Systemic inflammatory factors in non-alcoholic steatohepatitis-induced hepatocellular carcinoma: A bidirectional Mendelian randomization analysis

Background: Non-alcoholic fatty liver disease (NAFLD) has become the most prevalent chronic liver condition globally, with its severe form, cirrhosis, leading to liver failure, hepatocellular carcinoma (HCC), and mortality. The causal relationships among them remain unclear. Methods: This study employed a two-sample Mendelian randomization (MR) approach to systematically investigate the causal relationships among inflammatory proteins, non-alcoholic steatohepatitis (NASH), and HCC. We utilized NASH data and HCC data from the FinnGen biobank, and genomic data for 91 circulating inflammatory factors from GWAS database. Bidirectional MR analyses were conducted using the Inverse-Variance Weighted method as the primary analysis, supplemented by MR-Egger and weighted median methods for sensitivity analyses. Additionally, we performed heterogeneity, pleiotropy, and MR-PRESSO tests to assess the robustness of our findings. Results: Forward MR analysis identified five inflammatory factors significantly associated with NASH risk: CCL25 (OR = 0.8703, p = .0159), S100A12 (OR = 0.7582, p = .0204), IL18 (OR = 0.7412, p = .0026), and LIFR (OR = 0.6804, p < .001) showed negative correlations, while CXCL11 (OR = 1.2625, p = .0354) demonstrated a positive correlation. Reverse analysis revealed that NASH significantly influenced the levels of 10 circulating inflammatory factors, including Beta-nerve growth factor (OR = 1.0286, p = .0245), CCL20 (OR = 1.0382, p = .0012), CD5 (OR = 1.0268, p = .0466), CUBD1 (OR = 1.0371, p = .0078), CSF1 (OR = 1.0231, p = .0458), IL-12 beta (OR = 1.0261, p = .0381), IL-22R alpha 1 (OR = 0.9722, p = .0311), CCL3 (OR = 1.0372, p = .0013), OPG (OR = 1.0436, p = .0103), and uPA (OR = 1.0320, p = .0054). Further analysis revealed that IL-1 alpha levels were positively associated with HCC risk (OR = 1.5335, p = .0377), while SULT1A1 levels showed a negative correlation (OR = 0.7190, p = .0200). Pleiotropy analyses showed no significant heterogeneity or horizontal pleiotropy for most identified associations, except for OPG (heterogeneity p = .001, MR-PRESSO p < .001) and CUBD1 (MR-PRESSO p < .05) in the reverse analysis of NASH, which warrant cautious interpretation. Conclusion: This study unveils complex bidirectional causal relationships among inflammatory factors, NASH, and HCC. Intervention targeting IL-1 alpha in patients with liver cirrhosis will significantly impact the progression of HCC.