Prevalence and clinical impact of minority resistant variants in patients failing an integrase inhibitor-based regimen by ultra-deep sequencing

Prevalence and clinical impact of minority resistant variants in patients failing an integrase... Abstract Background Integrase strand transfer inhibitors (INSTIs) are recommended by international guidelines as first-line therapy in antiretroviral-naive and -experienced HIV-1-infected patients. Objectives This study aimed at evaluating the prevalence at failure of INSTI-resistant variants and the impact of baseline minority resistant variants (MiRVs) on the virological response to an INSTI-based regimen. Methods Samples at failure of 134 patients failing a raltegravir-containing (n = 65), an elvitegravir-containing (n = 20) or a dolutegravir-containing (n = 49) regimen were sequenced by Sanger sequencing and ultra-deep sequencing (UDS). Baseline samples of patients with virological failure (VF) (n = 34) and of those with virological success (VS) (n = 31) under INSTI treatment were sequenced by UDS. Data were analysed using the SmartGene platform, and resistance was interpreted according to the ANRS algorithm version 27. Results At failure, the prevalence of at least one INSTI-resistant variant was 39.6% by Sanger sequencing and 57.5% by UDS, changing the interpretation of resistance in 17/134 (13%) patients. Among 53 patients harbouring at least one resistance mutation detected by both techniques, the most dominant INSTI resistance mutations were N155H (45%), Q148H/K/R (23%), T97A (19%) and Y143C (11%). There was no difference in prevalence of baseline MiRVs between patients with VF and those with VS. MiRVs found at baseline in patients with VF were not detected at failure either in majority or minority mutations. Conclusions UDS is more sensitive than Sanger sequencing at detecting INSTI MiRVs at treatment failure. The presence of MiRVs at failure could be important to the decision to switch to other INSTIs. However, there was no association between the presence of baseline MiRVs and the response to INSTI-based therapies in our study. Introduction Integrase strand transfer inhibitors (INSTIs), which act by inhibiting the HIV integrase enzyme from inserting viral DNA genome into the host cell’s chromatin, are the most recent class of antiretroviral (ARV) drugs approved for treatment of HIV-infected individuals. Current practice guidelines (US Department of Health and Human Services, European AIDS Clinical Society and French guidelines) recommend the use of this class as first-line therapy in ARV-naive and -experienced patients.1–3 However, some studies have shown selection of raltegravir- and elvitegravir-resistant strains in patients experiencing virological failure (VF).4–6 The most frequent primary resistance pathways observed in vitro, in vivo and in clinical trials are Y143R, Q148H/R/K and N155H for raltegravir, and T66I, E92Q, Q148H/K/R and N155H for elvitegravir.7,8 Dolutegravir is known to create a higher genetic barrier to resistance, and thus an accumulation of multiple mutations, possibly selected under raltegravir or elvitegravir pressure, is required to reduce susceptibility to dolutegravir.9,10 In phenotypic studies, susceptibility to dolutegravir is reduced in the presence of Q148H/K/R mutations especially when combined with L74I, E138A/K/T, G140A/C/S or N155H.7,11 However, several studies have shown that raltegravir failures were also observed in the absence of detectable resistance. For example, a Spanish cohort treated with raltegravir showed that the viruses of 50/89 (56%) patients lacked emerging mutations on the integrase gene at failure.12 In a clinical setting, a study by Fourati et al.13 on 502 patients failing a raltegravir-based regimen showed a selection at failure of resistance mutations to the INSTI class in only 39% of cases. In clinical practice, VF and the emergence of resistance mutations have rarely been reported in patients receiving dolutegravir-based therapy.14,15 It should be noted that studies such as these are often performed by traditional Sanger sequencing and thus may underestimate the prevalence of drug-resistant variants. Therefore, our first objective was to investigate the prevalence at failure of resistant variants in patients failing