Advertisement

Hepatitis B surface antigen quantification: Why and how to use it in 2011 – A core group report

  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Henry Lik-Yuen Chan
    Correspondence
    Corresponding author. Address: Department of Medicine and Therapeutics, 9/F Prince of Wales Hospital, 30-32 Ngan Shing Street, Shatin, Hong Kong. Tel.: +852 26323593; fax: +852 26373852.
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Department of Medicine and Therapeutics and Institute of Digestive Disease, The Chinese University of Hong Kong, Hong Kong
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Alex Thompson
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Victorian Infectious Diseases Reference Laboratory, 10 Wreckyn St., North Melbourne, Victoria 3051, Australia
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Michelle Martinot-Peignoux
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    INSERM U773, Centre de Recherche Biomédicale Bichat-Beaujon (CRB3), Service d’Hépatologie, Université Paris-Diderot, Hopilal Beaujon, Clichy, France
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Teerha Piratvisuth
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Department of Medicine, Prince of Songkla University, NKC Institute of Gastroenterology and Hepatology, Hat Yai, Songkla, Thailand
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Markus Cornberg
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Department of Gastroenterology, Hepatology, and Endocrinology, Hannover Medical School, Hannover, Germany
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Maurizia Rossana Brunetto
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Hepatology Unit, University of Pisa, Italy
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Hans L. Tillmann
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Duke Clinical Research Institute, Duke University, Durham, NC, USA
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Jia-Horng Kao
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Graduate Institute of Clinical Medicine, National Taiwan University College of Medicine, Taipei, Taiwan
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Ji-Dong Jia
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Liver Research Center, Beijing Friendship Hospital, Capital Medical University, Beijing 100050, China
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Heiner Wedemeyer
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Department of Gastroenterology, Hepatology, and Endocrinology, Hannover Medical School, Hannover, Germany
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Stephen Locarnini
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Victorian Infectious Diseases Reference Laboratory, 10 Wreckyn St., North Melbourne, Victoria 3051, Australia
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Harry L.A. Janssen
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    Department of Gastroenterology & Hepatology, Erasmus MC University Hospital, Rotterdam, The Netherlands
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Patrick Marcellin
    Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
    Affiliations
    INSERM U773, Centre de Recherche Biomédicale Bichat-Beaujon (CRB3), Service d’Hépatologie, Université Paris-Diderot, Hopilal Beaujon, Clichy, France
    Search for articles by this author
  • Author Footnotes
    † for the Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)
Open AccessPublished:June 29, 2011DOI:https://doi.org/10.1016/j.jhep.2011.06.006
      Quantitative HBsAg had been suggested to be helpful in management of HBV, but assays were cumbersome. The recent availability of commercial quantitative assays has restarted the interest in quantitative serum hepatitis B surface antigen (HBsAg) as a biomarker for prognosis and treatment response in chronic hepatitis B. HBsAg level reflects the transcriptional activity of cccDNA rather than the absolute amount of cccDNA copies. Serum HBsAg level tends to be higher in hepatitis B e antigen (HBeAg)-positive than HBeAg-negative patients. Among patients with a low HBV DNA (<2000 IU/ml), HBsAg <1000 IU/ml in genotype D HBV infection and HBsAg <100 IU/ml in genotype B/C HBV infection is associated with inactive carrier state in HBeAg-negative patients. The HBsAg reduction by nucleos(t)ide analogues (NA) is not as pronounced as by interferon treatment. On peginterferon treatment, sustained responders tend to show greater HBsAg decline than the non-responders. The optimal on-treatment HBsAg cutoff to predict response needs further evaluation in HBeAg-positive patients, but an absence of HBsAg decline together with a <2 log reduction in HBV DNA at week 12 can serve as stopping rule in HBeAg-negative patients with genotype D HBV infection. A rapid serum HBsAg decline during NA therapy may identify patients who will clear HBsAg in the long-term. There are early reports among Asian patients that an HBsAg level of <100 IU/ml might predict lower risk of relapse after stopping NA treatment. In clinical practice, serum HBsAg level should be used together with, but not as a substitute for, HBV DNA.

      Abbreviations:

      Anti-HBe (antibodies to hepatitis B e antigen), ALT (alanine aminotransferase), cccDNA (covalently closed circular DNA), CHB (chronic hepatitis B), HBV (hepatitis B virus), HBeAg (hepatitis B e antigen), HBsAg (hepatitis B surface antigen), NA (nuclet(s)ide analogues), NPV (negative predictive value), ORF (open reading frame), PPV (positive predictive value)

      Keywords

      Introduction

      Since the discovery of hepatitis B surface antigen (HBsAg) by Blumberg in 1965, it has been used as the hallmark for the diagnosis of hepatitis B virus (HBV) infection [
      • Blumberg B.S.
      • Sutnick A.I.
      • London W.T.
      Hepatitis and leukemia: their relation to Australia antigen.
      ]. The importance of HBsAg quantification was recognized early on, but initial tests were labor intensive and thus restricted to the research setting [
      • Frosner G.G.
      • Schomerus H.
      • Wiedmann K.H.
      • et al.
      Diagnostic significance of quantitative determination of hepatitis B surface antigen in acute and chronic hepatitis B infection.
      ]. In addition, no treatment for HBV was available in the 1980s when those labor-intensive tests were developed and, therefore, these tests did not become clinically relevant. Although standard assays have been available for quantification of serum HBsAg for a number of years, it is the presence or absence of HBsAg that carries a clinical meaning to most physicians [
      • Deguchi M.
      • Yamashita N.
      • Kagita M.
      • et al.
      Quantitation of hepatitis B surface antigen by an automated chemiluminescent microparticle immunoassay.
      ,
      • Bonino F.
      • Brunetto M.R.
      • Colombatto P.
      • Moriconi F.
      • Torresani E.
      • Lunghi G.
      • et al.
      Use of the Elecysys HBsAg II assay for simple and accurate quantification of HBsAg levels in sera of patients infected with HBV.
      ]. The interest in HBsAg quantification started with the possible observation of its association with the level of covalently closed circular (ccc) DNA, the template for viral replication inside the nuclei of hepatocytes [
      • Werle-Lapostolle B.
      • Bowden S.
      • Locarnini S.
      • et al.
      Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy.
      ,
      • Chan H.L.
      • Wong V.W.
      • Tse A.M.
      • et al.
      Serum hepatitis B surface antigen quantitation can reflect hepatitis B virus in the liver and predict treatment response.
      ].
      In recent years, many studies have been performed, addressing the use of serum HBsAg level to monitor the natural history and predict treatment response in chronic hepatitis B (CHB). In light of the availability of the new data, a group of hepatologists met in September 2010 (in Asian Pacific Digestive Week, Kuala Lumpur, Malaysia), October 2010 (in the Annual meeting of American Association for the Study of the Liver, Boston, United States) and February 2011 (in the Annual meeting of Asian Pacific Association for the Study of the Liver, Bangkok, Thailand) with the unrestricted support of F. Hoffmann-la Roche to discuss the role of serum HBsAg quantification in clinical practice. An independent literature search was performed by the group members. Data were derived from the electronic databases and the abstracts of major international conferences. The review was based primarily on published full manuscripts and supplemented by new data from appropriate conference abstracts. The suggestions were based on solid evidence where possible, but on their own clinical experience and expert opinion where data were lacking. The industrial partner did not advise nor interfere with the content of this paper. The authors were responsible for writing the entire manuscript without the assistance of a medical writer. The manuscript is not meant to be a guideline or a position paper. The aim of this paper was to give insight to clinicians concerning the use of serum HBsAg quantification in clinical practice and to highlight the questions that need to be addressed by future studies.

      Virology of hepatitis B surface antigen

      Current evidence

      HBsAg is the glycosylated envelope protein of the mature HBV virion. There are three HBsAg proteins – small (S), medium (M), and large (L). The envelope open reading frame (ORF) contains three in frame “start” codons which further divide it into pre-S1, pre-S2, and S-ORF domains (Fig. 1). There are two HBV mRNA transcripts, Pre-S1 mRNA, and Pre-S2/S mRNA, from which the L (Pre-S1) and M/S (Pre-S2 + S/S) proteins are translated, respectively. HBV replication occurs via the pregenomic RNA (pgRNA), a separate RNA transcript, and, therefore, the HBsAg secretory pathway and the viral replication pathway can be considered as distinct processes within the hepatocyte (Fig. 2).
      Figure thumbnail gr1
      Fig. 1The HBV open reading frames (ORF), highlighting the overlapping relationship between the envelope ORF and the HBV polymerase ORF.
      Figure thumbnail gr2
      Fig. 2HVB lifecycle. The lifecycle of HBV, highlighting: (i) the nuclear reservoir, covalently closed circular (ccc) DNA, which is the transcriptional template for the virus; (ii) the HBsAg secretory pathways, and (iii) the viral replication pathways. Nucleos(t)ide analogue therapy, by targeting the HBV reverse transcriptase (RT), selectively inhibits virion production, but does not reduce HBsAg levels as cccDNA levels are preserved. IFN-based strategies, which can non-cytolytically clear hepatocytes of HBV cccDNA infection, may therefore induce larger reductions in HBsAg.
      As well as virions, the sera of viremic patients contain large numbers of two types of non-infectious particles: spherical particles and filamentous forms (Fig. 2). Both types of subviral particles are composed of HBsAg. The subviral particles are secreted at levels far in excess of mature virions, and are believed to play a role as a decoy for humoral immunity. HBsAg may also be produced from HBV DNA integrated into the host genome (Fig. 2). Although viral integration is not required for normal productive hepadnaviral infection, integration of HBV DNA occurs illegitimately through recombination mechanisms using host enzymes acting on the double-stranded linear DNA form of HBV [
      • Kimbi G.C.
      • Kramvis A.
      • Kew M.C.
      Integration of hepatitis B virus DNA into chromosomal DNA during acute hepatitis B.
      ,
      • Bill C.A.
      • Summers J.
      Genomic DNA double-strand breaks are targets for hepadnaviral DNA integration.
      ]. In HBV infection, viral integration seems to occur early in infection. Integrated sequences cannot provide a template for productive viral replication as a complete genome is not present. However, given that sequences of the S-ORF with enhancer I elements are often present in integrated segments, HBsAg may be produced, often as truncated subviral particles [
      • Thomas H.C.
      • Lemon S.M.
      Viral hepatitis.
      ].
      HBV is unique among DNA viruses for replicating its genome via an RNA intermediate. The transcriptional template of HBV is the cccDNA which exists in the cell nucleus as a viral minichromosome and serves as the intrahepatic reservoir for HBV (Fig. 2). The challenge of antiviral therapy for CHB is to clear the liver of cccDNA. HBsAg titer should theoretically reflect liver total cccDNA, as well as the transcriptional activity of the cccDNA (Fig. 2). Indeed, it has been shown that HBsAg titers can correlate with serum HBV DNA and intrahepatic cccDNA levels, but that this may vary in the different phases of disease [
      • Nguyen T.
      • Thompson A.J.
      • Bowden S.
      • et al.
      Hepatitis B surface antigen levels during the natural history of chronic hepatitis B: a perspective on Asia.
      ,
      • Jaroszewicz J.
      • Calle Serrano B.
      • Wursthorn K.
      • et al.
      Hepatitis B surface antigen (HBsAg) levels in the natural history of hepatitis B virus (HBV)-infection: a European perspective.
      ,
      • Thompson A.J.
      • Nguyen T.
      • Iser D.
      • et al.
      Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
      ]. A positive correlation has been noted between HBsAg titer and serum HBV DNA and liver cccDNA in most studies of HBeAg-positive patients. However, the regulation of HBsAg production and secretion appears to be disconnected from that of virion production in the HBeAg-negative state [
      • Thompson A.J.
      • Nguyen T.
      • Iser D.
      • et al.
      Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
      ,
      • Lin L.Y.
      • Wong V.W.
      • Zhou H.J.
      • et al.
      Relationship between serum hepatitis B virus DNA and surface antigen with covalently closed circular DNA in HBeAg-negative patients.
      ]. In HBeAg-negative patients, although serum HBsAg titer, serum HBV DNA level and liver cccDNA level are all reduced relative to patients with HBeAg-positive CHB, HBsAg titers have been noted to be preserved relative to serum HBV DNA and cccDNA [
      • Thompson A.J.
      • Nguyen T.
      • Iser D.
      • et al.
      Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
      ,
      • Lin L.Y.
      • Wong V.W.
      • Zhou H.J.
      • et al.
      Relationship between serum hepatitis B virus DNA and surface antigen with covalently closed circular DNA in HBeAg-negative patients.
      ,
      • Volz T.
      • Lutgehetmann M.
      • Wachtler P.
      • et al.
      Impaired intrahepatic hepatitis B virus productivity contributes to low viremia in most HBeAg-negative patients.
      ]. The reason for this observation is unclear, but may be related to the presence of integrated HBV surface genome into the host chromosome and/or a preferential control of the replicative pathway over HBsAg transcription/secretion, where virion production is inhibited but secretion of subviral particles is preserved [
      • Thompson A.J.
      • Nguyen T.
      • Iser D.
      • et al.
      Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
      ,
      • Volz T.
      • Lutgehetmann M.
      • Wachtler P.
      • et al.
      Impaired intrahepatic hepatitis B virus productivity contributes to low viremia in most HBeAg-negative patients.
      ]. Additional studies are needed to determine the utility of HBsAg measurement in HBeAg-negative patients.
      HBV has the added complexity of overlapping reading frames. The envelope ORF completely overlaps the polymerase gene (Fig. 2). Therefore point mutations in the polymerase ORF that are selected by nucleos(t)ide analogue (NA) therapy may result in critical changes, including stop codons, in the envelope proteins. For example, the mutations rtA181T and rtM204I, associated with NA resistance, can result in the stop codons sW172 and W196 in the overlapping S-ORF, respectively [
      • Warner N.
      • Locarnini S.
      The antiviral drug selected hepatitis B virus rtA181T/sW172 mutant has a dominant negative secretion defect and alters the typical profile of viral rebound.
      ]. It remains unclear whether these mutations, which arise as one population in a quasispecies, affect HBsAg titers. They have been associated with hepatocyte HBsAg retention and potential oncogenic risk [
      • Warner N.
      • Locarnini S.
      Can antiviral therapy for chronic hepatitis B enhance the progression to hepatocellular carcinoma?.
      ].
      Finally, from a diagnostic perspective, it is important to appreciate that HBsAg quantification detects all three forms of circulating HBsAg. The antibodies used in the quantitative enzyme immunoassays target epitopes in the S protein, and are therefore not capable of distinguishing between the different HBsAg proteins, nor can they distinguish between virion-associated HBsAg, subviral particles and HBsAg produced from integrated sequence. Currently, there are two commercialized assay that can measure the HBsAg quantification, the Architect QT assay (Abbott Laboratories) and the Elecsys HBsAg II Quant assay (Roche Diagnostic). The Architect assay can measure HBsAg level from 0.05 to 250 IU/ml, and manual dilution is needed for measurement of higher HBsAg levels. The Elecsys II assay has an automatic on-board dilution with a range of HBsAg measurement from 0.05 to 52,000 IU/ml. In a study among 611 chronic hepatitis B patients from 4 sites covering HBV genotype A–G, approximately 72% of samples could have HBsAg level measured on the first analysis by the Elecsys II assay without the need of further manual dilution [
      • Bonino F.
      • Moriconi F.
      • Bowden S.
      • et al.
      Multicenter evaluation of the Elecsys HBsAg II Quant Assay.
      ]. There is good correlation between the HBsAg measurements by these 2 assays [

