High correlation between the Roche COBAS(R) AmpliPrep/COBAS(R) TaqMan(R) HIV-1, v2.0 and the Abbott m2000 RealTime HIV-1 assays for quantification of viral load in HIV-1 B and non-B subtypes.

BACKGROUND: HIV-1 viral load assays are critical tools to monitor antiretroviral therapy efficacy in HIV-infected patients. Two assays based on real-time PCR are available, the Abbott Real-Time HIV-1 assay (Abbott assay) and the new Roche COBAS((R)) AmpliPrep/COBAS((R)) TaqMan((R)) HIV-1 test, v. 2.0 (TaqMan((R)) test v2.0). OBJECTIVES: We have compared the performance of the two assays in 546 clinical plasma specimens of group M strains from Luxembourg and Rwanda. STUDY DESIGN: Our analyses focused on subtype inclusivity and platforms accuracy for 328 low level viremia samples. RESULTS: Strong agreement and linear correlation were observed between the two assays (R(2) = 0.95) over a wide dynamic range. Bland-Altman analysis showed a mean difference of 0.04 log 10 indicating minimal overall viral load quantification differences between both platforms. One subtype C was severely underquantified by TaqMan((R)) test v2.0 for which sequence analysis revealed multiple mismatches between the viral sequence and the primer/probe regions. A non significant lower quantification of the Abbott assay was shown for subtype A1 with a mean log 10 difference of 0.24. For specimens under 200 cp/mL, the overall agreement was 90% at the cut-off of 50 cp/mL and 67% at assay's lower limit of detection of 20 and 40 cp/mL. 309 samples were retested by the COBAS((R)) AMPLICOR((R)) HIV-1 MONITOR Test, v. 1.5 and a lack of agreement between the three assays around their lower limit of quantification was revealed. CONCLUSIONS: Both real-time tests were closely comparable in the quantification of viral load specimens of ten HIV-1 subtypes and recombinant forms.