Background:  Potent antiretroviral therapy is effective in suppressing, but not eradicating, HIV-1 infection. Persistent viremia can be detected in patients on antiretroviral therapy despite suppression of plasma HIV-1 RNA to <50 copies/mL. The effect of baseline or on-treatment factors on the level of persistent viremia is unknown. To test the hypothesis that the level of persistent viremia is related to regimen potency, we used a real-time reverse transcriptase-polymerase chain reaction (RT-PCR) assay with single HIV RNA copy sensitivity to measure viremia in patients suppressed on different regimens.

Methods:  We tested plasma samples (3 to 7 mL) from 2 patient groups suppressed on therapy to <50 copies/mL for 4 to 333 weeks:  (1) 145 patients in the M98-863 trial (71 on stavudine/lamivudine/lopinavir/ritonavir (d4T/3TC/LPV/RTV) and 74 receiving d4T/3TC/nelfinavir (NFV); and (2) 50 patients enrolled in NIH protocols on different protease inhibitor (PI)- or non-nucleoside reverse trnscriptase inhisbitor (NNRTI)-containing regimens. Viral RNA was isolated from plasma spiked with an internal virion standard (RSV). After ultracentrifugation, the viral RNA was extracted and amplified using a 2-step RT-PCR procedure. Real-time PCR with gag-specific primers for either HIV-1 or RSV and fluorescence probe detection of product allowed quantification of specific viral cDNAs compared to a standard curve generated from RNA transcripts of known copy number. Samples were assayed in triplicate.

Results:  In all groups studied, about 75% of the patients had detectable low-level viremia ranging from 1 to 43 HIV RNA copies/mL. The distribution of HIV RNA levels was identical in all groups, with a mean of 3.2 copies/mL, and was independent of the treatment regimen. In the patients from the M98-863 trial, the similar levels of persistent viremia are in sharp contrast to the highly significant difference in virological outcomes between the 2 treatment arms in the overall study. Although no potency-related difference was observed, there was a highly significant association (p < 0.001) between viremia at week 60 and baseline RNA levels. A similar trend was seen with the samples from the NIH studies.

Conclusions:  The level of persistent viremia below 50 copies/mL was independent of the expected potency of the suppressive regimen (NFV vs LPV/RTV vs NNRTI), but correlated well with baseline viremia. These results imply that persistent viremia on treatment may result from virus production by cells that are infected prior to initiation of therapy.

Keywords: persistent viremia; antiretroviral potency; viral load