Comprehensive analysis of NSAIDs use and oncological outcomes in non-small cell lung cancer patients treated with immune checkpoint inhibitors.

Abstract

2530 Background: The effect of non-steroidal anti-inflammatory drugs (NSAIDs) on immune checkpoint inhibitors (ICIs) efficacy in non-small cell lung cancer (NSCLC) remains controversial. Although the COX-2/PGE2 pathway, a primary target of NSAIDs, has been implicated in diminished immunotherapy response, direct clinical association with NSAIDs and ICIs in real world has yet to be established. This study aims to evaluate the impact of NSAIDs use—considering types, duration, and timing—on ICI efficacy, alongside its effects on PGE2 and immune cell profiles. Methods: We included stage III-IV NSCLC patients receiving PD-1/PD-L1 antibodies in 5 centers. Blood and tumor samples were collected in perspective cohort. NSAIDs were categorized based on selectivity (non-selective COX inhibitors, selective COX-2 inhibitors) and chemical structure (salicylates, propionate derivatives, others). PGE2 and cytokines were measured in blood by ELISA. RNA sequencing data were obtained from databases. Tumor tissues were collected for immunohistochemical staining of immune cells. Multivariate Cox and logistic regression were used in analyses of progression-free survival (PFS) and objective response rate (ORR). Results: 883 patients were included, with 140 NSAIDs users and 743 non-users. 196 patients were enrolled prospectively with samples. Multivariate analysis showed that NSAIDs use was significantly associated with improved PFS (HR 0.67, 95% CI 0.51-0.88, P = 0.005) and ORR (OR 1.87, 95% CI 1.29-2.72, P = 0.001). Subgroup analyses indicated that non-selective COX inhibitors, salicylates, long-term use, and pre-ICI initiation were correlated with better outcomes. In contrast, selective COX-2 inhibitors, propionate derivatives, others, short-term use, and post-ICI initiation showed no effect on PFS or ORR. Blood analyses indicated that NSAIDs significantly lowered PGE2 levels, particularly salicylates and long-term use. Higher PGE2 was associated with worse outcomes. For immune cells, RNA sequencing revealed that COX-2 and mPGES-1 were significantly correlated with neutrophil enrichment and neutrophil-related cytokines. Single-cell RNA-seq showed high expression of COX-2 and mPGES-1 in neutrophils. Analysis of samples confirmed that NSAIDs use was associated with reduced neutrophils and neutrophil-related cytokines in blood and less neutrophil infiltration in tumor. Conclusions: NSAID use is an independent predictor of improved PFS and ORR in NSCLC patients receiving ICIs. Specifically, non-selective COX inhibitors, salicylates, long-term use, and pre-ICI initiation are associated with better clinical outcomes. NSAID use may enhance ICIs efficacy by reducing serum PGE2, which could serve as a predictive biomarker. Furthermore, NSAIDs decrease neutrophils in both blood and tumor, potentially contributing to the improvement in ICI efficacy.