Which risk model can better predict hepatocellular carcinoma in hepatitis B patients with an antiviral treatment duration of over 1 year?

Comparative performance of risk prediction models for hepatitis B-related hepatocellular carcinoma in the United StatesJournal of HepatologyVol. 76Issue 2PreviewGuidelines recommend hepatocellular carcinoma (HCC) surveillance in patients with chronic HBV infection. Several HCC risk prediction models are available to guide surveillance decisions, but their comparative performance remains unclear. Full-Text PDF This work was supported by grants from the National Natural Science Foundation of China (No. 82171834) and Key Research and Development Program of Social Development of Jiangsu Province (No. BE2022725). Ming-Cheng Guan (Conceptualization; Writing – original draft); Qian Ding (Writing – original draft); Hong Zhu (Conceptualization; Supervision; Funding acquisition; Writing – review & editing). The authors declare no conflicts of interest that pertain to this work. We read with great interest the article by Kim HS et al.[1]Kim H.S. Yu X. Kramer J. Thrift A.P. Richardson P. Hsu Y.C. et al.Comparative performance of risk prediction models for hepatitis B-related hepatocellular carcinoma in the United States.J Hepatol. 2022; 76: 294-301Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar The retrospective study estimated the performance of 10 models for predicting the risk of developing hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB) treated with entecavir/tenofovir in the United States. Generally, most models performed well in the large cohort of US-based patients, especially RWS-HCC, REAL-B, and AASL-HCC. Furthermore, model performance remained good in subgroups defined by race/ethnicity (White vs. African American patients). However, we would like to raise the following comments: Firstly, although all the patients included in the study were treated with entecavir or/and tenofovir, they received an average of 3.1 years (standard deviation 2.8 years) of antiviral treatment during the follow-up period, with more than one-fourth of patients on less than 1 year of treatment; in other words, the time span of treatment varied greatly among different patients. Current evidence demonstrates that long-term effective nucleos(t)ide analogue (NA) therapy (including entecavir and tenofovir) enables significantly lowering the chance of developing HCC in CHB patients.[2]Manne V. Gochanour E. Kowdley K.V. Current perspectives into the evaluation and management of hepatitis B: a review.Hepatobiliary Surg Nutr. 2019; 8: 361-369Crossref Google Scholar,[3]European Association for the Study of the LiverElectronic address: [email protected], European Association for the Study of the Liver. EASL 2017 Clinical Practice Guidelines on the management of hepatitis B virus infection.J Hepatol. 2017; 67: 370-398Abstract Full Text Full Text PDF PubMed Scopus (2367) Google Scholar Hence, the model performance might be affected by the duration of antiviral treatment. We suggest further comparison of these risk models, so as to identify which one can better predict HCC in CHB patients with an antiviral treatment duration of over one year. Secondly, the study excluded CHB patients with human immunodeficiency virus or hepatitis C virus co-infection, but hepatitis D virus (HDV) co-infection was not excluded, as described in Patients and Methods. HDV co-infection might dramatically increase the risk of HCC.[4]Farci P. Niro G.A. Clinical features of hepatitis D.Semin Liver Dis. 2012; 32: 228-236Crossref PubMed Scopus (121) Google Scholar This may confound the performance of risk models in this cohort. Thirdly, of the 10 models, 9 relied on baseline data at the time of initial antiviral treatment, but APA-B model was based on platelet counts and alpha-fetoprotein (AFP) levels at 12 months following initiation of HBV treatment. As shown in Table 1, only platelet and AFP levels at baseline were described, with the two parameters at 12 months missing. As we think, clarification of the aforementioned points would reinforce more validity and credibility for the conclusions presented in this research.