is often compounded by the short time between diagnosis and the need for initiation of CSI. In addition, a pediatric patient who has to relocate for treatment requires substantial family support; similar to adult cases, the large majority of pediatric treatment courses consist of fractionated 20 – 40 daily treatments. Many families however, are unwilling or unable to take time off of work and transplant the patient, parent(s) and siblings for an extended period of time. It is anticipated that with the proliferation of proton therapy centers to more areas geographically, there will likely be an increase in the proportion of patients who receive proton therapy.
Financial considerations can also be a challenging issue. From the breakdown of the patient tumor types and locations, as well as the support services required, the typical pediatric treatment case is considerably more complex in the planning and delivery of the radiation compared to the adult patient. This has significant implications for proton centers that have high patient volume requirements in order to meet debt obligations [19, 20]. For centers that are treating patients at capacity, this can lead to a conflict between treating the complex pediatric cases or maximization of revenue.
Another challenging aspect is that general radiation oncologists are not necessarily well-trained in pediatric oncology; with a requirement of only 12 pediatric patient simulations during the four-year residency – and often times at an external rotation – this had led to discussions about the feasibility of subspecialty certification for pediatrics . As it is, proton centers are tertiary referral centers for pediatric radiation treatments. There are currently only 3 pediatric RT fellowships with consistent funding in the US and each are affiliated with large proton centers. If pediatric RT continues the current trend of centralization at proton therapy centers, not only will there be fewer definitive cases that will be available to be seen by residents at their home institution, the treating physician at the proton centers will have a higher expectation of being adequately prepared to care for multiple complex pediatric patients. Combined with the nuances of treatment with PRT, this favors experience provided by subspecialty fellowship training following residency.
With the current concentration of pediatric proton therapy in only a few centers, it provides a unique opportunity to study these rare tumors of childhood. In the US, a national registry has been developed by the Pediatric Proton Consortium Registry with initial seed funding provided by the National Cancer Institute/Massachusetts General Hospital federal share funds to track pediatric patients treated with proton therapy in the US. As a longitudinal study that will continue to follow the outcomes of the patients, the goal will be that specific clinical outcomes can be determined. Specifics of the Pediatric Proton Consortium Registry are detailed in an accompanying article in this journal. As the rollout of this new registry to all the proton therapy centers in operation will continue to take time, we anticipate this survey to continue as a valuable tool in defining the pediatric patients that receive RT.
Charged particle therapy is a promising avenue that has the potential to spare a significant portion of non-target tissue within patients. The total number of children treated at US proton centers continues to increase, rising 33% since 2010, consistent with the international perception that pediatric patients derive a relative benefit from this rare technology. This makes up a significant percentage of non-palliative pediatric RT patients. The average patient is a child <10 years old with a curable brain tumor or axial sarcoma. At a minimum, affiliation with a children's hospital is necessary for pediatric treatments and the broad subspecialty support typically available in an academic medical center may be desirable due to the frequent need for anesthesia support and the delivery of concurrent chemotherapy in 3 of the 6 most common tumors treated.
ADDITIONAL INFORMATION AND DECLARATIONS
Acknowledgments: The authors would like to acknowledge David W. Shia from the University of California (Los Angeles, CA, USA).
Conflicts of Interest Disclosure: Dr. William Hartsell has ownership interest in CPTI, LLC (minority ownership interest in CDH Proton Center Chicago). Dr. Andrew Chang is a consultant for the Proton Center Development Corporation. Dr. Oren Cahlon has a stipend for medical directorship and minority ownership interest in ProCure New Jersey Proton Therapy Center. All other authors have no conflicts of interest to disclose.
Funding: This project was completed with support from the Pediatric Proton Foundation and The National Association for Proton Therapy