Discordance in the determination of non- or Iow-responders to HBV vaccine using IMx-AUSAB ® or AUSAB®-RIA Antibodies induced by vaccination using HBs antigen are the main protective factor conferring immunity to the pathogen. Non-responders to the vaccine are usually defined by a titre of anti-HBs antibodies (aHbs-Ab) <10IUI -t using AUSAB- RIA, and are considered unprotected against HBV. Although their degree of protection is difficult to assess, low responders, exhibiting < 100IU1-1 of aHbs-Ab using AUSAB-RIA, are often considered as needing new challenges to reach a safe level of aHbs-Ab 2. Comparing the new IMx-AUSAB to the conventional AUSAB-RIA detec- tion of aHbs-Ab, we were surprised by the considerable discrepancies be- tween these two quantitative estimations in a particular population: 16 healthy vaccinated subjects with poor antibody responses (< 200 IU 1-1 using conven- tional AUSAB-RIA) after at least three GenHevacB (Pasteur Vaccins, Paris, France) HBV vaccine injections at monthly intervals, followed by a chal- lenge 1 year later. As shown in Fioure 1, the quantification of aHbs-Ab using IMx-AUSAB was far higher in most sub- jects than using conventional AUSAB- RIA. All dosages were repeated at least twice and they yielded concordant results. To rule out trivial artefacts of dosage, four sera were subjected to ultracentri- fugation at 1000009 for 20min without any change in their aHbs-Ab titres. In addition, ten sera were tested for anti- HBc antibodies using IMx-AUSAB with negative results, suggesting that non- specific binding could not have been in- volved. Both IMx-AUSAB and AUSAB- RIA tests are heterogeneous phase sand- wich methods for antibodies from the same manufacturer (Abbott laboratories, Chicago, IL, USA), but use either iodin- ated or biotinylated HBs antigen 3''~. A better antibody detection using IMx- 1200 >1000 800 ~ 600 400 200 0 Anti-HBs antibodies Figure 1 Paired determination of aHbs-Ab using either IMx-AUSAB ( . ) or AUSAB-RIA (~1) in sera from poor responders to HBV vaccine AUSAB might be due to the fact that biotin labelling is a milder chemical procedure than radioactive iodination, and thus better preserves some antigenic determinants 5. Additional antibodies detected using IMx-AUSAB are probably either poor antibodies of low affinity or antibodies detecting fragile epitopes destroyed by radioactive iodination. Nevertheless, the protective value of these probable additional antibodies de- tected using IMx-AUSAB in poor re- sponders to HBV vaccine remains un- certain and, until now, unestablished. Thus, the threshold level of 10IU1 -x of aHBs-Ab established from conventional AUSAB-RIA should not be used for clinical purposes when estimates are obtained using another method of detec- tion of aHBs-Ab. Francine Bougy, V6ronique Neff-Continant, Christian Fessard and Jacques H.M. Cohen Laboratoire d'Immunolooie and Service de M~deeine Preventive, CHU R. Debr~, 51092 Reims Cedex, France References 1 Szmuness, W., Stevens, C.E., Harley, E.J. et al. Hepatitis B vaccine: demonstration of efficacy in a controlled clinical trial in a high risk population in the United States. New Engl. J. Med. 1980, :~1~, ~ 1 2 Jilg, W., Schmidt, M. and Deinhardt, F. Hepatitis B vaccination strategy for booster doses in high risk population groups. In: Progress in Hepatitis B Immunization (Eds Coursaget, P. and Tong, MJ.) Inserm Edition, Paris, 1990, pp.419~427 3 Courouc6, A.M. Quantification of post-vaccination humoral immunity. In: Progress in Hepatitis B Immunization (Eds Coursaget, P and Tong, MJ.) Inserm Edition, Paris, 1990, pp.67~2 4 Ostrow, D.H., Brook, E., Kimes, D. et al. Quanti- tation of hepatitis B surface antibody by an automated microparticle enzyme immuno-assay. J. Virol. Methods 1991, 3~, 265-276 5 Guesdon, J.L., Ternynck, T. and Avrameas, S. The use of avidin biotin interaction in immuno- enzymatic techniques. J. Histochem. Cytochem. 1979, 27, 1131 1139 0264~410X/93/04/0485~) 1 .~ 1993 Butterworth-Heinemann Ltd Vaccine, Vol. 11, Issue 4, 1993 485