Washington, DC, USAMay 22–24, 2005

A symposium on HIV variants and hepatitis B virus (HBV) surface antigen (HBsAg) mutants, sponsored by Abbott Laboratories, was held May 22–24, 2005, in Washington, DC. Eighteen speakers from 8 countries discussed the impact of HIV variants and HBsAg mutants on immunoassays.

HIV-1 is classified into 3 groups: group M, group O, and group N. Group M, representing most infections, is further subdivided into subtypes A–K. Recombination between subtypes and groups adds to the overall diversity of HIV-1. In some regions, recombinant strains, referred to as circulating recombinant forms (CRFs), have become the predominant form of HIV-1. The high degree of HIV-1 diversity presents a challenge to diagnostic assays.

Two presentations described surveillance programs designed to establish well-characterized panels that contain HIV-1 variants from diverse geographic locations. Since current serologic assays are based primarily on a single HIV-1 subtype B isolate, and because subtype B strains represent only ≈12% of infections globally, reference panels are vital in ensuring that tests are robust to ever-changing genetic and antigenic polymorphisms. Current panels include the known HIV-1 group M subtypes, recombinants, and group O viruses, and plans are under way to include HIV-1 group N viruses.

Another surveillance program is dedicated to searching for emerging HIV variants. Cameroon is an ideal location for discovering new HIV-1 variants because of the high HIV-1 diversity in that African country. Data on Cameroonian samples showed several new group N viruses and a high level of HIV-1 variation with 5 subtypes, 6 CRFs, numerous unique recombinant forms, and group O. Most HIV infections were recombinant HIV-1 strain, CRF02_AG, a mosaic of viral subtypes A and G.

Strain diversity like that observed in Cameroon is rapidly emerging in the industrialized world (e.g., France, the United Kingdom, the Middle East, and the United States). Variants once considered rare in industrialized countries now represent a substantial proportion of new infections. In the United Kingdom, 25% of new HIV-1 infections are of the non-B subtype (mostly A and C). In France, among new infections identified in the National Virological Surveillance program, half were non–subtype B, 0.2% dual HIV-1/HIV-2, and 1.9% HIV-2. In the United States, 0.8% of new HIV-1 infections in blood donors were non-B subtype from 1993 to 1996, rising to 1.8% from 1997 to 1998 and increasing to 3.1% from 1999 to 2000. Furthermore, >100 patients in a Minnesota medical center were identified as being infected with non–subtype B. These studies demonstrate that increasing numbers of non–subtype B strains are circulating in the United States. Since the prevalence of non–subtype-B infections in the United States may be underestimated, researchers noted at this meeting and in a recent publication (1) that additional studies on this issue are warranted.

Genetic diversity can affect the diagnosis and monitoring of HIV infections. The evidence presented included lack of detection of HIV variants by serologic and nucleic acid assays. In 1 case, amino acid sequence variations within the envelope immunodominant region resulted in the lack of detection of a subtype B strain. Genetic variation also influences nucleic acid assays, e.g., nucleotide polymorphisms within primer and probe sites can affect detection or accurate quantitation of divergent HIV-1 strains. The continued evolution and ever-changing global distribution of HIV will continue to challenge diagnostic assays.

The high rate of virus production, coupled with RNA-dependent replication, allows HBV to theoretically generate every possible single mutation daily. In addition, overlapping reading frames allow almost every mutation to have potentially pleiotropic effects.

A summation of a meeting held in 2004 in Germany (2) noted that HBsAg mutants probably occur more frequently than previously believed, that HBsAg assays should detect all major HBV variants, and that conditions favoring mutant outgrowth include failed treatment with HBV immune globulin, low-level persistent HBV infection, and antiviral therapy. Some HBsAg assays give false-negative results because they cannot detect HBsAg mutations in the immunodominant "a" determinant such as a threonine to leucine substitution at position 143 or a frequently reported glycine-to-arginine exchange at amino acid 145. False-negative results occurred with some assays for samples that contained multiple amino acid substitutions between positions 105 and 164.

The sensitivities of European CE-marked assays were discussed according to data on analytical sensitivity, seroconversion samples, and known HBsAg mutations (recombinant and native). Analytical sensitivity alone was insufficient to assess assay performance. Seroconversion panel data showed marked differences in the clinical sensitivity of CE-marked HBsAg assays. Some assays were unable to detect frequently occurring mutations in loop 2 of the "a" determinant. Detection of HBsAg mutant samples was recommended for inclusion in the European Common Technical Specifications (CTS) required for CE marking (mandatory for products sold in the European Union).

