Assessing causal relationships between iron level, disorders of iron metabolism, and hepatocellular carcinoma: a Mendelian randomisation study : The first author.

[BACKGROUND] Emerging evidence indicates that elevated systemic iron levels and dysregulated iron metabolism are closely associated with the pathogenesis of hepatocellular carcinoma (HCC). Observational studies have linked iron overload (e.g., hereditary hemochromatosis) to an increased risk of HCC, potentially mediated through iron-induced oxidative stress, DNA damage, and chronic inflammatory responses. However, the causal relationship between specific iron-related factors-such as organ-specific iron accumulation, systemic iron metabolism disorders, or exogenous iron exposure (dietary/drug iron)-and HCC development remains unconfirmed.

[MATERIALS AND METHODS] The data of genome-wide association studies (GWAS) involving a variety of human iron content indicators were collected and summarized, covering multiple dimensions such as liver iron content, spleen iron content, pancreas iron content, dietary iron content, drug iron content, and serum iron content, and the information related to genetic diseases of iron metabolism was also included. To explore the potential causal association between iron metabolism disorders and the risk of HCC using two-sample univariate and multivariate Mendelian randomization (MR) techniques. The inverse variance weighting (IVW) method was used as the main analysis method.

[RESULTS] No significant level of pleiotropy was detected according to the MR pleiotropy residual test. Cochran's Q test and MR-Egger regression analysis did not find evidence of heterogeneity and directional pleiotropy of single nucleotide polymorphisms (SNPs). The results of univariate MR Estimation based on the IVW. method show that Liver iron content (OR = 1.0009, CI = 1.0005-1.0013, p = 9.6499e-06) and disorders of iron metabolism (OR = 1.0000, CI = 1.0000-1.0001, P = 4.0417e-08) were the causes of liver cancer. There was no evidence of a causal relationship between drug iron(OR = 1.0176, CI = 0.9612-1.0773, p = 0.5486), dietary iron(OR = 0.9843, CI = 0.9683-1.0007, p = 0.0603) and HCC. Multivariate MR Estimation results based on the IVW method showed that no significant causal association was found between liver iron content, spleen iron content, pancreas iron content, serum iron content, iron metabolism disorder, and HCC risk after using the IVW method.

[CONCLUSIONS] The univariate results may reflect the complexity of the true association, which needs to be further explored in combination with other methods.