A Prospective Evaluation of Acute Toxicity from Proton Therapy for Targets of the Parotid Region

Cancers of the parotid region are often managed with surgery and postoperative radiotherapy depending on the pathologic findings and the risk of residual disease in the operative bed and neck [1–3]. Radiotherapy in this setting can deliver unintended dose to the mucosa of the oral cavity, pharynx, larynx, and contralateral salivary glands, leading to acute toxicities, including mucositis, dysguesia, xerostoma, and dysphagia. Such toxicities result in decreased quality of life (QOL) and longer duration of recovery due to malnutrition and dehydration; furthermore, with concurrent chemoradiotherapy, toxicities can lead to complications like renal insufficiency, hypotension, and hospitalization [4].

The parotid region represents an ideal target volume for harnessing the unique dosimetric properties of proton therapy. Because the depth of the target volume is shallow with these tumors, proximal non-conformity and entry radiation are less concerning, and all critical organs at risk, such as the oral cavity, pharynx, larynx, spinal cord, temporal lobe, brainstem, and contralateral salivary glands, lie distal to the target volume. Given the sharp distal dose fall-off generated by the properties of the proton's Bragg peak, these critical structures will receive either very little or no dose from most proton therapy plans.

We hypothesized that the dosimetric advantages of proton therapy would result in negligible acute mucosal toxicity, diet modifications, and weight loss. Herein we report prospectively collected data in these domains for patients treated with proton therapy to the parotid target.

Data were compiled prospectively in accordance with the Health Insurance Portability and Accountability Act and an institutional review board-approved prospective outcomes study. The eligibility criteria were as follows:

• Adults and pediatric patients

• Curative treatment including primary or postoperative proton therapy for any cancer such that the radiation target volume encompassed a single parotid region to a minimum of 50 Gy (RBE)

• No distant metastases

• No prior history of head and neck radiotherapy

• No active secondary malignancy other than squamous or basal cell skin cancer

• Patients could be treated with or without chemotherapy

Patient and treatment characteristics are shown in Table 1. Sixteen (65%) patients were treated with adjuvant RT, and 7 (35%) were treated with primary RT. The most common indication for proton therapy was skull base invasion or pediatric age at diagnosis. Passively scattered conformal proton therapy typically involving 3 fields was delivered to a median dose of 70 Gy (RBE) (range, 55.8-74.4 Gy [RBE]) using various fractionations (1.8-2 Gy [RBE]) once daily with 5 fractions per week (n=15); 1.2 Gy (RBE) twice daily with 10 fractions per week (n=7); and 2 Gy (RBE) once daily with 6 fractions per week [n=1]). Comparison color-wash dose distributions for intensity-modulated radiotherapy (IMRT) and 3-dimensional conformal proton therapy are shown in Figure 1. Acute toxicities were prospectively evaluated weekly during radiotherapy using Common Terminology Criteria for Adverse Events (CTCAE), version 3.0 [5]. Specifically, mucositis, dysphagia, xerostomia, dysgeusia, and weight loss were reported to assess mucosal toxicity. For some patients, pretreatment baseline scores were not obtained; therefore, baseline scores and week 1 scores were reported together as the baseline score.

Table 2 demonstrates the average minimum, mean, and maximum doses delivered to relevant organs at risk associated with acute mucosal toxicities. The results are summarized in Figure 2A-D. At baseline, no patient had mucositis, 13% had dysphagia (4% grade 1 and 9% grade 2), 8% had dysgeusia (4% grade 1 and 4% grade 2), and 19% had xerostomia (14% grade 1 and 5% grade 2). During radiotherapy, toxicities were minimal, typically emerging in weeks 2 and 3 and then stabilizing. Most patients experienced either no toxicity or grade1 toxicity throughout therapy. Weight loss (Figure 3) was minimal with a median percentage weight loss of 3% through therapy and holding steady at the first follow-up. Only 1 patient experienced greater than 10% weight loss, 74% experienced less than 5% weight loss, and 17% gained weight during radiotherapy. No patient required a feeding tube, intermittent intravenous hydration, or significant diet modification, such as the use of a soft/liquid diet or reliance on nutritional supplements. There were no occurrences of grade 4 toxicity and only 1 transient grade 3 dysphagia. This dysphagia occurred during only one week in a patient with baseline grade 2 dysphagia. Prior to radiotherapy, this patient underwent parotidectomy, mandible resection, and full-mouth extractions, and a percutaneous gastrostomy tube was placed. The grade 3 toxicity was recorded during week 4 of therapy and returned to baseline the following week. Upon completing therapy, the gastrostomy tube was removed when dysphagia improved to grade 1.

The worst mucositis recorded was grade 2 (patchy ulceration or pseudomembranes), ranging from 35 to 43% per week. Thus, most patients either experienced no mucositis (35-43%) or grade 1 mucositis (17-30%, mucosal erythema without ulceration or pseudomembranes) throughout their entire treatment course. Approximately 50% of patients had no dysphagia with 17-26% experiencing grade 1 dysphagia (symptomatic but able to eat a regular diet) per week and 26-30% experiencing grade 2 dysphagia (symptomatic with altered eating/swallowing requiring oral supplements, altered dietary habits, or less than 24 hours of intravenous hydration) per week. Dysgeusia was absent in 40-50% of patients each week throughout therapy; 22-40% had grade 1 dysgeusia (altered taste but no change in diet) each week and 9-26% developed grade 2 dysgeusia (altered taste requiring altered dietary habits) each week. Xerostomia was the most commonly reported toxicity. During the course of radiotherapy, the rate of xerostomia fluctuated slightly with 17-36% of patients experiencing no xerostomia each week, 55-64% experiencing grade 1 xerostomia (symptomatic with dry or thick saliva, but no change in diet) each week, and 9-25% experiencing grade 2 xerostomia (symptomatic and significant oral intake alteration) each week. Thus, throughout the entire treatment course, approximately 50% of patient had no dysgeusia, dysphagia, or mucostitis; approximately 25% had no xerostomia; the remaining patients reported minimally bothersome symptoms.

