Background:  HIV vaccines elicit humoral immune responses that confound serological diagnosis of HIV infection. The most promising products also induce cellular immune responses that are hoped will prevent infection or lower viral load set point. Thus, some proportion of both HIV infected and uninfected vaccine recipients will have positive serology results and undetectable viral RNA. This study compares several nucleic acid detection strategies in HIV-vaccine naïve individuals who are either HIV seropositive with suppressed viral load or seronegative but at risk for infection.

Methods:  Study participants were enrolled into 1 of 2 groups:  HIV seropositives having undetectable viral RNA (prior 3 months), and at-risk individuals without history of HIV infection. Enzyme immunoassay and Western blot were performed on serum to confirm HIV sero-status. HIV-1 Viral RNA load was determined by the Roche Amplicor HIV-1 Monitor Test V1.5, ultrasensitive method. The Procleix HIV-1 Discriminatory assay was used to detect HIV viral RNA in plasmas. Frozen whole blood, peripheral blood mononuclear cells (PBMC), and dried blood spots on 903 cards were tested in the Roche Amplicor HIV-1 DNA assay V1.5. Optical density cutoff value was determined (<0.8 = negative) and used to differentiate HIV DNA positive from negative specimens. Sensitivity and specificity calculations were made using enzyme immunoassay and Western blot as the “Gold Standard.”

Results:  The 338 volunteers enrolled, including 155 HIV-seropositive, and 183 seronegative subjects; 94% of the HIV-infected participants were taking antiretroviral medications. DNA polymerase chain reaction (PCR) sensitivity was 1.00 (95%CI, 0.9765 to 1.00) using frozen whole blood, 0.987 (0.954 to 0.998) for PBMC, and 0.942 (0.893 to 0.973) for dried blood spots. DNA PCR using all sample types were statistically superior to both Amplicor Monitor V1.5 (0.387, 0.310 to 0.469), and Procleix (0.548, 0.467 to 0.628). Assay specificity did not differ among assays: DNA PCR specificity was 1.00 (0.980 to 1.00), 1.00 (0.980 to 1.00), and .1.00 (0.980 to 1.00) for frozen whole blood, PBMC, and dried blood spots, respectively; Amplicor Monitor 1.5 and Procleix specificity was 0.984 (0.953 to 0.997), and 1.00 (0.980 to 1.00), respectively.

Conclusions:  DNA PCR, using any of the studied specimen types, is significantly more sensitive for qualitative HIV diagnosis than either of the studied RNA detection methods in HIV infected subjects having well-controlled disease. DNA detection assays merit strong consideration for infection status determination in HIV vaccine recipients.