Objective: The aim of the study was to evaluate the use of proviral DNA as a source of viral genetic material for genotypic coreceptor tropism testing (GTT).
Methods: GTT consisted of bulk V3 sequencing followed by geno2pheno interpretation with the interpretative cut-off [false positive rate (FPR)] set at 5 and 10%. GTT was performed for 165 patients with a viral load of >500 HIV-1 RNA copies/mL on simultaneously collected plasma RNA and proviral DNA, and for 126 patients with a viral load of <500 copies/mL on current proviral DNA and pretreatment plasma RNA. Phenotypic tropism testing (PTT) results were available for 142 samples.
Results: In the simultaneous RNA/DNA comparison, concordance in prediction was 95.2% (at FPR 10%) and 96.4% (at FPR 5%). Six RNA-R5/DNA-X4 and two RNA-X4/DNA-R5 discordances were observed at an FPR of 10%, and six RNA-R5/DNA-X4 discordances were observed at an FPR of 5%. In the longitudinal RNA/DNA comparison, concordance was 88.1% (at FPR 10%) and 90.5% (at FPR 5%). Eight RNA-X4/DNA-R5 and seven RNA-R5/DNA-X4 discordances were seen at an FPR of 10%, and 10 RNA-R5/DNA-X4 and two RNA-X4/DNA-R5 discordances at an FPR of 5%. The overall concordance of RNA GTT with PTT was 82% (at FPR 10%) and 83% (at FPR 5%). The overall concordance of DNA GTT with PTT was 85% (at both 10 and 5% FPRs).
Conclusions: GTT produced highly concordant tropism predictions for proviral DNA and plasma RNA. GTT on proviral DNA offers a promising approach for tropism prediction in clinical practice, particularly for the assessment of treated patients with low or suppressed viraemia.
© 2011 British HIV Association.