Propensity score matching analysis of patients with inoperable stage III NSCLC treated with chemoradio- vs. chemoradioimmunotherapy.

Abstract

e21087 Background: Chemoradiotherapy (CRT) followed by consolidation treatment with the PD-L1 Inhibitor durvalumab is the new standard of care for inoperable stage III NSCLC. The present study compares outcome of patients treated with CRT alone to those treated with additional concurrent and/or sequential immune check-point inhibition (CRT-IO) using propensity-score matching analysis (PSM). Methods: PSM was performed with retro- and prospectively collected data of patients treated with CRT or CRT-IO (consolidation with durvalumab/concurrent and consolidation with nivolumab). Overall survival (OS), progression free survival (PFS) and time to loco-regional recurrence (defined as progression in the mediastinum and ipsilateral lung) were calculated from last day of thoracic irradiation. Results: Sixty-two (37%) of 166 treated patients were successfully matched; 31 received CRT and 31 CRT-IO. 18F-FDG-PET/CT for treatment planning was performed in 97% and cranial contrast enhanced MRI in 81% of patients. PSM was based on age, gender, PTV volume, histology, T- and N-stage. 36 and 51% vs. 42 and 46% of patients had T4- and N3-disease in the CRT and CRT-IO cohorts, respectively. All patients were irradiated to a total dose of at least 60Gy (EQD2). 90% of patients received two cycles of concomitant platinum-based chemotherapy (CRT: 82%, CRT-IO 96%). The median follow-up for 62 patients was 17.3 (range: 1.7-96.0) months. Median PFS was 7.1 (95%CI 2.2-12.1) months in CRT vs. 13.8 (95%CI 13.1-14.5) in CRT-IO patients (p = 0.004). Twelve-month PFS rates were 30% and 55% in the CRT and the CRT-ICI cohort, respectively. Median time to loco-regional recurrence was 15.3 months for CRT vs. not reached for CRT-IO patients (p = 0.050). 12-month loco-regional recurrence rates were 43% vs. 22%; 6- and 12-month brain metastases rates after completion of radiotherapy in the CRT vs. CRT-ICI cohort were 8% and 26% vs. 0% and 20%, respectively. Median OS was 19.1 (8.4-29.8 95%CI) months for CRT and not reached for CRT-IO patients (p < 0.001). 12-month survival rates were 62% and 93% in the CRT and CRT-IO cohort, respectively. Conclusions: The addition of concurrent and/or sequential IO to CRT led to an impressive improvement of loco-regional control, PFS and OS in the matching cohorts.