Session 80Poster Presentations Diagnostic Tests for HIV Session Day and Time: Tuesday 1:30 - 3:30 pm Room: Hall A
659 Modification of a Commercial HIV-1 Enzyme Immunoassay for Identification of Recent HIV-1 Infections: Use of Differential Antibody Avidity J. Jenner*, M. Grazioplene, A. Kazianis, K. Phinney, B. Werner Massachusetts State Lab Inst, Boston
Background: The Serologic Testing Algorithm for Recent HIV Seroconversion (STARHS) uses a less-sensitive HIV-1 Enzyme Immunoassay (EIA) protocol to differentiate between recent and long-term infection. We sought to develop an alternative method to STAHRS, one that does not require a large sample dilution and is not affected by minor fluctuations in incubation temperatures and times. We modified the Genetic Systems rLAV EIA Human Immunodeficiency Virus Type I (BioRad) to obtain a measure of antibody avidity that allows for the differentiation of incident and prevalent infections using a single HIV-1 positive serum specimen.
Methods: The manufacturer's protocol was followed for the rLAV assay with a dissociation step added to a duplicate well. We tested several candidate dissociation reagents at varying concentrations and incubation times, as well as pH extremes to determine optimal conditions for disrupting low affinity antibody-antigen bonds. An avidity index (AI) was calculated by dividing the optical density (OD) of the well containing sample subjected to variable conditions by the OD of the corresponding untreated well. Fifty-eight (58) HIV-1 specimens obtained 2 to 146 days after seroconversion were tested under all dissociation conditions. Five (5) 8-member CDC Proficiency Panels (PT) available for STARHS quality control were tested using the most efficient dissociation reagent.
Results: All reagents, at an optimal concentration, were able to dissociate low-affinity bonds in the 58 seroconversion specimens, but a 30-min incubation with 0.1 M diethylamine (DEA) resulted in the lowest AI cutoff (55%) compared to those of the other reagents (60% to 90%). The Pearson correlation co-efficient between AI and time since HIV-1 seroconversion was high, demonstrating the progressive increase in antibody affinity over time (r = 0.84, p < 0.001). Using DEA, intra- and inter-assay CVs were = 10% and PT specimens (n = 40) were correctly scored as recent or long-term infections. All specimens used in this study that were indicative of recent infection based upon AI were similarly identified by STARHS.
Conclusions: This simple method based on differential serum antibody avidity can be used as an alternative to STARHS to identify recent HIV-1 infection. The modified rLAV EIA should be easier to operationalize and set up in laboratories performing routine HIV antibody testing.
