Assay of anti-HBs antibodies using a recombinant antigen and latex particle counting: comparison with five commercial tests.

An assay of anti-HBs antibodies based on agglutination of latex particles coated with recombinant HBs-antigen was compared with Abbott radioimmunoassay (Abbott-RIA), which uses a human plasma-derived antigen. The population examined consisted of 76 Abbott-RIA anti-HBs-negative prevaccinated subjects and 1044 serum samples anti-HBs found positive by Abbott-RIA, including 283 samples of subjects vaccinated either with a human plasma-derived vaccine (group A; n = 180) or with a recombinant vaccine (group B; n = 103). Correlation coefficients between the two techniques were respectively r = 0.89 for the whole population (n = 1044), r = 0.98 in group A and r = 0.74 in group B. Anti-HBs titres were higher with latex than with RIA in group B as shown by the regression slopes: latex = 508 + 1.11 RIA in group A and latex = -1138 + 3.97 RIA in group B, suggesting that some vaccinated subjects from group B produced antibodies against epitopes proper to the recombinant antigen. In the prevaccinated population and in group A, the latex results were compared with those of radioimmunoassays (Abbott, Sorin) and enzyme immunoassays (Behring, Roche, Pasteur). Only the Roche-EIA detected anti-HBs in the prevaccinated subjects. The correlation between the various immunoassays was r greater than 0.96 only for values higher than 100 IU/l.