Letter: More Precise Stratification Is Needed Within Chronic Hepatitis B Virus Low‐Level Viraemia

We read with great interest the article ‘Increased Risk for Liver Disease in Patients With Chronic Hepatitis B Virus Infection and Low-Level Viraemia’ by Paul Y. Kwo et al. [1]. By comparing the risk of liver-related events between patients with chronic hepatitis B virus low-level viraemia (HBV-LLV；HBV DNA ≤ 2000 IU/mL) and controls without evidence of HBV, this study found that the HBV-LLV group had significantly higher risks of cirrhosis, decompensation, hepatocellular carcinoma (HCC) and the composite endpoint. This study demonstrates that HBV-LLV is not a benign condition. However, we wish to raise a methodological and clinical concern: dichotomising HBV-LLV as a single category (≤ 2000 IU/mL) versus ‘no HBV’ may obscure clinically meaningful risk gradients within the LLV range. With reference to international guidelines [2-4], the study defined HBV-LLV as HBV DNA ≤ 2000 IU/mL and accordingly set this as the enrollment criterion, yet it did not perform further stratification within this range according to viral load gradients (e.g., < 20, 20–200, 200–2000 IU/mL). Therefore, it remains unclear whether the observed elevated risk is primarily driven by those with higher viral loads within the HBV-LLV range (e.g., 500–2000 IU/mL), and whether the risk profile of patients with very low viremia (e.g., < 20 IU/mL) more closely resembles that of the control population without HBV infection. Indeed, this distinction carries direct clinical significance: for example, when managing patients with HBV DNA levels of 1900 IU/mL versus 15 IU/mL, clinicians may adopt different surveillance frequencies and consider different treatment initiation thresholds. Simply grouping both into the broad category of ‘HBV-LLV’ may lead to over-surveillance of the very low-risk subgroup or inadequate recognition of risk in the subgroup with higher viral loads. Moreover, existing evidence has already demonstrated an elevated trend in the risk of cirrhosis, decompensation, HCC and the composite endpoint among patients with HBV-LLV [5, 6]. Therefore, we suggest that subsequent analyses based on the same dataset should stratify the HBV-LLV cohort into finer subgroups according to baseline HBV DNA levels (e.g., < 20, 20–200, 200–2000 IU/mL) and calculate the event rates for each stratum. Additionally, HBV DNA could be modelled as a continuous variable or using restricted cubic splines to capture nonlinear risk gradients, ideally with adjustment for time-updated viral loads. Furthermore, the very low viremia subgroup (e.g., < 20 IU/mL) should be compared with the control group without HBV infection to assess whether even minimal viremia confers a residual risk. In summary, the study by Kwo and colleagues is of great value, as it clearly confirms that an HBV DNA level ≤ 2000 IU/mL is still associated with a significant liver-related risk and is not a ‘risk-free state’. Although current guidelines do not yet routinely recommend antiviral therapy for such patients, finer risk stratification within HBV-LLV should still be performed to guide individualised surveillance and identify high-risk individuals who may require early intervention. Qun Yang: conceptualization, validation, writing – original draft. Jie Zheng: writing – original draft. Jinheng Cao: writing – review and editing. The authors have nothing to report. The authors have nothing to report. The authors declare no conflicts of interest. This article is linked to Kwo et al. paper. To view this article, visit https://doi.org/10.1111/apt.70613. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.