an INSTI-based regimen by ultra-deep sequencing (UDS). Furthermore, to better document the VF under INSTI treatment, pharmacological determinations were performed concomitantly to investigate treatment adherence in patients failing INSTI-based regimen. Our second objective was to investigate the clinical impact of baseline minority resistant variants (MiRVs) on an INSTI-based regimen, one of the major questions still under discussion. Indeed, the detection of pre-existing MiRVs was correlated with a higher risk of VF for the first generation of NNRTIs,16,17 while few other studies have failed to establish this association for NRTIs and PIs.18–20 For INSTIs, several small-scale studies have shown that MiRVs have little impact on the virological response to a raltegravir-based regimen.21,22 Methods Study population and design Sanger sequencing and UDS were performed on plasma samples at failure of 134 patients failing an INSTI-based regimen (65 failed under raltegravir, 20 under elvitegravir and 49 under dolutegravir), which were collected between January 2014 and March 2017 at Pitié-Salpêtrière, Saint-Antoine and Bichat hospitals, Paris, France. Patients were defined as having failure under an INSTI-based regimen if two consecutive viral loads were >50 copies/mL during treatment. To evaluate the clinical impact of baseline MiRVs on INSTI response, we sequenced the integrase gene by UDS and compared the MiRV presence at baseline of patients having VF with those having virological success (VS) under INSTI treatment in the same period. For the VF group, of 134 patients failing INSTIs who were enrolled, we selected 34 patients whose samples were available prior to INSTI initiation. For the VS group, we selected 31 patients initiating a first-line INSTI-based regimen, with samples available prior to INSTI initiation and undetectable viral load (<50 copies/mL) for at least 6 consecutive months. All patients signed an informed consent for the anonymous use of their clinical and biological data. RNA extraction, PCR and Sanger sequencing Samples had been previously sequenced in our laboratory during the clinical monitoring of patients. Briefly, 80 μL of HIV RNA was extracted from 1 mL of plasma (NucliSENS® easyMAG®, bioMérieux Clinical Diagnostics). The extracted RNA was reverse transcribed into cDNA and a fragment of the integrase gene (corresponding to amino acids 49–286) was amplified by PCR in a two-round process. The first round was performed with forward primer (4339–4359) 5′-TAG TAG CCA GCT GTG ATA AAT GTC-3′ and reverse primer (5082–5102) 5′-TTC CAT GTT CTA ATC CTC ATC CTG-3′ using a Transcriptor One Step RT-PCR kit (Roche Life Science, Germany) following the manufacturer’s protocol. PCR products were diluted to 1/10 and subjected to a nested round with forward primer (4374–4389) 5′-GAA GCC ATG CAT GGA CAA G-3′ and reverse primer (5072–5090) 5′-ATC CTC ATC CTG TCT ACT TGC C-3′ using the Q5® High-Fidelity PCR Kit (New England Biolabs, USA) following the manufacturer’s protocol. The PCR products were purified by Sephadex gel and sequenced using the Sanger method (BigDye Terminator, Applied Biosystems, Foster City, CA, USA). UDS and computational method Integrase amplicons were deep-sequenced using the Illumina MiSeq platform. The nested PCR products were purified by SPRIselect beads (Beckman Coulter, France), ‘tagmentated’ (fragmented and tagged) and prepared for libraries using Nextera® DNA Sample Preparation and Index Kit (Illumina, USA) according to the manufacturer’s protocol. Resulting libraries were quantified on a 2100 Bioanalyzer (Agilent Technologies), normalized and pooled equimolarly. Pooling libraries were subjected to standard Illumina paired-end sequencing at 2 × 150 bp on the MiSeq platform. UDS analyses were performed using a fully automated analysis pipeline commercially available via SmartGene (www.smartgene.com; SmartGene, Zug, Switzerland). Briefly, paired-end reads are merged and quality-filtered to remove noise. Alignment is performed using a target-specific profile and a consensus is produced based on a user-selected ambiguity threshold. Mutations are called by frame-aware alignment with reference sequence HXB2 (Los Alamos, accession number AF033819) above a user-selected threshold (range = 0.5%–30%) at a