Establishing a histology-specific biologically effective dose threshold for lung stereotactic ablative radiotherapy (SABR): Is ≥100 Gy10 enough?

Abstract

Objectives: Squamous cell carcinoma (SCC) is associated with worse local control and overall survival (OS) compared to adenocarcinoma (ADC) in patients with early stage non-small cell lung cancer (ES-NSCLC). Biological effective dose (BED) escalation above 100 Gy10 improves tumor control, yet SCC and ADC may respond differentially to BED beyond 100 Gy10. Materials and Methods: We queried the National Cancer Database for ES-NSCLC (T1-2N0, Stage I-IIA) patients with SCC or ADC treated with stereotactic ablative radiotherapy (SABR). Receiver operator characteristic (ROC) curve analysis was used to identify the optimal dose threshold for SCC and ADC. Patients were stratified by histology and BED (≥122 Gy10 vs <122 Gy10). Univariable and multivariable analyses identified characteristics predictive of OS. Cox proportional hazard ratios with inverse probability weighting (IPW) were used to mitigate indication bias between the two dose arms. Results: Ultimately 11,084 ES-NSCLC patients with either ADC (n = 6476) or SCC (n = 4608) were eligible for analysis. Calculated optimal BED threshold for both SCC and ADC was 122 Gy10. Univariable analysis demonstrated a median (36 months vs 32 months), 3-year (51% vs 43%), and 5-year (27% vs 22%) OS advantage in SCC patients receiving BED escalation ≥122 Gy10 (p = 0.002). No survival difference was observed in the ADC dose escalation arm (p = 0.650). BED escalation ≥122 Gy10 remained an independent predictor of improved survival on IPW multivariable comparison (p < 0.0001). Conclusion: Escalation of BED ≥ 122 Gy10 was an independent prognosticator of improved survival in patients with SCC of the lung post-SABR. No survival benefit was observed for ADC, suggesting a differential response to BED escalation.