Salvage Reirradiation with Proton Beam Therapy for Locoregionally Recurrent Non-Small Cell Lung Cancer.

Background/objectives: This retrospective study evaluates outcomes of 66 patients who underwent reirradiation (re-RT) with proton beam therapy (PBT) for recurrent non-small cell lung cancer.

Methods: Toxicity was scored via the CTCAE v5.0, and outcomes estimated using the Kaplan-Meier method, with associations evaluated via Cox proportional hazards and logistic regression analyses.

Results: Patients were treated to a median re-RT prescription of 66 Gy/33 fxs (BED10 = 79 Gy; IQR: 71-84 Gy) at an interval of 1.4 years from prior RT. Half (50%) received concurrent chemotherapy. At 14 months follow-up, the median OS and PFS were 5 months (95%CI: 13-17) and 12.5 months (95%CI: 10-15), respectively. On multivariable analysis, a higher RT dose (BED10 > 70 Gy) [HR0.37; 95%CI: 0.20-0.68, = 0.001] and concurrent chemotherapy (HR0.48; 95%CI: 0.28-0.81, = 0.007) were associated with improved PFS, while treatment site overlap was adversely associated (HR1.78; 95%CI: 1.05-3.02, = 0.031). The median PFS for definitive RT with concurrent chemotherapy ( = 28), definitive RT alone (BED10 > 70 Gy) [ = 22], and lower prescription RT (BED10 < 70 Gy) [ = 16] was 15.5 months (95%CI: 7.3-23.7), 14.1 months (95%CI: 10.9-17.3), and 3.3 months (95%CI: 0-12.3), respectively (log-rank, = 0.006), with corresponding 2-year estimates of 37% (±9), 18% (±8), and 12.5% (±8), respectively. The incidence of Grade 3+ toxicity was 10.5% (6% pulmonary; 3% esophageal; and 1.5% skin), including one Grade 4 bronchopulmonary hemorrhage but no Grade 5 events. Cases with central site overlap had higher composite Dmax to the esophagus (median 87 Gy [IQR:77-90]), great vessels (median 120 Gy [IQR:110-138]), and proximal bronchial tree (median 120 Gy [IQR:110-138]) as compared to other cases ( ≤ 0.001 for all). However, no significant associations were identified with Grade 3+ events.

Conclusions: Thoracic re-RT with PBT is an option for recurrent NSCLC with acceptable outcomes and toxicity for select patients. When feasible, higher prescription doses (BED10 > 70 Gy) should be delivered for definitive intent, and concurrent chemotherapy may benefit individual cases.