predetermined minimum coverage of reads. Resistance interpretation Interpretation of resistance was in accordance with the ANRS resistance algorithm version 27 (updated September 2017, www.hivfrenchresistance.org). Briefly, integrase gene mutations A49G, T66I/A/K, L74M/I, E92Q, T97A, G118R, F121Y, E138A/K/T, G140A/C/S, Y143A/C/G/H/R/S, P145S, S147G, Q148E/G/H/K/R, V151L, S153Y/F, N155H/S/T, E157Q, S230G/R and R263K were analysed. Mutational load was calculated based on the frequency of mutations detected by UDS and plasma viral load. Mutations were defined as majority if they accounted for ≥20% of viral population and were detected by both techniques, and as minority if they accounted for <20% of viral population and were detected only by UDS. The susceptibility of virus at baseline to ARVs received including INSTIs was evaluated by calculating the genotypic susceptibility score (GSS) based on Sanger genotypic resistance test results at baseline. Results of ‘susceptible’, ‘possible resistance’ and ‘resistance’ to an ARV were interpreted into scores of 1, 0.5 and 0, respectively. We classified patients into two groups: those with baseline GSS equal to 1 or 1.5 (considered as functional INSTI monotherapy) and those with GSS of ≥2. ARV plasma concentration measurement To evaluate each patient’s adherence to ART, trough plasma concentrations of INSTI (Ctrough at 12 or 24 h) were determined using ultra-performance LC coupled with tandem MS (Acquity UPLC/TQ24H, Waters Corp., Milford, MA, USA) as described previously.23 INSTI plasma concentrations were interpreted according to the respective in vitro protein-adjusted IC95 for WT HIV-1 (raltegravir, 15 ng/mL;24 elvitegravir, 45 ng/mL25) and according to the pharmacokinetic/pharmacodynamic relationship of the SAILING trial26 (dolutegravir, 1000 ng/mL). Results Patient characteristics Of 134 patients failing an INSTI-based regimen, 65, 20 and 49 patients failed raltegravir-, elvitegravir- and dolutegravir-based regimens, respectively. Patient characteristics are described in Table 1. Table 1. Characteristics of patients failing an INSTI-based regimen (n = 134) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; RAL, raltegravir; /r, ritonavir. a Patients received and failed their first-line INSTI-based regimen. b Patients received INSTIs previously but developed failure on their last INSTI-based regimen. Table 1. Characteristics of patients failing an INSTI-based regimen (n = 134) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; RAL, raltegravir; /r, ritonavir. a Patients received and failed their first-line INSTI-based regimen. b Patients received INSTIs previously but developed failure on their last INSTI-based regimen. Viruses at baseline of 34 patients with VF were compared with those of 31 patients with VS under INSTI treatment. Significant differences between both groups were observed for gender, viral load at baseline, nadir CD4 and use of NRTI class in association with INSTIs. Patient characteristics are shown in Table 2. Table 2. Characteristics of patients with VS and VF under INSTI treatment Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — RAL, raltegravir; EVG/c, elvitegravir/cobicistat; DTG, dolutegravir; —, not applicable. Table 2. Characteristics of patients with VS and VF under INSTI treatment Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — RAL, raltegravir; EVG/c, elvitegravir/cobicistat; DTG, dolutegravir; —, not applicable. Read coverage and technical validation of UDS A median of 62 250 reads per amplicon (IQR = 55 147–100 483) was obtained. The cellular clone 8E5, harbouring a single HIV provirus, was also amplified and sequenced following the same procedure as a control for error rate. The mean ± SD of error rate per bp (%) was 0.004 ± 0.0766. The maximal error rate obtained at one nucleotide position was 1.6% but was only observed once in the 5′ extremity where the reads coverage diminished significantly, far fewer than 5000 reads. Few data are available regarding the threshold to interpret the impact of MiRVs in a clinical context. Therefore, thresholds of 1% and 5% with a minimum coverage of 5000 reads at one nucleotide position were used to retain MiRVs for resistance interpretation. Sanger sequences were submitted to GenBank with accession