Minor viral population with drug-resistant mutation and risk of persistent low-level viremia or ‘blips’ in HIV-1 subtype C

The recent study from the Swiss HIV cohort described a limited clinical impact of minority K103N and Y181C mutants [1], which contradicts earlier studies showing that low-frequency HIV-1 drug-resistance mutations (DRMs), specifically non-nucleoside reverse transcriptase inhibitor (NNRTI) mutations, were associated with increased risk of viral failure and decreased treatment efficacy [2–4]. However, in most of the studies, the minor mutations were detected with allele-specific PCR. In this study, sensitive high-throughput next-generation sequencing technology using Illumina MiSeq Bench-top sequencer with multiplexed amplicon sequencing was used, as described by us recently [5]. A total of 44 plasma samples obtained from patients followed at the Karolinska University Hospital were included. The patients were randomly chosen from a larger cohort of HIV-1 subtype B (HIV-1B) and subtype C (HIV-1C)-infected treatment-naïve patients at Karolinska, where all newly diagnosed patients are analysed for DRM by routine ViroSeq HIV-1 Genotyping System. No primary DRM was identified. Patients with less than years of follow-up (n = 4), treatment interruptions (n = 3), multiple treatment regimens (n = 3), lost to follow-up (n = 1), and use of ribavirin prior to therapy (n = 1) were excluded. Among the remaining 32 patients, 14 were initiated with NNRTI + two nucleoside reverse transcriptase inhibitors (NRTIs) [12 with efavirenz (EFV), one with rilpivirine (RPV) and one started with EFV and switched to RPV] and 18 with protease inhibitors + two NRTIs. The choice of the treatment was done on the discretion of the physician. Among the NNRTI-treated patients, seven each were infected with HIV-1B and HIV-1C. Seven HIV-1C and 11 HIV-1B patients, respectively, were given protease inhibitor. At baseline, primary minor NNRTI mutations (K103N, V108I, Y181C, and Y188C) were identified in three patients with HIV-1C and one with HIV-1B among those who initiated NNRTI therapy. One HIV-1C patient had the M230I mutation. Two HIV-1C patients also had M184V mutation. At follow-up, all four HIV-1C-infected patients had either low-level persistence viraemia or several viral ‘blips’ (viral load 20–1000 copies/ml). The HIV-1B-infected patient was suppressed successfully. However, four out of the nine NNRTI-treated patients without any minor mutation also had single viral ‘blips’ (two each in HIV-1B and HIV-1C). Thus, six out of the seven NNRTI-treated HIV-1C patients had at least one viral blip during a median duration of treatment of 51 months, whereas only two out of the seven HIV-1B-infected patients had viral ‘blips’ with no significant difference in the median duration of treatment (51 versus 44 months). In the protease inhibitor group, four out of the seven HIV-1C-infected patients had viral ‘blips’ or low-level viraemia, whereas four out of the 11 HIV-1B-infected patients had viral ‘blips’ (either single episode or twice). Altogether 10 out of the 14 HIV-1C patients had detectable viral RNA versus six out of the 18 HIV-1B patients during follow-up (P = 0.07). Our results are in line with the data from the Swiss cohort as none of our patients with minor mutations failed virologically (defined as viral load >1000 copies/ml) and the number of viral blips seemed not to be increased. However, the extent of viral ‘blips’ was comparatively high in HIV-1C patients. An earlier study has shown that virological and immunological responses are independent of HIV-1 subtypes, but viral rebound has been claimed to be more rapid in HIV-1C [6]. Therefore, larger and longer follow-up studies to understand the synergistic effects of viral ‘blips’ and minor primary DRMs should be prioritized, which could be feasible using cost-effective methods for detection of minor mutations such as the method we have recently developed [5]. Acknowledgements Funding: The study was partially funded by Swedish International Developing Agency, the Swedish Civil Contingencies Agency (SWE-2009–151) and the Swedish Research Council (521–2012–3476). Conflicts of interest There are no conflicts of interest.