Establishment of a multi-parameter prediction model for the functional cure of HBeAg-negative chronic hepatitis B patients treated with pegylated interferonα and decision process based on response-guided therapy strategy

This study aimed to establish multivariate prediction models according to a response-guided therapy (RGT) based strategy at baseline and week 12 and 24 of follow-up to predict the functional cure for HBeAg-negative patients with chronic hepatitis B (CHB) treated with pegylated interferonα (PEG-IFNα). A total of 242 HBeAg-negative patients with CHB were treated with PEG-IFNα for 52 weeks and followed up for 24 weeks. Responses at the end of follow-up (EOF) were defined as hepatitis B surface antigen (HBsAg) loss, and patients were defined as either responders or non-responders. The three most meaningful predictors were an age ≤ 40 years, alanine aminotransferase (ALT) levels ≤ 40 U/L, and HBsAg levels ≤ 100 IU/mL at baseline; ALT levels ≥ 80 U/L, anti-HBc levels ≤ 8.42 S/CO, and HBsAg levels ≤ 50 IU/mL at week 12; and ALT levels ≥ 40 U/L, anti-HBc levels ≤ 8.46 S/CO, and HBsAg levels ≤ 0.2 IU/mL at week 24. The response rates of patients with a score of 0-1 and 4-5 at baseline, week 12, and 24 were 13.5%, 7.8%, and 11.7%; and 63.6%, 68.1%, and 98.1%, respectively. At week 12, the cumulative scores were 0-2, 3-4, 5-7, and 8-10 (response rates 5.0%, 18.9%, 41.3%, and 71.4%, respectively). At week 24, the cumulative scores were 0-3, 4-6, 7-10, and 11-15 (response rates: 1.3%, 12.3%, 37.0%, and 92.5%, respectively). At baseline, patients with scores of 0-1 were slightly recommended; at week 12, patients with 0-1 or 0-2 cumulative scores were recommended to stop treatment. At week 24, patients with a score of 0-1 or a cumulative score of 0-6 were recommended to stop treatment. We established a multi-parameter prediction model for the functional cure of HBeAg-negative patients with CHB treated with PEG-IFNα.