numbers MH143121–MH143242 and UDS sequences were submitted to GenBank with accession number SRP137063. INSTI plasma concentration results INSTI plasma concentrations at time of treatment failure were available for 102/134 patients. Eighty-five patients (79.4%) had an effective Ctrough of the INSTIs used, while 6 (5.6%) and 11 (10.2%) patients had low level and undetectable levels of the INSTIs used, respectively. The median (IQR) Ctrough values for raltegravir q12h, dolutegravir q24h, dolutegravir q12h and elvitegravir/cobicistat on available dosage results were 107 (53.5–500), 1868 (1310–2877), 2002 (787–3171) and 472 (258–826) ng/mL, respectively (Figure 1). Figure 1. View largeDownload slide Measurement of INSTI Cmin at time of failure to INSTI based on available dosage results. DTG BID, dolutegravir q12h; DTG QD, dolutegravir q24h; EVG, elvitegravir; RAL BID, raltegravir q12h. Figure 1. View largeDownload slide Measurement of INSTI Cmin at time of failure to INSTI based on available dosage results. DTG BID, dolutegravir q12h; DTG QD, dolutegravir q24h; EVG, elvitegravir; RAL BID, raltegravir q12h. GSS at baseline based on Sanger genotypic resistance test results Sanger genotypic resistance test results were available for 130/134 patients failing INSTI-based regimens. Thirteen patients (10%) had a total GSS at baseline of 1 or 1.5, while 117 patients (90%) had a GSS of 2 (44/130, 33.8%) or >2 (73/130, 56.2%). Prevalence of majority resistant variants (MaRVs) at failure detected by both Sanger sequencing and UDS techniques Overall, viruses harboured no MaRVs and were then considered as fully susceptible to all INSTIs in 60.4% cases (n = 81/134). INSTI Ctrough measurement was available for 60 of them. Fifty of these 60 patients (83.3%) had an effective Ctrough of the INSTIs used, while 4 (6.7%) and 6 (10.0%) patients had a low level and an undetectable level of the INSTIs used, respectively. Ctrough measurement was available for 42 of 51 patients whose virus was resistant genotypically to at least one INSTI. Effective, low-level and undetectable Ctrough values were determined in 34 (81.0%), 3 (7.1%) and 5 (11.9%) patients, respectively. Genotypically, viruses of 35.8% (n = 48), 38.8% (n = 52), 24.6% (n = 33) and 4.5% (n = 6) of patients were considered to be resistant to raltegravir, elvitegravir, dolutegravir q24h and dolutegravir q12h, respectively. Regarding INSTI mutation patterns, among 53 patients detected at failure with at least one MaRV, the most dominant pathways of resistance are N155H and Q148H/K/R, which were observed in 24 (45.2%) and 12 (22.6%) patients, respectively, whereas Y143C was detected in 6 (11.3%) patients. The other INSTI mutations detected were T66A/K/I (n = 3), L74M/I (n = 12), E92Q (n = 6), T97A (n = 10), E138A/K (n = 4), G140A/C/S (n = 7), S147G (n = 4), S153Y (n = 1) and E157Q (n = 3). Q148H/K/R was accompanied by G140A/C/S in 7 cases among the 12 patients harbouring this resistance profile. No patients had the R263K mutation. No statistical difference was observed in the prevalence of MaRVs at failure in INSTI-naive and -experienced patients (37.8% and 43.2%, respectively; P = 0.729). Prevalence of MiRVs at failure additionally detected by UDS UDS additionally detected MiRVs in viruses of 9% (12/134) of patients at the 5% threshold and in 18% (24/134) of patients at the 1% threshold. Among those 24 patients, 13, 3 and 8 failed a raltegravir-, a dolutegravir- and an elvitegravir-based regimen, respectively. Ten out of the 24 patients were previously treated with an INSTI-based regimen. Considering the 1% threshold, among the 13 patients failing a raltegravir-based regimen (6 were INSTI-experienced patients), we detected the presence of MiRVs at failure, including T66A/I (3/13), E92Q (3/13), T97A (3/13), E138A/K (4/13), G140S (1/13), Y143C (2/13), S147G (1/13), N155H (3/13) and S230G (1/13). In particular, the virus of one patient carried mutations E92Q at 21.5% and N155H at 25.7%, which were detected only by UDS. Among the three patients failing an elvitegravir-based regimen, R263K was detected at a non-negligible frequency (7.9% and 20.4%) in viruses of two patients who were previously treated with raltegravir. Furthermore, in the virus of another patient naive to INSTI treatment, S230R was