      Sonneveld MJ, Rijckborst V, Boucher CAB, et al. A comparison of two assays for quantification of hepatitis B surface antigen in serum of patients with chronic hepatitis B. J Hepatol 2011;54:Abstract 750.

      ,
      • Wursthorn K.
      • Jaroszewicz J.
      • Zacher B.J.
      • et al.
      Correlation between the Elecsys HBsAg II assay and the Architect assay for the quantification of hepatitis B surface antigen (HBsAg) in the serum.
      ].

      Natural history

      HBeAg-positive chronic hepatitis B

      Current evidence

      In perinatally acquired HBV infection from infected mothers with positive hepatitis B e antigen (HBeAg), immune tolerance will be induced by the HBeAg that crosses the placenta. Immune tolerance phase usually occurs in the initial 2–3 decades of life characterized by positive HBeAg, very high HBV DNA, normal alanine aminotransferase (ALT) levels, and minimal histologic damage [
      • Chan H.L.
      • Wong G.L.
      • Wong V.W.
      A review of the natural history of chronic hepatitis B in the era of transient elastography.
      ]. It is followed by an immune clearance phase, which may lead to HBeAg seroconversion. However, some patients have prolonged but unsuccessful immune clearance with persistently elevated HBV DNA and ALT levels, and these patients have a high risk of developing liver cirrhosis.
      Based on the results of two cross-sectional studies, the serum HBsAg level was generally higher among patients in the immune tolerance phase than in the immune clearance phase [
      • Nguyen T.
      • Thompson A.J.
      • Bowden S.
      • et al.
      Hepatitis B surface antigen levels during the natural history of chronic hepatitis B: a perspective on Asia.
      ,
      • Jaroszewicz J.
      • Calle Serrano B.
      • Wursthorn K.
      • et al.
      Hepatitis B surface antigen (HBsAg) levels in the natural history of hepatitis B virus (HBV)-infection: a European perspective.
      ]. In the European study comparing 30 patients in the immune tolerance phase versus 48 patients in the immune clearance phase, the mean serum HBsAg level was 4.96 log IU/ml vs. 4.37 log IU/ml, respectively [
      • Jaroszewicz J.
      • Calle Serrano B.
      • Wursthorn K.
      • et al.
      Hepatitis B surface antigen (HBsAg) levels in the natural history of hepatitis B virus (HBV)-infection: a European perspective.
      ]. In the Asian study, the mean serum HBsAg level was 4.53 log IU/ml among patients in the immune tolerance phase (n = 32) vs. 4.03 log IU/ml among patients in the immune clearance phase (n = 55) [
      • Nguyen T.
      • Thompson A.J.
      • Bowden S.
      • et al.
      Hepatitis B surface antigen levels during the natural history of chronic hepatitis B: a perspective on Asia.
      ]. One possible reason for the lower serum HBsAg levels among patients in the immune tolerance phase in the Asian study was the inclusion of patients with ALT at 1–2 times upper limit of laboratory normal, which might actually represent early immune clearance. Another explanation may be the difference in HBsAg expression related to different HBV genotypes. In an experiment using HBV infected cell lines, HBsAg expression was highest among HBV subgenotype A2/Ae, followed by A1/Aa and B2/Ba, and distantly by B1/Bj as well as C, and the least for D [
      • Sugiyama M.
      • Tasuhito T.
      • Kato T.
      • et al.
      Influence of hepatitis B virus genotypes on the intra- and extracellular expression of viral DNA and antigens.
      ].
      In a longitudinal study including untreated CHB patients followed up for 99 ± 16 months, serial HBsAg levels among chronic hepatitis B patients at different stages of disease were compared [
      • Chan H.L.
      • Wong V.W.
      • Wong G.L.
      • Tse C.H.
      • Chan H.Y.
      • Sung J.J.
      A longitudinal study on the natural history of serum HBsAg changes in chronic hepatitis B.
      ]. HBsAg level was persistently high at approximately 5 log IU/ml among patients in the immune tolerance phase, and HBsAg level tended to be very stable with a median annual decline of −0.006 log IU/ml. The HBsAg level among patients in the immune clearance phase was approximately 4 log IU/ml, and no difference could be detected between patients who had persistently active disease and those who underwent HBeAg seroconversion. The HBsAg levels also tended to be very steady among HBeAg-positive patients with active disease with a median annual decline of 0.021 log IU/ml.
      The ratio of HBsAg to HBV DNA, which reflects the proportion of subviral particles to virions was very similar among all HBeAg-positive patients. In various studies, the median HBsAg to HBV DNA ratio ranged between 0.5 and 0.6 in all HBeAg-positive patients, with individual patients ranging between 0.1 and 2.0 [
      • Nguyen T.
      • Thompson A.J.
      • Bowden S.
      • et al.
      Hepatitis B surface antigen levels during the natural history of chronic hepatitis B: a perspective on Asia.
      ,
      • Jaroszewicz J.
      • Calle Serrano B.
      • Wursthorn K.
      • et al.
      Hepatitis B surface antigen (HBsAg) levels in the natural history of hepatitis B virus (HBV)-infection: a European perspective.
      ,
      • Chan H.L.
      • Wong V.W.
      • Wong G.L.
      • Tse C.H.
      • Chan H.Y.
      • Sung J.J.
      A longitudinal study on the natural history of serum HBsAg changes in chronic hepatitis B.
      ].

      Suggestions

      Serum HBsAg level is higher in the immune tolerance phase than in the immune clearance phase. A very high HBsAg level of approximately 100,000 IU/ml can be a supportive evidence of immune tolerance. Hence HBsAg can provide additional information to differentiate immune tolerance and immune clearance when the HBV DNA is high and ALT is normal or minimally elevated. HBsAg level or HBsAg/HBV DNA ratio should not be used to predict the chance of spontaneous HBeAg seroconversion.

      Challenges

      More research is required to explore the rate of HBsAg reduction on the possibility of successful immune clearance in HBeAg-positive patients. There is preliminary data suggesting lower HBsAg level is associated with more severe liver fibrosis in HBeAg-positive patients [

      Cheng PN, Tsai HW, Chang TT. Serum hepatitis B surface antigen level is associated with liver fibrosis in patients with HBeAg-positive chronic hepatitis B. J Hepatol 2011;54:Abstract 362.

      ]. This is an important area as HBV DNA level has no association with the severity of liver fibrosis in HBeAg-positive patients [
      • Wong G.L.
      • Wong V.W.
      • Choi P.C.
      • et al.
      Clinical factors associated with liver stiffness in hepatitis B e antigen-positive chronic hepatitis B.
      ].

      HBeAg-negative chronic hepatitis B

      Current evidence

      Later in life, most of the HBV infected population evolves into the immune clearance phase, losing HBeAg and seroconverting to antibodies to HBeAg (anti-HBe). Because of the aging of the infected population, the prevalence of HBeAg-negative CHB seems to be increasing worldwide [
      • Gaeta G.B.
      • Stornaiuolo G.
      • Precone D.F.
      • et al.
      Epidemiological and clinical burden of chronic hepatitis B virus/hepatitis C virus infection. A multicenter Italian study.
      ,
      • Hadziyannis S.
      • Papatheodoridis G.V.
      Hepatitis B e antigen-negative chronic hepatitis: natural history and treatment.
      ,
      • Zarski J.P.
      • Marcellin P.
      • Leroy V.
      • et al.
      Characteristics of patients with chronic hepatitis B in France: predominant frequency of HBe antigen negative cases.
      ]. HBeAg-negative CHB is characterized by wide fluctuation of serum HBV DNA levels and biochemical activity [
      • Brunetto M.R.
      • Oiveri F.
      • Coco B.
      • et al.
      Outcome of anti-HBe positive chronic hepatitis B in alpha-interferon and untreated patients: a long term cohort study.
      ]. In patients with HBeAg-negative CHB, viremia may fall below 2000 IU/ml (the threshold identifying inactive HBV infection) even if transiently, thus for the differential diagnosis between HBeAg negative chronic hepatitis B and “inactive HBV carrier”, serial determinations of serum ALT and HBV DNA levels are needed [
      • Sung J.J.
      • Chan H.L.
      • Wong M.L.
      • et al.
      Relationship of clinical and virological factors with hepatitis activity in hepatitis B e antigen-negative chronic hepatitis B virus-infected patients.
      ,
      • Martinot-Peignoux M.
      • Boyer N.
      • Colombat M.
      • et al.
      Serum hepatitis B virus HBV DNA levels and liver histology in inactive HBsAg carriers.
      ]. The inactive carriers have no or mild liver histological lesions, with an excellent survival prognosis and low incidence of cirrhosis and HCC, while patients with HBeAg-negative CHB with fluctuating activity have more severe progression of the disease with frequent development of cirrhosis [
      • Martinot-Peignoux M.
      • Boyer N.
      • Colombat M.
      • et al.
      Serum hepatitis B virus HBV DNA levels and liver histology in inactive HBsAg carriers.
      ,
      • Brunetto M.R.
      • Oliveri F.
      • Rocca G.
      • et al.
      Natural course and response to interferon of chronic hepatitis B accompanied by antibody to hepatitis B e antigen.
      ,
      • Papatheodoridis G.V.
      • Manesis E.K.
      • Manalakopoulos S.
      • et al.
      Is there a meaninful serum hepatitis B virus DNA cutoff level for therapeutic decisions in hepatitis B e antigen-negative chronic hepatitis B infection?.
      ,
      • McMahon B.J.
      The natural history of chronic hepatitis B virus infection.
      ]. According to the American and European guidelines [
      • Lok A.S.
      • Heathcote E.J.
      • Hoofnagle J.H.
      Management of hepatitis B: 2000 – summary of a workshop.
      ,

      EASL clinical practice guidelines: management of chronic hepatitis B. European Association for the Study of the Liver. J Hepatol 2009;50:227–242.

