Proton therapy for thymic malignancies: multi-institutional patterns-of-care and early clinical outcomes from the proton collaborative group and the university of Florida prospective registries

Abstract

Objective: Thymic malignancies (TM) are rare tumors with long-term survivorship, causing concerns for radiotherapy-related late side effects. Proton therapy (PT) reduces the radiation dose to organs at risk, potentially decreasing long-term toxicities while preserving disease control. We report patterns-of-care and early clinical outcomes after PT for thymoma and thymic carcinoma.

Methods: Between January 2008 and March 2017, 30 patients with TMs enrolled on one of two IRB-approved prospective protocols and received postoperative or definitive PT. Clinical outcomes, pathology, treatment dose, toxicities, and follow-up information were analyzed.

Results: Twenty-two thymoma patients with a median age of 52.1 years (range, 23–72) received a median RT dose of 54 Gy (RBE) (range, 45–70) either postoperatively (91%) or definitively (9%); 23% received adjuvant chemotherapy. Among eight thymic carcinoma patients, the median age was 65.5 years (range, 38-88) and median RT dose was 60 Gy (RBE) (range, 42–70) delivered postoperatively (75%) or definitively (25%); 50% received chemotherapy.

Median follow-up for all patients was 13 months (range, 2–59 months). Five patients relapsed, one locally (3%). Three patients died of disease progression, including two thymomas and one thymic carcinoma patient; a fourth died of intercurrent disease. One patient with thymic carcinoma and 1 with thymoma are alive with disease. No patients treated with PT for their initial disease (de novo) experienced grade ≥3 toxicities. The most common grade 2 toxicities were dermatitis (37%), cough (13%), and esophagitis (10%).

Conclusion: Adjuvant and definitive PT are being used in the treatment of TMs. Early results of the largest such cohort reported to date demonstrates an acceptable rate of recurrence with a favorable toxicity profile. Longer follow-up and a larger patient cohort are needed to confirm these findings.

Introduction

Thymoma and thymic carcinoma are rare thymic malignancies (TMs) arising from the thymus in the anterior mediastinum. The typical age of diagnosis is relatively young, from the fourth to sixth decade of life [1]. Primary treatment for patients with operable TM consists of surgical resection of the tumor. Patients with close or positive surgical margins after surgery or advanced-stage disease receive adjuvant external-beam radiation therapy (EBRT) ± systemic therapy [2]. If a patient has an an inoperable disease, definitive chemo-radiation is the treatment of choice.

Patients who receive treatment for thymomas experience 10-year overall survival rates of 80–90% [3,4]. Given this long survivorship and the relatively young age at diagnosis for patients with TMs, mitigating long-term treatment side effects are of paramount importance, similar to that of patients with mediastinal lymphoma [5].

EBRT delivered to the mediastinum increases the risk of cardiac disease and the development of secondary malignancies [6,7]. Studies of locally advanced non-small cell lung cancer and mediastinal lymphoma have demonstrated the correlation between patient overall survival or cardiac survival and heart dose [8–10]. In addition, the NCCN guidelines for managing TMs suggest that the mean total dose to the heart should be as low as reasonably achievable to potentially maximize survival [11]. Secondary to its dosimetric advantages, proton therapy (PT) allows for the reduction of radiation dose to the organs at risk around the mediastinum (i.e., heart, lungs, and esophagus) [12,13]. In comparison to conventional EBRT, PT potentially decreases the long-term toxicities of radiotherapy, while achieving comparable local control [14].

Although PT is being utilized in the management of TMs, data regarding clinical outcomes are sparse [14]. This study reports on the patterns-of-care and early clinical outcomes of patients treated with PT for thymoma and thymic carcinoma and prospectively enrolled on the Proton Collaborative Group (PCG) Registry and at the University of Florida Health Proton Therapy Institute (UFHPTI) Registry.

