Real world incidence, severity and timing of adverse events (AEs) among patients with metastatic non-small cell lung cancer (NSCLC) receiving second-line (2L) immuno-oncology (IO) therapy vs chemotherapy (C)

Abstract

Background: Randomized clinical trials (RCTs) in metastatic NSCLC have demonstrated superior responses and lower rates of AEs with 2L IO compared with 2L C. Realworld data (RWD) are lacking to confirm tolerability of IO reported in the controlled environment of RCTs. This study examines RWD on AEs among patients (pts) with metastatic NSCLC receiving either 2L IO or C. Methods: Cota database electronic health records (EHR) of 55 oncologists at 13 US centers (2 academic; 11 community) were reviewed to identify pts with stage IV NSCLC who received 2L therapy following prior first‐line C (March 2015 ‐ December 2017). EHR documented AEs were graded per CTCAE v4 criteria by trained oncology nurses. Results: Of 206 pts identified, 152 received 2L IO and 54 2L C. No differences were noted between cohorts in age (median [years]; IO: 70; C: 65; p=0.21), sex (IO: 55% male, C: 41%; p=0.08), or age‐adjusted Charlson comorbidity index (IO: 4; C: 4; p=0.92); more IO pts had squamous cell histology (IO: 23%; C: 9%; p=0.03). Median duration of 2L therapy [mo (range)] was 2.3 (0‐19.1) with IO and 1.7 (0‐21.6) with C. AEs of grade 3‐4 or any grade resulting in treatment change or discontinuation occurred in 39 (19%) pts; less frequently with IO (13%) vs C (35%; p<0.01); with 87% of AEs occurring by 3 mo. Cumulative%of pts experiencing AEs by 1, 2, 3, and 4mo of treatment was significantly lower (p<0.05) at all time‐points with IO vs. C (5.3, 8.6, 10.5, 10.5 vs. 25.9, 25.9, 33.3, 35.2, respectively). Unadjusted survival data were similar. Conclusions: This retrospective study of RWD shows 2L therapy with IO, compared with C, is associated with a lower frequency of AEs of grade 3‐4 or any grade leading to treatment discontinuation, over 1 to 4 months of therapy. A limitation is that spontaneous reporting of AEs in RWD likely captures fewer AEs compared with RCTs. (Table Presented).