      ], the differentiation between the inactive and active phases of HBeAg-negative CHB is based on an HBV DNA cut-off set up at 2000 IU/ml. This cut-off has led to several controversial reports [
      • Martinot-Peignoux M.
      • Boyer N.
      • Colombat M.
      • et al.
      Serum hepatitis B virus HBV DNA levels and liver histology in inactive HBsAg carriers.
      ,
      • Chu C.J.
      • Hussain M.
      • Lik A.S.F.
      Quantitative serum HBV DNA levels during different stages of chronic hepatitis B infection.
      ,
      • Seo Y.
      • Yoon S.
      • Truong B.X.
      • et al.
      Serum HBV DNA levels differentiating inactive carriers from patients with chronic hepatitis B.
      ,
      • Feld J.J.
      • Ayers M.
      • El-Ashry D.
      • et al.
      Hepatitis B virus DNA prediction rules for Hepatitis B e antigen-negative chronic hepatitis B.
      ].
      Based on several longitudinal studies, HBsAg levels are higher in patients with active HBeAg-negative CHB than in “inactive carriers”. A longitudinal study in Hong Kong reported, on 68 HBeAg-negative CHB patients predominantly infected with genotypes B and C HBV and followed for over 8 years, that the patients with an inactive disease tended to have lower HBsAg levels than those who had active disease; 2.24 ± 1.61 log IU/ml vs. 2.98 ± 0.88 log IU/ml, respectively (p = 0.054) [
      • Chan H.L.
      • Wong V.W.
      • Wong G.L.
      • Tse C.H.
      • Chan H.Y.
      • Sung J.J.
      A longitudinal study on the natural history of serum HBsAg changes in chronic hepatitis B.
      ]. In an extended report among 103 HBeAg-negative CHB patients followed up for a median of 11 years, serum HBsAg of ⩽100 IU/ml had 75% sensitivity and 91% specificity to predict spontaneous HBsAg seroclearance [
      • Chan H.L.
      • Wong G.L.
      • Tse C.H.
      • Chan H.Y.
      • Wong V.W.
      Viral determinants of hepatitis B surface antigen seroclearance in hepatitis B e antigen-negative chronic hepatitis B patients.
      ].
      In Taiwan where genotype B and C HBV are prevalent, a cross-sectional study on 251 chronic hepatitis B patients showed that mean serum HBsAg level was higher in the immune clearance phases (3.81 log IU/ml), significantly decreased in the inactive carrier state (2.25 log IU/ml) and re-elevated in the reactivation phase (2.77 log IU/ml) [
      • Su T.H.
      • Hsu C.S.
      • Chen C.L.
      • et al.
      Serum hepatitis B surface antigen concentration correlates with HBV DNA level in patients with chronic hepatitis B.
      ]. In another report including 390 patients who experienced spontaneous HBeAg seroconversion, serum HBsAg level at 1 year post HBeAg seroconversion was inversely associated with the probability of HBsAg loss in a dose–response manner. Compared to patients with HBsAg levels ⩾1000 IU/ml, the rate of HBsAg loss was higher for those with HBsAg levels of 100–999 and <100 IU/ml, with hazard ratios of 4.4 (95% confidence interval, 1.1–17.0) and 24.3 (8.7–67.5), respectively. In patients with serum HBV DNA levels <200 IU/ml, serum HBsAg level <100 IU/ml predicted HBsAg loss within 6 years with a positive predictive value (PPV) of 45.5% and negative predictive value (NPV) of 98.6% [

      Tseng TC, Liu CJ, Su TH, et al. Serum Hepatitis B surface antigen levels predict surface antigen loss in hepatitis B e antigen seroconverters. Gastroenterology, 2011[Epub ahead of print].

      ].
      A study performed in Italy investigated serum HBsAg levels and their diagnostic role in genotype D HBeAg-negative HBV carriers: 209 carriers where prospectively followed-up for a median period of 34.5 (6–110) months and classified as having active or inactive infection according to the virologic profile during 1 year of monthly monitoring [
      • Brunetto M.R.
      • Oliveri F.
      • Colombatto P.
      • et al.
      Hepatitis B surface antigen serum levels help to distinguish active from inactive hepatitis B virus genotype D carriers.
      ]. Serum HBsAg levels were significantly lower in inactive (56 carriers) than in active infection (153 carriers): median values 62.12 (0.1–4068) IU/ml vs. 3029 (0.5–82,480) IU/ml, respectively (p <0.001). Among inactive carriers, HBsAg serum levels were lower in 31 asymptomatic carriers whose viremia remained persistently below 20,000 IU/ml than in 122 asymptomatic carriers with fluctuations >20,000 IU/ml: 883 (0.5–7838) vs. 4233 (164–82,480) IU/ml (p = 0.002). The combination of a single measurement of HBsAg <1000 IU/ml and HBV DNA <2000 IU/ml allowed identifying “inactive carriers” with a PPV 87.9% and NPV 96.7%.
      Similar results were observed in a French study performed in 122 HBeAg-negative CHB (genotypes A–E) [
      • Martinot-Peignoux M.
      • Lada O.
      • Cardoso A.C.
      • et al.
      Quantitative HBsAg: A new specific marker for the diagnosis of HBsAg inactive carriage.
      ]. One hundred and two were inactive carriers (three normal ALT measurements in 1 year) and 20 were patients with HBeAg-negative CHB. HBsAg levels were lower in the 102 inactive carriers than in the 20 patients with HBeAg-negative CHB; 3.30 ± 0.97 vs. 3.77 ± 0.11 log IU/ml, respectively (p <0.001). At baseline, 50 of 54 (92.5%) patients who had HBsAg ⩽2000 IU/ml and 53 of 57 (93%) patients who had HBV DNA ⩽2000 IU/ml were inactive carriers. Overall, 32 patients had HBsAg ⩽2000 IU/ml and HBV DNA ⩽2000 IU/ml, and all were “inactive carriers” (PPV 100%). Similarly, one recent Turkish study found that an HBsAg level of 2040 IU/ml best defined inactive carriers from chronic hepatitis patients during a 3-year follow-up (sensitivity 87.2%, specificity 75.3%) [

      Yakut M, Bektas M, Seven G, et al. Characterization of the inactive HBsAg carrier state with 3 year follow-up. J Hepatol 2011;54:Abstract 398.

      ].

      Suggestions

      The combination of HBsAg and HBV DNA measurements at a single time-point might allow an accurate identification of “true inactive carriers” and prediction of HBsAg loss in the HBeAg-negative CHB population with distinct HBV genotype infection. It seems that HBsAg <1000 IU/ml is necessary but may not be sufficient to identify inactive carriers with HBV genotype D infection, and HBsAg <100 IU/ml can predict HBsAg loss over time in genotype B or C HBV-infected patients, respectively.

      Challenges

      The utility of this new marker needs further validation, for the management of HBeAg-negative CHB patients, in order to better identify those that could benefit from therapy and increased survival. Better identification of “true inactive carriers” or patients with a high likelihood of HBsAg loss, who have a good survival prognosis and need less frequent monitoring, will reduce the costs of medical expenses. However, HBsAg may also decrease with age and disease progression, and other clinical parameters including HBV DNA level must be taken into consideration to define inactive carrier state [

      Jang JW, Yoo SH, Ho BS, et al. Distribution patterns of serum hepatitis B surface antigen levels over the natural course of chronic hepatitis B: the role of age and immune phase. J Hepatol 2011;54:Abstract 373.

      ].

      Peginterferon treatment

      HBeAg-positive patients

      Current evidence

      Serum HBsAg levels show some correlation with intrahepatic cccDNA levels in patients with HBeAg-positive chronic hepatitis B, and decline in HBsAg may represent decline in cccDNA [
      • Thompson A.J.
      • Nguyen T.
      • Iser D.
      • et al.
      Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
      ,
      • Chan H.L.
      • Wong V.W.
      • Wong G.L.
      • Tse C.H.
      • Chan H.Y.
      • Sung J.J.
      A longitudinal study on the natural history of serum HBsAg changes in chronic hepatitis B.
      ,
      • Wursthorn K.
      • Lutgehetmann M.
      • Dandri M.
      • et al.
      Peginterferon alpha-2b plus adefovir induce strong cccDNA decline and HBsAg reduction in patients with chronic hepatitis B.
      ]. The degree of decline in HBsAg level during peginterferon treatment may reflect the efficacy of peginterferon and may provide useful information in prediction of treatment response. In an early study by Janssen et al., serum HBsAg level has been shown to decrease dramatically among HBeAg-positive patients who responded to interferon therapy [
      • Wong V.W.
      • Wong G.L.
      • Yan K.K.
      • et al.
      Durability of Peginterferon alfa-2a treatment at 5 years in patients with hepatitis B e antigen-positive chronic hepatitis B.
      ]. HBsAg level at the end of peginterferon treatment has also been shown to be significantly lower in patients with sustained response (defined as HBeAg seroconversion and HBV DNA <2000 IU/ml) 5 years post-treatment [
      • Janssen H.L.A.
      • Kerhof-Los C.J.
      • Heijtink R.A.
      • Schalm S.W.
      Measurement of HBsAg to monitor hepatitis B viral replication in patients on α-interferon therapy.
      ].
      Current data indicate that lower baseline serum HBsAg levels are present in patients achieving sustained response post-peginterferon treatment compared to non-responders [
      • Chan H.L.
      • Wong V.W.
      • Chim A.M.
      • et al.
      Serum HBsAg quantification to predict response to peginterferon therapy of e antigen positive chronic hepatitis B.
      ,
      • Tangkijvanich P.
      • Komolmit P.
      • Mahachai V.
      • et al.
      Low pretreatment serum HBsAg level and viral mutations as predictors of response to Peginterferon alfa-2b therapy in chronic hepatitis B.
      ]. Early serological response, defined as low HBsAg level or greater HBsAg decline early during treatment, has been associated with higher HBeAg seroconversion and HBV DNA suppression 6 months post-treatment [
      • Lau G.
      • Marcellin P.
      • Brunetto M.
      On treatment monitoring of HBsAg levels to predict response to peginterferon alfa-2a in patients with HBeAg-positive chronic hepatitis B.
      ]. Chan et al. reported that sustained response (defined as HBeAg seroconversion and HBV DNA ⩽2000 IU/ml at the end of peginterferon and/or lamivudine treatment, and that was sustained at 12 months post-treatment) was higher in patients who had HBsAg ⩽300 IU/ml at month 6 during treatment (62% vs. 11% in those who had HBsAg >300 IU/ml at month 6, p<0.001) [
      • Chan H.L.
      • Wong V.W.
      • Chim A.M.
      • et al.
      Serum HBsAg quantification to predict response to peginterferon therapy of e antigen positive chronic hepatitis B.
      ]. Patients having both HBsAg decline by greater than 1 log and serum HBsAg level ⩽300 IU/ml at month 6 during treatment achieved a sustained response rate of 75% compared with 15% in those who did not have this combined response (p <0.001). The PPV and NPV for achieving sustained response of this combined HBsAg response were 75% and 85%, respectively.
      In the phase III registration trial on peginterferon alfa-2a treatment in HBeAg-positive patients, HBsAg level <1500 IU/ml at week 12 and week 24 during treatment generated PPVs of 57% and 54% and NPVs of 72% and 76%, respectively, for HBeAg seroconversion 6 months post-treatment [
      • Lau G.
      • Marcellin P.
      • Brunetto M.
      On treatment monitoring of HBsAg levels to predict response to peginterferon alfa-2a in patients with HBeAg-positive chronic hepatitis B.
      ]. This study also demonstrated an HBsAg clearance rate of 17.6% at 6 months post- treatment in patients with HBsAg level <1500 IU/ml at week 12 and HBeAg seroconversion 6 months post-peginterferon treatment. The association of HBsAg and sustained HBeAg seroconversion was confirmed by the results of the NEPTUNE study [

