Letter on ‘Long-Term Dynamic Changes of Alanine Aminotransferase Levels Are Associated With Liver-Related Events in Nucleos(t)ide Analogue-Treated Chronic Hepatitis B Patients in China’—Authors' Reply

We are grateful for the insightful letter by Drs. Dong Hyun Kim and Hye Won Lee [1] on our recent study regarding the prognostic significance of longitudinal alanine aminotransferase (ALT) dynamics in nucleos(t)ide analogue-treated chronic hepatitis B (CHB) patients [2]. We sincerely appreciate the opportunity to address the issues raised.

First, the Search-B cohort, from which the ALT thresholds were derived, originated from a well-designed prospective study with standardised follow-up protocols and systematic data collection, ensuring high data quality and reliability for threshold determination. In contrast, the Hong Kong cohort was constructed from retrospective electronic medical records, in which the frequency of patient follow-up showed significant fluctuations according to the clinical condition of patients. These characteristics make it more suitable as a real-world validation set. Moreover, the ALT thresholds were selected based on LRE risk discrimination and defined using the 95th percentile of the statistical reference range—a widely adopted approach for determining clinical cutoffs. When compared with the established ALT cutoffs [3-5], patients with ALT ≤ 23/16 U/L had the lowest 7-year LRE incidence (5.8%), which increased with ALT ≤ 30/19 (6.8%), or ≤ 35/25 (6.9%) (Figure 1A–C). Notably, the cutoffs of 23/16 U/L can further stratify the risk of LRE in patients with ALT below these established thresholds (Figure 1D). These findings further support the notion that lower ALT thresholds provide superior prognostic discrimination. While histologic validation would strengthen our findings, concordant findings across two independent cohorts underscore the robustness of sex-specific thresholds.

Second, we fully acknowledge that ALT levels can be influenced by various factors beyond viral activity. While accounting for these determinants would enhance precision for specific populations, it introduces great challenges and substantial complexity for widespread clinical implementation. Therefore, sex-specific ALT thresholds remain a practical and feasible strategy for risk stratification. Additionally, our analyses confirmed that the cutoffs of 23/16 U/L demonstrated consistent prognostic value across varied clinical profiles, suggesting their broad applicability. Nevertheless, as noted in the limitations, residual confounding from alcohol use, medications and lifestyle factors cannot be fully excluded in this retrospective design study. Future research should incorporate these elements to validate our findings.

Third, we fully agree that ALT fluctuations—which may reflect viral replication and host immune responses [6, 7]—have important clinical implications and warrant further investigation. As shown in table S5 of the original manuscript [2], patients with fluctuating trajectories had a higher 7-year LRE incidence (13.1% in men and 6.8% in women) than those in trajectory group 1 (7.5% in men and 0.0% in women). However, as our study focused on exploring the optimal ALT threshold associated with long-term outcomes, further analyses of fluctuation patterns were not conducted. Future well-designed clinical and mechanistic studies are needed to elucidate the prognostic and therapeutic implications of ALT fluctuation patterns in on-treatment CHB patients.

In summary, we appreciate the constructive feedback, which further emphasises the complexity of ALT interpretation in CHB. Our findings consistently indicate that maintaining ALT below lower, sex-specific thresholds is strongly associated with reduced risk of LRE, highlighting the clinical value of longitudinal ALT monitoring.

Ning Yu: writing – original draft, formal analysis. Rong Fan: writing – review and editing.

The authors' declarations of personal and financial interests are unchanged from those in the original article [2].

This article is linked to Fan et al. papers. To view these articles, visit https://doi.org/10.1111/apt.70195 and https://doi.org/10.1111/apt.70347.