In recent years, domestic and international chronic hepatitis B (CHB) management guidelines have recommended that the goal for CHB patients should be functional cure to improve prognosis (1–3). Recently, studies related to hepatitis B surface antigen (HBsAg) clearance have increased, but there are still fewer studies related to recurrence after HBsAg clearance. Our previous study and a meta-analysis found that the recurrence rate after functional cure was 6.19%–9.66%, including cases with HBsAg positivity and/or hepatitis B virus (HBV) DNA positivity (4, 5). Due to the few large or even sufficient sample relapse cohorts are available for analysis, the effectiveness of retreatment and clinical outcomes after relapse and the correlation between immune function and relapse are even rarer. Therefore, we conducted a long-term follow-up of CHB who achieved functional cure in the previous study cohort. And we evaluated the clinical characteristics, retreatment effect, and outcome of relapsers after functional cure, as well as the predictors of relapse.

HBsAg seroclearance or seroconversion has been considered the ideal treatment endpoint for CHB patients (1–3). Some studies showed that the long-term prognosis of patients with CHB could be improved after functional cure (4, 8, 9, 12). However, relapse still occurs in some patients, and the types of relapse are diverse. In addition, relapse cohorts are difficult to obtain. There is a lack of evidence-based medical knowledge on the effects of retreatment after recurrence and the relationship between recurrence and disease progression. Moreover, there are few studies on immunological predictors to predict relapse. Therefore, we studied the issues as mentioned above in 32 relapsed patients with functional cure observed since 2005.

In the long-term follow-up of 32 relapsed patients, 20 patients received retreatment, 13 of them achieved functional cure, and the functional cure rate was 65.0% (13/20). Of a total of 15 patients treated with Peg-IFN, 13 patients were cured, and the recovery rate of patients retreated with Peg-IFN was 86.7% (13/15). A second recurrence occurred in 3 patients who were followed up from 33 to 244 weeks after drug withdrawal. The 5 patients in the NA group did not achieve a second cure. However, among the untreated, two cases had spontaneous clearance of HBV DNA and one case had spontaneous clearance of HBsAg, resulting in functional cure of hepatitis B again. Few studies reported on the treatment outcome of relapsers in clinically cured hepatitis B. Kim et al. (13) and Chu and Liaw (8) reported that HBV DNA was spontaneously cleared in patients with HBV DNA reversion, and HBsAg was spontaneously cleared in patients with HBsAg reversion. In 2020, Lok et al. (14) reported 10 relapsed patients; two of them retreated, but the treatment effect was not reported. Similar studies have been reported by Wu et al. (15), who reported a functional cure rate of 81.0% (17/21) in patients with HBsAg reversion after a median (mean) of 24 (28) weeks of Peg-IFN retreatment and observed recurrence in 2 patients. These results suggest that Peg-IFN-based retreatment can obtain higher HBsAg clearance in relapsed patients. The results of the study by Wu et al. are similar to those of our study. These results suggest that Peg-IFN-based retreatment can obtain higher HBsAg clearance.

Previous reports of relapsers after functional cure of hepatitis B were mainly HBsAg positive and both HBsAg and HBV DNA positive (16–18). In 2019, our team reported a new type of relapse, namely, “HBsAg negative, anti-HBs positive, HBV DNA positive,” and sequencing results show that the mechanism may be related to the mutation of the HBV S region (5). Lok et al. (14) also observed 2 cases of recurrence with HBsAg negative but HBV DNA positive. In this study, we observed that different recurrence types had different recovery rates, and the recovery rate of patients with type I recurrence was significantly higher than that of patients with other recurrence types. In addition, in type II and III relapsers, S-region mutations were detected in 3 patients, and HBV DNA was continuously below the lower limit of detection after NA treatment, which must be maintained for the long-term according to current treatment guidelines. These patients have difficulty in obtaining functional cures. One case was resistant to NAs; despite long-term treatment with Peg-IFN combined with NAs, HBsAg conversion was not achieved. This finding, combined with the results of previous research by our team (16), suggests that NA-resistant patients are not only prone to recurrence after achieving functional cure but are also difficult to cure again. It is suggested that such patients should be discontinued with caution and closely followed after obtaining HBsAg clearance.

Twelve relapsed patients (37.5%, 12/32) did not receive drug intervention, and their follow-up period was 42–269 weeks. In 6 of 12 cases, HBsAg and HBV DNA fluctuated at low levels; and in 3 out of 12 cases, the HBsAg and/or HBV DNA fluctuations were significant (>3 log10 IU/ml), but no liver function abnormalities were observed. However, the follow-up period is limited, and the long-term outcome of untreated patients needs to be further observed.

