|
|
|
|
|
Session 81
Poster Presentations Isolation and Quantification of HIV-1 in Biological Specimens Session Day and Time: Tuesday 1:30 - 3:30 pm Room: Hall A |
Background: HIV-1 evolution and changing strain distribution
presents a challenge to nucleic acid-based assays. Reliable patient monitoring
of viral loads should ensure the detection and accurate quantification of
genetically diverse HIV-1 strains. In this study, a well-characterized panel of
genetically diverse group M, group O, and recombinant viruses was used to
evaluate performance of four commercially available quantitative viral load
tests.
Methods: A panel of 97 HIV seropositive plasma specimens was
collected in Cameroon, South Africa and Brazil. The gag p24, pol
integrase, and env gp41 region sequences of each sample were
characterized to assign group/subtype and to assess genetic diversity at
primer/probe binding sites. HIV RNA concentrations were compared using 4
ultrasensitive viral load assays: LCx HIV RNA Quantitative (LCx) (Abbott
Laboratories), AMPLICOR HIV-1 MONITOR version 1.5 (Monitor v1.5) (Hoffman La
Roche), Versant HIV-1 RNA version 3.0 (bDNA) (Bayer Diagnostics) and NucliSens
HIV-1 QT (NucliSens) (bioMerieux).
Results: Based on phylogenetic analysis, the panel included
group M subtypes A, B, C, D, F, G, and mosaics (CRF01_AE, CRF02_AG, G/A, F/B,
and H/A) as well as group O. In this study, the LCx HIV assay quantified 97 (100%)
of the samples. In comparison, bDNA, Monitor v1.5 and NucliSens quantified 94
(96.9%), 92 (94.8%), and 86 (88.6%) of the samples, respectively. Monitor v1.5,
bDNA and NucliSens failed to detect 1 subtype C and 2 group O samples. Monitor
v1.5 and NucliSens also failed to detect 2 CRF02_AG samples while NucliSens did
not detect 6 additional samples: 1 A, 1 C, 1 G, and 3 mosaics. Relative to LCx,
Monitor v1.5 underquantified 2 CRF02_AG and 1 subtype A by more than 1 log10
RNA copies/ml. These 2 AG mosaics were underquantified by Monitor v1.5 compared
to NucliSens. Relative to Monitor v1.5 and bDNA, LCx underquantified 1 G/A
mosaic sample by more than 1 log10 RNA copies/ml. For group M
strains, mean total mismatches at primer/probe sites were 1.4 for LCx, 4.1 for
Monitor v1.5 and 7.9 for NucliSens.
Conclusions: Overall, the highest correlation for group
M-infected plasma specimens was observed between bDNA and LCx with
progressively lower correlations for Monitor v1.5 and NucliSens. However, only
the LCx assay quantified the group O samples. Of the tests evaluated,
performance of the LCx was the most subtype- and group-independent.