Phenotypic or genotypic resistance testing for choosing antiretroviral therapy after treatment failure: a randomized...

To assess the respective value of phenotype versus genotype versus standard of care for choosing antiretroviral therapy in patients failing protease inhibitor-containing regimens.

Patients with plasma HIV-1 RNA exceeding 1000 copies/ml were randomly allocated to phenotyping, genotyping, or standard of care.

Five-hundred and forty-one patients were randomized, 190 to phenotyping, 192 to genotyping and 159 to standard of care. The baseline median CD4 cell count (280 x 106 cells/l), the plasma HIV-1 RNA level (4.3 log10 copies/ml), and the number of drugs previously received (n = 6) were similar in the three arms. More patients in the standard-of-care arm received at least three new drugs (55% versus 20% in the other arms; P < 0.001) and a regimen containing drugs from the three different classes. Plasma HIV-1 RNA was < 200 copies/ml at week 12 in 35% of patients in the phenotyping arm, 44% in the genotyping arm and 36% in the standard-of-care arm (phenotyping versus standard of care, P = 0.918; genotyping versus standard of care, P = 0.120). In a secondary analysis of 179 patients experiencing a first protease inhibitor failure, the percentage of patients achieving HIV-1 RNA < 200 copies/ml was significantly higher in the genotyping arm (65%) than in the phenotyping (45%) and the standard-of-care arms (45%) (genotyping versus standard of care, P = 0.022).

Overall, resistance assays did not demonstrate benefit over standard of care. In patients with the most limited protease inhibitor experience, a significant benefit was observed in the genotyping arm.