Article Text

Download PDFPDF
A unique monoclonal antibody for therapeutic use against chronic hepatitis B: not all antibodies are created equal
  1. Camille Sureau
  1. Correspondence to Dr Camille Sureau, Laboratoire de Virologie Moléculaire, INSERM-U1134, INTS, 6 rue Alexandre-Cabanel, Paris 75739, France; csureau{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Even though current antiviral treatments based on pegylated interferon, or nucleos(t)ide analogues, in mono or combination therapies, have demonstrated clinical benefit to HBV-infected patients, chronic hepatitis B remains a difficult-to-cure disease. An absolute cure, defined as total elimination of HBV DNA, is currently not achievable, and a functional cure characterised by sustained virological response and HBV surface antigen (HBsAg) clearance off-treatment remains a challenge. Pegylated interferon treatment can lead to a sustained virological response and HBsAg clearance off-treatment in only 3–7% of patients,1 and it is too often associated with contraindications and adverse effects. Long-term treatment with nucleos(t)ide analogues is better tolerated, but the chance to achieve a sustained virological response and HBsAg loss remains low,2 and the drugs do not eradicate intrahepatic HBV DNA.

The risk of disease progression and development of hepatocellular carcinoma in chronically HBV-infected patients is clearly associated with high levels of serum HBV DNA and HBsAg.3 ,4 Since high levels of serum HBsAg may be responsible for exhaustion of HBsAg-specific T-cell response in chronically infected individuals, an antiviral strategy aimed at suppressing circulating HBsAg could restore virus-specific immune response and promote viral clearance.5

Zhang et al 6 investigated the potential of anti-HBsAg monoclonal antibodies (mAbs) infusion for clearance of circulating HBsAg in chronic infection. The study was based on (i) the previous observation that anti-HBsAg mAbs treatments could have short-term antiviral effects,7 ,8 and (ii) the growing interest in mAbs-based immunotherapy for viral diseases, including infections with HIV, HCV and Ebola virus.9 The therapeutic use of antiviral mAbs was initially based on the mAbs activity in virus neutralisation or direct infected-cell elimination, but mAbs-based treatment may in addition induce protective immunity by recruiting the immune system. Antiviral mAbs are now being developed for an increasing number of viral diseases in humans, and some are being tested in clinical trials.9

To assess the potential of anti-HBV mAbs, the authors conducted a study using different mouse models mimicking persistent HBV infection in different immunological settings. They first selected 11 mAbs directed against various epitopes in the pre-S1 or S domains of the HBV surface proteins. Both pre-S1 and S epitopes overlap with well-characterised infectivity determinants at the surface of HBV virions,10 and hence, all anti-pre-S1 and anti-S mAbs demonstrated a neutralising activity in an in vitro infection assay, which, however, did not directly correlate with their affinity to HBV surface proteins.

Then the therapeutic potential of selected mAbs was evaluated in HBV-transgenic (HBV-Tg) mice—a model mimicking persistent HBV infection in an immunotolerant phase—by a single intravenous infusion at a dose of 20 mg/kg. All mAbs directed to the immunodominant a-determinant (HBsAg) in the HBV surface proteins had a significant effect on reducing serum HBsAg levels and viraemia. However, the effect was most pronounced with one mAb, E6F6, specific for the conserved linear epitope glycine-proline-cysteine-arginine (GPCR)/lysine-threonine-cysteine-threonine (KTCT) at position 119–126 in the S domain of HBV surface proteins. Interestingly, this motif is essential to HBV infectivity, presumably at the step of initial attachment of HBV virion to cell surface heparan sulfate proteoglycans.11 ,12 No HBV suppression was observed after infusion of the most neutralising anti-pre-S1 mAbs known to interfere directly with binding of the virus to its cell surface receptor, the sodium taurocholate cotransporting polypeptide. There was no correlation between HBV-clearance activity in HBV-Tg mice and neutralisation activity in vitro. Note, however, that a neutralisation effect in HBV-Tg mice could not be observed because mouse liver cells are not susceptible to HBV infection.

A single infusion of E6F6 mAb in HBV-Tg mice lead to a robust suppression of circulating HBsAg and HBV DNA in a dose-dependent manner, reaching a maximum effect when infused at 1.2 mg/kg. Among intrahepatic HBV markers, HBsAg proteins were suppressed after mAb infusion, but HBcAg, HBV DNA and RNA remained unchanged. In prolonged treatment, there was no change of serum HBeAg or alanine aminotransferase levels.

