METLUNG: plasma metabolomic profiling identifies metabolomic signatures associated with immunotherapy outcomes in non-small cell lung cancer.

backgroundIdentifying reproducible biomarkers associated with treatment outcomes remains a major unmet clinical need in advanced non-small cell lung cancer (NSCLC). METLUNG evaluated whether plasma metabolomic profiling, alone or combined with clinical variables, could identify biomarkers associated with clinical response and survival outcomes in patients with NSCLC treated with immune checkpoint inhibitors.

methodsMETLUNG is a prospective observational cohort (ISRCTN98848959) that enrolled 128 patients with stage III-IV NSCLC eligible for immune checkpoint inhibitors between October 2020 and December 2022 at Hospital Universitari Sant Joan de Reus, Reus (Spain). Clinical response classification was based on Immune Response Evaluation Criteria in Solid Tumors criteria, and subsequent clinical and radiological course. Baseline plasma metabolomic profiling was followed by univariate analyses and multivariate modeling. Survival models were built using clinical variables, metabolomic variables, significant metabolites, and an integrated clinical-metabolomic composite model. Predictive performance was evaluated using integrated area under the receiver operating characteristic curve and concordance index (C-index).

resultsAmong 127 evaluable patients, 52.8% achieved a clinical response. Most had adenocarcinoma (n=94), followed by squamous carcinoma (n=28). Treatment included first-line immunotherapy monotherapy (n=39), first-line chemo-immunotherapy (n=47), or second-line immunotherapy (n=41). Clinical response differed by sex and Eastern Cooperative Oncology Group performance status but showed no association with smoking history or PD-L1 expression.Univariate analyses identified 41 baseline metabolites associated with response, and 69 and 71 metabolites associated with progression-free survival and overall survival (OS), respectively. Five metabolites (histidine, citric acid, uracil, lactic acid, and sarcosine) remained consistently significant across all endpoints. Higher histidine and citric acid levels were associated with improved outcomes, whereas uracil, lactic acid, and sarcosine correlated with worse outcomes. The integrated clinical-metabolomic model showed the highest prognostic performance compared with clinical or metabolomic prognostic models alone, providing better early survival discrimination. In the immunotherapy monotherapy subgroup, only the integrated model was significant for OS, while metabolomic signatures showed stronger prognostic value.

conclusionsMetabolomic profiling identified metabolic signatures associated with clinical response and survival outcomes in NSCLC patients treated with immune checkpoint inhibitors. An integrated clinical-metabolomic model showed improved prognostic stratification, supporting the potential value of plasma metabolomics for patient risk stratification. These findings remain exploratory and require validation in independent, ideally multicenter, cohorts before clinical implementation. TRIAL REGISTRATION NUMBER: ISRCTN98848959.