Objective To study the changes in the serum markers in chronic hepatitis B patients who have had previous treatment with long-acting interferon therapy of nucleoside and those who have not and to assess the value of the serum markers for clinical prognosis evaluation.

Methods The clinical data of 411 cases of chronic hepatitis B were collected. All cases were given the additional treatment of long-acting interferon between October 2019 to April 2022. The cases were divided into two groups, a previously treated group consisting of patients who had been treated with nucleoside and nucleotide analogues (NAs) for more than 6 months after they became infected with hepatitis B virus (HBV) for over 6 months and an initial treatment group, or treatment naïve group, consisting of patients who had HBV infection for over 6 months and received no treatment or patients who have stopped NAs therapy for more than 6 months. The serum marker levels of the previously treated group and the initial treatment group, i.e., the previously treatment-naïve patients, were compared, and the receiver operating characteristics (ROC) curve was used to evaluate the value of the baseline levels of hepatitis B surface antigen (HBsAg) and HBV pregenomic RNA (pgRNA) for predicting the rate of cured cases in the two groups.

Results There was no significant difference in the rate of cured cases between the previously treated group and the initial treatment group. The baseline HBV DNA, HBsAg, and hepatitis B e antigen (HBeAg) levels of the cured cases in both groups were significantly lower than those in the uncured cases (P<0.0001). After 48 weeks of treatment, the serum HBsAb levels (mIU/mL) of the cured cases in both the previously treated and initial treatment groups were significantly higher than those of the uncured cases in the two groups (previously treated group: 78.97±22.57 vs. 0.99±0.38, P<0.0001; initial treatment group: 235.50±175.00 vs. 1.32±0.88, P<0.0001). The serum HBsAb levels (mIU/mL) of the cured cases in the initial treatment groups were significantly higher than that of cured cases in the previously treated group (235.50±175.00 vs. 78.97±22.57, P<0.0001). Within 0 to 60 weeks of treatment, HBV pgRNA levels of cured cases in both groups were significantly lower than those of the the uncured cases in both groups (P<0.0001). Multivariate logistic regression and ROC curve analysis showed that baseline serum HBsAg was the influencing factor and predictor of interferon efficacy in both the previously treated cases and the initial treatment cases, with the area under the curve (AUC) being 0.80 (95% confidence interval CI: 0.7423-0.8615, P<0.0001) and 0.74 (95% CI: 0.6283-0.8604, P=0.0079), respectively, and the optimal cut-off values being 244.60 IU/mL and 934.40 IU/mL, respectively. However, the baseline serum HBV pgRNA level of under 1340.00 copies/mL in the initial treatment cases led to better sensitivity and better specificity in efficacy prediction, with the AUC of the baseline HBV pgRNA being 0.9649 (95% CI: 0.9042-1.0000, P<0.0001).

Conclusion Among the previously treated cases and the initial treatment cases, patients who achieve clinical cure have lower levels of HBV DNA, HBsAg, and HBeAg at baseline, lower level of HBV pgRNA over the course of their treatment, and higher level of HBsAb at week 48. Baseline HBsAg levels can be used to effectively predict the clinical cure outcomes in previously treated cases and initial treatment cases. Baseline HBV pgRNA levels also exhibit a high predictive value for treatment outcomes in initial treatment cases.