Clinical predictors of efficacy for immune checkpoint inhibition in lung cancer patients.

Abstract

e20600 Background: Immune checkpoint inhibitors (ICIs) have changed the treatment paradigm of advanced NSCLC and SCLC. The populations in clinical trials generally were composed of patients (pts) with excellent performance status (PS), a minimal number of prior lines of therapy, and no history of (h/o) autoimmune disease (AD). The aim of this retrospective study was to investigate the predictors of survival in less fit and more heavily pretreated lung cancer pts at our institution. Methods: Medical records of lung cancer pts who started single-agent ICI from 2015 to 2018 were reviewed for data collection. Progression-free survival (PFS) and overall survival (OS) were compared by age, PS, smoking, PD-L1 expression, brain metastasis, prior lines of systemic therapy, h/o AD, and immune-related adverse events (irAE). Results: We included 150 pts who received at least 1 dose of nivolumab, pembrolizumab, or atezolizumab (median age 66, range 36-92, 56% female, 89% NSCLC, 97% stage 4, 13% never smoker, 34% ECOG ≥ 2, 37% ≥ 2 prior lines of therapy, 29% developed irAE). The overall response rate was 30% and clinical benefit rate 51% (3% CR, 27% PR, 21% SD). The median PFS was 3.7 months (m) overall and 12.3 m for those with ≥ 4 cycles of therapy. The median OS was 12.4 m overall and 26.0 m for those with ≥ 4 cycles. The median PFS and OS were not significantly different by age, PS, smoking, PD-L1, brain metastasis, prior lines of therapy, or h/o AD. The median OS was longer for pts with ECOG 0-1 than those ≥ 2 (14.6 vs 7.5 m, p < .001). The median PFS and OS for pts with irAE were longer compared to those without irAE (12.3 vs 2.6 m, p < .001 for PFS and 28.9 vs 9.1 m, p = .001 for OS). Multiple Cox regression analysis identified ECOG 0-1 (HR 0.54, p = .007) and irAE (HR 0.48, p = .003) to be independently associated with improved OS. 27 pts (18%) required steroids for irAE. There were no treatment-related deaths. Of 11 pts with h/o AD, only 1 experienced disease flare. Conclusions: The clinical benefit of ICIs persists in older or heavily pretreated pts. ECOG 0-1 and incident irAE predicted for improved survival. Survival for pts with ECOG ≥ 2 is very limited, suggesting ICIs should be used judiciously in this group. Therapy was well tolerated, with a low risk for flare of previous autoimmune disease.