Clnical study Detection of anti-HIV antibody and p24 antigen Retrospective evalua,tion qf the Vtdos ID// DUO testfor simultaneous detection qf anti-IDV antibody and p24 antigen M. Poljak, D. Babič and K. Seme SUMMARY A fourth generation screening assay Vidas HIV DUO (bioMerieux, Lyon, France) which permits a simul- taneous detection of anti-HIV-1/2/O antibodies and HIV-1 p24 antigen was evaluated on 927 serum samples with previously determined anti-HIV-1/2 status in a retrospective manner. The Vidas HIV DUO yielded sensitivity, specificity, positive and negative predictive value of 100%, 99.27%, 94.69% and 100%, respectively. The Vidas HIV DUO is a highly specific and sensitive anti-HIV screening assay, suitable far the implementation in the routine laboratory diagnostics of HIV infection. Introduction Current laboratory diagnosis of human immunode- ficiency virns (HIV) infection is mainly based on the cletection of anti-HIV antiboclies (1). Since the intro- cluction of anti-HIV screening assays in 1985, their per- formance has continuecl to improve. Recently, fourth generation screening assays which permit the simulta- neous cletection of anti-HIV antibody and HIV-1 p24 antigen have been clevelopecl. They are reportecl to reduce the cliagnostic winclow by at least 6 days com- parecl to third generation assays in cases with symp- tomatic infections (2-5). Five commercially available fourth generation anti-HIV screening assays are avail- able at the moment: Vidas HIV DUO (HIV4) (bioMerieux, Lyon, France), Vironostika HIV Uni-Form II Ag/ Ab (Organon Teknika, Boxtel, Netherlands), Enzymun-Test® HIV Combi (Roche Boehringer Mann- heim, Penzberg, Germany), Enzygnost HIV Integral (Dade Behring, Mannheim, Germany) and Genscreen plus HIV Ag-Ab (Biorad, Hercules, USA). Accorcling to the results of the first evaluations of the fourth genera- tion HIV screening assays, they represent an important step towards earlier serologic diagnosis-of HIV infec- tion (2-5). Because of an extreme genome variability ofHIV-1 and HIV-2, each new anti-HIV screening test or even its variant should be evaluated in local settings before ap- proval is granted by the government and the test is introduced in a routine work (1). Thus, the aim of the Acta Dermatoven APA Vol 11, 2002, No 1 ------------ - -----11 Detection o.f anti-H!V antibody and p24 antigen present study was to evaluate a fourth generation anti- HIV screening assay Vidas HIV DUO (bioMerieux, Lyon, France) on serum samples with previously cletermined anti-HIV-1/2 status, which were carefully selectecl from the collection of the Slovenian AIDS Reference Center. Materials and methods A total of 927 serum sam ples with previously cleter- mined anti-HIV-1/2 status were retrievecl from the files of the Slovenian AIDS Reference Center. Serum samples were collectecl between 1985 ancl 1998, testecl rou- tinely for the presence ofHIV-1/2 antiboclies and stored at -20 °c. All serum samples were previously tested for the presence of anti-HIV-1 and anti-HIV-2 antibodies using at least two different screening enzyme immuno as- says in a routine manner. All repeatedly reactive serum samples were supplementa1y tested with confirmato1y Western blot tests. The Western blot results were inter- preted according to American Red Cross standards (6). 726 samples were anti-HIV-1 / 2 negative (they were ali obtained from subjects with high risk of HIV infection, none of these samples was obtained from bloocl donors), 98 samples were anti-HIV-1 positive (51 serotyped as subtype B, five as subtype C, four as sub- type A, one as subtype D and one as subtype E, 36 samples were not serotyped) and 9 samples were anti- HIV-2 positive (all obtained from subjects arriving from Central Africa). The most important group of samples consistecl of 94 samples which testecl anti-HIV reactive in one or more screening assays and anti-HIV-1/2 inde- terminate using confirmatory Western blot tests. All these 94 samples were finally determined as anti-HIV- 1/ 2 negative in follow-up testing. All 927 serum samples included in a retrospective evaluation were tested in August 1998 using the fourth generation assay Vidas HIV DUO (bioMerieux, Lyon, Table 1. Results of the retrospective evaluation ofVidas HIV DUO assay. Nonreactive Reactive Tota! 814 6" 820 o 107 107 814 113 927 "All 6 serum samples belonged to the group of94 anti- HIV-1/ 2 indeterminate serum samples which were fi- nally determined as anti-HIV-1/ 2 negative in follow-up testing (VIDAS HIV DUO false positive results). France), according to manufacturer's instrnctions as de- scribed previously (6). Results A total of 113 serum samples tested initially Vidas HIV DUO positive and 814 tested Vidas HIV DUO nega- tive . After afina! evaluation 107 serum samples were clefined as Vidas HIV DUO trne positive, 6 samples as false positive and 814 as trne negative, yielding a sensi- tivity of Vidas HIV DUO of 100% and specificity of 99.27% (Table 1). All 6 Vidas HIV DUO false positive serum sam ples belonged to the group of 94 anti-HIV- 1/ 2 Western blot indeterminate serum samples which were finally determined as anti-HIV-1/ 2 negative in follow-up testing. The positive predictive value ofVidas HIV DUO was 94.69% and the negative predictive value 100%. Discussion Recently, the fourth generation of HIV screening assays in which HIV p24 antigen detection is combinecl with anti-HIV-1 , anti-HIV-2 and anti-HIV-1 group O detection have been developed. Comparative evalua- tions of the fourth generation HIV screening assays performed on the seroconversion panels have shown that the fourth generation HIV screening assays reduced the diagnostic window of HIV infection by an average of 7 days in comparison to the third generation HIV screening assays and that they permit an earlier diagno- sis ofHIV infection than thircl generation screening as- says, by detecting p24 antigen which may be present in samples from individuals with recent HIV infection prior to seroconversion (2-5, 6-10). The results of our retrospective evaluation showed 100% sensitivity and a very high specificity of 99.27% of Vidas HIV DUO assay, which is in accorclance with the results of earlier evaluations (2, 3, 5, 8). The posi- tive predictive value of the Vidas HIV DUO was ve1y high (94.69%), which was the result of a low number of false positive results and a high prevalence of HIV in- fection (11.54%) in the artificially created population that was tested in our study. Namely, the number of false positive results decreases with the increase of prevalence of infection and accordingly the positive predictive value of the assay increases. Since the fourth generation HIV screening assays combine two different test principles in one assay, the potencial for nonspecific reactivity might be expected to be higher than with the third generation antibody assays. However, the results of our study did not con- Clinical study .12 ------ ------- - --- ----- ------- - - ----Acta Dermatoven APA Vol 11, 2002, No 1 Clinical study Detection