The impact of prior chemotherapy (PrC) on immunotherapy outcomes in non-small cell lung cancer (NSCLC) patients: Real data from a mono-institutional study

e21639 Background: Immunotherapy has changed the paradigm on NSCLC treatment. No widespread and reproducible predictive biomarkers have been established. We analyzed how PrC could affect ICI outcomes. Methods: We conducted a retrospective observational study and analyzed all NSCLC patients (pts) treated with ICI (nivolumab or atezolizumab or pembrolizumab) in our institution from 2015 to 2018 and followed until January 2020. All pts had received at least one PrC. We recorded clinical features of pts both before and after ICI treatment. Results: A total of 83 pts were included, with a median age of 69 (range, 47-82). Sixty pts (73%) were male, 74 (89%) were smokers and 81 (97%) had ECOG PS 0-1. Thirteen (15.7%) pts had a immune-related toxicity (iRT, G1-G2 76.9% and G3-G4 23.1%) and 21 (25.3%) continued ICI therapy beyond progression disease (PD). Patients with a PD as best response to PrC had more probability to reach a PD with ICI (p = 0.06). Median Progression-Free Survival (PFS) on ICI was 2.9 months (interquartile range, 1.9-9.4) with no statistical difference between pts with oligo- or diffuse progression (≤ or > 5 metastasis) during PrC (p = 0.42). Corticosteroids use was associated with worse PFS (p < 0.01). Median Overall Survival (OS) was 9.1 months (interquartile range, 3.3-26.5), with a benefit for pts with a stable disease (SD) or partial response (PR) to PrC (p = 0.02), for pts who experimented iRT (p = 0.04) and who didn’t receive corticosteroids (p < 0.01). In multivariate analysis liver or brain metastases (HR 8.8, 95% CI 2.9-26.8 and HR 2.2, 95% CI 0.6-8.1), coticosteroids use (HR 3.7, 95% CI 1.3-10.5) or previous cisplatin-based chemotherapy (HR 8.4, 95% CI 2.3-30.1) were associated with worse OS. Instead, pts whit iRT (HR 0.4, 95% CI 0.1-1.3) and PR as best response to PrC (HR 0.8, 95% CI 0.3-2.0) had a better OS. Conclusions: Our study confirms literature data and suggests that PrC could affect ICI outcomes. Tumor burden seems not to influence ICI outcomes, unlike response rate to PrC. Systemic corticosteroids use and iRT predicts ICI outcomes.