关键词: 血清乙型肝炎病毒 RNA, 慢性乙型肝炎, 核苷（酸）类似物, 乙型肝炎e抗原, 血清学转换

Abstract:

Objective To evaluate the predictive capacity of the baseline serum HBV RNA level for HBeAg seroconversion in patients with chronic hepatitis B （CHB） undergoing nucleos（t）ide analogues （NAs） treatment. Methods A total of 317 HBeAg-positive CHB patients who received nucleos（t）ide analogs （NAs） treatment in the Department of Hepatology， Kunming Third People's Hospital from July 2023 to September 2024 were recruited as the study subjects. These patients were categorized into the seroconversion group and the non-seroconversion group according to the occurrence of HBeAg seroconversion at week 48 of treatment. A retrospective analysis was carried out on the baseline serum HBV RNA levels and other relevant indicators. Variables that showed statistical significance in the univariate analysis were subjected to a multicollinearity test， followed by a multivariate Logistic regression analysis to identify the independent influencing factors for HBeAg seroconversion. The "rms" package in R software was utilized to construct a nomogram risk prediction model. Bootstrap sampling （B = 1 000） was applied to generate the calibration curve of the model. The Hosmer - Lemeshow test was adopted to assess the model's goodness-of-fit， and the receiver operating characteristic （ROC） curve was plotted with the calculation of the area under the curve （AUC）. Results At the 48th week of NA treatment， 23.97% （76/317） of the 317 HBeAg-positive CHB patients achieved HBeAg seroconversion. Baseline HBV RNA （odds ratio ［OR］ = 12.630， 95% confidence interval ［CI］： 6.096 ~ 26.167， P < 0.001）， HBcAb （OR = 0.110， 95% CI： 0.041 ~ 0.298， P < 0.001）， AFP （OR = 1.231， 95% CI： 1.072 ~ 1.413， P = 0.003）， and GGT （OR = 1.010， 95% CI： 1.001 ~ 1.019， P = 0.034） were identified as independent predictors of HBeAg seroconversion. A nomogram prediction model was developed based on these factors. The calibration curve generated through Bootstrap sampling （B = 1 000） demonstrated a high level of consistency between the predicted probability and the actual observed probability of HBeAg seroconversion. The Hosmer-Lemeshow goodness-of-fit test （χ² = 4.939， P = 0.764） further verified that the model had good calibration. The area under the curve （AUC） of the combined predictors was 0.894. Among them， the AUC of HBV RNA was 0.786， with a sensitivity of 0.913 and a specificity of 0.645. Conclusion The baseline serum HBV RNA level holds a specific predictive value for HBeAg seroconversion in chronic hepatitis B （CHB） patients undergoing nucleoside analogues （NAs） treatment.

Key words: serum HBV RNA, chronic hepatitis B, nucleos（t）ide analogues, HBeAg, seroconversion