In this study, patients who relapsed after achieving functional cure with Peg-IFN treatment did not have disease progression or the occurrence of liver cirrhosis or liver cancer during the follow-up period, regardless of whether they received retreatment. However, the follow-up period is limited, and the long-term prognosis of patients with recurrence needs to be further observed. There is no report on whether clinical cure can also improve long-term prognosis in patients with recurrence. However, several studies have shown that the incidence of hepatocellular carcinoma (HCC) in patients with CHB after obtaining HBsAg clearance ranges from 0% to 4.8% (12, 17–20). In contrast, the 5-year risk rate of HCC in patients who were HBsAg positive and without HBV DNA inhibition was 18.8% (21). The 5-year incidence of HCC in patients who achieved HBV DNA suppression and HBeAg seroconversion with NAs still ranged from 6% to 7.7% (21, 22). This indicates that functional cure for hepatitis B patients can significantly improve the prognosis. Given the low HBsAg and HBV DNA levels at the time of relapse and the high rate of functional cure in the majority of relapsers in this study, we recommend that relapsed patients should retreat if conditions permit to achieve functional cure as early as possible.

It is generally accepted that anti-HBs is a protective antibody that is produced by the body after HBV infection. Previous studies reported that anti-HBs levels at drug withdrawal were associated with recurrence; the recurrence rate in anti-HBs-positive patients was lower than that in anti-HBs-negative patients (5, 13, 23). This study compared the quantitative levels of anti-HBs between the recurrence and non-recurrence groups, and we found a significant difference between the two groups at the time of drug withdrawal and after drug discontinuation. Patients with high anti-HBs levels at drug withdrawal were less likely to relapse (OR = 0.525, p < 0.001). In addition, studies have shown that anti-HBc level can not only reflect the immune response (7) but also serve as a predictor of antiviral efficacy or relapse after NA withdrawal (24, 25). In this study, anti-HBc levels of the recurrence group were significantly lower than those of the non-recurrence group at both drug discontinuation and relapse (48 weeks of drug discontinuation) (all p < 0.05). In addition, anti-HBc levels were below the lower limit of detection in patients who relapsed again. The predictive value of anti-HBc levels at drug discontinuation for relapse after functional cure was evaluated by the ROC curve, which had an AUROC of 0.724, higher than anti-HBs levels at drug discontinuation (0.679). To more accurately predict relapse after drug withdrawal, we combined anti-HBs and anti-HBc at drug withdrawal. The combined predictors had an AUROC of 0.833, which was a better predictor of relapse than a single indicator. The combined predictors had good sensitivity (86.67%) and specificity (90.62%) for predicting relapse, which could compensate for the lower specificity of the single predictor. Humoral immunity plays an important role in the clearance of HBV infection. Anti-HBs and anti-HBc are produced by B lymphocytes specific for HBsAg and HBcAg, respectively. Therefore, the high anti-HBs and anti-HBc levels at drug discontinuation may reflect that patients have a high adaptive immune status, which may be associated with low relapse after treatment cessation. Anti-HBs and anti-HBc are noninvasive indicators, which are convenient for detection, clinical application, and observation. Therefore, they can be used as predictors of drug withdrawal and recurrence in functionally cured patients.

Overall, the recovery rate of relapsed patients receiving retreatment is 65%, and the recovery rate of Peg-IFN retreatment is relatively high, reaching higher than 80%. However, the recurrence and special types of recurrence could still occur. Conversely, it is more difficult to obtain functional cure after relapse with NA retreatment and untreated patients. In this study, patients who relapsed after achieving functional cure with Peg-IFN treatment did not have disease progression or the occurrence of liver cirrhosis or liver cancer during the follow-up period, regardless of whether they received retreatment. However, the long-term prognosis of patients with recurrence needs to be further observed. In addition, both anti-HBs and anti-HBc levels at the time of treatment cessation can be used as predictors of recurrence, and the combined predictive value of the two indicators is better. They are convenient for detection and clinical application. However, this is an exploratory study; the sample size is small and limited to one center. So, the results should be further confirmed by large, multicenter, prospective clinical trials.

The original contributions presented in the study are included in the article/Supplementary File, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by Beijing Youan Hospital, Capital Medical University. The patients/participants provided their written informed consent to participate in this study.

XL, AS, and XC designed the study. XL, AS, JL, LM, ZC, HL, and YZ collected the data. XL and SR tested serum samples. XL and AS analyzed the data. SR, JL, SZ, and ZH guided statistical analysis. XL, AS, and XC drafted the article. XC contributed to the interpretation of the results and critical revision of the article for important intellectual content. XC, SZ, ZH, and SR provided the financial support for the project leading to this publication. All authors read and approved the final article. All authors contributed to the article and approved the submitted version.

This work was supported by the Capital Health Research and Development Projects (2020–1–2181), the Capital Clinical Diagnostic Techniques and Translational Application Projects (Z211100002921059), the Beijing Municipal Administration of Hospitals’ Youth Program (QML20211702), the Beijing Municipal Administration of Hospitals Clinical medicine Development of special funding support (ZYLX202125), the National Science and Technology Key Project on “Major Infectious Diseases such as HIV/AIDS, Viral Hepatitis Prevention and Treatment” (2017ZX10302201-004, 2017ZX10202203-006).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.