Evaluation of an alternative HIV diagnostic algorithm using specimens from seroconversion panels and persons with established HIV infections

Abstract

Background: The current algorithm for HIV diagnosis in the US involves screening with an immunoassay (IA) and supplemental testing with Western blot (WB) or immunofluorescence assay. Because of existence of more sensitive and specific FDA-approved assays that would also reduce the cost and turn-around time of testing compared to WB, several alternative algorithms have been evaluated. Recently, an alternative algorithm using a sensitive 3rd or 4th generation IA followed by an HIV-1 and HIV-2 discriminatory supplemental test on the initial IA-positive specimens was proposed. Concordant positive results indicate HIV-positive specimens and discordant results are resolved by nucleic acid amplification testing (NAAT).

Objectives: To evaluate the sensitivity of assays during acute HIV infection and the performance of the current and an alternative algorithm using samples from HIV-1 seroconversion panels and persons with established HIV infections.

Study design: To evaluate the algorithms in early infections, 26 HIV-1 seroconverters from the US were tested with three 3rd generation and one 4th generation IA, six rapid tests (RTs), one NAAT, and WB. Sensitivity and specificity of the algorithms were calculated by testing an additional 416 HIV-positive and 414 uninfected control samples with one 3rd generation and one 4th generation IA, four RTs, one NAAT, and WB.

Results: The individual assays evaluated became positive 5 (RT) to 26 days (NAAT) before WB was positive. Among seroconverters, the alternative algorithm detected significantly more infections than the current algorithm (103-134 versus 56, p<0.0001). Furthermore, the use of a 4th generation IA instead of a 3rd generation assay as the screen resulted in significantly higher detection of acute infections (p<0.0001). In contrast, the algorithms performed equally among specimens from established HIV-1 infections.

Conclusions: This study demonstrated improved sensitivity of the alternative algorithm for detecting acute HIV-1 infections, while maintaining the ability to accurately detect established HIV-1 infections. Early detection is important as individuals can be highly infectious during acute infection. In addition, the alternative algorithm should reduce turn-around time by using a RT as the supplemental test has the potential to increase the number of test results returned.