HBV surface antigen loss prediction: A challenging yet achievable goal

Document Type : Original article

Authors

1 Endemic Medicine Faculty of Medicine, Helwan Univ., Cairo, Egypt

2 Hepatology and Gastroenterology Unit, Internal Medicine dept., Faculty of Medicine, Mansoura Univ., El-Mansoura, Egypt

3 Egyptian Liver Research Institute and Hospital (ELRIAH), Sherbin, El-Mansoura, Egypt

4 Internal Medicine dept., Albaha Univ., Albaha, KSA

5 Tropical Medicine dept., Faculty of Medicine, Port Said University, Port Said, Egypt

Abstract

Introduction and aim: Seroconversion of hepatitis B surface (HBs Ag) antigen is an infrequent event of cure during treatment of chronic hepatitis B virus (HBV) patients. It was estimated to be about 1-2% annually in Asian and Western cases. The aim of our study is to assess factors that could predict HBV surface (HBs) antigen loss in chronic HBV patients treated by nucleos(t)ide analogues. Methods: A total of 510 chronic HBV patients treated by (lamivudine, entecavir and tenofovir) with a median treatment time of 3.5 years (2.1-4.2 years). HBs antigen levels, HBV DNA levels, liver stiffness by FibroScan, liver functions tests, complete blood counts and prothrombin time were assessed at baseline and during follow up visits every 6 months. Results: Out of the 510 treated chronic HBV patients; 34 patients achieved HBs antigen loss. Multivariant regressive analysis showed that HBs antigen levels and HBV DNA levels were significantly lower in patients achieving HBs antigen loss compared to whom don’t achieve with median [IQR] values of (2.53 [2.15-2.89] log10 IU/ml and 3.73 [3.66-3.83] log10 IU/ml) versus (3.24 [2.81-3.68] log10 IU/ml and 3.88 [3.69-4.22] log10 IU/ml), with p 1000 IU/ml (p value <0.001) and there were only 5 (14.7%) cases of cirrhosis (liver stiffness > 13.4 kPa) in HBs antigen loss group and 29 (85.3%) had no cirrhosis (p value 0.057). The cut-off points of basal HBs antigen quantitation to detect on-treatment HBs antigen loss was found to be 489 IU/ml with sensitivity 0.876 (95% CI 0.843 to 0.904), specificity 0.618 (95% CI 0.436 to 0.778), positive predictive value 0.970 (95 CI 0.939 to 0.977) and negative predictive value 0.262 (95% CI 0.213 to 0.436). The HBs antigen clearance group showed log10 reduction of HBs antigen quantitation by 0.42 after 24 weeks of treatment in comparison to 0.04 in the failure group (p value <0.001). Regarding trea-tment type in HBs antigen loss group; 19 cases was treated by entecavir and 15 by tenofovir and none of the Lamivudine treated patients achieved HBs antigen loss. Conclusions: Low HBs antigen, HBV DNA levels, and absence of liver cirrhosis could be considered as baseline predictors of HBs antigen loss. Rapid decline of HBs antigenemia with treatment and absence of treatment relapses are on-treatment indicators of HBs antigen clearance.

Keywords

Files

Share

How to cite

Statistics