additionally detected by UDS at 41%, which was in association with the major mutation N155H found at 100% by both techniques. Four out of eight patients failing a dolutegravir-based regimen were previously treated with a raltegravir-containing regimen. Only two among them harboured viruses carrying MaRVs detected by both techniques, such as E138K, G140S and Q148H. UDS additionally detected the T66A as MiRVs in these two patients. Six out of 8 patients’ viruses (75%) carried solely MiRVs detected by UDS at failure, such as T66A/I (2/6), Y143C/H (2/6), S147G (1/6), Q148R (2/6) and S153F (1/6). No MiRVs detected by UDS at frequencies of ≥20% were identifiable from the Sanger sequencing chromatograms even as background noise. Overall, the presence of MiRVs led to changes in resistance interpretation of the INSTI class in 13% (17/134) of patients when taking into account mutations at the 1% threshold and in 7% (10/134) of patients at the 5% threshold. No statistical difference was observed in the prevalence of MiRVs at failure in INSTI-naive and -experienced patients (55.6% and 61.3%, respectively; P = 0.527). To verify if these MiRVs existed prior to treatment or were selected under treatment pressure, we were able to sequence viruses at baseline of 10/24 patients who carried MiRVs at failure; this sequencing was not possible for the 14 other patients due to the absence of samples or low viral load prior to INSTI treatment. Finally, no MiRVs were detected in viruses at baseline of the 10 patients for whom viral sequences were obtained. MiRVs and MaRVs at failure are shown in Table S1 (available as Supplementary data at JAC Online). Presence of MiRVs at baseline in patients with VF and in those with VS To further evaluate the impact of MiRVs on response to INSTI treatment, samples at baseline of patients with VF (n = 34) and those with VS (n = 31) were sequenced by UDS. No MaRVs were detected at baseline in either group. Among the 34 patients failing INSTI treatment, 17 (50%) harboured at failure viruses carrying INSTI MaRVs. We detected MiRVs in 5/34 (14.7%) viruses at baseline, but none of these mutations was detected either as majority or minority mutations at failure. Regarding the 31 patients who achieved VS under INSTI treatment, MiRVs were found at baseline in viruses of 4 (12.9%) patients, with one in particular harbouring Q148H at 13.1%. Statistical analysis showed no difference between both groups in terms of presence of MiRVs at baseline (P = 0.817). Detected mutations and clinical data of patients are presented in Tables 3 and 4. Table 3. Prevalence of resistant variants at baseline in patients who achieved treatment success Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; RPV, rilpivirine; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Table 3. Prevalence of resistant variants at baseline in patients who achieved treatment success Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; RPV, rilpivirine; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Table 4. Prevalence of resistant variants at baseline in patients failing an INSTI-based regimen Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective ATV, atazanavir; DRV/r, darunavir/ritonavir; DTG, dolutegravir; ETR, etravirine; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; MVC, maraviroc; RAL, raltegravir; T20, enfuvirtide; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Table 4. Prevalence of resistant variants at baseline in patients failing an INSTI-based regimen Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective ATV, atazanavir; DRV/r, darunavir/ritonavir; DTG, dolutegravir; ETR, etravirine; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; MVC, maraviroc; RAL, raltegravir; T20, enfuvirtide; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Discussion In this study, we analysed the prevalence and impact of INSTI-resistant variants on treatment outcome by a sensitive technique, UDS, compared with Sanger sequencing. In accordance with results from other studies, we detected by both techniques the presence of INSTI MaRVs at failure in only 39.6% of INSTI non-responders. Indeed, Fourati et al.13 found the same prevalence (39%) of resistant variants in their study of 502 patients failing a raltegravir-based regimen. Moreover, both