Elsevier

Journal of Hepatology

Volume 58, Issue 6, June 2013, Pages 1089-1095
Journal of Hepatology

Research Article
Hepatitis B surface antigen serum level is associated with fibrosis severity in treatment-naïve, e antigen-positive patients

https://doi.org/10.1016/j.jhep.2013.01.028Get rights and content

Background & Aims

Little is currently known about the association between serum HBsAg or HBV DNA levels and the severity of liver disease in chronic hepatitis B (CHB) patients. Therefore, we investigated these relationships in a large cohort of unselected, well-characterized, treatment-naïve CHB patients.

Methods

CHB patients were assessed at the Hôpital Beaujon in Paris, France, between 2000 and 2008. Serum samples and liver biopsies were obtained on the same day. HBsAg, HBV DNA, and HBV genotype were investigated using commercial diagnostic assays and liver histology was scored using the METAVIR system.

Results

406 patients were included in this cross-sectional study. Serum HBsAg and HBV DNA levels in hepatitis B e antigen-positive (HBeAg[+]) patients showed strong correlation (r = 0.44, p <0.0001), as did serum HBsAg levels and fibrosis severity (r = 0.43, p <0.0001). HBeAg(+) patients with moderate to severe fibrosis exhibited significantly lower serum HBsAg and HBV DNA levels compared with patients with no or mild fibrosis. Modeling analysis suggested a serum HBsAg cut-off of 3.85 log IU/ml would provide a theoretical sensitivity of 100% (95% CI: 0–100), theoretical specificity of 86% (95% CI: 50–100), and a negative predictive value of 100% (95% CI: 67–100) in HBeAg(+) patients infected with HBV genotype B or C.

Conclusions

We found an association between low serum HBsAg levels and moderate to severe fibrosis in HBeAg(+) CHB patients. Furthermore, we described a serum HBsAg cut-off for the prediction of fibrosis severity in CHB patients infected with HBV genotype B or C.

Introduction

Chronic infection with hepatitis B virus (HBV) remains a considerable global healthcare problem despite the greatly reduced rates of infection observed since the introduction of a vaccine in the early 1980s. The disease is estimated to be responsible for 0.5–1.2 million deaths per year, with the Asia Pacific region accounting for 75% of the estimated 350 million individuals living with chronic hepatitis B (CHB) [1], [2]. The disease has a relatively low prevalence in Europe and the USA, but the rates of infection within high-risk groups, such as injection drug users, sex workers, and immigrant populations from areas with high prevalence, underlines the need for additional measures that reduce the burden of the disease [2], [3].

HBV infection is not cytopathogenic and the virus persists within the nuclei of infected hepatocytes as covalently closed circular DNA (cccDNA) minichromosomes and viral DNA sequences integrated into the host genome [2]. The quantification of serum HBV DNA is an established measure of viral load and a standard diagnostic tool for evaluating disease progression, candidacy for antiviral therapy, monitoring of treatment response, and for distinguishing patients with active disease from ‘inactive carriers’ [2]. A strong association between serum HBV DNA level and risk of hepatocellular carcinoma (HCC) has also been demonstrated [4], [5]. Hepatitis B surface antigen (HBsAg), which forms the protein envelope of the virus, is generally considered diagnostic for HBV infection [2]. HBsAg is secreted from infected cells and circulates within the serum as non-infectious spherical and filamentous structures, which can outnumber complete virions from 100- to 100,000-fold [6]. Seroclearance of HBsAg is commonly regarded as the clinical end point closest to cure and is associated with reduced progression to more serious liver disease [7], [8], [9]. There is growing interest in the use of quantitative HBsAg (qHBsAg) as a prognostic marker in CHB patients [10]. Serum HBsAg levels have been demonstrated to be clinically useful for identifying the stage of disease [11], [12], for distinguishing true inactive carriers from patients with hepatitis B e antigen-negative (HBeAg[−]) disease [13], [14], [15], [16], and for predicting response to interferon therapy [17], [18]. Recent research has also demonstrated that HBsAg levels are associated with the risk of progression to HCC, especially in patients with low HBV DNA levels [19], [20]. Little is currently known about the association between HBsAg or HBV DNA levels and the severity of liver disease at any specific time point. In contrast to HBV DNA [4], [5], only one published study has investigated the potential correlation between HBsAg levels and the stage of fibrosis [21]. This study was restricted to selected, hepatitis B e antigen-positive (HBeAg[+]) CHB patients and did not include analysis of viral genotype. The aim of this cross-sectional study was to therefore investigate the potential relationship between serum HBsAg and HBV DNA levels and the severity of liver lesions in a large cohort of unselected, consecutive, well-characterized, treatment-naïve CHB patients infected with different HBV genotypes.

