Early on-treatment LSM reliably predicts liver-related events in CHB patients with significant fibrosis and cirrhosis

Background: Evidence comparing longitudinal liver stiffness measurements (LSM) dynamics to on-treatment LSM for predicting clinical outcomes in patients receiving etiology therapy is limited. Objective: This study aimed to assess the prognostic value of on-treatment LSM in patients with chronic hepatitis B (CHB) receiving antiviral therapy. Design: Patients with CHB and significant fibrosis or cirrhosis were included in this prospective cohort. Liver-related events (LREs) were defined as hepatic decompensations, liver transplantation, or liver-related death. Association between LREs and baseline, on-treatment, and dynamic changes in LSM were analyzed. Results: A total of 1116 patients with CHB, including 875 (78.4%) diagnosed with cirrhosis, were followed for a median of 7.5 (2.5–9.5) years. On-treatment LSM was the most reliable predictor of 3- and 5-year outcomes (AUROC: 0.72–0.78) after 1–3 years of antiviral therapy, outperforming baseline LSM (AUROC: 0.59–0.65) and LSM changes (AUROC: 0.42–0.65). Patients with LSM <10 kPa at 1, 2, or 3 years of antiviral therapy have a much lower risk of LREs, with a 5-year cumulative incidence of 2.2%, 2.6%, and 2.7%, respectively. This finding held true in cirrhosis subgroup, in validation cohort, and for predicting decompensations alone. Notably, patients with on-treatment LSM <10 kPa showed better restoration of lobular architecture assessed by liver biopsies. Conclusions: On-treatment LSM measured 1 to 3 years after antiviral therapy offers superior predictive accuracy for LREs compared to baseline or LSM changes, with LSM <10 kPa indicating a significantly lower risk, likely due to improved lobular architecture.