Molatoli, Mhlekazi Cathrine2020-03-122020-03-122019https://hdl.handle.net/10539/29137A dissertation submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science, January 2019The development of antiretroviral (ARV) drug resistance to Protease (PR) Inhibitors (PIs) requires the emergence of both primary and secondary/ compensatory mutations in the PR enzyme. Several studies have shown that most patients virologically failing a PI based ARV drug regimen do not harbour PI drug resistant mutations (DRMs). The presence of minority HIV-1 PI variants undetected by population-based Sanger sequencing (sensitivity limitation of ≥ 15- 20%), have been hypothesised to possibily contribute to the development of PI resistance in these patients. A total of 188 participant samples previously collected from six provinces across South Africa and virologically failing a PI based second-line regimen were available for this study. Population based Sanger sequences and associated DRMs were available for all the participants. Following viral RNA extraction using the EasyMag Extraction Kit (Biomerieux, INC, France) and the NucliSeNS® EasyMag instrument (Biomerieux INC, France), RT-PCR amplification and sequencing-by-synthesis on the next generation sequencing (NGS) platform, Illumina Miseq (California, USA), was successful for 158 of the participants. Identification of all HIV-1 DRM variants was performed using both Deepchek® and Geneious® analysis tools. Following a comparison of DR variants obtained using Deepchek® and Geneious®, a minimal cut-off of ≥ 4.5% was used for the remainder of the study. NGS detected virus containing PI DRMs in 26 of the participants versus 24 participants detected by Sanger sequencing. Minority HIV-1 variants as defined at ≥ 4.5- 20.0%, were identified in viral PR for five of the 26 participants sequences by NGS and resulted in PI regimen drug resistance for only two of the five participants. Overall, NGS detected PI DRMs resulting in PI resistance for 20 participants versus 18 participants by Sanger sequencing. Thus, using NGS based genotyping results, an additional two participants would be referred to switch to a third-line ART regimen. Furthermore, the detection of minority RT inhibitor variants in three of the abovementioned 20 participant sequences resulted in a further change in the recommended third-line RT backbone. Our findings confirm previous studies and show that the majority (83.5%) of study participants that were failing a PI based regimen do so in the absence of detectable, known PI DRMs. It is likely that non-adherence to the PI based regimen may explain the observed virological failures in the absence of relevant DRMs. Alternatively; an as of yet unidentified mechanism of DR may contribute to changes in PI drug resistance or susceptibility. Future work must focus on elucidating whether additional viral mechanisms or patient non-adherence are responsible for virologic failure to a PI based regimen in the absence of PI DRMs.enThesis