Methods

Between January 2008 and March 2017, 30 patients with a TM were treated with postoperative or definitive PT and their medical records were retrospectively reviewed. Patients were treated under a prospective institutional review board-approved outcomes protocol at the University of Florida or a multicenter prospective registry study conducted across six institutions: Procure PT Center New Jersey (Somerset, NJ, USA), Procure PT Oklahoma City (Oklahoma City, OK, USA), Northwestern Medicine Chicago Proton Center (Warrenville, IL, USA), Seattle Cancer Care Alliance PT Center (Seattle, WA, USA), Mayo Clinic Arizona (Scottsdale, AZ, USA), and the University of Maryland (Baltimore, MD, USA).

Clinical outcomes, pathology, treatment dose, acute toxicities using the CTCAE version 4.0, and follow-up information were analyzed. Baseline patient, disease, and treatment characteristics are listed in Table 1. Follow-up was weekly during PT treatment and then at the discretion of the treating physician. Patient follow-up time was calculated from the PT start date to the date of the last follow-up.

Table 1. Patient characteristics (N = 30).

Patient Characteristic	Thymic carcinoma	Thymoma
	n = 8 (%)	n = 22 (%)
Male/Female	3 (37.5%)/5 (62.5%)	9 (40.9%)/13 (59.1%)
Median age (range)	65 years (38–88)	53 years (23–74)
Masaoka stage
I	0 (0%)	1 (5%)
II	1 (12.5%)	7 (32%)
III	4 (50%)	10 (45%)
IV	1 (12.5%)	2 (9%)
Recurrent	0 (0%)	2 (9%)
Unknown	2 (25 %)	0 (0%)
Definitive PT	2 (25%)	2 (9%)
Postoperative PT	6 (75%)	20 (91%)
Surgical margins for Post-op patients
Negative (R0 resection)	0 (0%)	3 (14%)
Close (≤1mm)	1 (12.5%)	4 (18%)
Positive (R1 resection)	3 (37.5%)	7 (32%)
Positive (R2 resection)	0 (0%)	5 (23%)
Unknown	2 (25%)	1 (5%)
Median RT dose	60 Gy(RBE) (range, 45–70)	54 Gy(RBE) (range, 45–70)
Adjuvant chemotherapy	4 (50%)	5 (23%)

Patients were treated with passive-scatter, uniform-scanning, and pencil-beam scanning PT techniques. Two patients were treated with mixed photon/proton plans. Limited information on radiation treatment technique for the PCG registry patients was available. The UF treatment planning details have previously been reported [12].

Results

Patterns of proton therapy use

Of the 30 patients, 22 were treated with PT for a thymoma. The median age at diagnosis of these patients was 53 years (range, 23–74). Six of these patients (27%) were younger than 40 years old. Masaoka staging distribution was as follows: stage I, 1 (5%) patient; stage II, 7 (32%) patients; stage III, 10 (45%) patients; stage IV, 2 (9%) patients; and 2 (9%) patients were treated for recurrent disease. Most patients were treated postoperatively (91%), although 2 patients were treated with definitive PT and chemotherapy without surgery. One patient with recurrent disease was re-irradiated with PT after receiving prior radiation with photons. Over half of the thymoma cohort had positive margins after an R1 (n = 7) or R2 (n = 5) resection. The median postoperative PT dose was 54 Gy(RBE) (range, 45–70 Gy[RBE]). PT doses for the two definitively treated patients were 54 Gy(RBE) (re-irradiation case) and 70 Gy(RBE). The range of PT treatment doses can be attributed to patient differences such as tumor location, the extent of surgery, and pathological findings, the need to meet normal tissue constraints, and physician discretion. Overall, 23% received adjuvant chemotherapy, either combined cisplatin/doxorubicin/cyclophosphamide (n = 3), carboplatin (n = 1), or cisplatin (n = 1).

The remaining eight patients were treated for thymic carcinoma. The median age for these patients at diagnosis was 65 years (range, 38–88). One patient (12.5%) had Masaoka stage II disease, 4 (50%) had stage III disease, 1 (12.5%) had stage IV disease, and staging information was unknown for 2 (25%) patients. None of the thymic carcinoma patients were treated for a recurrent disease or with re-irradiation. Of these patients, 75% received postoperative RT while 25% underwent definitive PT. Four patients had a positive or close margin (≤1 mm) and margin status was unavailable for the other two patients whose tumors were resected. The median postoperative PT dose was 60 Gy(RBE) (range, 45–70 Gy[RBE]). PT doses for the two definitively treated patients were 62 Gy(RBE) and 70 Gy(RBE). Half of these patients received adjuvant chemotherapy.

