Immunoassay of hepatitis b surface antigen by particle counting after pepsin digestion

Particle counting immunoassay is based on latex agglutination, the reaction being measured by instrument counting of the particles remaining unagglutinated. Most interference which generally affects latex agglutination can be avoided by pepsin digestion of the sample, provided the antigen (Ag) of interest resists pepsin, which is the case of the hepatitis B surface antigen (HBsAg). Pepsin treatment has the additional advantage of inactivating antibodies and so releasing the Ag from immune complexes. We have set up an assay of HBsAg, proceeding in a prototype of Impact Instrument (Acade Diagnostic Systems, Belgium) at a rate of 60 samples· h −1 and a total running time of 2 or 4 h. This assay was compared with Abbott radioimmunoassay (RIA) in 706 consecutive patients (A) and 31 selected sera for which values close to the cut-off had been obtained by RIA (B). In A, 38 sera were found positive and 668 negative by both methods. In B, RIA after neutralization classified the samples as positive ( n = 14), negative ( n = 14), or dubious ( n = 3). Complete agreement between latex and RIA was achieved for nine positive, 12 negative, and two dubious samples. Of five RIA-positive samples, two were classified as latex-negative and three as dubious in the latex assay. One sample dubious in RIA was found latex-positive and two RIA-negative samples were found, respectively, latex-positive and dubious; when retested after pepsin digestion, the first of them became RIA-positive. That pepsin is useful to release antigen from immune complexes was also shown by addition of rabbit anti-HBs antibodies to five positive serum samples and by the recovery of the antigen both in RIA and latex agglutination after pepsin digestion.