Evaluation of Proton Therapy Reirradiation for Patients With Recurrent Head and Neck Squamous Cell Carcinoma

IMPORTANCE Use of proton therapy reirradiation (PT-ReRT) for head and neck cancer is increasing; however, reports are heterogenous and outcomes can be difficult to interpret. OBJECTIVE To evaluate outcomes and toxic effects following PT-ReRT in a uniform and consecutive cohort of patients with head and neck squamous cell carcinoma. DESIGN, SETTING, AND PARTICIPANTS This retrospective cohort study included patients with recurrent primary head and neck squamous cell carcinoma who were treated with PT-ReRT from January 1, 2013, to December 31, 2020, at a single institution. Patient, clinical, and treatment characteristics were obtained, and multidisciplinary review was performed to record and grade early and late toxic effects. EXPOSURES Proton therapy reirradiation. MAIN OUTCOMES AND MEASURES Follow-up was defined from the start of PT-ReRT. The Kaplan-Meier method was used for outcomes of interest, including local control (LC), locoregional control, distant metastatic control, progression-free survival, and overall survival (OS). Cox proportional hazards regression modeling was used to assess associations of covariates with OS. RESULTS A total of 242 patients (median [range] age, 63 [21-96] years; 183 [75.6%] male) were included. Of these patients, 231 (95.9%) had a Karnofsky performance status score of 70 or higher, and 145 (59.9%) had at least a 10-pack-year smoking history. Median (range) follow-up was 12.0 (5.8-26.0) months for all patients and 24.5 (13.8-37.8) months for living patients. A total of 206 patients (85.1%) had recurrent disease vs second primary or residual disease. The median (range) interval between radiation courses was 22 (1-669) months. Median PT-ReRT dose was 70 cobalt gray equivalents (CGE) for the fractionated cohort and 44.4 CGE for the quad shot cohort. For the fractionated cohort, the 1-year LC was 71.8%(95% CI, 62.8%-79.0%) and the 1-year OS was 66.6% (95% CI, 58.1%-73.8%). For the quad shot cohort, the 1-year LC was 61.6%(95% CI, 46.4%-73.6%) and the 1-year OS was 28.5%(95% CI, 19.4%-38.3%). Higher Karnofsky performance status scores (hazard ratio [HR], 0.50; 95% CI, 0.25-0.99; P =.046) and receipt of salvage surgery prior to PT-ReRT (HR, 0.57; 95% CI, 0.39-0.84; P =.005) were associated with improved OS, whereas receipt of quad shot (HR, 1.97; 95% CI, 1.36-2.86; P <.001) was associated with worse OS. There were a total of 73 grade 3 and 6 grade 4 early toxic effects. There were 79 potential grade 3, 4 grade 4, and 5 grade 5 late toxic effects. CONCLUSIONS AND RELEVANCE The findings of this cohort study suggest that, compared with previous reports with photon-based reirradiation, patients are living longer with aggressive PT-ReRT; however, surviving patients remain at risk of early and late complications.