Abstract
The causal links between serum amino acids (AAs) and hepatobiliary neoplasms remain unclear. This study aimed to systematically investigate these associations using Mendelian randomization (MR). Summary-level data on 20 serum AAs were obtained from publicly available genome-wide association studies. Genome-wide association studies data on hepatobiliary neoplasms - including primary liver cancer (PLC), hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC), gallbladder and extrahepatic bile duct carcinoma, secondary liver cancer, benign liver tumors, and benign tumors of the extrahepatic bile ducts - were derived from FinnGen and 2 UK Biobank-based studies. A meta-analysis was conducted to calculate pooled effect sizes. Inverse variance weighting was the primary method, supplemented by MR-Egger, weighted median, MR.RAPS, maximum likelihood, and MR-PRESSO methods for sensitive analyses. Higher serum methionine was associated with lower risks of PLC (OR = 0.84, 95% CI: 0.72-0.97, P = .021) and HCC (OR = 0.87, 95% CI: 0.80-0.94, P < .001), but not ICC. Alanine increased PLC risk (OR = 1.19, 95% CI: 1.00-1.42, P = .047), with no significant effect on HCC or ICC. No AAs were linked to gallbladder and extrahepatic bile duct carcinoma or secondary liver cancer. For benign tumors, aspartate (OR = 1.13, 95% CI: 1.01-1.26, P = .037), cysteine (OR = 0.72, 95% CI: 0.57-0.92, P = .008), and lysine (OR = 1.49, 95% CI: 1.15-1.93, P = .003) were significantly associated with benign liver tumors or benign tumors of the extrahepatic bile ducts. This study offers robust evidence of causal associations between specific serum AAs and hepatobiliary neoplasms, emphasizing their potential as biomarkers and modifiable targets for early intervention.