studies concluded that, when resistant variants were detected, the most dominant resistant pathways evidenced were N155H and Q148H/K/R.13 In this study, the INSTI plasma concentration measurement results available (102/134 patients) showed good adherence to ART in 79% of patients. No difference in patient adherence was observed between patients whose viruses carried at least one INSTI MaRV and those without an MaRV. Effective INSTI Ctrough values in the former and the latter were 83% and 81%, respectively. To explore other factors, such as mutations in the reverse transcriptase and protease at baseline inducing resistance to associated drugs, we evaluated the susceptibility of virus at baseline to ARVs received by calculating the GSS. We found that only 10% of patients had a total GSS of 1–1.5, while 90% of patients had a total GSS of 2 (33.8%) or >2 (56.2%). This suggests that patients were treated with a solid backbone of ARVs and therefore resistance to other drug classes at baseline could not explain failure of the INSTI-based regimen. Therefore, in addition to drug resistance testing and adherence follow-up, other factors might be considered should an INSTI-based regimen fail. A recent in vitro study by Malet et al.27 has interestingly described a new mechanism of resistance to INSTI with the emergence of mutations located in the nef region of the virus. To understand and better explain therapeutic failure to INSTI-based regimen, we investigated by UDS the prevalence at failure of MiRVs in 134 INSTI non-responders. We additionally detected the presence of MiRVs in 24/134 viruses (18%) at the 1% threshold, leading to changes in INSTI resistance interpretation in 17 patients. The resistance mutations detected most frequently only by UDS were T66A/I (7/24, 29.2%), E92Q and E138A/K (both at 4/24, 16.7%), T97A, Y143C and N155H (all at 3/24, 12.5%), and S230G and R263K (both at 2/24, 8.3%). Interestingly, R263K, which was reported to be selected in vitro by elvitegravir,28,29 was detected by UDS at a non-negligible frequency (7.9% and 20.4%) in viruses of two patients failing an elvitegravir-based regimen. It should be noted that several mutations present at frequencies >20% but only detected by UDS have been described previously30,31 but the contrary is rarely reported. In this study, one mutation (E92Q) was detected by UDS at 11% but was also detected by Sanger sequencing. This may be explained either by sequencing technique or PCR bias. In our study, some low viral loads (<1000 copies/mL) may have led to possible PCR bias and hence inaccurate quantification by UDS. However, a more important question here is whether MiRVs at any frequency could, given INSTI selection pressure, turn into MaRVs and therefore contribute to INSTI VF. In fact, several studies have tried to answer this question but the clinical significance of INSTI MiRVs is still debated.18,32–34 To the best of our knowledge, few studies have conclusively established the relationship between the presence of integrase gene MiRVs and the risk of failure under INSTI treatment. A study by Charpentier et al.18 using UDS at a detection threshold of 1% reported a prevalence of 9% of raltegravir-resistant variants at baseline but no association with VF was established. Another study by Charpentier et al.34 using UDS on 27 patients receiving dolutegravir plus lamivudine treatment showed that 26% of patients harboured INSTI MiRVs in baseline DNA genotypes, although all maintained virological suppression. To better characterize this association, we used UDS to search for the presence at baseline of integrase gene MiRVs in patients with VF and in those with VS receiving INSTI treatment. We found that MiRVs at baseline were present in both groups with similar prevalence (14.7% versus 12.9% in treatment failure and treatment success groups, respectively). Moreover, MiRVs detected at baseline in 5/34 patients failing an INSTI-based regimen were not found at failure as either majority or minority mutations regardless of selection of other MaRVs at failure. These results suggest that the presence of INSTI MiRVs at baseline might not be associated with risk of VF in patients receiving INSTI treatment. However, the clinical significance of MiRVs at baseline is somehow still