Metabolic-Associated Fatty Liver Disease, Hepatitis B Surface Antigen Seroclearance, and Long-Term Risk of Hepatocellular Carcinoma in Chronic Hepatitis B

SIMPLE SUMMARY: Concomitant hepatic steatosis is common in patients with chronic hepatitis B (CHB), but data regarding its impact on the natural history of CHB remain sparse and conflicting. Recently, metabolic-associated fatty liver disease (MAFLD) was defined as hepatic steatosis in the presence of metabolic abnormality. The utility of MAFLD and its ability to assess hepatocellular carcinoma (HCC) risk for CHB requires further investigation. In this population-based cohort study of CHB, MAFLD was not associated with an increased risk of HCC. However, it was associated with a higher probability of functional cure for CHB in terms of HBsAg loss, and the MAFLD–HCC association was largely mediated by the development of HBsAg seroclearance driven by hepatic steatosis. On the contrary, the burden of metabolic risk abnormality was robustly associated with an increased risk of HCC in CHB, and this association was observed for participants both with and without hepatic steatosis and for participants both with and without HBsAg seroclearance. ABSTRACT: The value of metabolic-associated fatty liver disease (MAFLD) and its ability to assess hepatocellular carcinoma (HCC) risk remains uncertain for chronic hepatitis B (CHB). We evaluated the impacts of MAFLD and its coincidental metabolic abnormalities and related genetic predisposition on HCC incidence and mortality outcomes in CHB. We analyzed data from 1453 HBsAg-positive men (median age = 49.2 years at baseline) from a cohort of civil servants recruited from 1989–1992. MAFLD was defined as hepatic steatosis on ultrasound with obesity, diabetes, or metabolic dysfunction at baseline. During follow-up (median = 19.3 years), 105 HCC events occurred. MAFLD was not associated with HCC (adjusted hazard ratio (aHR) = 1.02) but was associated with a higher HBsAg seroclearance rate (aHR = 1.43). In mediation analysis, HBsAg seroclearance driven by hepatic steatosis explained 31.6% of the association between MAFLD and HCC. Antiviral treatment or fatty liver disease-associated genetic variants did not influence the MAFLD–HCC association. In contrast, even after adjustment for MAFLD and the other metabolic abnormalities, diabetes (aHR = 2.28), obesity (aHR = 1.72), and metabolic dysfunction (aHR = 3.30) increased the risk of HCC (all p < 0.030). The risk of HCC increased with the number of metabolic abnormalities (vs 0: aHR = 2.05 and 5.72 for 2 and ≥ 3 metabolic abnormalities, respectively), and the cumulative effect of metabolic abnormalities was found across subgroups categorized by hepatic steatosis as well as in participants both with and without HBsAg seroclearance. In conclusion, MAFLD was not associated with increased HCC incidence in CHB. A more informative assessment of HCC risk can be obtained by taking into account the number of metabolic abnormalities.