Integration of serum metabolomics into clinical assessment to improve outcome prediction of metastatic soft tissue sarcoma patients treated with trabectedin

Soft tissue sarcomas (STS) are a group of rare and heterogeneous cancers with few diagnostic or prognostic biomarkers. This metabolomics study aimed to identify new serum prognostic biomarkers to improve the prediction of overall survival in patients with metastatic STS. The study enrolled 24 patients treated with the same trabectedin regimen. The baseline serum metabolomics profile, targeted to 68 metabolites encompassing amino acids and bile acids pathways, was quantified by liquid chromatography-tandem mass spectrometry. Correlations between individual metabolomics profiles and overall survival were examined and a risk model to predict survival was built by Cox multivariate regression. The median overall survival of the studied patients was 13.0 months (95% CI, 5.6–23.5). Among all the metabolites investigated, only citrulline and histidine correlated significantly with overall survival. The best Cox risk prediction model obtained integrating metabolomics and clinical data, included citrulline, hemoglobin and patients’ performance status score. It allowed to distinguish patients into a high-risk group with a low median overall survival of 2.1 months and a low-to moderate-risk group with a median overall survival of 19.1 months (p < 0.0001). The results of this metabolomics translation study indicate that citrulline, an amino acid belonging to the arginine metabolism, represents an important metabolic signature that may contribute to explain the high inter-patients overall survival variability of STS patients. The risk prediction model based on baseline serum citrulline, hemoglobin and performance status may represent a new prognostic tool for the early classification of patients with metastatic STS, according to their overall survival expectancy.