High-throughput sequencing (HTS) has unique advantages and significantly improves sensitivity in quantifying the minority HIV drug-resistant variants within HIV quasispecies [1]. Increased identification of pre-treatment minority drug resistance mutations (DRMs) compared to Sanger-based sequencing genotypic resistance testing (GRT) was reported from both resource-rich and resource-limited settings [2, 3]. The role of minority drug-resistant variants and their clinical consequences in the failure of combination antiretroviral therapy (cART) is debatable [4, 5, 14–19, 6–13]. However, the presence of the minor variants remains unclear but clinical consequences cannot be ignored.
Studies have shown that even in adherent patients, those with pre-existing Y181C mutants have a triple higher risk of virological failure on an efavirenz-based cART regimen [20]. Several studies have shown that minority pre-treatment drug resistance was associated with reduced treatment efficacy for first-generation non-nucleoside reverse transcriptase inhibitors (NNRTIs), but not for rilpivirine and integrase inhibitors (INIs) [3, 18, 21]. In contrast, other studies have indicated that in a population with a relatively low prevalence of DRM, the use of deep sequencing to detect minority HIV-1 DRM has limited clinical benefit [22]. However, in a recent conducted by Inzaule et al., reported that incorporating the minor DRMs might improve the predictive value of GRT, but that very low thresholds of minority mutations can compromise the test specificity [23]. Data on acquired minority mutations on treatment-failure patients are limited.
In sub-Saharan Africa, South Africa has 23.6% pre-treatment drug resistance to efavirenz or NVP, followed by Namibia with 13.8%, while Zimbabwe has 10.9% resistance to NVP [8]. HIV-1 subtype C (HIV-1C) is the major HIV-1 subtype in South Africa, responsible for more than 90% of infections. The recommended second-line cART consists of the nucleoside reverse transcriptase inhibitors (NRTIs) zidovudine or tenofovir and lamivudine and a ritonavir-boosted (/r) protease inhibitor (PI), usually lopinavir (LPV/r) [24, 25]. Earlier studies from South Africa and Sweden reported that despite good adherence, there is an increased risk of virological failure in patients with HIV-1C on bPI-based regimens [21, 26]. Ex vivo and in vitro experiments also indicated large variations in susceptibility of HIV-1C viruses in the absence of PI resistance-associated mutations (RAMs) [27].
Studies have reported that the rates of virological failure on second-line cART are high in resource-limited settings, including South Africa, and are associated with the duration of exposure to previous drug regimens and poor adherence [28], mostly without any protease RAM [29]. In South Africa, with more than 4.5 million HIV-infected individuals accessing cART, approximately 145 000 (∼4%) are accessing second-line cART [30]. However, the drug resistance pattern in patients failing on bPIs is limited and often described by GRT through Sanger sequencing [31]. An earlier study with only seven patients indicated the presence of PI RAMs in bPI-failure patients, which was missed by bulk Sanger sequencing [32]. In clinical settings, Sanger bulk sequencing is the most common and widely used for HIV drug resistance testing. The limitation of this method is that it can only detect variants with prevalence >20% which is well known [33–35]. Studies have described the presence of minority HIV-1 drug resistance mutations in treatment-naïve patients which could potentially impact treatment outcome [4, 14, 17, 20, 36]. Next-Generation Sequencing (NGS) method have the unique advantage of detecting of minority variants with a threshold as low as 1%; however, this method can also generate errors, so when reporting low-frequency, caution should be exercised [37, 38]. Therefore, the primary aim of the present study was to determine the level and pattern of HIV-1 drug resistance in minor (<20%) and major viral populations in patients receiving bPIs.