Blood monocyte and lymphocyte counts patient outcomes in non-small cell lung cancer after immune checkpoint inhibitor therapy: Real-world data of 12,416 patients.

Abstract

e21023 Background: Immune checkpoint inhibitors (ICPi) are now commonly used in non-small cell lung cancer (NSCLC) treatment. Beyond PD-L1 status, no clinically validated predictive biomarkers guide therapeutic decision making regarding their use. Blood monocyte, lymphocyte counts, or their ratio might inform clinical utility of these expensive, potentially toxic agents. Methods: Adults with advanced/metastatic NSCLC receiving any ICPi after diagnosis from January 2015 to November 2020 were included. Treatment outcomes were followed until 31 May 2021. Data were sourced from structured fields of electronic health records from The US Oncology Network and non-Network practices. The primary end point was overall survival (OS), and time to treatment discontinuation (TTD) was secondary. Baseline absolute monocyte count (AMC), absolute lymphocyte count (ALC), and monocyte-to-lymphocyte ratio (MLR) were subdivided into quintiles (Q) with Q1 lowest and Q5 highest. Patient (pt) and treatment characteristics were described. Time-to-event outcomes were assessed by Kaplan-Meier methods. Univariate and multivariate Cox regression analyses described associations among baseline biomarkers, OS, and TTD. Results: For 12,416 pts with NSCLC, median OS and TTD were 13.6 months (mos) and 4.5 mos, respectively. For AMC, patients in Q2 and Q3 had the longest median OS of 15.9 mos. Median TTD by AMC was longest for Q1 and Q2 (5.1 mos), declining sequentially to 3.3 mos for Q5. For ALC, Q5 had longest median OS of 19.2 mos, as well as longest median TTD of 5.8 mos. For MLR, Q1 had the longest median OS and TTD of 22.3 and 6.2 mos., respectively. In univariate Cox regression, AMC, ALC, and MLR each associated with OS and TTD (p < 0.0001). Pts in AMC Q5, compared with AMC Q1, had an OS hazard ratio (HR) of 1.48 (95% CI 1.38-1.60). For ALC, each successive increasing quintile, compared to ALC Q1, demonstrated a lower risk of death (p < 0.0001). Pts in MLR Q5, compared with MLR Q1, had an OS HR of 2.21 (95% CI 2.04-2.39). Conclusions: This data set is the largest to date to examine interactions among baseline immune cell counts, ICPi use, and pt outcome in NSCLC. These data inform efforts to describe immune-based prognostic risk groups among NSCLC pts receiving immunotherapy.[Table: see text]