A Novel Method to Determine HIV Incidence Among Youth

HIV infection incidence estimates are important, not only for determining specific populations where community HIV education sessions can have the most benefit or where changes in infection patterns are occurring, but also to target these populations for therapeutic interventions and measure their effect in the community. It is estimated that half of all new HIV infections in the United States occur in persons under the age of 25 each year, translating to approximately 20,000 new HIV-infected young people annually.

Incidence can be estimated by testing a cohort of individuals for antibody at two different time periods and observing the number of new infections, or by demonstrating the presence of HIV p24 or viral RNA in antibody-negative persons. Since these strategies are logistically difficult, expensive, and/or require significant laboratory infrastructure, new laboratory-based strategies were devised that can classify individuals as recently infected or with established infection. These serologic assays are known as "sensitive/less sensitive (S/LS) assays". They are based on either the concept of antibody titer or the concept of antibody avidity. There are several S/LS tests available but all of them are performed on serum collected via venipuncture. Based on several studies showing that obtaining saliva is a more desirable method of collection for adolescents receiving HIV testing, it is anticipated that the collection of saliva and its use for determining HIV incidence by S/LS methods among adolescents would be a valuable adjunct for population studies and for increasing compliance for testing and enrollment.

The aim of this study is to prove equivalence between the serum S/LS reference method (CDC validated serum S/LS Dilutional Vironostika (DV) from Organon Teknika) and the experimental method that uses an FDA approved saliva collection device (OraSure Technologies). Our rationale follows a syllogism, i.e., if the serum S/LS EIA (DV) is the best serologic method to determine incidence estimates and a saliva S/LS method is found to perform equivalently, then it follows that the saliva S/LS method can be used effectively to determine incidence. Therefore, our purpose is to show equivalence between the two assays. To address the issue of accuracy, further studies are planned to assess the saliva S/LS test against seroconversion panels where the exact time of acquisition of HIV infection is known.

This is a cross-sectional, laboratory-based study designed to develop and evaluate a saliva-based S/LS assay for differentiating individuals with recent HIV infection (less than 133 days) from those with established HIV infection. A total of 440 subjects will be recruited in the study. A serum and two saliva samples will be obtained from every subject. The first 40 patient samples will be used to develop and calibrate the saliva-based S/LS method. The remaining patient samples will be tested using the calibrated saliva based S/LS assay and the reference serum S/LS assay and concordance between the two methods will be analyzed.