The dosimetric advantages of advanced radiotherapy techniques such as proton therapy can improve the therapeutic ratio in one of two ways: by improving the probability of tumor control through treatment intensification or by reducing treatment-associated toxicities. When considering the potential advantages of proton therapy, physicians and patients often focus on reducing late toxicities, such as blindness, brainstem injury, cardiac injury, or second malignancies; however, reducing acute toxicities is also an important goal. In regards to head-and-neck cancer patients, acute mucosal toxicities represent a major burden for patients and caregivers, often requiring aggressive pain management, intravenous hydration, expensive nutritional support, and potential hospitalizations. Following therapy, these toxicities can linger, severely impacting QOL, time to functional recovery, and lost productivity in working-age patients. Weight loss in particular has been associated with decreased survival [6]. The parotid target volume is ideal for proton therapy, which can nearly eliminate the exit radiation to organs at risk, such as the mucosa of the oral cavity and pharynx, contralateral major salivary glands, and swallowing musculature, thereby reducing the risk of acute toxicity. These advantages can facilitate the use of concurrent chemotherapy, which is currently under investigation in these tumors (NCT01220583) [7].

In the conventional radiotherapy literature, little is reported on the rate of acute toxicity with treatment of the parotid target. Alterio et al. reported outcomes of 43 patients treated with 3-dimensional conformal radiotherapy for parotid tumors (range, 60-66 Gy) [8]. Toxicities were reported using the European Organisation for Research and Treatment of Cancer/Radiation Therapy Oncology Group scoring system. Overall, 28% of patients experienced severe (grade 3 - confluent) mucositis; an additional 28% had grade 2 (patchy) mucositis, 30% had grade 1 mucositis (mucosal erythema), and only 14% had no evidence of mucositis. Al-Mamgani et al. reported outcomes and QOL data in 186 patients treated with radiotherapy for parotid cancers [9]. While acute toxicity was not reported, patient-reported QOL assessment demonstrated that most patients deteriorated in the domains of swallowing, sensation, sticky saliva, dry mouth, and social eating, although these scores rapidly returned to baseline in most domains.

To date, there have been no direct prospective studies comparing proton and conventional radiotherapy for salivary gland cancers. It is not likely that such a comparison would ever be available given the rarity of these cancers, limited proton resources, and ethical dilemmas associated with such studies. Thus, we are reliant on indirect comparative data to inform our treatment approach. Our study demonstrated that proton therapy essentially eliminates the risk of any severe acute mucosal toxicity, and that most patients experience either no toxicity or only grade 1 toxicities. Our findings are supported by a recently published comparison of patients treated with proton therapy to those treated with IMRT for head-and-neck tumors requiring ipsilateral radiation [10]. In this study, 23 patients were treated with IMRT and 18 with proton therapy during the same time range at two institutions. The two groups were balanced in terms of tumor size, age, radiotherapy dose, use of concurrent chemotherapy, and additional pathologic and clinical features. Dose-volume histogram comparisons demonstrated significantly reduced doses to normal tissues like the oral cavity, contralateral salivary glands, larynx, spinal cord, and brainstem. A comparison of acute toxicities demonstrated that proton therapy resulted in significantly lower rates of grade 2 or higher acute mucositis (16.7% versus 52.2%; P=0.019), nausea (11.1% versus 56.5%; P=0.003), and dysgeusia (5.6% versus 65.2%; P<0.001), without comprising tumor control. Only the rate of acute dermatitis increased with proton therapy (100.0% versus 73.9%; P=0.019).

Our study has several limitations, including the lack of a comparative group of patients treated with conventional radiotherapy, the lack of patient-reported QOL data, and incomplete data regarding both toxicities in the sub-acute period and time to recovery/resolution of toxicities. Nevertheless, we are able to demonstrate very minimal acute mucosal toxicity in a patient population that is expected to develop these symptoms when treated with conventional radiotherapy techniques. The overall impact of this toxicity reduction with regards to QOL, recovery of potentially lost productivity, or cost-effectiveness is yet to be determined.

In our cohort of 23 patient treated with ipsilateral radiotherapy to tumors of the parotid region, proton therapy resulted in very minimal acute mucosal toxicity. The majority of patients experienced either no or only grade 1 dysgeusia, mucositis, dysphagia, and xerostomia. The median percent weight loss was 3% throughout therapy and at first follow-up. Only 1 patient experienced transient grade 3 dysphagia. These findings should be validated in a larger prospective multi-institutional cohort with comparisons to similar patients treated with conventional radiotherapy and additional focus on the impact on patient-reported QOL, functional recovery, and cost-effectiveness.

Conflicts of Interest: The authors have no conflicts to disclose.