Plasma proteomic profiling reveals shared biomarkers for MASLD-related hepatocellular carcinoma and gastrointestinal cancers in the UK Biobank

Background: Metabolic dysfunction-associated steatotic liver disease (MASLD) is a major global health challenge and markedly elevates the risk of hepatocellular carcinoma (HCC) and extrahepatic gastrointestinal (GI) cancers. Reliable biomarkers for early detection in MASLD populations remain lacking. We leveraged large-scale plasma proteomic profiling from the UK Biobank to identify circulating protein markers for MASLD-related HCC and GI cancers. Methods: We performed a nested case-control study within the UK Biobank. Differentially expressed plasma proteins were identified and evaluated using Gene Ontology, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment, and protein-protein interaction (PPI) network analyses. Overlapping dysregulated proteins between cancer types were mapped to determine shared pathways. Results: The UK Biobank cohort included 19,341 MASLD patients with proteomics data (mean age 61.1 ± 6.3 years), comprising 40 MASLD-HCC cases and 589 MASLD-GI cases. After excluding samples with protein data missing rates exceeding 30%, the study included 33 MASLD-HCC cases and 516 MASLD-GI cancer cases, matched 1:3 by age and sex to 99 and 1,548 MASLD controls, respectively.Quantitative proteomics identified 45 significantly dysregulated (44 upregulated and 1 downregulated) proteins in MASLD-HCC compared to MASLD, including established HCC-related keratins (e.g., KRT8, KRT18) and novel candidates (e.g., SPINT3, ADGRG1). Enriched pathways highlighted lipid metabolic dysfunction and xenobiotic metabolism. IL-6 and AGXT emerged as central hub genes in the PPI network. In MASLD-GI cancers, 24 proteins were significantly altered. Notably, 14 proteins were consistently dysregulated across both MASLD-HCC and MASLD-GI cancers, including CDHR2, INSL3, GPRC5C, FGF21, MME, CES1, KRT18, ADGRG1, HAO1, KLK3, GAST, and FOLR3. Conclusion: This large prospective proteomic study delineates distinct and overlapping plasma protein signatures of MASLD-related cancers. The shared biomarkers map to a unified "metabolism-damage-inflammation" axis, suggesting common mechanisms underlying carcinogenesis in MASLD. These findings provide a promising biomarker panel with potential utility for multi-cancer early detection and risk stratification in MASLD populations.