Exploring the causal role of multiple metabolites on ovarian cancer: a two sample Mendelian randomization study Original paper

What was studied?

This study used metabolites and ovarian cancer Mendelian randomization to test causal links between 486 serum metabolites and overall epithelial ovarian cancer (OC) and four histotypes. The authors performed a two-sample Mendelian randomization (MR) using inverse-variance weighted (primary), MR-Egger, and weighted median estimators, with sensitivity tests for heterogeneity, horizontal pleiotropy, and leave-one-out stability. They also ran reverse MR and pathway enrichment. The goal was to identify metabolites and pathways that may contribute to OC etiology and inform biomarker discovery and prevention.

Who was studied?

Genetic instruments for 486 circulating metabolites came from a metabolomics GWAS of 7,824 individuals of European ancestry; 309 were known metabolites categorized across eight KEGG classes. Outcomes were from OCAC GWAS summary data: 25,509 OC cases and 40,941 controls of European ancestry, with subtype datasets for serous (14,049 cases), endometrioid (2,810), clear cell (1,366), and mucinous (2,566).

Most important findings

The analysis yielded 112 suggestive metabolite–phenotype associations; stringent consistency across ≥3 MR methods reduced these to 18 associations spanning 14 known metabolites, with eight risk-linked and six protective signals. Notably for overall OC, asparagine associated with a lower risk (OR 0.65), while 4-acetamidobutanoate, alpha-hydroxyisovalerate, 3-(3-hydroxyphenyl)propionate, and X-13183 stearamide are associated with a higher risk. For subtypes, betaine is associated with reduced clear cell risk, whereas estrone-3-sulfate is associated with increased clear cell risk. For endometrioid OC, higher risk signals included 3-(3-hydroxyphenyl)propionate, 1,5-anhydroglucitol, 1-linoleoyl-GPE, and the fibrinogen-derived peptide ADpSGEGDFXAEGGGVR; protective signals for endometrioid included arachidonate (20:4n6) and stearidonate (18:4n3). For serous OC, DSGEGDFXAEGGGVR and salicylurate (2-hydroxyhippurate) are associated with a lower risk. Pathway analysis linked mucinous OC to caffeine metabolism, arginine biosynthesis, and the TCA cycle, and linked endometrioid OC to caffeine and alpha-linolenic acid metabolism. After FDR correction, associations remained suggestive rather than confirmatory, but heterogeneity and pleiotropy checks were negative and leave-one-out analyses were stable; reverse MR did not support reverse causation. From a microbiome perspective, several signals involve xenobiotics and aromatic metabolites that are microbially influenced in humans, suggesting plausible gut–tumor metabolic axes, although the study did not profile microbiota directly.

Key implications

For clinicians, these findings highlight metabolic pathways that may intersect with host–microbe chemistry in OC. The caffeine, arginine, and TCA axes in mucinous disease, and alpha-linolenic acid metabolism in endometrioid disease, point to diet- and microbiome-modifiable routes worth validation. Metabolites with suggestive risk signals and protective signals could seed candidate biomarker panels for risk stratification or recurrence surveillance. Yet none survived FDR correction, and all data were European; apply caution and seek replication in diverse cohorts with integrated microbiome and metabolomics profiling before clinical use.