Branson BM; National HIV Prevention Conference (2003 : Atlanta, Ga.).
Abstr Book 2003 Natl HIV Prev Conf July 27 30 2003 Hyatt Regency Atlanta Hotel Atlanta Ga Natl HIV Prev Conf 2003 Atlanta Ga. 2003; abstract no. T1-B0103.
CDC, Atlanta, GA
BACKGROUND: STARHS (the Serologic Testing Algorithm for Recent Seroconversion) currently uses the Vironostika less-sensitive EIA to classify whether an HIV infection is recent. The technique is useful for estimating HIV incidence in populations, but may be less accurate for providing results to individuals. Alternative techniques based on different principles are being investigated to refine the accuracy and improve the applicability of the STARHS concept. This presentation will summarize the status of current research.METHODS: Assay techniques based on four different principles are under investigation: "detuned" assays using modifications of commercial EIAs; an antibody capture assay using a synthetic gp41 peptide representing antigens from HIV subtypes B, E, and D that measures the ratio of HIV antibody to total IgG; an avidity index that compares the strength of antigen-antibody bonds as the immune response matures; and an EIA that detects a specific IgG subclass antibody which wanes soon after infection. Specimens collected during follow-up of persons with known dates of seroconversion were tested with each assay to determine their ability to distinguish recent HIV infections, and to calculate the length of time after seroconversion ("window period") that each classified infections as recent.RESULTS: With the Vironostika less-sensitive EIA, the mean window period is 170 days for estimating incidence in populations with subtype B infections. For providing results to individuals, the window period is 1 year. The Vironostika window period varied markedly with different non-B subtypes of HIV. The BED assay demonstrates a mean population window period of 160 days which remained consistent with subtype B, C, and E specimens. Preliminary results indicate the other assays show promise for classifying recent HIV infection, but data are not yet sufficient to provide accurate estimates of their window periods.CONCLUSIONS: STARHS results with Vironostika are more reliable for estimating incidence in populations than for providing information about individuals, and are less useful with non-B subtypes of HIV. Validation of alternative serologic approaches based on different principles and calibrating them for use in estimating incidence is underway. The clinical relevance of STARHS results for individuals has not yet been determined.
Publication Types:
Keywords:
- AIDS Vaccines
- Acquired Immunodeficiency Syndrome
- Algorithms
- HIV Antibodies
- HIV Infections
- HIV Seropositivity
- Immunoenzyme Techniques
- Incidence
- Research Design
- Serologic Tests
- methods
- organization & administration
Other ID:
UI: 102261935
From Meeting Abstracts