      Gane E, Jia J, Han K, et al. NEPTUNE study: on-treatment HBsAg level analysis confirms prediction of response observed in phase 3 study of peginterferon alfa-2a in HBeAg-positive patients. J Hepatol 2011;54:Abstract 69.

      ].
      Sonneveld et al. showed a more vigorous decline in HBsAg level starting at week 4 of peginterferon alfa-2b treatment in responders defined as HBeAg loss with HBV DNA <2000 IU/ml at 26 weeks post-treatment [
      • Sonneveld M.J.
      • Rijckborst V.
      • Boucher C.A.
      • Hansen B.E.
      • Janssen H.L.
      Prediction of sustained response to peginterferon alfa-2b for hepatitis B e antigen-positive chronic hepatitis B using on-treatment hepatitis B surface antigen decline.
      ]. Patients who did not have any decline in HBsAg at week 12 achieved response in only 3%. Therefore, the NPV of the absence of any decline in HBsAg at week 12 is 97% for prediction of response 6 months post-treatment. However, the retrospective analysis from the phase III trial of peginterferon alfa-2a did not confirm a high NPV as in the Sonneveld study; 18% of patients who did not have any HBsAg decline at week 12 achieved HBeAg loss and HBV DNA <2000 IU/ml at 6 months post-treatment (NPV 82%) [
      • Piratvisuth T.
      • Marcellin P.
      Further analysis is required to identify an early stopping rule for peginterferon therapy that is valid for all HBeAg-positive patients.
      ]. It is possible that differences in study population and HBV genotypes could explain why the high NPV in the Sonneveld study was not confirmed. The Sonneveld study included mainly Caucasian patients (72%) infected with genotype A and D HBV (35% and 39%, respectively), whilst 87% of patients in the peginterferon alfa-2a study were Asians mainly infected with genotype B and C HBV (23% and 71%, respectively). On the other hand, week 12 HBsAg > 20000 IU/ml has an NPV of 84% in the peginterferon alfa-2a phase III study and 100% in the NEPTUNE study for sustained HBeAg seroconversion [
      • Lau G.
      • Marcellin P.
      • Brunetto M.
      On treatment monitoring of HBsAg levels to predict response to peginterferon alfa-2a in patients with HBeAg-positive chronic hepatitis B.
      ,

      Gane E, Jia J, Han K, et al. NEPTUNE study: on-treatment HBsAg level analysis confirms prediction of response observed in phase 3 study of peginterferon alfa-2a in HBeAg-positive patients. J Hepatol 2011;54:Abstract 69.

      ].

      Suggestions

      The decline in serum HBsAg level at weeks 12 and 24 during peginterferon treatment can be used as a surrogate marker to predict sustained response in patients with HBeAg-positive CHB, and to identify patients who are unlikely to have sustained response despite continued treatment. In general, poor HBsAg response at week 12 can predict non-responders and good HBsAg response at week 24 can predict good responders to peginterferon therapy (Table 1).
      Table 1Proportion of HBeAg-positive patients with sustained virological response (SVR) as predicted by serum HBsAg at week 12 and 24 of treatment. In general, poor HBsAg response at week 12 can predict non-responders and good HBsAg response at week 24 can predict good responders to peginterferon therapy.
      NA = not available.

      Challenges

      Further studies to identify and validate optimal cutoff levels of HBsAg are needed. Furthermore, HBsAg levels and their declines during peginterferon treatment appear to differ across HBV genotypes, suggesting genotype-specific algorithms are required [
      • Brunetto M.R.
      • Moriconi F.
      • Bonino F.
      • et al.
      Hepatitis B virus surface antigen levels: a guide to sustained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
      ,
      • Brunetto M.R.
      • Bonino F.
      • Marcellin P.
      • et al.
      Kinetic of HBsAg decline in patients with HBeAg-negative chronic hepatitis B treated with peginterferon alfa-2a according to genotype and its association with sustained HBsAg clearance 4 years post treatment.
      ,
      • Brunetto M.R.
      • Cavallone D.
      • Moriconi F.
      • et al.
      Kinetics of HBsAg decline during and following treatment of CHB: early and rapid HBsAg decline during peginterferon-2a in predictive of HBsAg clearance.
      ]. More research on the management of patients with sub-optimal HBsAg response will be needed before making recommendations.

      HBeAg-negative patients

      Current evidence

      Response rates of HBeAg-negative patients to peginterferon are low, and, therefore, baseline predictors of response are important when considering peginterferon based treatment in these patients. Unfortunately, only one study has so far addressed this clinical problem [
      • Bonino F.
      • Marcellin P.
      • Lau G.K.
      • et al.
      Predicting response to peginterferon alpha-2a, lamivudine and the two combined for HBeAg-negative chronic hepatitis B.
      ]. One other report of preliminary data from a small cohort of patients treated with peginterferon alfa-2a and adefovir showed that lower baseline HBsAg levels were associated with a sustained virological response (HBV DNA <2000 IU/ml at 6 months post-treatment), but clinically useful cut-off values were not reported [

      Takkenberg B, Zaaijer HL, De Niet A, et al. Baseline HBsAg level and on-treatment HBsAg and HBV DNA decline predict sustained virological response in HBeAg-negative chronic hepatitis B patients treated with peginterferon alfa-2a (Pegasys) and Adefovir (Hepsera); an interim analysis. Hepatology 2009;50:Abstract 491.

      ], and other studies were unable to confirm these findings [
      • Rijckborst V.
      • Hansen B.E.
      • Cakaloglu Y.
      • et al.
      Early on-treatment prediction of response to peginterferon alfa-2a for HBeAg-negative chronic hepatitis B using HBsAg and HBV DNA levels.
      ,
      • Moucari R.
      • Mackiewicz V.
      • Lada O.
      • et al.
      Early serum HBsAg drop: a strong predictor of sustained virological response to pegylated interferon alfa-2a in HBeAg-negative patients.
      ].
      On-treatment monitoring of HBsAg kinetics in HBeAg-negative patients treated with peginterferon may add considerably to HBV DNA monitoring alone. This is especially relevant since many HBeAg-negative patients treated with peginterferon achieve undetectable HBV DNA levels during therapy, but relapse after discontinuation [
      • Rijckborst V.
      • Ter Borg M.J.
      • Cakaloglu Y.
      • et al.
      A randomized trial of peginterferon alpha-2a with or without ribavirin for HBeAg-negative chronic hepatitis B.
      ,

      Marcellin P, Bonino F, Lau GK, et al. Sustained response of hepatitis B e antigen-negative patients 3 years after treatment with peginterferon alpha-2a. Gastroenterology 2009;136:2169–79 [e1–4].

      ,
      • Marcellin P.
      • Lau G.K.
      • Bonino F.
      • et al.
      Peginterferon alfa-2a alone, lamivudine alone, and the two in combination in patients with HBeAg-negative chronic hepatitis B.
      ]. One relatively small study by Moucari et al. reported on 48 HBeAg-negative CHB patients treated with peginterferon alfa-2a for 48 weeks [
      • Moucari R.
      • Mackiewicz V.
      • Lada O.
      • et al.
      Early serum HBsAg drop: a strong predictor of sustained virological response to pegylated interferon alfa-2a in HBeAg-negative patients.
      ]. Similar to HBeAg-positive disease, patients who achieved an off-treatment sustained response (defined as undetectable HBV DNA 6 months after treatment discontinuation) experienced the most pronounced HBsAg declines during treatment. The authors also showed that patients who achieved a decline of more than 0.5 log IU/ml through the first 12 weeks of therapy had a very high probability of response (89%), compared to a probability of only 10% in those who failed to achieve such a decline. Similar findings were reported for a 1 log decline at week 24.
      In a retrospective analysis of HBsAg levels during treatment in 120 patients who participated in the registration trial for peginterferon alfa-2a, patients who failed to achieved a decline of at least 10% in serum HBsAg from baseline through 12 weeks of treatment (n = 67) had a lower probability of response (16%, vs. 47% in those with a decline >10%, p <0.01) [
      • Marcellin P.
      • Piratvisuth T.
      • Brunetto M.
      • et al.
      On-treatment decline in serum HBsAg levels predicts sustained immune control 1 year post-treatment, subsequent HBsAg clearance in HBeAg-negative hepatitis B virus-infected patients treated with peginterferon alfa-2a.
      ]. However, a considerable proportion of patients who did not achieve a 10% decline in HBsAg levels achieved HBsAg loss, limiting the clinical utility of this cut-off as a stopping rule for peginterferon based treatment in HBeAg-negative CHB (Table 2).
      Table 2Prediction of treatment response by HBsAg decline at different phases of peginterferon therapy for HBeAg-negative chronic hepatitis B in the phase III trial of peginterferon alfa-2a
      • Brunetto M.R.
      • Bonino F.
      • Marcellin P.
      • et al.
      Kinetic of HBsAg decline in patients with HBeAg-negative chronic hepatitis B treated with peginterferon alfa-2a according to genotype and its association with sustained HBsAg clearance 4 years post treatment.
      ,

      Takkenberg B, Zaaijer HL, De Niet A, et al. Baseline HBsAg level and on-treatment HBsAg and HBV DNA decline predict sustained virological response in HBeAg-negative chronic hepatitis B patients treated with peginterferon alfa-2a (Pegasys) and Adefovir (Hepsera); an interim analysis. Hepatology 2009;50:Abstract 491.

      .
      Another study among HBeAg-negative patients, predominantly infected with HBV genotype D and treated with peginterferon ± ribavirin, showed that monitoring of both HBV DNA and HBsAg levels during treatment may provide more optimal prediction of response than either marker alone [
      • Rijckborst V.
      • Hansen B.E.
      • Cakaloglu Y.
      • et al.
      Early on-treatment prediction of response to peginterferon alfa-2a for HBeAg-negative chronic hepatitis B using HBsAg and HBV DNA levels.
      ]. In this study, patients who failed to achieve both a decline in HBsAg levels as well as a decline of >2 log in HBV DNA did not respond (response defined as HBV DNA <10,000 copies/ml and normal ALT at 6 months post-treatment). Importantly, this possible stopping-rule was recently validated in other cohorts of HBeAg-negative genotype D patients treated for either 1 or 2 years with peginterferon alfa-2a, and its utility was confirmed [

      Rijckborst V, Hansen B, Ferenci P, et al. Early on-treatment HBsAg and HBV DNA levels identify HBeAg-negative patients not responding to 48 or 96 weeks of peginterferon alfa-2a therapy. Hepatology 2010;52:Abstract 479.

