Editorial: Predicting Hepatocellular Carcinoma Development in Chronic Hepatitis B Patients With Metabolic Dysfunction—Do Existing Scores Clear the Bar?

Chronic hepatitis B (CHB) associated hepatocellular carcinoma (HCC) remains a significant global health concern, leading to significant morbidity and mortality. While most CHB patients undergo standard surveillance programmes, there is substantial research interest in developing predictive models to risk stratify patients and individualise surveillance schedules, as well as to support decision-making regarding the initiation of antiviral therapy [1, 2]. Metabolic dysfunction-associated steatotic liver disease (MASLD) has become a well-recognised risk factor for HCC development, yet its effect on carcinogenesis as a cofactor in CHB remains contentious [3, 4]. Furthermore, whether existing HCC predictive models for CHB can be used in patients with concurrent MASLD is unclear. In a large multicentre retrospective cohort of 1922 CHB patients with metabolic dysfunction, including 732 with concurrent MASLD, Patmore et al. investigated risk factors for HCC and explored the use of existing HCC prediction models [5]. Key factors that emerged include older age, male sex, lower platelet count, lower albumin, higher viral load, and cirrhosis status, defined as liver stiffness measurement > 12 kPa. Unsurprisingly, however, liver steatosis was not found to be associated with HCC development, which underscores the complex interplay between CHB and liver steatosis. At a basic level, the steatotic liver appears to be an unfavourable environment for the hepatitis B virus (HBV), as evidenced by a suppressed rate of HBV replication and accelerated HBsAg seroconversion in patients with liver steatosis [6, 7]. However, the evidence for whether concurrent MASLD accelerates or decelerates carcinogenesis and fibrogenesis in CHB remains notably mixed [3, 4, 8, 9]. Recently, numerous CHB prediction models for HCC development have been derived and tend to incorporate host factors (such as age and sex), liver factors (such as transaminases, albumin, platelet count, liver stiffness and cirrhosis status) or viral factors (such as e-antigen status and viral load) [10]. In derivation studies, the PAGE-B score comprising platelet count, age and sex appeared to discriminate HCC risk well, with a c-index of 0.82 [2], and was able to stratify patients into distinct risk categories: low-risk patients had a 0% cumulative HCC incidence by 5 years, compared to 17% in the high-risk group. When applied to the present study, low, medium and high-risk CHB patients with concurrent MASLD had similar 5-year cumulative HCC incidences of 0%, 2.8% and 11.1% respectively. However, receiver operator characteristic (ROC) curve analysis was notably absent, and whether the discriminatory power of PAGE-B, as evidenced by c-indices or area under the ROC curve, holds across the full spectrum of score values remains unclear. Furthermore, whether risk stratification by such models improves patient outcomes in practice remains unknown. With a rising incidence of metabolic dysfunction in the general population, clinicians will increasingly encounter CHB patients with concurrent MASLD. This timely study contributes to the expanding body of literature regarding the diverse interplay between CHB and MASLD. Whether MASLD confers additional risk of HCC development in CHB remains unclear, though PAGE-B appears to be a robust index that could additionally be applied to CHB patients with concurrent MASLD. Danny Con: writing – original draft, writing – review and editing. D.C. has received educational support from Viatris and is supported by an NHMRC Postgraduate Scholarship. This article is linked to Patmore et al papers. To view these articles, visit https://doi.org/10.1111/apt.70360 and https://doi.org/10.1111/apt.70410. The author has nothing to report.