To test the therapeutic efficacy of E6F6 in comparison to commercial anti-HBV immunoglobulin (HBIG) preparations, a human-Fc chimeric E6F6 mAb was constructed to substitute its mouse counterpart in E6F6 with the human IgG1 Fc region. When tested in parallel with HBIG by infusion in HBV-Tg mice after normalisation for anti-HBsAg antibodies by ELISA, the chimeric E6F6 demonstrated superior and longer-lasting clearance effect than did HBIG.

Mouse E6F6 mAb infusion in the setting of an established infection in immunodeficient human-liver-chimeric mice could also suppress HBV for an average 14-day period post-infusion before rebound. But consecutive infusions lead to a sustained effect. As expected, the neutralising E6F6 mAb was also effective in preventing infection when infused to mice 1 day prior to HBV inoculation.

When the hydrodynamic injection mouse model was used to mimic the tolerance phase of chronic infection in an immmunocompetent setting, successive infusions of E6F6 lead to a decrease in serum HBsAg levels and to an increase of HBcAg-specific interferon-γ-secreting T-cells and HBsAg-specific and HBcAg-specific CD8+T-cells, suggesting a restoration of anti-HBV T-cell response. The E6F6 mAb-mediated phagocytosis was clearly the most potent and rapid process of viral clearance: in HBV-Tg mice, HBsAg decline after E6F6 infusion occurred as soon as 10 min post infusion. The effector cells involved in viral clearance were liver Kupffer cells, peripheral blood lymphocytes (PBLs) neutrophils and phagocytes, but they did not include PBLs natural killer (NK) cells. The predominant mechanism of E6F6-mediated immunoclearance was the Fcγ receptor-dependent opsonophagocytosis. It suggested that the affinity between mAb-Fc and Fcγ-receptor could constitute an important factor in immunoclearance, but the results of in vitro binding assays argued against this hypothesis. It was then speculated that the E6F6 epitope position—or its binding properties—and the characteristics of E6F6-viral particle immune complexes (ICs) might instead be the essential determinants. The minimal binding sequence of the E6F6 epitope was mapped to amino acid residues 122–124 (-CK/RTC-) in the HBV surface proteins S domain, analogous to the H166 linear epitope previously described.13 It is noteworthy that H166 and E6F6 are reactive with both adw (CKTC) and ayw (CRTC) serotypes. Then, upon examination of mAb-viral particle ICs by electron microscopy, and their size characterisation by low-speed centrifugation, it appeared that E6F6, in contrast to other anti-HBsAg mAbs, did not aggregate viral particles. As an explanation to the superior performance of E6F6 in HBsAg clearance, the authors speculated that the small size of E6F6-viral particles ICs—or lack of particles aggregation—was key to an efficient phagocytosis that would not occur with larger ICs formed by other mAbs.14 Perhaps future selection of mAbs with therapeutic potential should not be based on their ability to neutralise virions at viral entry, or their affinity for HBsAg, but rather on the nature, or position of the epitope at the surface of HBV particles, and on the ability of the related mAb to form small ICs upon binding to HBsAg.

It is likely that the conserved E6F6 epitope is not unique in eliciting antibodies with potent HBV-clearance ability. Additional mAbs specific for linear or conformational epitopes in the HBV envelope proteins could demonstrate similar characteristics and thereby be valuable to constitute a panel of anti-HBsAg mAbs for tailor-made treatments of patients chronically infected with HBV variants. In this regard, the clearance activity of the many anti-HBsAg mAbs that have developed over the years could be evaluated for such characteristics as correlation between clearance activity, epitope position/structure, mAb-HBsAg affinity and mAb-viral particle ICs size.

The E6F6 mAb displays potent neutralising activity that blocks viral entry and propagation of HBV in the liver, but its predominant activity in chronic infection therapy resides in viral particles clearance by antibody-mediated phagocytosis. The ensuing reduction of circulating HBsAg would then stimulate the restoration of HBsAg-specific T-cell response. While these data were generated in the mouse models, their implication in humans is certainly encouraging. Such mAb-based immunotherapy, alone or in combination with current antiviral therapies, therefore represents a promising avenue for treatment of chronic hepatitis B, and E6F6-like mAbs could eventually substitute for HBIG in the current immunoprophylaxis regimen.



  • Competing interests CS is a CNRS scientist and he is supported by Agence Nationale pour la Recherche sur le SIDA et les hépatites virales (ANRS).

  • Provenance and peer review Commissioned; internally peer reviewed.

Linked Articles