      ]. However, the stopping-rule has less discriminatory capabilities in HBeAg-negative patients infected with other HBV genotypes. A possible explanation for this discrepancy is the influence of HBV genotype on HBsAg kinetics during treatment, as recently shown in a small study [
      • Moucari R.
      • Martinot-Peignoux M.
      • Mackiewicz V.
      • et al.
      Influence of genotype on hepatitis B surface antigen kinetics in hepatitis B e antigen-negative patients treated with pegylated interferon-alpha2a.
      ], suggesting a need for genotype specific prediction rules.
      In addition to on-treatment prediction of response, one study has investigated the use of end-of-treatment HBsAg levels as predictors of sustained off-treatment response and subsequent HBsAg clearance [
      • Brunetto M.R.
      • Moriconi F.
      • Bononi F.
      • et al.
      Hepatitis B virus surface antigen levels: a guide to sutained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
      ] (Table 2). In this study, patients with HBsAg levels <10 IU/ml (n = 23) at week 48 had a 52% probability of HBsAg clearance through 3 years of post-treatment follow-up, compared to only 2% in all other patients. Importantly, the predictive capabilities of HBsAg levels at week 48 were higher than that of HBV DNA levels; of those patients with undetectable HBV DNA at week 48 (n = 161), only 15% achieved HBsAg clearance. Overall, patients with a HBsAg level >19 IU/ml (or a decline from baseline <0.46 log) had such a low probability of a sustained response through 3 years of post-treatment follow-up that retreatment in these patients with nucleos(t)ide analogues (NA) seems warranted [
      • Marcellin P.
      • Piratvisuth T.
      • Brunetto M.
      • et al.
      On-treatment decline in serum HBsAg levels predicts sustained immune control 1 year post-treatment, subsequent HBsAg clearance in HBeAg-negative hepatitis B virus-infected patients treated with peginterferon alfa-2a.
      ].

      Suggestions

      Combination of HBsAg and HBV DNA decline at week 12 of peginterferon therapy could well be used as a stopping rule in clinical practice for HBeAg-negative patients, particularly among those infected by genotype D HBV (Fig. 3).
      Figure thumbnail gr3
      Fig. 3Flow chart on the use of HBV DNA and HBsAg to predict response to peginterferon treatment in the PARC study.

      Challenges

      The early identification of sustained response to peginterferon therapy (high positive predictive value) needs to be improved in HBeAg-negative disease. The use of HBV DNA and HBsAg to guide stopping of peginterferon in genotype B and C HBV infected patients needs to be validated in future studies.

      Treatment with nucleos(t)ide analogues

      Current evidence

      General trend of HBsAg decline on NA therapy

      Until recently only few studies have analyzed HBsAg levels during therapy with NA [
      • Kohmoto M.
      • Enomoto M.
      • Tamori A.
      • et al.
      Quantitative detection of hepatitis B surface antigen by chemiluminescent microparticle immunoassay during lamivudine treatment of chronic hepatitis B virus carriers.
      ,
      • Manesis E.K.
      • Hadziyannis E.S.
      • Angelopoulou O.S.
      • Hadziyannis S.J.
      Prediction of treatment-related HBsAg loss in HBeAg-negative chronic hepatitis B: a clue from serum HBsAg levels.
      ,
      • Wiegand J.
      • Wedemeyer H.
      • Finger A.
      • et al.
      A decline in hepatitis B virus surface antigen (HBsAg) predicts clearance, but does not correlate with quantitative HBeAg or HBV DNA levels.
      ,
      • Borgniet O.
      • Parvaz P.
      • Bouix C.
      • et al.
      Clearance of serum HBsAg and anti-HBs seroconversion following antiviral therapy for chronic hepatitis B.
      ,
      • Wursthorn K.
      • Jung M.
      • Riva A.
      • et al.
      Kinetics of hepatitis B surface antigen decline during 3 years of telbivudine treatment in hepatitis B e antigen-positive patients.
      ,
      • Cai W.
      • Xie Q.
      • An B.
      • et al.
      On-treatment serum HBsAg level is predictive of sustained off-treatment virologic response to telbivudine in HBeAg-positive chronic hepatitis B patients.
      ]. Most studies included heterogeneous patient populations (e.g. transplanted patients [
      • Wiegand J.
      • Wedemeyer H.
      • Finger A.
      • et al.
      A decline in hepatitis B virus surface antigen (HBsAg) predicts clearance, but does not correlate with quantitative HBeAg or HBV DNA levels.
      ] or focused only on patients with anti-HBs seroconversion or assessed only few time-points during NA treatment [
      • Brunetto M.R.
      • Moriconi F.
      • Bononi F.
      • et al.
      Hepatitis B virus surface antigen levels: a guide to sutained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
      ].
      Some studies have compared HBsAg kinetics between patients treated with interferon alfa and NA [
      • Brunetto M.R.
      • Moriconi F.
      • Bononi F.
      • et al.
      Hepatitis B virus surface antigen levels: a guide to sutained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
      ,
      • Manesis E.K.
      • Hadziyannis E.S.
      • Angelopoulou O.S.
      • Hadziyannis S.J.
      Prediction of treatment-related HBsAg loss in HBeAg-negative chronic hepatitis B: a clue from serum HBsAg levels.
      ,
      • Reijnders J.G.
      • Rijckborst V.
      • Sonneveld M.J.
      • et al.
      Kinetics of hepatitis B surface antigen differ between treatment with peginterferon and entecavir.
      ]. The data suggest that HBsAg decline during NA therapy is slower and less pronounced compared to interferon treatment, despite a higher effect on HBV DNA suppression. For example, Manesis et al. compared HBsAg decline during a median of 33 months of lamivudine treatment and compared it with that observed with interferon [
      • Manesis E.K.
      • Hadziyannis E.S.
      • Angelopoulou O.S.
      • Hadziyannis S.J.
      Prediction of treatment-related HBsAg loss in HBeAg-negative chronic hepatitis B: a clue from serum HBsAg levels.
      ]. Patients treated with interferon showed a stronger and faster HBsAg decline than lamivudine treated patients. Based on HBsAg kinetics, the authors estimated that the time to HBsAg loss was around 5 years for patients with sustained response to interferon compared with >10 years for lamivudine virological responders. Other studies suggested that the predicted median time to HBsAg loss in NA treated patients is more than 30 years [

      Zoutendijk R, Hansen BE, Van Vuuren AJ, Boucher CA, Janssen HL. Prediction of HBsAg loss using HBsAg decline after long-term virological response to nucleos(t)ide analogue therapy for chronic hepatitis B. Hepatology 2010;52:Abstract 381.

      ,

      Chevaliez S, Hezode C, Grare M, Pawlotsky JM. Long-term monitoring of HBsAg kinetics and prediction of HBsAg clearance in patients with chronic hepatitis B treated with nucleoside/nucleotide analogues. Hepatology 2010;52:Abstract 374.

      ]. Brunetto et al. analyzed 386 HBeAg-negative patients treated either with peginterferon alfa-2a and/or lamivudine. HBsAg decrease was stronger with peginterferon (alone or combined with lamivudine; mean decline at end of treatment, −0.71 and −0.67 log IU/ml, respectively) compared to treatment with lamivudine alone (−0.02 log IU/ml) [
      • Brunetto M.R.
      • Moriconi F.
      • Bononi F.
      • et al.
      Hepatitis B virus surface antigen levels: a guide to sutained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
      ]. The reason for the slow decline of HBsAg in NAs treatment is probably because NAs inhibit only the reverse transcription of the pregenomic RNA but do not target the cccDNA directly. Thus, changes at transcriptional levels, particularly in the HBsAg secretory pathway, are not expected. On the other hand, IFN has both direct antiviral and immune mediated effects. It is likely that the immune modulation by interferon leads to a more dramatic decline in HBsAg production and secretion.
      All studies investigating quantitative HBsAg during NA therapy suggest that HBsAg decline is slower than that of serum HBV DNA [
      • Moucari R.
      • Martinot-Peignoux M.
      • Mackiewicz V.
      • et al.
      Influence of genotype on hepatitis B surface antigen kinetics in hepatitis B e antigen-negative patients treated with pegylated interferon-alpha2a.
      ], even in patients who subsequently clear HBsAg [
      • Wiegand J.
      • Wedemeyer H.
      • Finger A.
      • et al.
      A decline in hepatitis B virus surface antigen (HBsAg) predicts clearance, but does not correlate with quantitative HBeAg or HBV DNA levels.
      ,
      • Borgniet O.
      • Parvaz P.
      • Bouix C.
      • et al.
      Clearance of serum HBsAg and anti-HBs seroconversion following antiviral therapy for chronic hepatitis B.
      ]. This may reflect the remaining cccDNA in the infected hepatocytes in patients with long-term established chronic HBV infection. This is different to the situation during acute HBV infection where HBV DNA and HBsAg decline in parallel [

      Jaroszewicz J, Ho H, Deterding K, et al. Prediction of HBsAg loss by quantitative HBsAg kinetics during long-term treatment with nucleos(t)ide analogues. Hepatology 2010;52:Abstract 395.

      ]. Borgniet et al. analyzed HBsAg kinetics in 16 patients who cleared HBsAg during antiviral therapy. In most patients HBsAg persisted after HBV DNA elimination from the serum for 1–5 years, with a progressive decline in titer observed. The average time to HBsAg loss after undetectable levels of HBV DNA were achieved was 30 months [
      • Borgniet O.
      • Parvaz P.
      • Bouix C.
      • et al.
      Clearance of serum HBsAg and anti-HBs seroconversion following antiviral therapy for chronic hepatitis B.
      ].

      HBeAg-positive patients

      A very important question is whether it would be possible to predict HBsAg loss using quantitative HBsAg values during NA treatment. Wursthorn et al. analyzed quantitative HBsAg in 162 HBeAg positive patients treated with telbivudine for at least 3 years [
      • Wursthorn K.
      • Jung M.
      • Riva A.
      • et al.
      Kinetics of hepatitis B surface antigen decline during 3 years of telbivudine treatment in hepatitis B e antigen-positive patients.
      ]. All patients included had a maintained HBV DNA <60 IU/ml after two years of therapy. Nine patients (6%) developed HBsAg loss through the follow-up of three years. A rapid HBsAg decline of more than 1 log after 1 year of treatment was predictive for HBsAg loss, emphasizing the value of quantitative HBsAg for monitoring response to NA therapy. Similarly, in the pivotal trial of tenofovir, patients who had HBsAg loss, tended to have higher baseline HBsAg level and a more rapid decline in HBsAg level as compared to those who failed to lose HBsAg on 4-year tenofovir treatment [
      • Heathcote E.J.
      • Marcellin P.
      • Buti M.
      • et al.
      Three-year efficacy and safety of tenofovir disoproxil fumarate treatment for chronic hepatitis B.
      ,

      Marcellin P, Heathcote EJ, Buti M, et al. HBsAg kinetics in patients with chronic hepatitis B (CHB) treated with tenofovir disoproxil fumarate (TDF) for up to 4 years. J Hepatol 2011;54:Abstract 740.

      ]. Genotype A and D HBV had higher baseline HBsAg and more continuous decline in HBsAg than genotype B and C HBV. In a small study in China, among 11 HBeAg positive patients who were treated with telbivudine for 2 years, HBsAg <100 IU/ml at the end of treatment predicted for sustained response (defined as undetectable HBV DNA, normal ALT and HBeAg seroconversion) for 2 years after stopping treatment [
      • Cai W.
      • Xie Q.
      • An B.
      • et al.
      On-treatment serum HBsAg level is predictive of sustained off-treatment virologic response to telbivudine in HBeAg-positive chronic hepatitis B patients.
      ].

