Immune Microenvironment Differences Between Squamous and Non-squamous Non–small-cell Lung Cancer and Their Influence on the Prognosis

Abstract

IntroductionCheckpoint blockades have entered routine clinical use for non–small-cell lung cancer (NSCLC). However, there were some differences in efficacy and response predictors for anti-programmed cell death protein 1 (PD-1) antibodies between squamous (SQ) and nonsquamous (non-SQ) NSCLC. The study aims to elucidate the possible difference in immune microenvironment between SQ-NSCLC and non-SQ-NSCLC and their influence on the prognosis. Patients and MethodsA total of 197 patients with stages I to III NSCLC were included. cluster of differentiation 8 (CD8), cluster of differentiation 4 (CD4), transcription factor forkhead box P3 (FOXP3), and programmed death-ligand 1 (PD-L1) expression were examined in cancer nest and stroma on 85 SQ-NSCLC and 112 non-SQ-NSCLC samples using immunohistochemistry. ResultsMore CD8+ tumor infiltrating lymphocytes (TILs) were detected in the cancer nests (cCD8) from patients with SQ-NSCLC than those with non-SQ-NSCLC (56% vs. 34%; P = .002). There were no significant differences between the SQ and non-SQ groups in terms of other TIL markers or PD-L1 expression. Multivariate analysis showed that the degree of cCD8+ TIL infiltration was an independent positive predictor for overall survival (OS) in the SQ-NSCLC group (P = .003) and in the non-SQ-NSCLC group (P = .024). In the univariate analysis, CD8+ TILs in the stroma, CD4+ TILs in the cancer nest and stroma, and FOXP3+ TILs in the cancer stroma associated with different prognoses for patients with either non-SQ-NSCLC or SQ-NSCLC. Using a 10% cutoff, PD-L1 expression was a poor prognostic factor in total NSCLC (P = .011), stage I (P = .037), SQ-NSCLC (P = .097), and non-SQ-NSCLC (P = .051). ConclusionThe different cCD8+ TIL profile and different prognostic value with certain TILs indicates that SQ-NSCLC and non-SQ-NSCLC are likely different cancer types with respect to their immune microenvironments.