Long-term Liver Stiffness Dynamics and Residual Risk of Liver-Related Events in HBV-Related Compensated Cirrhosis During Entecavir Therapy: A 13-Year Prospective Cohort Study

Technical Summary: Long-term Liver Stiffness Dynamics and Residual Risk of Liver-Related Events in HBV-Related Compensated Cirrhosis During Entecavir Therapy

Problem Statement
While nucleos(t)ide analogue (NA) therapy, particularly entecavir (ETV), effectively suppresses hepatitis B virus (HBV) replication and slows the progression of HBV-related cirrhosis, the long-term clinical trajectory beyond a decade remains incompletely characterized. Specifically, there is a lack of data regarding the very long-term dynamics of liver stiffness measurements (LSM) and the persistence of liver-related events (LREs), such as hepatic decompensation and hepatocellular carcinoma (HCC), in patients with compensated cirrhosis. Furthermore, the predictive value of dynamic, on-treatment biomarkers—specifically the Albumin-Bilirubin (ALBI) score—for long-term outcomes in this population requires further validation.

Methodology
This study was a prospective, single-center cohort study conducted at Peking University People's Hospital.

• Cohort: 50 treatment-naïve adults (aged 18–70) with HBV-related compensated cirrhosis were enrolled between July 2012 and July 2015. Inclusion required confirmed cirrhosis (histological, endoscopic, or clinical criteria including LSM >12.4 kPa) and specific HBV DNA thresholds. Exclusion criteria included decompensated cirrhosis, coinfections, other chronic liver diseases, and active malignancy.
• Intervention: All patients received ETV monotherapy.
• Follow-up: Assessments occurred every 6 months until the study concluded in July 2025, covering a median follow-up of 11.0 years (up to 13 years). Evaluations included laboratory testing (HBsAg, HBV DNA, liver biochemistry, AFP), abdominal ultrasound, and transient elastography (LSM).
• Endpoints: The primary endpoint was the composite occurrence of LREs (hepatic decompensation, HCC, or liver-related death).
• Statistical Analysis:
  • Longitudinal LSM changes were analyzed using linear mixed-effects models (LMM).
  • Cumulative incidence of LREs and LSM-defined regression (LSM <10 kPa) was estimated via Kaplan-Meier analysis.
  • A prespecified 1-year landmark analysis was performed to identify on-treatment predictors of subsequent LREs, minimizing immortal time bias.
  • Cox proportional hazards regression (univariable and multivariable) was used to assess associations between baseline and year-1 variables and subsequent LREs.

Key Results

• Clinical Outcomes: Over a median follow-up of 11.0 years, 12 patients (24.0%) developed LREs. Of these, 9 (18.0%) developed HCC. The cumulative incidence of LREs was 0% at year 1, 11.0% at year 5, 22.6% at year 9, and 31.0% at year 13.
• Liver Stiffness Dynamics: LSM showed a sustained decline over time. The linear mixed-effects model estimated an average annual decrease of 0.50 kPa (β = −0.5009, P < 0.001). Among the 36 patients with baseline LSM ≥10 kPa, the cumulative probability of LSM-defined regression (dropping below 10 kPa) reached 100% by year 10. The median time to regression was 2.0 years.
• Predictors of Events:
  • Baseline characteristics (including age, sex, HBV DNA, LSM, and standard scores like Child-Pugh and MELD) were not significantly associated with subsequent LREs in univariable analysis.
  • In the 1-year landmark analysis, the year-1 ALBI score was significantly associated with subsequent LREs (Unadjusted HR: 16.41; P = 0.011).
  • In a multivariable model adjusting for age and the year-1 FIB-4 index, the year-1 ALBI score remained significantly associated with LREs (Adjusted HR: 11.96; 95% CI, 1.19–120.69; P = 0.035).
  • An exploratory ROC analysis suggested a cut-off of −3.04 for the year-1 ALBI score, stratifying patients into high-risk (ALBI ≥ −3.04) and low-risk groups with significantly different event-free survival (Log-rank P = 0.005).

Key Contributions

1. Long-term Trajectory Data: The study provides one of the longest prospective datasets (up to 13 years) characterizing the longitudinal decline of liver stiffness in HBV-related compensated cirrhosis under ETV therapy, demonstrating that while LSM regression is common and sustained, it does not eliminate clinical risk.
2. Residual Risk Quantification: It quantifies the substantial residual risk of LREs, particularly HCC, persisting despite sustained viral suppression and biochemical normalization, with a 31% cumulative incidence at 13 years.
3. Dynamic Risk Stratification: The study identifies the year-1 ALBI score as a potential early on-treatment predictor of long-term adverse outcomes, suggesting that dynamic assessment of liver functional reserve offers prognostic value beyond static baseline characteristics.

Significance and Claims
The authors conclude that prolonged entecavir therapy is associated with sustained reductions in liver stiffness and improvement in liver function markers. However, they emphasize that these improvements do not equate to a "reset" of carcinogenic risk; a significant residual risk of HCC and other LREs persists, necessitating lifelong surveillance.

Regarding the ALBI score, the paper claims a preliminary association between the year-1 ALBI score and subsequent LREs. However, the authors explicitly state that this finding is exploratory. They note limitations including the small sample size (n=50), the limited number of events (n=12), the wide confidence intervals in the adjusted model, and the lack of internal or external validation for the derived ALBI cut-off. Consequently, the paper asserts that while the ALBI score shows promise for risk stratification, its incremental clinical utility and specific decision thresholds require validation in larger, multicenter studies before they can justify changes to surveillance protocols. The study does not claim that the ALBI score should currently replace standard surveillance but rather highlights its potential as a dynamic prognostic tool.