      HBeAg-negative patients

      Among entecavir and tenofovir treated patients, the decline in HBsAg is less pronounced in HBeAg-negative patients than in HBeAg-positive patients [
      • Reijnders J.G.
      • Rijckborst V.
      • Sonneveld M.J.
      • et al.
      Kinetics of hepatitis B surface antigen differ between treatment with peginterferon and entecavir.
      ,
      • Heathcote E.J.
      • Marcellin P.
      • Buti M.
      • et al.
      Three-year efficacy and safety of tenofovir disoproxil fumarate treatment for chronic hepatitis B.
      ]. In a study in Hong Kong among 53 HBeAg-negative patients on lamivudine treatment for a mean of 19 months, the end-of-treatment HBsAg levels could predict sustained viral suppression (HBV DNA ⩽200 IU/ml) [

      Chan HL, Wong GL, Chim AM, Chan HY, Chu SH, Wong VW. Serum HBsAg quantification can predict sustained response to lamivudine in patients with negative HBeAg: a long-term post-treatment study. Hepatol Int 2011;5:Abstract PP05-102.

      ]. All five patients who had HBsAg reduction by >1 log to ⩽100 IU/ml (PPV 100%) and 4 of the 8 patients who achieved either HBsAg ⩽100 IU/ml or a reduction of HBsAg by >1 log (PPV 50%) had sustained viral suppression at 12 months post-lamivudine treatment. On the other hand, all 40 patients who had HBsAg reduction by ⩽1 log to a level >100 IU/ml could not achieve sustained viral suppression at 12 months post-treatment (NPV 100%). The end of treatment HBsAg response could also predict sustained viral suppression and HBsAg loss up to 5 years after stopping lamivudine.
      However, in another German study with predominantly HBeAg-negative patients (67%) on NA therapy, early on-treatment decline did not necessarily predict HBsAg loss [

      Jaroszewicz J, Ho H, Deterding K, et al. Prediction of HBsAg loss by quantitative HBsAg kinetics during long-term treatment with nucleos(t)ide analogues. Hepatology 2010;52:Abstract 395.

      ]. Some patients with early HBsAg decline during NA therapy might have just changed the phase of HBV infection (i.e. transition from immune clearance phase to HBeAg-negative hepatitis), which is associated with different HBsAg levels [
      • Kimbi G.C.
      • Kramvis A.
      • Kew M.C.
      Integration of hepatitis B virus DNA into chromosomal DNA during acute hepatitis B.
      ,
      • Bill C.A.
      • Summers J.
      Genomic DNA double-strand breaks are targets for hepadnaviral DNA integration.
      ]. Thus, it may represent an already ongoing natural HBsAg decline in some patients. In contrast to interferon therapy [
      • Sonneveld M.J.
      • Rijckborst V.
      • Boucher C.A.
      • Hansen B.E.
      • Janssen H.L.
      Prediction of sustained response to peginterferon alfa-2b for hepatitis B e antigen-positive chronic hepatitis B using on-treatment hepatitis B surface antigen decline.
      ,
      • Brunetto M.R.
      • Moriconi F.
      • Bonino F.
      • et al.
      Hepatitis B virus surface antigen levels: a guide to sustained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
      ,
      • Rijckborst V.
      • Hansen B.E.
      • Cakaloglu Y.
      • et al.
      Early on-treatment prediction of response to peginterferon alfa-2a for HBeAg-negative chronic hepatitis B using HBsAg and HBV DNA levels.
      ,
      • Moucari R.
      • Mackiewicz V.
      • Lada O.
      • et al.
      Early serum HBsAg drop: a strong predictor of sustained virological response to pegylated interferon alfa-2a in HBeAg-negative patients.
      ], early HBsAg kinetics alone during NA therapy may not reliably predict later HBsAg loss. In this German cohort, HBsAg decline of more than 0.5 log in the 2 years after virological response (defined as HBV DNA <100 IU/ml) was associated with later HBsAg loss [

      Jaroszewicz J, Ho H, Deterding K, et al. Prediction of HBsAg loss by quantitative HBsAg kinetics during long-term treatment with nucleos(t)ide analogues. Hepatology 2010;52:Abstract 395.

      ]. However, patients who received no treatment also demonstrated similar HBsAg kinetics during the natural course of HBV infection in this study. This suggests that NA therapy may have only a limited impact on HBsAg suppression, explaining that HBsAg loss remains a rare event during NA therapy. Thus, most patients treated with NA will require life-long maintenance therapy. Taking all currently available data into account, HBsAg loss during NA therapy may be limited to patients with more active immune responses also supported by data showing than NA induced HBsAg loss is associated with higher baseline ALT [

      Zoutendijk R, Hansen BE, Van Vuuren AJ, Boucher CA, Janssen HL. Prediction of HBsAg loss using HBsAg decline after long-term virological response to nucleos(t)ide analogue therapy for chronic hepatitis B. Hepatology 2010;52:Abstract 381.

      ,

      Jaroszewicz J, Ho H, Deterding K, et al. Prediction of HBsAg loss by quantitative HBsAg kinetics during long-term treatment with nucleos(t)ide analogues. Hepatology 2010;52:Abstract 395.

      ].

      Suggestions

      Overall, serum HBsAg decline is slow and does not correlate with HBV DNA levels during treatment with NAs. However, a rapid serum HBsAg decline during NA therapy, ideally after virological response has been achieved, may identify patients who will clear HBsAg in the long-term. A 6–12 monthly assessment of quantitative HBsAg level can be considered to monitor NA therapy. Among Asian patients (genotype B and C HBV infected patients), an HBsAg level of <100 IU/ml might predict lower risk of relapse and stopping treatment can be considered. This is based on limited evidence and further validation by larger scaled, multi-centered studies will be needed [
      • Cai W.
      • Xie Q.
      • An B.
      • et al.
      On-treatment serum HBsAg level is predictive of sustained off-treatment virologic response to telbivudine in HBeAg-positive chronic hepatitis B patients.
      ,

      Chan HL, Wong GL, Chim AM, Chan HY, Chu SH, Wong VW. Serum HBsAg quantification can predict sustained response to lamivudine in patients with negative HBeAg: a long-term post-treatment study. Hepatol Int 2011;5:Abstract PP05-102.

      ,

      Jiang NJ, Liang XY, Su HM, et al. Study of influence factors on HBV relapse after NAS withdrawal in the CHB patients who met NAS cessation criteria. J Hepatol 2011;54:Abstract 722.

      ].

      Challenges

      It will be important to identify a level of serum HBsAg reduction or a cut-off that is associated with subsequent HBsAg loss. It will also be of interest to identify an HBsAg level that is associated with effective immune control where termination of antiviral therapy is possible with low risk of reactivation. Whether this cut-off is in accordance with the one proposed for the identification of low replicative inactive HBsAg carrier remains to be determined [
      • Brunetto M.R.
      • Oliveri F.
      • Colombatto P.
      • et al.
      Hepatitis B surface antigen serum levels help to distinguish active from inactive hepatitis B virus genotype D carriers.
      ,
      • Martinot-Peignoux M.
      • Lada O.
      • Cardoso A.C.
      • et al.
      Quantitative HBsAg: A new specific marker for the diagnosis of HBsAg inactive carriage.
      ,

      Manesis EK, Papatheodoridis GV, Hadziyannis E. Significance of serum HBsAg levels for the definition of the inactive hepatitis B carrier. Hepatology 2010;52:Abstract 483.

      ]. It is also unknown if viral breakthrough due to drug resistance affects quantitative HBsAg level as mutations in the HBV polymerase can also involve changes in the HBsAg [
      • Kohmoto M.
      • Enomoto M.
      • Tamori A.
      • et al.
      Quantitative detection of hepatitis B surface antigen by chemiluminescent microparticle immunoassay during lamivudine treatment of chronic hepatitis B virus carriers.
      ]. As certain mutants at the surface region, which overlaps with the polymerase region, might influence the amount of HBsAg level detected by the Architect (Abbott) and Elecsys (Roche) assays, it appears advisable to use the same assay for follow-up of individual patients on treatment [

      Verheyen J, Neumann-Fraune M, Berg T, Kaiser R, Obermeier M. Mutations in the HBs-antigen influence the results of HBsAg quantification assays. J Hepatol 2011;54:Abstract 394.

      ,

      Bruce M, Horner M, Shang D, et al. Mutation RTA181T/SW172 is associated with lower HBsAg levels than other mutations in HBV polymerase codon 181 an overlapping surface codon 172 – implications for vaccination? J Hepatol 2011:54:Abstract 706.

      ].

      Unanswered questions and perspectives

      In the recent years, increasing evidence shows that in chronic HBV infection, serum HBsAg levels are indirectly correlated with the control of the infection: the higher the control, the lower serum HBsAg levels. Accordingly, serum HBsAg levels are lower in inactive carriers than in HBeAg-negative CHB patients and decline significantly during effective antiviral treatment. The information provided by serum HBsAg quantification is different but complementary to that of serum HBV DNA. Hence, HBsAg quantification should not be used as a substitute for HBV DNA measures in clinical practice. In particular, among patients on NA therapy, HBV DNA is used to monitor response to treatment and emergence of drug resistance, while HBsAg level may have a role in identifying HBeAg-negative patients suitable for one treatment cessation.
      However, until now, a precise biological understanding of these clinical findings is missing: an overall correlation has been suggested between serum HBsAg levels and the intrahepatic amount of cccDNA, but this may not hold true in all the phases of HBV infection, particularly in the HBeAg-negative phase. We may speculate that HBsAg is the indirect expression of cccDNA amounts and transcriptionally active cccDNA, but we need to better understand the complex equilibrium between intracellular HBV biology and the host’s immune system and its relations with transcription of HBsAg specific mRNAs, HBsAg production and secretion. Furthermore, we need to know whether HBV heterogeneity in terms of genotypes and “S” gene variability, either naturally occurring or therapy induced, influences serum HBsAg levels in clinically relevant terms. This information is needed to identify correctly HBsAg thresholds to be used in clinical practice.
      Substantial variations of serum HBsAg levels in the different phases of HBV infection propose quantitative HBsAg as a new diagnostic tool for the characterization of HBV carriers. However, in spite of the different cut-offs so far identified, larger multicenter studies have to be performed for their appropriate clinical validation. In addition, further work is required to understand whether monitoring serum HBsAg levels at the time of HBeAg to anti-HBe seroconversion and their delta variations over time might help to distinguish patients with durable seroconversion from those with higher risk of HBeAg seroreversion. Accordingly, HBsAg monitoring could provide a useful tool for tailoring the duration of antiviral treatment in HBeAg positive patients. It also remains to be analyzed if HBsAg quantification might serve as an indicator of viral breakthrough in patients who have become HBV DNA negative in serum but remain HBsAg positive. In addition to the issue of successful immune clearance, in HBeAg positive patients, data are missing on the relationship between serum HBsAg level and the severity of histologic grading and staging. Overall, to improve the diagnostic accuracy in the characterization of HBV carriers, more studies must be performed on the combined use of quantitative HBV DNA and HBsAg as their complementary information on the phases of HBV infection may be very useful in clinical practice to define the specific condition of the single HBV carrier during the highly dynamic phases of chronic HBV infection. Finally it will be important to identify levels of HBsAg reduction or cut-offs that are associated with the effective immune control that brings to subsequent HBsAg loss, therefore, allowing termination of NA therapy with the lowest risk of reactivation. Whether this cut-off is in accordance with the one proposed for the identification of low replicative inactive HBsAg carriers or peginterferon treated patients remains to be determined.

