All IHCs met the criteria defined in the prevention and treatment guidelines for chronic hepatitis B (2019 edition). (i) the history of HBsAg positive was longer than 6 months and the level of HBsAg lower than 200 IU/mL, HBeAg negative, anti-HBe negative/positive; (ii) HBV DNA < 20 IU/mL, Alanine aminotransferase (ALT) normal (male < 50 IU/L, female < 40 IU/L); (iii) white blood cell count > 4×10⁹/L, platelet count >150×10⁹/L; (iv) Total bilirubin (TBil) <34 μmol/L, albumin > 40 g/L. Exclusion criteria: (i) history of autoimmune diseases; (ii) human immunodeficiency virus (HIV) or hepatitis C virus (HCV) or hepatitis E virus(HEV) coinfected; (iii) pregnant, lactating women and those who are preparing to be pregnant; (iv) liver cirrhosis or liver cancer; (v) history of severe heart disease; (vi) history of mental illness or psychiatric disorders. (vii) uncontrolled epilepsy. (viii) alcohol or drug abuser. (ix) uncontrolled diabetes mellitus, hypertension, thyroid disease, retinopathy; (x) contraindication to interferon.

All patients signed an informed consent form before enrollment. The protocol and the consent form for the study were approved by the research ethics committee of Beijing Youan Hospital, Capital Medical University, China ([2017]24).

This is a prospective cohort study. Patients in treatment group were given PEG-IFN180ug subcutaneously once a week for 96 weeks. Patients in the control group were observed for 96 weeks without anti-viral treatment. Patients in the treatment group were divided into HBsAg clearance group (Responders, R group) and non-HBsAg clearance group (Non-responders, NR group) according to whether HBsAg clearance was achieved within 96 weeks. Patients were divided into four groups according to controlled attenuation parameter (CAP)value at baseline: no hepatic steatosis (CAP < 238dB/m), mild steatosis (238dB/m < CAP < 250dB/m), moderate steatosis (250dB/m ≤ CAP < 292dB/m) and severe steatosis (CAP ≥ 292dB/m).

All patients were tested with HBsAg, anti-HBs, HBV DNA, liver function, blood count, thyroid function, thyroid antibodies and autoantibodies at baseline, and at the time point of week 12, 24, 36, 48, 60, 72, 84 and 96. CAP and liver stiffness measure (LSM) were evaluated at baseline and week 96. HBsAg was quantified with HBsAg quantification kit (Roche, Mannheim, Germany), the lower limit of detection of HBsAg was 0.05 IU/mL. HBV DNA was analyzed by cobas^® AmpliPrep/cobas^® Taqman automated nucleic acid isolation and purification and PCR system (Roche Diagnostics GmbH, Mannheim, Germany), the lower limit of detection of HBV DNA was 20 IU/ml. ALT was measured by OLYMPUS-AU5400 (Japan), normal value < 50 IU/L for men and < 40 IU/L for women. Thyroid antibodies were detected using RocheCobase801, with normal reference values: anti-thyroglobulin antibody (TGAb) < 115IU/ml, anti-thyroid peroxidase antibody (TPOAb) < 34IU/ml, thyrotropin receptor antibody (TRAb) < 1.75IU/ml. Thyroid function was measured by Architecti2000 (Abbott, Mannheim, USA) with normal reference ranges: FT3 ranged from 3.6 to 6.5pmol/L, TT3 ranged from 0.92-2.79 nmol/L, FT4 ranged from 11.5 to 22.7pmol/L, TT3 ranged from 58.1-140.6 nmol/L, and TSH ranged from 0.55-4.78 mIU/L. CAP and LSM were measured by transient elastic scanner (Fibroscan‐502, M probe, frequency 3.5M Hz, Echosens, France).

Statistical software package SPSS (SPSS Inc., USA) 25.0 software was used for statistical analyses. Normally distributed parameters were expressed as mean ± standard deviation and compared by t-test. (ANOVA for more than 2 groups) Data that were not normally distributed were expressed as median (25th, 75^th range) and non-parametric tests were used for comparison between groups. The Kaplan-Meier survival analysis was used to calculate the cumulative HBsAg clearance rate. P-value <0.05 was considered statistically significant. In this study, the subjects who completed 96-week trial after enrolment were analyzed without regard of dropped off or loss of follow-up, i.e. using per-protocol analysis (PP analysis).

