Disentangling the Effect of BMI on Hepatocellular Carcinoma From Cirrhosis With Multivariable Mendelian Randomization

Background & Aims: While cirrhosis is a primary risk factor for hepatocellular carcinoma (HCC), a significant proportion of HCC cases attributed to metabolic dysfunction-associated steatotic liver disease (MASLD) develop in the absence of cirrhosis. MASLD is strongly linked to obesity, a known risk factor for multiple cancers. Whether the effect of obesity on HCC is mediated via cirrhosis or other factors is unknown. Methods: We used univariable Mendelian randomization (MR) to test the total effect of a higher body mass index (BMI), a proxy for obesity, on HCC, and multivariable MR to test the direct effect. Results: We estimated that the effect of BMI was a 1.65-fold higher risk of HCC per standard deviation increase (95% confidence interval (CI): 1.28–2.12, p-value = 1.0 × 10⁻⁵). The BMI effect became indistinguishable from zero when taking liability to cirrhosis into account with multivariable MR (odds ratio = 1.12, 95% CI: 0.84–1.50, p-value = 0.44). We investigated additional potential pathways linking BMI to HCC—such as inflammation and type 2 diabetes—and explored the direct effect of childhood obesity on the risk of HCC. We found no direct effect of inflammation or type 2 diabetes (p-values > 0.05). Childhood body size increased the risk of HCC (odds ratio = 1.78, 95% CI: 1.27–2.49, p-value = 8 × 10⁻⁴), but the effect disappeared when we took adult body size into account using multivariable MR. Conclusions: Cirrhosis liability is the primary mediator of the causal effect of obesity on HCC.