Subtype-specific problems with quantification of plasma HIV-1 RNA

To determine whether two commercial assays for quantification of plasma HIV-1 RNA levels detect different genetic subtypes of HIV-1 with equal efficiency.

Blind testing of stored plasma samples from 95 individuals infected with different genetic subtypes of HIV-1 (27 subtype A, 24 B, 18 C, 18 D, two E, two G, two H, and two J). The HIV-1 subtype had previously been determined by direct sequencing of the V3 domain of the env gene.

One plasma sample from each individual was tested once by the Roche HIV monitor assay and once by the Organon nucleic acid sequence-based amplification (NASBA) HIV-1 RNA quantitative assay, according to the manufacturers' recommendations. Information about CD4+ lymphocyte counts and antiretroviral treatment was available.

The results from the two assays were strongly correlated with each other for subtypes B, C and D, but not for subtype A because many samples had RNA levels close to or below the lower detection limit of the assays. Thus, 15 out of 27 (56%) subtype A samples were negative by the HIV monitor assay and 12 (44%) were negative by the NASBA assay. These frequently occurring negative results among subtype-A-infected individuals were not due to better immunological status, more aggressive antiretroviral treatment, or differences in sample storage conditions.

The HIV monitor assay and, possibly to slightly lesser degree, the NASBA assay appear unable to accurately quantify HIV-1 RNA levels in plasma samples from many subtype-A-infected individuals. These problems are likely to be due to primer mismatches and they limit the possibility of using these assays for routine monitoring of HIV-1-infected individuals in many parts of the world.