The relationship between the serum lipid profile and hepatocellular carcinoma in east Asian population: A mendelian randomization study.

Background: Although several studies have found that the serum lipid profile may be correlated with hepatocellular carcinoma (HCC), the causal relationships between the serum lipid profile and HCC have not been determined due to potential confounder. Here, Mendelian randomization (MR) analysis was performed to identify the relationship between the serum lipid profile and HCC in the East Asian population. Method: Our study made a MR analysis with the validation of two data sets. We obtained genome-wide association study (GWAS) data related to the serum lipid profile from Asian Genetic Epidemiology Network (AGEN). Then, the data from a recent large GWAS of the East Asian ancestry in Japan (BioBank Japan, BBJ) were extracted. Summary-level statistical data for HCC were obtained from a large GWAS of the East Asian ancestry in Japan. Univariable MR analysis were performed to identify whether the genetic evidence of serum lipid profile was significantly associated with HCC risk. Multivariable MR analysis was conducted to estimate the independent effects of exposures on HCC. Results: Univariable and multivariable MR analyses indicated that the serum lipid profile was not a risk factor for HCC incidence in either data set based on the East Asian population. Multivariable MR analysis revealed that the hazard ratios of the probability of HCC in AGEN were 1.134 (95% confidence interval (CI), 0.903-1.424) for TG, 1.010 (95% CI: 0.824-1.237) for HDL-C, 0.974 (95% CI: 0.746-1.271) for TC, 0.918 (95% CI: 0.734-1.147) for LDL-C, while the results in BBJ were also non-significant: 1.111 (95% CI: 0.869-1.419) for TG, 0.957 (95% CI: 0.790-1.158) for HDL-C, 0.917 (95% CI: 0.643-1.308) for TC, 0.932 (95% CI: 0.699-1.243) for LDL-C. Conclusion: Our MR study with the validation of two data sets found no strong evidence to support causal associations between the serum lipid profile and HCC risk.