Abstract
In view of recent revised recommendations for human immunodeficiency virus (HIV) confirmatory testing, the performance of 3 HIV confirmatory assays was compared. Using the HIV Blot 2.2 (MP-WB), the INNO-LIA HIV I/II Score (INNO), and the Geenius HIV 1/2 Confirmatory Assay (Geenius), we tested 199 HIV-1 positive, 161 HIV negative, 65 HIV western blot indeterminate, 26 HIV seroconversion, 34 early HIV infection and 4 HIV-2 positive archived specimens. We show that all 3 assays had comparable test sensitivity in the detection of HIV-1 positive cases. However, less non-specific reactivity was observed with the INNO and Geenius assays, where both of them were able to resolve MP-WB indeterminate cases. When early HIV cases were considered, INNO and Geenius were more likely to confirm an early-stage infection as positive. Nevertheless, overall poor sensitivity (25.5% - 44.7%) of these assays for the detection of early cases was observed, likely because these cases had very low or non-detectable levels of HIV antibodies. Hence, further testing by a nucleic acid test or a p24 antigen test of specimens reactive on screening with a fourth generation Ag/Ab assay that are negative on confirmatory testing for HIV-specific antibody, may be useful. In conclusion, INNO and Geenius had comparable test performance, although the ease of use and shorter assay time for Geenius may make it the preferred choice for laboratories. In that regard, of note is our observation of non-specific reactivity of lipaemic specimens to the HIV-2 gp140 band in the Geenius assay, which should prompt caution when interpreting results of such specimens.