Lack of association between early on-treatment HBeAg seroclearance and development of hepatocellular carcinoma or decompensated cirrhosis

Background & Aims The association between hepatitis B envelope antigen (HBeAg) seroclearance during long-term nucleos(t)ide analogue (NA) treatment and the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB) remains unclear. Here, we aimed to investigate the association of HBeAg seroclearance during potent NA treatment with the development of HCC and decompensated cirrhosis. Methods Using a multicenter historical cohort including 2,392 non-cirrhotic adult patients with HBeAg-positive CHB who initiated NA treatment with tenofovir or entecavir, the risk of HCC and decompensated cirrhosis was compared between patients who achieved HBeAg seroclearance within 36 months of NA treatment (the HBeAg-loss group) and those who did not (the HBeAg-maintained group), using inverse probability of treatment weighting. Results Over a median of 6.6 years of NA treatment, 1,077 patients achieved HBeAg seroclearance (HBeAg loss rate = 6.0 per 100 person-years), 64 patients developed HCC (HCC incidence rate = 0.39 per 100 person-years), and 46 patients developed decompensated cirrhosis (decompensation incidence rate = 0.28 per 100 person-years). The HBeAg-loss and HBeAg-maintained groups had a similar risk of developing HCC (hazard ratio 0.89; 95% CI 0.47–1.68; p = 0.72) and decompensated cirrhosis (hazard ratio 0.98; 95% CI 0.48–1.81; p = 0.91). Compared with delayed HBeAg seroclearance beyond 10 years of NA treatment, the risk of HCC was comparable in those who achieved earlier HBeAg seroclearance at any time point within 10 years, regardless of baseline age and fibrotic burden. Conclusions Early HBeAg seroclearance during NA treatment was not associated with a reduced risk of development of HCC or decompensated cirrhosis in non-cirrhotic HBeAg-positive patients with CHB. Impact and implications The association between hepatitis B envelope antigen (HBeAg) seroclearance during long-term nucleos(t)ide analogue treatment and the risk of hepatocellular carcinoma in patients with chronic hepatitis B remains unclear. Our findings indicate that early on-treatment HBeAg seroclearance within 3 years was not associated with the development of hepatocellular carcinoma or decompensated cirrhosis. Achieving HBeAg seroclearance may not be an appropriate surrogate endpoint for preventing the development of liver-related outcomes in non-cirrhotic patients with HBeAg-positive chronic hepatitis B treated with nucleos(t)ide analogues.


HCC, hepatocellular carcinoma
The cumulative incidences of HCC were compared by Kaplan-Meier curves, and P values were derived from log-rank test.Table S1.Baseline characteristics of the subgroup of platelet count of more than 150×10 3 /μL, and the HBeAg-maintained and HBeAg-loss groups before and after inverse probability of treatment weighting.

Fig. S2 .
Fig. S2.Risk of (A) decompensated liver cirrhosis, and (B) liver-related outcomes according to HBeAg-seroclearance at 36 months after initiation of antiviral treatment.Weight analysis balanced by inverse probability of treatment weighting.

Fig. S3 .
Fig. S3.Risk of HCC according to HBeAg-seroclearance at 60 months after initiation of antiviral treatment.(A) Unweighted analysis.(B) Weight analysis balanced by inverse probability of treatment weighting.

Fig. S4 .
Fig. S4.Risk of HCC according to HBeAg-seroclearance at 36 months after initiation of antiviral treatment.(A) Unweighted analysis.(B) Weight analysis balanced by inverse probability of treatment weighting in the subgroup of platelet count of more than 150×10 3 /μL.

Fig. S7 .
Fig. S7.Risk of HCC according to HBV DNA suppression (i.e., serum HBV DNA ≤ 2,000 U/mL) at 4 months and baseline HBV DNA level after initiation of antiviral treatment.

Table S2 .
Baseline characteristics of the stratified subgroup of the FIB-4 index ≥ 2.0, and the HBeAg-maintained and HBeAg-loss groups before and after inverse probability of treatment weighting.
IPTW, inverse probability of treatment weighting; HBeAg, hepatitis B virus envelope antigen; SMD, standardized mean difference; TDF, tenofovir disoproxil fumarate; AST, aspartate aminotransferase; ALT, alanine aminotransferase; FIB-4, Fibrosis-4; mPAGE-B, modified PAGE-B All variables were compared between the HBeAg-loss and HBeAg-maintained groups before and after IPTW.Sex, and the type of antiviral treatment were analyzed by Fisher's exact test.Other variables were analyzed by Kruskal-Wallis test and expressed in median (IQR).
All variables were compared between the HBeAg-loss and HBeAg-maintained groups before and after IPTW.Sex, and the type of antiviral treatment were analyzed by Fisher's exact test.Other variables were analyzed by Kruskal-Wallis test and expressed in median (IQR).

Table S4 .
Baseline characteristics of the stratified subgroup of age ≥ 45, and the HBeAg-maintained and HBeAg-loss groups before and after inverse probability of treatment weighting.
All variables were compared between the HBeAg-loss and HBeAg-maintained groups before and after IPTW.Sex, and the type of antiviral treatment were analyzed by Fisher's exact test.Other variables were analyzed by Kruskal-Wallis test and expressed in median (IQR).

Table S5 .
Baseline characteristics of the stratified subgroup of age < 45, and the HBeAg-maintained and HBeAg-loss groups before and after inverse probability of treatment weighting.

Table S7 .
Univariable and multivariable time-dependent Cox regression analyses for the predictive factors of decompensated liver cirrhosis occurrenceMultivariable Cox regression analysis was performed on covariates associated with HCC occurrence identified in univariable analysis.The results were expressed in terms of HR, 95% CI, and P values.This analysis included a time-dependent covariate, HBeAg seroclearance, which was the variable of interest.

Table S9 .
Baseline characteristics of the Delayed-suppression and Early-suppression groups before and after inverse probability of treatment weighting.
, inverse probability of treatment weighting; SMD, standardized mean difference; TDF, tenofovir disoproxil fumarate; AST, aspartate aminotransferase; ALT, alanine aminotransferase; FIB-4, Fibrosis-4; mPAGE-B, modified PAGE-B All variables were compared between the HBeAg-loss and HBeAg-maintained groups before and after IPTW.Sex, and the type of antiviral treatment were analyzed by Fisher's exact test.Other variables were analyzed by Kruskal-Wallis test and expressed in median (IQR). IPTW