A comparative evaluation of the performance of the Viroseq HIV-1 genotyping system v2.0 against an in-house assay using specimens from the national HIV drug resistance monitoring survey, in Zimbabwe.

Abstract

This study was carried out to assess the performance of the Viroseq HIV-1 genotyping system v2.0 (Celera, Alameda, California, USA) on Zimbabwean samples at the National Microbiology Reference Laboratory (NMRL) in Harare and to compare the results obtained with those obtained at a World Health Organization accredited genotyping laboratory in Entebbe, Uganda, which genotyped the same samples using an in-house method. Study samples: Thirty eight (38) plasma samples were used in this study. They were selected from samples stored at NMRL, which had been collected from patients eligible for commencement on antiretroviral therapy just prior to taking antiretroviral drugs at sentinel sites for the National HIV Drug Resistance Monitoring Survey in Zimbabwe. Results: Viral RNA extraction and RT-PCR reactions were successful at first attempt for all the 38 study samples. The success rates of cycle sequencing reactions using the 7 proprietary Viroseq sequencing primer mixes on the 38 samples, at first attempt were: Primer A (81.6%), primer B (86.8), primers C and G (84.2%), Primer D (10.5%), Primer F (73.7%) and Primer H (94.7%). Thirty seven out of the 38 samples (97.4%) were successfully genotyped. However, 14 (37.8%) of these pol region consensus sequences obtained were unidirectional. Thirty two out of the 37 (86.5%) Viroseq genotyped samples clustered in pairs with their corresponding sample sequences generated in Uganda with bootstrap support values ≥70 (median bootstrap support value, 97.5: range 31-99) on neighbour joining tree phylogenetic analysis. All of the genotyped study samples were found to be HIV-1 subtype C (n=37). Nineteen of the 37 (48.6%) genotyped samples had at least one drug resistance mutation detected by either Viroseq or the Ugandan in-house assay. The detected mutations as reported by the two genotyping systems were fully concordant in 11/19 (57.9%), partially concordant in 2/19 (10.5%) and discordant in 6/19 (31.6%) of these specimens with drug resistance mutations. Overall, all the reports generated by the two systems with or without any mutations were 78.4% fully concordant, 16.2% partially concordant and 5.4% discordant. Conclusions: The Viroseq system can be reliably used to genotype Zimbabwean samples by adequately trained operators at NMRL. Sequencing Primer D is not suitable for sequencing Zimbabwean subtype C isolates due to its high sequencing reaction failure rate.