HBsAg mutant detection is influenced by assay format and monoclonal antibody binding sites. Assays with combinations of different monoclonal or polyclonal antibodies appear less sensitive to amino acid changes in the "a" determinant.

Mutant strains have been isolated from blood donors, in breakthrough HBV infections after vaccination, in reactivated HBV infection due to immunosuppressive therapy, in liver transplant patients who have received hepatitis B immune globulin, in dialysis patients, and in patients with chronic hepatitis. A study of HBV chronic carriers in Spain found that an estimated 6.6% had vaccine-escape mutants, and 9.2% had immunotherapy-resistant variants.

Because of overlapping reading frames for the polymerase and HBsAg genes, drug-resistant mutations in the pol region may potentially result in changes in the HBsAg amino acid sequence. Although the known major antiviral resistance mutations that affect HBsAg are located outside the "a" determinant and are unlikely to affect HBsAg tests, these mutations warrant continued monitoring.

Studies of infants who received HBV vaccination and hepatitis immune globulin showed that mutations in the HBsAg "a" determinant developed in 0.6% to 4%. In liver transplant patients treated with hepatitis B–specific immune globulin, mutations in the "a" determinant have been observed in 11% to 66% of cases. Data are lacking on the prevalence of HBsAg mutations in the general population in Europe and North America. The prevalence of HBsAg mutants is likely underestimated, and prevalence studies in different regions of the world are needed.

Not all cases of HBsAg-negative hepatitis B infection are due to mutants. These cases may occur during early acute infection before HBsAg becomes detectable, in carriers with subdetectable levels of wild-type HBsAg, in occult wild-type HBV infection, and in cases with HBsAg bound within immune complexes. Similarly, HBV DNA-negative results have been observed in some HBsAg-positive chronic carriers with replication below the detection limits of sensitive nucleic acid assays.

Surveillance studies to monitor the emergence of HBsAg mutants are vital in the design of next-generation HBsAg assays to ensure detection of all mutants. Reference panels with recombinant HBsAg used in conjunction with native samples are useful tools in evaluating HBsAg tests.

Recommendations for improved HBsAg assays included substantial increases in sensitivity for wild-type antigen, particularly in detecting early acute and occult infection, and in assay recognition of the most frequent HBsAg mutations. Suggested approaches for assay improvements included more sensitive assay signal detection methods, monoclonal antibodies with increased avidity, polyvalent antibodies for all major variants, and monoclonal antibodies directed against the HBsAg pre-S region.

Clearly, variations in HIV and HBV are critical considerations in the design of assays for these viruses. Meeting participants agreed that continued worldwide surveillance programs are vital in identifying new HIV and HBV variants and in understanding how changing patterns of viral variation affect current and future diagnostic assays.

List of speakers and panel discussants: M. Alter, Centers for Disease Control and Prevention, Atlanta, GA, USA; F. Barin, Centre National de Reference VIH, Universite F. Rabelais, Tours, France; A. Bartholomeusz, Victorian Infectious Diseases Reference Laboratory, North Melbourne, Victoria, Australia; R. Biswas, Food and Drug Administration (FDA), Rockville, MD, USA; C. Brennan, Abbott Laboratories, Abbott Park, IL, USA; M. Busch, Blood Systems Research Institute, San Francisco, CA, USA; C. Cartwright, Hennepin County Medical Center, Minneapolis, MN, USA; P. Coleman, Abbott Laboratories; J-M. Echevarria, Instituto des Salud Carlos III, Madrid, Spain; W. Gerlich, Justus Liebig University, Giessen, Germany; J. Hackett, Abbott Laboratories; I. Hewlett, FDA; F.B. Hollinger, Baylor College of Medicine, Houston, TX, USA; S. Kleinman, University of British Columbia, Vancouver, British Columbia, Canada; F. McCutchan, Henry M. Jackson Foundation, Rockville, MD, USA; S. Nick, Paul Ehrlich Institute, Langen, Germany; C. Osiowy, Public Health Agency of Canada, Winnipeg, Manitoba, Canada; J. Parry, Health Protection Agency Centre for Infections, London, UK; J-M Pawlotsky, University of Paris XII, Creteil, France; E. Tabor, FDA; B. Weber, Laboratoires Reunis, Junglinster, Luxembourg.