Abstract
BACKGROUND: Metabolomic dysregulation contributes to prostate cancer (PCa) pathogenesis, and studies suggest that circulating metabolites have strong clinical potential to act as biomarkers. However, evidence of circulating metabolite associations has not been quantitively aggregated. METHODS: Systematic searches were performed in PubMed and Embase (October 17(th), 2024) to identify pre-diagnostic untargeted serum metabolomic studies of PCa risk. After harmonizing metabolite names across studies, restricted maximum likelihood was used to conduct meta-analyses to quantify associations between metabolites and risk of overall PCa, low- to intermediate-risk PCa, high- to very high-risk PCa and lethal PCa, as defined by the NCCN. Statistical significance was defined as FDR-adjusted P<0.05. Enrichment analyses were conducted on significant metabolites to identify biologically relevant pathways. Correlation of effect estimates between PCa outcomes was assessed via Pearson correlation. RESULTS: We identified 12 untargeted pre-diagnostic circulating metabolomic studies in a systematic review and meta-analyzed associations between up to 408 metabolites with four PCa outcomes. Three, eleven and nineteen metabolites were significantly associated with risk of overall, high/very high-risk and lethal PCa, respectively. Metabolites associated with high/very high-risk PCa were significantly enriched for lipids. Limited evidence of correlation between metabolite effects across outcomes was identified, highlighting potentially unique metabolite drivers of high-risk and lethal PCa. Follow-up analyses found that 13 of the significant metabolites were drug and/or dietary modifiable. CONCLUSIONS: These findings suggest the strong potential for metabolites to inform risk of lethal PCa, which could inform risk-stratified screening strategies and facilitate the identification of targets for PCa prevention.