Clinical outcomes

The median follow-up for the entire cohort was 13 months (range, 2–59 months). At last follow-up, three patients had died of disease, two patients were alive with disease and one patient had died of intercurrent disease. Four patients were treated with PT for their initial disease (de novo disease) had disease progression: three of whom experienced a distant recurrence and one a local in-field and distant progression. One patient who was re-irradiated with PT for recurrent disease presented with oligometastatic disease and had continual distant progression.

Patient characteristics for those with recurrences are detailed in Supplementary Table 1. One patient with unresectable thymic carcinoma who was treated definitively with concurrent chemotherapy and PT had a local-in field recurrence 13 months after the start of treatment and died of disease. Two thymoma patients, one treated postoperatively and one treated definitively, recurred distantly at 33 months in the lung and 44 months in the lung and liver, respectively; both died of disease. One thymic carcinoma patient treated postoperatively developed a distant recurrence in the lung at six months and was alive with disease at last follow-up.

Toxicities are reported in Supplementary Table 2. At baseline, patients most commonly experienced grade 1 fatigue (27%) followed by a grade 1 cough (20%). Two patients presented at baseline with grade 3 dyspnea. At follow-up during or after PT treatment, the most common grade 1 side effects were dermatitis (43%), fatigue (37%), and skin pain (40%). The most common grade 2 side effects were dermatitis (37%), cough (13%), and esophagitis (10%). No patients with de novo disease experienced grade ≥3 toxicities after PT. The patient who was treated with reirradiation for a recurrent thymoma developed grade 3 radiation pneumonitis two months after completing PT. Although the extended treatment course for this patient is unknown, the patient did require hospitalization and was alive with disease at last follow-up.

Discussion

This study reports on the use of PT for the treatment of TMs across multiple institutions. Patients with TM stand to benefit from PT, given the relatively young median age of diagnosis (40–60 years old) and long-term survivorship. Patterns of care, however, demonstrate that older individuals are those receiving PT, which is counterintuitive to who would benefit the most from this treatment modality.

Compared with published reports of patients with TM who receive postoperative photon-based radiotherapy, our overall study population was slightly older [15,16]. In addition, these reports suggest that younger patients with TMs are not being considered for radiotherapy. Modh et al. reported a median age of 53.2 years in their series of patients treated with locally advanced thymoma and thymic carcinoma at MD Anderson (Houston, TX) and Memorial Sloan Kettering (New York, NY) [16]. Perri et al. reported a median age of 51 years [15]. The median age for our cohort was 58 years old. The difference in age may reflect differences in insurance coverage and corresponding PT insurance approval rates among older versus younger patients with TM [15,17]. Yet, further investigation is needed to better understand why younger patients are not receiving PT.

Furthermore, a higher proportion of patients on our study were treated for unresectable disease than typically seen in other radiotherapy series [3,4,15], which may suggest that more challenging patients are referred for PT. Given the higher radiation doses, these patients require in the definitive setting (60–70 Gy), they stand to benefit most from PT in terms of maximizing disease control while minimizing side effects. Both in our study and in the cohort treated at the University of Pennsylvania [14], some patients were treated for the recurrent or oligometastatic disease. Several studies have recently demonstrated how PT allows for safe re-irradiation in lung cancer [18–20]; therefore, PT may also be beneficial in the setting of re-irradiation for TMs. Although there was a case of grade 3 pneumonitis in our study, it occurred in the re-irradiated patient and no patient with de novo disease experienced grade 3 pneumonitis. We hypothesize that photon-based radiotherapy would have caused the same or a worse outcome for this patient.