A total of 174 IHCs were enrolled, 84 patients received PEG-IFN, 90 patients did not receive anti-viral treatment. Two patients in the treatment group discontinued PEG-IFN at week 24 due to hyperthyroidism and one patient stopped treatment due to hypothyroidism. At week 48, one patient stopped treatment due to hyperthyroidism and one patient discontinued treatment because of an increase of TPOAb (355.3 IU/ml); One patient stopped PEG-IFN due to hyperthyroidism at week 60. 78 patients completed the 96-week course of treatment. 88 patients in the control group were followed up to week 96 and two subjects were lost follow-up at week 24 ( Figure 1 ).

There were no statistical differences in gender, age and HBsAg quantification between the two groups ( Table 1 ).

In the treatment group, the HBsAg clearance rate was 30.77% (24/78) and HBsAg seroconversion rate was 23.08% (18/78) at week 48. HBsAg clearance rate increased to 57.69% (45/78) and HBsAg seroconversion rate was 55.13% (43/78) at week 96. In the control group, spontaneous HBsAg clearance occurred in two patients (one with baseline HBsAg 0.202 IU/mL and HBsAg cleared at week 72, the other with baseline HBsAg 3.67 IU/mL and cleared at week 96, with a clearance rate of 2.27% (2/88) and no patient achieved seroconversion. HBsAg clearance rate was significantly higher in the treatment group than that in the control group (p < 0.001) ( Figure 2 ).

A total of 45 patients in the treatment group achieved HBsAg clearance and were classified as responder group (R group). 33 patients did not achieve HBsAg clearance and were classified as non-Responder group (NR group). At baseline, there was no statistically significant difference in HBsAg levels between R and NR group (p = 0.602), and HBsAg quantification was significantly lower in R group than that in NR group at 12, 24, 36, 48, 60, 72, 84 and 96 weeks of PEG-IFN treatment (p = 0.003, p = 0.001, p < 0.001, p < 0.001, p<0.001, p < 0.001, p < 0.001, p < 0.001 respectively). In addition, patients in R group had a significantly greater decrease of HBsAg level at each time point during PEG-IFN treatment than those in NR group (p = 0.031, p = 0.019, p = 0.124, p = 0.002, p = 0.059, p = 0.017, p = 0.002, p < 0.001, respectively). ( Table 2 ; Figure 3 ).

The level of ALT in R group showed significantly higher at week 12 [48.0 (27.0, 72.0)] than baseline [26.0 (15.5, 38.5), p<0.001]. ALT level in NR group also showed significantly higher at week 12 [38.0 (26.5, 64.8)] than baseline [27.0 (18.0, 39.0), p=0.001].

In spite of that the ALT level in R and NR group were not diverse at each time point, the proportion of patients with elevated ALT [1.0-1.5 upper limit of normal (ULN) and ALT > 1.5 (ULN)] was slightly higher in R group than NR group, especially at week 12. ( Table 3 ; Figure 4 ).

At baseline, 78 patients in treatment group were divided into four groups according to CAP value, 27 with no hepatic steatosis, 8 with mild steatosis, 26 with moderate steatosis and 17 with severe steatosis. The proportion of patients with moderate steatosis in R group (42.2%) was significantly higher than that in NR group (21.2%) (p=0.033). The cumulative clearance rate was 62.47% at week 96 in IHCs with hepatic steatosis (mild, moderate, and severe), compared to 48.15% of those without steatosis (p=0.058). The cumulative HBsAg clearance rate was 71.42%, 74.07% and 47.05% at week 96 in patients with mild, moderate, and severe steatosis, respectively. IHCs with moderate steatosis was most sensitive to PEG-IFN, the HBsAg clearance rate in moderate steatosis was significantly higher compared with that in no-steatosis group (P=0.008). ( Table 4 ; Figure 5 )

Six of 84 individuals who received PEG-IFN treatment stopped the medication because of side effects. Four patients suffered hyperthyroidism and one patient developed hypothyroidism. The level of TPOAb was increased by at least 10-fold in one patient. HBsAg clearance was obtained in two patients with hyperthyroid when PEG-IFN was discontinued. During the follow-up period, thyroid function and thyroid antibodies were gradually back to normal in six individuals with aberrant thyroid function ( Table 5 ).