      Conflict of interest

      H.L.Y.C. is a consultant and advisor of Abbott, Bristol Myers Squibb, F. Hoffmann-la Roche, Merck and Novartis Pharmaceutical. A.T. has served as an advisor to Merck, Roche, Gilead and Janssen-Cilag and has received speaker’s fees from Merck and Roche. T.P. served on the advisory board and is on the speaker bureaus for Roche, Schering-Plough Corporation, Novartis, GlaxoSmithKline and Bristol-Myers Squibb. M.C. received lecture fees and/or consult fees from GlaxoSmithKline, Gilead, Novartis, Bristol-Myers-Squibb, Roche, and Merck. M.R.B. is speaker bureau/advisor of Abbott, BMS, Gilead, Hoffmann-la Roche and Merck. H.L.T. received travel support, lecturer fees, grants and consultant fees by Novartis, Roche and/or Abbott, and his wife is an employee of Abbott and holds also stocks in Abbott. J.H.K. is a consultant and advisor of Abbott, Bristol Myers Squibb, GlaxoSmithKline, Merck, Novartis, Omrix, and F Hoffmann-la Roche. J.D.J. is an advisory board member of Bristol Myers Squibb, Glaxo-Smith-Kline, Novartis Pharmaceutical, Merck, and Roche China. H.W. received lecturer/consultant fees, and grant support from Abbott, BMS, Gilead, Merck, Novartis, and Roche. H.L.A.J. received grants from and is consultant for Bristol Myers Squibb, Gilead Sciences, Novartis, Roche, and Merck. P.M. is a consultant for, and is on the speakers’ bureau of Roche, Schering-Plough, Gilead, Bristol-Myers Squibb, GlaxoSmithKline, and Idenix-Novartis; a consultant for and advises Vertex, Valeant, Human Genome Sciences, Cythesis, Intermune, Wyeth, and Tibotec; and an advisor for Coley Pharma. M.M.-P. and S.L. reported no potential conflict of interest.

      Financial support

      F. Hoffmann-la Roche has provided unrestricted financial support for the meeting of the group but did not advise nor interfere with the content of this paper.

      Appendix. Good Practice in using sAg in Chronic Hepatitis B Study Group (GPs-CHB Study Group)

      Henry Lik-Yuen Chan (The Chinese University of Hong Kong, Hong Kong), Alex Thompson and Steven Locarnini (Victorian Infectious Diseases Research Laboratory, Australia), Michelle Martinot-Peignoux and Patrick Marcellin (INSERM, France), Teerha Piratvisuth (Prince of Songkla University, Thailand), Markus Cornberg and Heiner Wedemeyer (Hanover Medical School, Germany), Maurizia Rossana Brunetto (University of Pisa, Italy), Hans L. Tillmann (Duke University, USA), Jia-Horng Kao (National University of Taiwan, Taiwan), Ji-Dong Jia (Capital Medical University, China), Yun-Fan Liaw (Chang Gung Memorial Hospital and University, Taiwan), Harry L.A. Janssen (Erasmus MC University Hospital, Netherlands).

      References

        • Blumberg B.S.
        • Sutnick A.I.
        • London W.T.
        Hepatitis and leukemia: their relation to Australia antigen.
        Bull NY Acad Med. 1968; 44: 1566-1586
        • Frosner G.G.
        • Schomerus H.
        • Wiedmann K.H.
        • et al.
        Diagnostic significance of quantitative determination of hepatitis B surface antigen in acute and chronic hepatitis B infection.
        Eur J Clin Microbiol. 1982; 1: 52-58
        • Deguchi M.
        • Yamashita N.
        • Kagita M.
        • et al.
        Quantitation of hepatitis B surface antigen by an automated chemiluminescent microparticle immunoassay.
        J Virol Methods. 2004; 115: 217-222
        • Bonino F.
        • Brunetto M.R.
        • Colombatto P.
        • Moriconi F.
        • Torresani E.
        • Lunghi G.
        • et al.
        Use of the Elecysys HBsAg II assay for simple and accurate quantification of HBsAg levels in sera of patients infected with HBV.
        Hepatol Int. 2009; 3: 108
        • Werle-Lapostolle B.
        • Bowden S.
        • Locarnini S.
        • et al.
        Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy.
        Gastroenterology. 2004; 126: 1750-1758
        • Chan H.L.
        • Wong V.W.
        • Tse A.M.
        • et al.
        Serum hepatitis B surface antigen quantitation can reflect hepatitis B virus in the liver and predict treatment response.
        Clin Gastroenterol Hepatol. 2007; 5: 1462-1468
        • Kimbi G.C.
        • Kramvis A.
        • Kew M.C.
        Integration of hepatitis B virus DNA into chromosomal DNA during acute hepatitis B.
        World J Gastroenterol. 2005; 11: 6416-6421
        • Bill C.A.
        • Summers J.
        Genomic DNA double-strand breaks are targets for hepadnaviral DNA integration.
        Proc Natl Acad Sci USA. 2004; 101: 11135-11140
        • Thomas H.C.
        • Lemon S.M.
        Viral hepatitis.
        3rd ed. Blackwell Publishing, Massachusetts, USA2005
        • Nguyen T.
        • Thompson A.J.
        • Bowden S.
        • et al.
        Hepatitis B surface antigen levels during the natural history of chronic hepatitis B: a perspective on Asia.
        J Hepatol. 2010; 52: 508-513
        • Jaroszewicz J.
        • Calle Serrano B.
        • Wursthorn K.
        • et al.
        Hepatitis B surface antigen (HBsAg) levels in the natural history of hepatitis B virus (HBV)-infection: a European perspective.
        J Hepatol. 2010; 52: 514-522
        • Thompson A.J.
        • Nguyen T.
        • Iser D.
        • et al.
        Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
        Hepatology. 2010; 51: 1933-1944
        • Lin L.Y.
        • Wong V.W.
        • Zhou H.J.
        • et al.
        Relationship between serum hepatitis B virus DNA and surface antigen with covalently closed circular DNA in HBeAg-negative patients.
        J Med Virol. 2010; 82: 1494-1500
        • Volz T.
        • Lutgehetmann M.
        • Wachtler P.
        • et al.
        Impaired intrahepatic hepatitis B virus productivity contributes to low viremia in most HBeAg-negative patients.
        Gastroenterology. 2007; 133: 843-852
        • Warner N.
        • Locarnini S.
        The antiviral drug selected hepatitis B virus rtA181T/sW172 mutant has a dominant negative secretion defect and alters the typical profile of viral rebound.
        Hepatology. 2008; 48: 88-98
        • Warner N.
        • Locarnini S.
        Can antiviral therapy for chronic hepatitis B enhance the progression to hepatocellular carcinoma?.
        Antivir Ther. 2009; 14: 139-142
        • Bonino F.
        • Moriconi F.
        • Bowden S.
        • et al.
        Multicenter evaluation of the Elecsys HBsAg II Quant Assay.
        Hepatol Int. 2011; 5: 80
      1. Sonneveld MJ, Rijckborst V, Boucher CAB, et al. A comparison of two assays for quantification of hepatitis B surface antigen in serum of patients with chronic hepatitis B. J Hepatol 2011;54:Abstract 750.

        • Wursthorn K.
        • Jaroszewicz J.
        • Zacher B.J.
        • et al.
        Correlation between the Elecsys HBsAg II assay and the Architect assay for the quantification of hepatitis B surface antigen (HBsAg) in the serum.
        J Clin Virol. 2011; 50: 292-296
        • Chan H.L.
        • Wong G.L.
        • Wong V.W.
        A review of the natural history of chronic hepatitis B in the era of transient elastography.
        Antivir Ther. 2009; 14: 489-499
        • Sugiyama M.
        • Tasuhito T.
        • Kato T.
        • et al.
        Influence of hepatitis B virus genotypes on the intra- and extracellular expression of viral DNA and antigens.
        Hepatology. 2006; 44: 915-924
        • Chan H.L.
        • Wong V.W.
        • Wong G.L.
        • Tse C.H.
        • Chan H.Y.
        • Sung J.J.
        A longitudinal study on the natural history of serum HBsAg changes in chronic hepatitis B.
        Hepatology. 2010; 52: 1232-1241
      2. Cheng PN, Tsai HW, Chang TT. Serum hepatitis B surface antigen level is associated with liver fibrosis in patients with HBeAg-positive chronic hepatitis B. J Hepatol 2011;54:Abstract 362.

        • Wong G.L.
        • Wong V.W.
        • Choi P.C.
        • et al.
        Clinical factors associated with liver stiffness in hepatitis B e antigen-positive chronic hepatitis B.
        Clin Gastroenterol Hepatol. 2009; 7: 227-233
        • Gaeta G.B.
        • Stornaiuolo G.
        • Precone D.F.
        • et al.
        Epidemiological and clinical burden of chronic hepatitis B virus/hepatitis C virus infection. A multicenter Italian study.
        J Hepatol. 2003; 39: 1036-1041
        • Hadziyannis S.
        • Papatheodoridis G.V.
        Hepatitis B e antigen-negative chronic hepatitis: natural history and treatment.
        Semin Liver Dis. 2006; 26: 130-141
        • Zarski J.P.
        • Marcellin P.
        • Leroy V.
        • et al.
        Characteristics of patients with chronic hepatitis B in France: predominant frequency of HBe antigen negative cases.
        J Hepatol. 2006; 45: 355-360
        • Brunetto M.R.
        • Oiveri F.
        • Coco B.
        • et al.
        Outcome of anti-HBe positive chronic hepatitis B in alpha-interferon and untreated patients: a long term cohort study.
        J Hepatol. 2002; 36: 263-270
        • Sung J.J.
        • Chan H.L.
        • Wong M.L.
        • et al.
        Relationship of clinical and virological factors with hepatitis activity in hepatitis B e antigen-negative chronic hepatitis B virus-infected patients.
        J Viral Hepat. 2002; 9: 229-234
        • Martinot-Peignoux M.
        • Boyer N.
        • Colombat M.
        • et al.
        Serum hepatitis B virus HBV DNA levels and liver histology in inactive HBsAg carriers.
        J Hep. 2002; 36: 543-546
        • Brunetto M.R.
        • Oliveri F.
        • Rocca G.
        • et al.
        Natural course and response to interferon of chronic hepatitis B accompanied by antibody to hepatitis B e antigen.
        Hepatology. 1989; 10: 198-202
        • Papatheodoridis G.V.
        • Manesis E.K.
        • Manalakopoulos S.
        • et al.
        Is there a meaninful serum hepatitis B virus DNA cutoff level for therapeutic decisions in hepatitis B e antigen-negative chronic hepatitis B infection?.
        Hepatology. 2008; 48: 1451-1459
        • McMahon B.J.
        The natural history of chronic hepatitis B virus infection.
        Hepatology. 2009; 49: S45-S55
        • Lok A.S.
        • Heathcote E.J.
        • Hoofnagle J.H.
        Management of hepatitis B: 2000 – summary of a workshop.
        Gastroenterology. 2001; 120: 1828-1853
      3. EASL clinical practice guidelines: management of chronic hepatitis B. European Association for the Study of the Liver. J Hepatol 2009;50:227–242.

        • Chu C.J.
        • Hussain M.
        • Lik A.S.F.
        Quantitative serum HBV DNA levels during different stages of chronic hepatitis B infection.
        Hepatology. 2002; 36: 1408-1415
        • Seo Y.
        • Yoon S.
        • Truong B.X.
        • et al.
        Serum HBV DNA levels differentiating inactive carriers from patients with chronic hepatitis B.
        Eur J Gastroenterol Hepatol. 2005; 17: 753-757
        • Feld J.J.
        • Ayers M.
        • El-Ashry D.
        • et al.
        Hepatitis B virus DNA prediction rules for Hepatitis B e antigen-negative chronic hepatitis B.
        Hepatology. 2007; 46: 1070-1507
        • Chan H.L.
        • Wong G.L.
        • Tse C.H.
        • Chan H.Y.
        • Wong V.W.
        Viral determinants of hepatitis B surface antigen seroclearance in hepatitis B e antigen-negative chronic hepatitis B patients.
        J Infect Dis. 2011; 204: 408-414
        • Su T.H.
        • Hsu C.S.
        • Chen C.L.
        • et al.
        Serum hepatitis B surface antigen concentration correlates with HBV DNA level in patients with chronic hepatitis B.
        Antivir Ther. 2010; 15: 1133-1139
      4. Tseng TC, Liu CJ, Su TH, et al. Serum Hepatitis B surface antigen levels predict surface antigen loss in hepatitis B e antigen seroconverters. Gastroenterology, 2011[Epub ahead of print].

