P09.06 Metabolomic-derived biomarkers predict immunotherapy efficacy in advanced non-small cell lung cancer

Background While immunotherapy improves survival in advanced Non-small cell lung cancer (NSCLC), predictive biomarkers are still needed. Emerging evidence suggests gut microbiota-related metabolites may influence treatment response, but their role in NSCLC remains unclear.Methods This prospective study enrolled previously untreated wild-type NSCLC patients scheduled to receive immune checkpoint inhibitors. Clinicopathological features were recorded. Baseline fecal and blood samples were collected. Fecal and serum metabolites were analyzed through untargeted metabolomics.Results 29 wild-type NSCLC patients were enrolled in the study. 808 differential metabolites between the two groups, of which 62 differential metabolites could be annotated. Responder enriched in L-Urobilin, Pentadecanoic acid, among others, while the non-responder enriched in L-Leucine, Glyceraldehyde, among others. Higher L-Urobilin levels were significantly associated with improved immunotherapy outcomes (P=0.0031). Multivariate analysis established L-Urobilin as an independent prognostic factor (HR=0.88, 95% CI 0.78-0.99, p=0.0389). Plasma metabolomics further validated the predominant accumulation of L-Urobilin in the responder cohort.Conclusion L-Urobilin may serve as a novel predictive biomarker for immunotherapy response in NSCLC. This study identifies a potential target for metabolomics-guided personalized immunotherapy. Qing Zhou Received honoraria from: AstraZeneca, Lilly, Roche, Pfizer, Boehringer Ingelheim, MSD Oncology, Bristol Myers Squibb, Hengrui, BeOne Medicines Ltd, outside the submitted work. Other authors declare no competing interests.