Propensity-matched analysis of concurrent/sequential versus sequential immune checkpoint inhibition in inoperable stage III NSCLC patients treated with chemoradiotherapy.

Abstract

e20589 Background: Durvalumab maintenance treatment after completion of concurrent chemoradiotherapy (CRT) in patients with inoperable stage III NSCLC is the international standard. In this propensity-matched analysis, we prospectively evaluated the role of concurrent/sequential versus sequential immune checkpoint inhibition. Methods: Between 10/1/2016 and 12/31/2020, 39 NSCLC patients in stage IIIA/B/C were analyzed. 11 (28%) patients received concurrent/sequential PD-1 inhibition (nivolumab) and 28 (72%) patients received sequential PD-L1 inhibition (durvalumab) up to one year after the end of CRT. A 1:2 propensity score matching (PSM) using age, gender, T category and histology was performed to reduce selection bias and address for confounding factors (n = 33, 11 patients receiving nivolumab (SIM-cohort), 22 patients receiving durvalumab (SEQ-cohort). Treatment-related adverse events were assessed weekly during the CRT and at 6 weeks, 3,6,9, and 12 months after the end of CRT. Results: The median follow-up time of the overall cohort, SIM-I cohort, and SEQ-I cohort was 27.6, 33.3, and 26.5 months after CRT, respectively. For the entire cohort median survival (OS) was not reached; median progression-free survival (PFS) achieved 26.3 months. In the SIM-cohort, median PFS was 22.8 months and median OS was not reached. In the SEQ-cohort, neither median PFS nor OS was reached. PFS at 12/24 months was 82/44% in the SIM-cohort and 57/57% in the SEQ-cohort (p = 0.714), respectively. In the SIM-cohort, 18.2% of patients showed grade III pneumonitis; in the SEQ-cohort 13.6% (p = 0.735). Grade 4 and 5 toxicities were not observed. Conclusions: Both concurrent/sequential and sequential immune checkpoint inhibition shows a favorable side effect profile and promising survival in terated patients. Concurrent immunotherapy did not result in an improved outcome (PFS, OS) compared to sequential approach. However, our propensity-matched analysis found that concurrent immunotherapy was associated with a non-significant moderate increase in grade III pneumonitis.