        • Brunetto M.R.
        • Oliveri F.
        • Colombatto P.
        • et al.
        Hepatitis B surface antigen serum levels help to distinguish active from inactive hepatitis B virus genotype D carriers.
        Gastroenterology. 2010; 139: 483-490
        • Martinot-Peignoux M.
        • Lada O.
        • Cardoso A.C.
        • et al.
        Quantitative HBsAg: A new specific marker for the diagnosis of HBsAg inactive carriage.
        Hepatology. 2010; 52: 992A
      5. Yakut M, Bektas M, Seven G, et al. Characterization of the inactive HBsAg carrier state with 3 year follow-up. J Hepatol 2011;54:Abstract 398.

      6. Jang JW, Yoo SH, Ho BS, et al. Distribution patterns of serum hepatitis B surface antigen levels over the natural course of chronic hepatitis B: the role of age and immune phase. J Hepatol 2011;54:Abstract 373.

        • Wursthorn K.
        • Lutgehetmann M.
        • Dandri M.
        • et al.
        Peginterferon alpha-2b plus adefovir induce strong cccDNA decline and HBsAg reduction in patients with chronic hepatitis B.
        Hepatology. 2006; 44: 675-684
        • Wong V.W.
        • Wong G.L.
        • Yan K.K.
        • et al.
        Durability of Peginterferon alfa-2a treatment at 5 years in patients with hepatitis B e antigen-positive chronic hepatitis B.
        Hepatology. 2010; 51: 1945-1953
        • Janssen H.L.A.
        • Kerhof-Los C.J.
        • Heijtink R.A.
        • Schalm S.W.
        Measurement of HBsAg to monitor hepatitis B viral replication in patients on α-interferon therapy.
        Antivir Res. 1994; 23: 251-257
        • Chan H.L.
        • Wong V.W.
        • Chim A.M.
        • et al.
        Serum HBsAg quantification to predict response to peginterferon therapy of e antigen positive chronic hepatitis B.
        Aliment Pharmacol Ther. 2010; 32: 1323-1325
        • Tangkijvanich P.
        • Komolmit P.
        • Mahachai V.
        • et al.
        Low pretreatment serum HBsAg level and viral mutations as predictors of response to Peginterferon alfa-2b therapy in chronic hepatitis B.
        J Clin Virol. 2009; 46: 117-123
        • Lau G.
        • Marcellin P.
        • Brunetto M.
        On treatment monitoring of HBsAg levels to predict response to peginterferon alfa-2a in patients with HBeAg-positive chronic hepatitis B.
        J Hepatol. 2009; 50: S333
      7. Gane E, Jia J, Han K, et al. NEPTUNE study: on-treatment HBsAg level analysis confirms prediction of response observed in phase 3 study of peginterferon alfa-2a in HBeAg-positive patients. J Hepatol 2011;54:Abstract 69.

        • Sonneveld M.J.
        • Rijckborst V.
        • Boucher C.A.
        • Hansen B.E.
        • Janssen H.L.
        Prediction of sustained response to peginterferon alfa-2b for hepatitis B e antigen-positive chronic hepatitis B using on-treatment hepatitis B surface antigen decline.
        Hepatology. 2010; 52: 1251-1257
        • Piratvisuth T.
        • Marcellin P.
        Further analysis is required to identify an early stopping rule for peginterferon therapy that is valid for all HBeAg-positive patients.
        Hepatology. 2011; 53: 1054-1055
        • Brunetto M.R.
        • Moriconi F.
        • Bonino F.
        • et al.
        Hepatitis B virus surface antigen levels: a guide to sustained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
        Hepatology. 2009; 49: 1141-1150
        • Brunetto M.R.
        • Bonino F.
        • Marcellin P.
        • et al.
        Kinetic of HBsAg decline in patients with HBeAg-negative chronic hepatitis B treated with peginterferon alfa-2a according to genotype and its association with sustained HBsAg clearance 4 years post treatment.
        Hepatology. 2008; 48 ([Abstract]): 965A
        • Brunetto M.R.
        • Cavallone D.
        • Moriconi F.
        • et al.
        Kinetics of HBsAg decline during and following treatment of CHB: early and rapid HBsAg decline during peginterferon-2a in predictive of HBsAg clearance.
        Hepatology. 2008; 48 ([Abstract]): 717A
        • Bonino F.
        • Marcellin P.
        • Lau G.K.
        • et al.
        Predicting response to peginterferon alpha-2a, lamivudine and the two combined for HBeAg-negative chronic hepatitis B.
        Gut. 2007; 56: 699-705
      8. Takkenberg B, Zaaijer HL, De Niet A, et al. Baseline HBsAg level and on-treatment HBsAg and HBV DNA decline predict sustained virological response in HBeAg-negative chronic hepatitis B patients treated with peginterferon alfa-2a (Pegasys) and Adefovir (Hepsera); an interim analysis. Hepatology 2009;50:Abstract 491.

        • Rijckborst V.
        • Hansen B.E.
        • Cakaloglu Y.
        • et al.
        Early on-treatment prediction of response to peginterferon alfa-2a for HBeAg-negative chronic hepatitis B using HBsAg and HBV DNA levels.
        Hepatology. 2010; 52: 454-461
        • Moucari R.
        • Mackiewicz V.
        • Lada O.
        • et al.
        Early serum HBsAg drop: a strong predictor of sustained virological response to pegylated interferon alfa-2a in HBeAg-negative patients.
        Hepatology. 2009; 49: 1151-1157
        • Rijckborst V.
        • Ter Borg M.J.
        • Cakaloglu Y.
        • et al.
        A randomized trial of peginterferon alpha-2a with or without ribavirin for HBeAg-negative chronic hepatitis B.
        Am J Gastroenterol. 2010; 105: 1762-1769
      9. Marcellin P, Bonino F, Lau GK, et al. Sustained response of hepatitis B e antigen-negative patients 3 years after treatment with peginterferon alpha-2a. Gastroenterology 2009;136:2169–79 [e1–4].

        • Marcellin P.
        • Lau G.K.
        • Bonino F.
        • et al.
        Peginterferon alfa-2a alone, lamivudine alone, and the two in combination in patients with HBeAg-negative chronic hepatitis B.
        N Engl J Med. 2004; 351: 1206-1217
        • Marcellin P.
        • Piratvisuth T.
        • Brunetto M.
        • et al.
        On-treatment decline in serum HBsAg levels predicts sustained immune control 1 year post-treatment, subsequent HBsAg clearance in HBeAg-negative hepatitis B virus-infected patients treated with peginterferon alfa-2a.
        Hepatol Int. 2010; 4: 151
      10. Rijckborst V, Hansen B, Ferenci P, et al. Early on-treatment HBsAg and HBV DNA levels identify HBeAg-negative patients not responding to 48 or 96 weeks of peginterferon alfa-2a therapy. Hepatology 2010;52:Abstract 479.

        • Moucari R.
        • Martinot-Peignoux M.
        • Mackiewicz V.
        • et al.
        Influence of genotype on hepatitis B surface antigen kinetics in hepatitis B e antigen-negative patients treated with pegylated interferon-alpha2a.
        Antivir Ther. 2009; 14: 1183-1188
        • Brunetto M.R.
        • Moriconi F.
        • Bononi F.
        • et al.
        Hepatitis B virus surface antigen levels: a guide to sutained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B.
        Hepatology. 2009; 49: 1141-1150
        • Kohmoto M.
        • Enomoto M.
        • Tamori A.
        • et al.
        Quantitative detection of hepatitis B surface antigen by chemiluminescent microparticle immunoassay during lamivudine treatment of chronic hepatitis B virus carriers.
        J Med Virol. 2005; 75: 235-239
        • Manesis E.K.
        • Hadziyannis E.S.
        • Angelopoulou O.S.
        • Hadziyannis S.J.
        Prediction of treatment-related HBsAg loss in HBeAg-negative chronic hepatitis B: a clue from serum HBsAg levels.
        Antivir Ther. 2007; 12: 73-82
        • Wiegand J.
        • Wedemeyer H.
        • Finger A.
        • et al.
        A decline in hepatitis B virus surface antigen (HBsAg) predicts clearance, but does not correlate with quantitative HBeAg or HBV DNA levels.
        Antivir Ther. 2008; 13: 547-554
        • Borgniet O.
        • Parvaz P.
        • Bouix C.
        • et al.
        Clearance of serum HBsAg and anti-HBs seroconversion following antiviral therapy for chronic hepatitis B.
        J Med Virol. 2009; 81: 1336-1342
        • Wursthorn K.
        • Jung M.
        • Riva A.
        • et al.
        Kinetics of hepatitis B surface antigen decline during 3 years of telbivudine treatment in hepatitis B e antigen-positive patients.
        Hepatology. 2010; 52: 1611-1620
        • Cai W.
        • Xie Q.
        • An B.
        • et al.
        On-treatment serum HBsAg level is predictive of sustained off-treatment virologic response to telbivudine in HBeAg-positive chronic hepatitis B patients.
        J Clin Virol. 2010; 48: 22-26
        • Reijnders J.G.
        • Rijckborst V.
        • Sonneveld M.J.
        • et al.
        Kinetics of hepatitis B surface antigen differ between treatment with peginterferon and entecavir.
        J Hepatol. 2011; 54: 449-454
      11. Zoutendijk R, Hansen BE, Van Vuuren AJ, Boucher CA, Janssen HL. Prediction of HBsAg loss using HBsAg decline after long-term virological response to nucleos(t)ide analogue therapy for chronic hepatitis B. Hepatology 2010;52:Abstract 381.

      12. Chevaliez S, Hezode C, Grare M, Pawlotsky JM. Long-term monitoring of HBsAg kinetics and prediction of HBsAg clearance in patients with chronic hepatitis B treated with nucleoside/nucleotide analogues. Hepatology 2010;52:Abstract 374.

      13. Jaroszewicz J, Ho H, Deterding K, et al. Prediction of HBsAg loss by quantitative HBsAg kinetics during long-term treatment with nucleos(t)ide analogues. Hepatology 2010;52:Abstract 395.

        • Heathcote E.J.
        • Marcellin P.
        • Buti M.
        • et al.
        Three-year efficacy and safety of tenofovir disoproxil fumarate treatment for chronic hepatitis B.
        Gastroenterology. 2011; 140: 132-143
      14. Marcellin P, Heathcote EJ, Buti M, et al. HBsAg kinetics in patients with chronic hepatitis B (CHB) treated with tenofovir disoproxil fumarate (TDF) for up to 4 years. J Hepatol 2011;54:Abstract 740.

      15. Chan HL, Wong GL, Chim AM, Chan HY, Chu SH, Wong VW. Serum HBsAg quantification can predict sustained response to lamivudine in patients with negative HBeAg: a long-term post-treatment study. Hepatol Int 2011;5:Abstract PP05-102.

      16. Jiang NJ, Liang XY, Su HM, et al. Study of influence factors on HBV relapse after NAS withdrawal in the CHB patients who met NAS cessation criteria. J Hepatol 2011;54:Abstract 722.

      17. Manesis EK, Papatheodoridis GV, Hadziyannis E. Significance of serum HBsAg levels for the definition of the inactive hepatitis B carrier. Hepatology 2010;52:Abstract 483.

      18. Verheyen J, Neumann-Fraune M, Berg T, Kaiser R, Obermeier M. Mutations in the HBs-antigen influence the results of HBsAg quantification assays. J Hepatol 2011;54:Abstract 394.

      19. Bruce M, Horner M, Shang D, et al. Mutation RTA181T/SW172 is associated with lower HBsAg levels than other mutations in HBV polymerase codon 181 an overlapping surface codon 172 – implications for vaccination? J Hepatol 2011:54:Abstract 706.