Patient characteristics at baseline
This cohort of 203 patients with CHB showing NA-induced HBeAg seroclearance was followed-up for up to 8 years. The mean age was 37 years (range, 18-82) at baseline and 40 years (range, 20-84) at the time of HBeAg seroclearance. The mean follow-up duration was 5 years (range, 2-11). Baseline characteristics of patients are shown in Table 1. All patients received oral antiviral therapy before HBeAg seroclearance. A total of 173 (85.2%) patients received entecavir or tenofovir as the first-line treatment. The remaining 30 (14.8%) patients received tenofovir only or tenofovir plus entecavir combination therapy as a rescue therapy. One hundred seventy-five (86.2%) patients had normal alanine aminotransferase (ALT) levels at the time of HBeAg seroclearance, while 28 (13.8%) had abnormal ALT levels (range, 41-110 IU/L). One hundred fifty-two (74.9%) patients showed undetectable HBV DNA (< 12 IU/mL) at the time of HBeAg seroclearance, while 51 (25.1%) had sustained detectable HBV DNA (range, 1.3-4.9 IU/mL).
Cumulative incidence of HCC
A total of 16 (7.9%) patients developed HCC during a mean follow-up period of 5 years (range, 2-11). The cumulative incidence of HCC was 1.5%, 6.2%, 10.0%, and 11.5% at 1, 3, 5, and 8 years after HBeAg seroclearance, respectively (Figure 1A). According to the presence of cirrhosis, the cumulative incidence of HCC ranged from 4% at 1 year to 28.0% at 8 years after HBeAg seroclearance and was significantly higher in patients with cirrhosis than in those without cirrhosis (P < 0.001) (Figure 1B). In patients that achieved HBeAg seroclearance before 30, 31-40, 41-50, and > 50 years of age, the cumulative incidence of HCC was 0%, 5.2%, 14.1%, and 34.3%, respectively, after 8 years from HBeAg seroclearance (P = 0.001; Figure 1C). The mean age for HCC development was 52 years (range, 43-75).
Based on the ALT level at the time of HBeAg seroclearance, no significant difference was observed in the incidence of HCC between elevated ALT and normal ALT groups (10.1% versus 11.8%, P = 0.784). In addition, HBV DNA level analysis at the time of HBeAg seroclearance revealed no significant difference in the incidence of HCC between sustained detectable HBV DNA and undetectable HBV DNA groups (13.3% versus 8.3%, P = 0.821).
In the multivariate Cox proportional hazard analysis, age, male gender, cirrhosis, detectable HBV DNA, elevated ALT, and type of antiviral agent (entecavir and tenofovir) at the time of HBeAg seroclearance were included. The presence of cirrhosis was the only significant factor associated with HCC development (hazard ratio [HR], 24.651; confidence interval [CI], 3.018 to 201.365; P = 0.003) (Table 2).
Cumulative incidence of HBsAg seroclearance
A total of 16 patients achieved HBsAg seroclearance, and 13 of these patients achieved HBsAg seroconversion with anti-HBs. The cumulative incidence of HBsAg seroclearance was 3.5%, 4.7%, 10.2%, and 18.7% at 1, 3, 5, and 8 years after HBeAg seroclearance, respectively (Figure 2A). For patients that achieved HBeAg seroclearance before the age of 30, 31-40, 41-50, and > 50 years, the cumulative rate of HBsAg seroclearance was 0%, 5.1%, 41.7%, and 22.9%, respectively, after 8 years from HBeAg seroclearance (P = 0.005; Figure 2B). The mean age at HBsAg seroclearance was 48 years (range, 35-70).
As per the ALT level at the time of HBeAg seroclearance, no significant difference was reported in the incidence of HBsAg seroclearance between elevated ALT and normal ALT groups (14.3% versus 21.1%, P = 0.448). The HBV DNA level at the time of HBeAg seroclearance also revealed the absence of any significant difference in the incidence of HBsAg seroclearance between sustained detectable HBV DNA and undetectable HBV DNA groups (15.6% versus 17.6%, P = 0.834).
In addition, no significant difference was observed in the cumulative incidence of HCC between HBsAg seroclearance and sustained HBsAg-positive groups (22.5% versus 10.5%, P = 0.069; Figure 3).
In the multivariate Cox proportional hazard analysis, age, male gender, cirrhosis, detectable HBV DNA, elevated ALT, and type of antiviral agent (entecavir and tenofovir) at the time of HBeAg seroclearance were included, and no factor was found to be significantly related to HBsAg seroclearance (Table 2).