Survival stratification using a baseline inflammatory physiology based scoring system in advanced non-small cell lung cancer (NSCLC) treated with anti-programmed cell death-1 (anti-PD-1) therapy.

Abstract

152 Background: Immune checkpoint blockade (ICB) has shown promise in NSCLC with improved survival and durability in disease control. Despite these advances, the response to ICB remains variable. Thus identifying easily available biomarkers that can assist in the optimal selection of patients for ICB holds paramount importance. Methods: Retrospectively we identified 87 stage III/IV NSCLC patients initiated on anti-PD-1 therapy from April 2015 to March 2017 after progressing on a platinum doublet. These patients were part of an ongoing prospective biomarker-based study at our institution. Follow up cutoff for survival analysis was set at October 1, 2017. Enrolled participants had inflammatory markers (C-reactive protein, absolute neutrophil count, absolute lymphocyte count, serum albumin) measured on the day of first dose of anti-PD-1 administration as well as subsequent doses. Using multivariate Cox analysis, factors demonstrating an association with overall survival after immunotherapy (OSI) were used to develop a composite score to stratify patient survival. Results: The median age was 64 years with predominant histology being adenocarcinoma in 46.0 % followed by squamous cell carcinoma (43.7 %). Stage IV disease was present in 70.1%, with skeletal involvement (54.1 %) and liver (27.9 %) being the most common metastatic sites. In the multivariate Cox regression with backward elimination, factors independently associated with OSI were noted to be: CRP, neutrophil-lymphocyte ratio, and prognostic nutritional index. A composite inflammatory biomarker score was developed using the B-coefficients from the Cox multivariate regression. A score > 1 demonstrated inferior OSI compared to a score of ≤ 1 [1.7 vs. 9.3 months; P < 0.001, HR 4.00, 95% CI (2.21-7.25)]. Conclusions: This study provides preliminary evidence in favor of a composite inflammation based score that can aid in survival stratification of these patients. Validation of this score in prospective NSCLC trials to elucidate its potential utility as a predictive or prognostic tool in facilitating optimal patient selection for ICB is required.