Integrating Programmed Cell Death Ligand 1 (PD-L1) and Neutrophil to Lymphocyte Ratio (NLR) As Predictive Panel of Response to Nivolumab in Non-Small Cell Lung Cancer (NSCLC).

Background: Cost-effectiveness and viability of new predictive markers for immunotherapy limit their application in clinical practice. This retrospective study explores PD-L1 expression and NLR as an easy feasible panel to predict benefit to Nivolumab. Methods: Out of 108 pre-treated NSCLC pts included in the Nivolumab expanded access program, 71 (66%) were evaluable for PD-L1 expression on archival tissue samples by immunohistochemistry (clone E1L3N, Cell Signalling Technology) and NLR on baseline blood cell count. PD-L1 positivity was defined as expression on ≥ 1% of tumor cells, while NLR was considered as high (NLR>3) or low (NLR<3). Progression free survival (PFS) was estimated by Kaplan Meyer method; the combined predictive value of PD-L1/NLR for PFS and overall survival (OS) was assessed using multivariate Cox proportional hazard models. Results: Pts were mostly males (63%), smokers (86%), with adenocarcinoma (76%); EGFR + (10%), K-RAS + (33%). Median PFS and OS to Nivolumab were 5.52 months (mo) (95% [CI] 4.5-8.05) and NR (95% [CI] 7.4-NR), respectively. 1-y OS was 51%. PFS to Nivolumab was not affected by PD-L1 status (PD-L1 >1% 13.2 mo, 95% [CI] 3.4-NR vs PDL1 <1% 6 mo, 95% [CI] 3.35-14.21; p= 0.51). High NLR significantly predicted poorer PFS than low NLR (NLR>3: 2.5 mo, 95% [CI] 2-5.29 vs NLR<3: 8.61 mo, 95% [CI] 4.86-23.02, p<0.001). Although median OS was not reached at the time of analysis, 25th percentile survival rate favored low NLR (NLR<3: 6.84 mo, 95% [CI] 4.98-7.88 vs NLR>3: 2.89 mo, 95% [CI] 2.5-4.3, p<0.001). Patients were divided in two cohorts according to combined PD-L1/NLR (cohort1: PD-L1+/low NLR, cohort 2: PD-L1-/high NLR). In the multivariate analysis, PFS to Nivolumab was significantly longer in cohort 1 (Cohort 1: 13.26 mo, 95% [CI] 3.42-NR vs Cohort 2: 2.5 mo, 95% [CI] 1.45-6.08, p < 0.0001). OS is available only for cohort 2 (OS 3.29 mo 95% [CI] 2.76-5.16). Conclusions: This retrospective study supports NLR as predictive biomarker for survival to immunotherapy and shows how the combined use of PD-L1/NLR can predict longer PFS to nivolumab treatment. This feasible panel may be routinely applied to select pts for immunotherapy avoiding more complex and expensive methods.