The dosimetric advantages of PT potentially mitigate the long-term increased risk of second cancer and cardiac disease (like cardiomyopathy, myocardial infarction, valvular disease, and pericarditis) as can be seen in patients who receive mediastinal EBRT [6,21]. Vogel et al. estimated a 45% relative reduction in the risk of major cardiac events with PT compared to intensity-modulated radiation therapy (IMRT) based on a predictive model developed for breast cancer patients [22]. In a retrospective thymoma study comparing the dosimetric plans of patients treated with PT and an analogous IMRT plan, Parikh et al. reported a statistically significant reduction in heart, lung, and esophagus doses. The relative dose reductions by PT to the heart, lungs, and esophagus for the plans analyzed were 42.5, 43.3, and 73.9%, respectively [13]. At the University of Florida, when comparing IMRT and PT plans for TM patients, PT on average reduced the mean heart dose by 36.5%, mean lung dose by 33.5%, and lung V20 by 27.7%, all of which were statistically significant [12]. However, in female patients, no significant relative dose reductions were observed when comparing the breast dose.

Early clinical outcomes with PT demonstrate an acceptable rate of disease control. In the present study, one patient who received definitive PT recurred locally, whereas distant progression was more common. Similarly, in a single-institution prospective study from the University of Pennsylvania, the three-year regional control rate was 96% [14].

Favorable acute toxicity profiles from PT were reported in two recent studies. Parikh et al. found no grade 2 toxicities other than dermatitis among four patients treated with PT. In another prospective study of 27 patients, only one patient with a history of prior thoracic radiotherapy experienced grade 2 radiation pneumonitis [14]. Other grade 2 toxicities included dermatitis (37%), fatigue (11%), and esophagitis (7%) [14]. In comparison, grade 2 toxicities in the present study were similar, with patients experiencing dermatitis (37%), cough (13%), and esophagitis (10%). The one re-irradiated patient experienced grade 3 pneumonitis.

Late toxicity assessment in patients with TM has been limited by relatively short follow-up, but Vogel et al. reported on an established organ-specific cancer incidence model to estimate the rate of secondary malignancies based on doses delivered to normal tissue. Overall, the study estimated that five additional second malignancies could be avoided in 100 patients by treating them with PT rather than photon therapy. Late cardiac toxicities from mediastinal irradiation have been well-studied in lymphoma [21], breast cancer [23], and lung cancer [8]. For example, a recent prospective multi-institutional study on locally advanced non-small cell lung cancer reported a 7% increase of grade 3 cardiotoxicities for every 1 Gy increase in mean heart dose, translating into an absolute risk of 10% for grade 3 cardiac events with a mean heart dose of 23 Gy among patients with no cardiac history [24]. In comparison, a typical photon-based plan for thymoma can easily exceed the mean heart dose of 23 Gy. Several dosimetric studies have shown a significant dose reduction to the heart using PT for thymoma [6,13,25]. Therefore, it is possible that, with longer follow-up, cardiac toxicities may begin to lessen with the increasing use of PT in this patient population.

Although this is the largest study to date assessing PT for patients with TMs, the present study was limited by its relatively small patient population, short follow-up compared to studies of more common intrathoracic malignancies and the inherent limitations of using a multi-institution registry database. The database lacks information on RT delivery techniques, tumor volume or location, and specifics on chemotherapy dosing. In addition, there was heterogeneity in radiation doses delivered to patients. Given the rarity of TMs and the known difficulty in accruing patients with intrathoracic disease for PT trials [17], future proton-based treatment for TMs may best be managed through a multi-institutional protocol similar to the International Thymic Malignancies Interest Group. Finally, we want to note that all recent publications on PT for thymic malignancies were based on three-dimensional conformal double-scattered proton therapy. As pencil-beam scanning becomes more widely available at proton centers and used to treat thoracic malignancies [26], we may see a further reduction in normal tissue toxicity with PT, particularly with regard to breast dose in female patients.

Conclusion

Adjuvant and definitive PT is being used in the treatment of thymic malignancies. Early results demonstrate an acceptable rate of recurrence with a tolerable toxicity profile. Longer follow-up and a larger patient cohort are needed to confirm these findings.

Acknowledgments

This research was made possible by James E. Lockwood, Jr., Professorship. We would also like to thank Robin Cacchio for research administration; Keri Hopper, Lana Cook, Rossio Rodriquez, and Tisha Adams for patient care; and Jessica Kirwan and Christopher Stich for editorial assistance.

Disclosure statement

The authors have no conflicts of interest to disclose.