Section snippets

Patients

Unselected, consecutive, treatment-naïve CHB patients were assessed at the Hôpital Beaujon in Paris, France, between 2000 and 2008. Inclusion criteria for patients were: HBsAg-positive, known HBeAg status, scheduled liver biopsy, and absence of HCV, HDV, or HIV co-infection. Serum samples and liver biopsies were obtained on the same day in all cases.

Diagnostic tests

Serum ALT levels were expressed as IU/ml. HBsAg levels in patient serum samples were measured using the Elecsys® HBsAg II quant assay (Roche

Study population

406 unselected, consecutive, treatment-naïve CHB patients were included in the study: 101 patients (25%) were HBeAg(+) and 305 patients (75%) were HBeAg(−) (Table 1). There were more men than women in the study population, with 75% of patients in both HBeAg(+) and HBeAg(−) groups being male. Mean alanine aminotransferase (ALT) levels were significantly higher in HBeAg(+) patients and there was a significantly greater proportion of patients with normal ALT in the HBeAg(−) group. Both the

Discussion

Our study population represents a well-characterized CHB cohort from the multicultural city of Paris, which had not been pre-selected in any way and had not received previous treatment for HBV infection. All patient serum samples were taken on the same day as liver biopsies and therefore represent an accurate ‘snapshot’ of the state of disease occurring within each patient at that specific time. We observed a strong correlation between HBsAg and HBV DNA levels in the serum of treatment-naïve

Financial support

The study was funded in part by Roche Diagnostics.

Conflict of interest

Michelle Martinot-Peignoux has acted as a speaker and expert for Roche Diagnostics.

Patrick Marcellin has received research grants from and acted as an investigator, speaker, and expert for Roche Diagnostics. Richard Batrla and Friedemann Krause are employees of Roche Diagnostics. All other authors declare no conflict of interest.

Acknowledgements

The authors thank Roche Diagnostics for financial support and David Wateridge, associated with Elements Communications (Westerham, UK), for providing editorial assistance with the preparation and submission of the final draft of the manuscript, which was funded by Roche Diagnostics.

References (25)

  • U.H. Iloeje et al.

    Risk and predictors of mortality associated with chronic hepatitis B infection

    Clin Gastroenterol Hepatol

    (2007)
  • D. Lavanchy et al.

    Virus epidemiology, disease burden, treatment, and current and emerging prevention and control measures

    J Viral Hepat

    (2004)
  • Cited by (101)

    • Quantitative HBcrAg and HBcAb versus HBsAg and HBV DNA in predicting liver fibrosis levels of chronic hepatitis B patients

      2020, Gastroenterologia y Hepatologia
      Citation Excerpt :

      Within the frameworks of HBeAg-positive and HBeAg-negative patients, respectively, we investigated comparatively the correlation of HBsAg, HBcrAg, HBV DNA and HBcAb of quantitative HBV markers with liver pathological grade and stage; and evaluated comparatively the performance of these HBV markers, and determined clinically valuable single tradeoff cutoffs and practical cutoffs of these HBV markers in not only affirming significant fibrosis but also screening cirrhosis. Previous studies showed that, HBsAg, HBcrAg and HBV DNA are significantly negatively correlated with pathological grade and stage of HBeAg-positive patients, and significantly positively correlated with pathological grade and stage of HBeAg-negative patients;9–22 while HBcAb is significantly positively correlated with pathological grade and stage of both HBeAg-positive and HBeAg-negative patients.23–28 This study also displayed similar results.

    View all citing articles on Scopus
    View full text