Comparison of Methods to Detect Recent HIV Type 1 Infection in Cross-Sectionally Collected Specimens from a Cohort of Female Sex Workers in the Dominican Republic

Abstract

Interest in estimating HIV-1 incidence using specimens obtained as part of cross-sectional surveys has led to the development of new methods to detect recent HIV-1 infection through the testing of a single anti-HIV-positive specimen. These assays are based on quantitative and qualitative differences in anti-HIV-1 antibodies between recent and long-standing infections. An ongoing vaccine preparedness study enrolled female sex workers in the Dominican Republic. Specimens from women found to be HIV positive at baseline were tested for recent HIV-1 infection using the detuned assay, avidity index, and BED-CEIA assay. An unweighted kappa statistic in pairwise comparisons was used to estimate the correlation of recent HIV-1 infection detection by the three methods. Nineteen (3.9%) of 482 women were positive for HIV-1 infection. The incidence of HIV infection was 1.4% [95% confidence interval (CI): 0.2, 5.3], 0.9%(95% CI: 0.1, 4.4), and 1.0%(95% CI: 0.1, 4.4) using detuned assay, avidity index, and BED-CEIA techniques, respectively. The overall agreement between both detuned assay and avidity index and detuned assay and BED-CEIA was 94%(kappa = 0.8, 95% CI; 0.3, 1.0). The correlation was highest between BED-CEIA and avidity index methods (100%; kappa = 1.0). All three methods performed similarly in detecting recent HIV-1 infection in this region dominated by clade B HIV-1 infection. Although incidence estimates were slightly higher using the detuned assay method, they were not significantly different. These assays may be of value in both clinical research and practice. The utility of individual assays for recent infection detection will depend upon operating characteristics, HIV-1 subtype limitations, and selection of appropriate assay cutoff values.