Background: The emergence and spread of antiretroviral drug resistant HIV-1 variants is one of the major factors associated with therapeutic failure in persons living with HIV (PLWH) as it jeopardizes the efforts to reduce the progression to AIDS. Whereas Sanger sequencing is the most appropriate conventional method for HIV drug resistance testing, it has limited capacity to detect low-abundance variants. This study assessed the suitability of next generation sequencing (NGS) to reveal low-abundance HIV-1 drug resistance mutations amongst patients experiencing virological failure at the time of therapy switching in Uganda. Methods: Archived blood samples previously collected from 60 PLWH were used in this study. Briefly HIV viral RNA was extracted and performed targeted NGS of portions of both the HIV protease and reverse transcriptase genes on the illumina MiSeq. For performance comparison, Sanger sequencing was also performed for all the samples targeting the highlighted genes. The sequence data generated was analyzed using HyDRA bioinformatics pipeline, accompanied by the Stanford HIV drug resistance database, to annotate and report drug resistance mutations/variants. Results: Out of the 60 samples, 58 passed preliminary quality control and were considered for subsequent analysis—of which 38/58 (65.5%) registered low-abundance HIV drug resistance variants. Overall, 757 variants from the NGS data and 90 variants from the Sanger data were identified. The most prevalent minority variants included; K65R (65.5%), K14R (63.8%), K45R (63.8%), L63P (63.8%), I15V (63.8%), K70R (60.3%), V77I (60.3%), L283I (60.3%), G16E (58.6%) and L282C (58.6%). Conclusion: An estimated 65.5% of the sampled population harbors low-abundance HIV-1 variants, most of which are associated with virological failure, and consequently antiviral drug resistance. NGS suitably detects drug resistance mutations even at frequencies below 20% of the viral quasi species that are occasionally missed by Sanger sequencing.
