T-cell effector cytokine signature as predictor of survival and toxicity in metastatic NSCLC patients treated with immunotherapy.

Abstract

8555 Background: Immune checkpoint inhibitors (ICIs) have led to significant prolongation of survival in patients with advanced non-small cell lung cancer (NSCLC). However, less than 20% have durable long-term survival. Presence of a pre-existing tumor immune response that can subsequently be unleashed by ICIs predicts success to ICI. Specifically, a T- cell effector replete tumor immune microenvironment predicts for improved survival, following ICI. Given intra-tumor heterogeneity, we asked whether baseline and on treatment T cell effector mediated serum cytokines predict benefit from ICI in metastatic NSCLC. We also asked whether cytokines predicted immune related adverse events (iRAEs) from ICI in these patients. Methods: To assess whether serologic markers were associated with response and toxicity to immunotherapy, we conducted a multiplex ELISA for a panel of 70 innate and adaptive immune cytokines and chemokines in 100 NSCLC patients treated with PD-1 inhibitors, either as monotherapy or in combination with chemotherapy. Cytokine expression was correlated with outcomes (progression- free and overall survival, PFS/OS) and occurrence of any grade 3 or greater iRAE. Kaplan Meier analyses were performed for survival analyses, with the median value used to stratify cytokines. We then compared the predictive value of this cytokine signature with other predictive signatures for outcomes from ICIs. Results: Increased concentrations of baseline TRAIL correlated with improved PFS and OS; conversely, decreased baseline IL-3, IL-6, IL-8, APRIL, IL20, IL33, MCP4, IL-7, and TARC correlated with improved PFS. Increased baseline SDF1 and TRAIL correlated with OS; conversely, decreased baseline IL-6, IL-8, TPO, I-TAC, IL-3, APRIL, IL-20, IL-33, MCP4, MCP2, IL-15, MCSF, PDGFa, VEGFA, and MIP1d correlated with improved OS. Increased concentrations of TRAIL correlated with increased concentrations of SDF1, Perforin, and GzmB (T effector cell, Teff signature). A high Teff signature was associated with a statistically significant improvement in OS, in univariate and multivariate analysis. IL1b and IL-17a were statistically higher in on treatment samples taken from in patients who developed IRAE, whereas Eotaxin3 and MIP1d were lower in on treatment samples taken from in patients who developed IRAE. Conclusions: We report that an increase in a Teff cell cytokine signature at baseline predicts long term survival from ICI and may reflect the presence of a pre-existing immune response characterized by increased T-effector cells that are then subsequently unleashed by ICI therapy. Conversely, higher baseline levels of IL-1β and IL-17 indicate the presence of a heightened inflammatory state that predisposes to the occurrence of iRAEs, following ICI therapy.