questionable because the presence of certain pre-existing MiRVs could probably facilitate selection of other MaRVs at failure, although the mechanism for this has not been demonstrated. Moreover, if switching to other INSTIs is considered after failure to an INSTI, the presence of MiRVs at failure could be important for choice of switch especially if patients are not treated with a fully active backbone or if their virus replicates for a long time under INSTI treatment. This study has some limitations. We determined the detection threshold of UDS based on an external control, the cellular clone 8E5 harbouring a provirus instead of a viral RNA template. The error rate could be higher than measured if the reverse transcription step is taken into account and if samples are of low viral load creating insufficient templates for sequencing. However, the two rounds of PCR were performed with highly effective and high-fidelity enzymes, which are expected to induce a very low error rate as described elsewhere (<1%).35,36 Nevertheless, based on this limitation, two thresholds (1% and 5%) of MiRV detection were presented in this study. Another limitation in this study was the low number of patients selected in the two groups of patients with VF and VS, making it difficult to draw a robust conclusion about the clinical impact of baseline MiRVs. Importantly, to ensure that MiRVs detected in patients with VS would not be affected by previous exposures to ARVs or INSTIs, we included only those initiating their first-line ART via an INSTI-based regimen. Several differences were observed in the two groups of patients, but these were coherent with our selection criteria. The same prevalence of baseline MiRVs detected in patients with VS compared with those with VF can therefore be presumed to enhance our conclusion about the small impact of baseline MiRVs on INSTI response. In conclusion, the present study showed a prevalence of 39.6% of INSTI-resistant variants by Sanger sequencing and 57.5% by UDS at the 1% threshold among 134 patients failing an INSTI-containing regimen, without significant impact on the virological outcome of an INSTI-based regimen containing raltegravir, elvitegravir or dolutegravir. Acknowledgements Orally presented in part at the Conference on Retroviruses and Opportunistic Infections, Boston, MA, USA, 2018 (Abstract 545).  We would like to thank Doris Loutan and Stefan Emler (SmartGene IDNS) for technical assistance with UDS data analysis. Funding This work was supported by the Agence Nationale de Recherches sur le SIDA et les Hépatites virales (ANRS). Transparency declarations None to declare. Supplementary data Table S1 is available as Supplementary data at JAC Online. References 1 Conseil national du sida et des hépatites virales . Prise en charge du VIH-Recommandations du groupe d’experts. Mis à jour mai 2017 . https://cns.sante.fr/actualites/prise-en-charge-du-vih-recommandations-du-groupe-dexperts/. 2 US Department of Health and Human Services . Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. 2016 . https://aidsinfo.nih.gov/guidelines/htmL/1/adult-and-adolescent-treatment-guidelines/0/. 3 European AIDS Clinical Society . European Guidelines for Treatment of HIV-Positive Adults in Europe. 2017 . http://www.eacsociety.org/guidelines/eacs-guidelines/eacs-guidelines.html. 4 Sax PE , DeJesus E , Mills A et al. Co-formulated elvitegravir, cobicistat, emtricitabine, and tenofovir versus co-formulated efavirenz, emtricitabine, and tenofovir for initial treatment of HIV-1 infection: a randomised, double-blind, phase 3 trial, analysis of results after 48 weeks . Lancet 2012 ; 379 : 2439 – 48 . Google Scholar CrossRef Search ADS PubMed 5 Marcelin A-G , Delaugerre C , Beaudoux C et al. A cohort study of treatment-experienced HIV-1-infected patients treated with raltegravir: factors associated with virological response and mutations selected at failure . Int J Antimicrob Agents 2013 ; 42 : 42 – 7 . Google Scholar CrossRef Search ADS PubMed 6 Messiaen P , Wensing AMJ , Fun A et al. Clinical use of HIV integrase inhibitors: a systematic review and meta-analysis . PLoS One 2013 ; 8 : e52562. Google Scholar CrossRef Search ADS PubMed 7 Geretti AM , Armenia D , Ceccherini-Silberstein F. Emerging patterns and implications of HIV-1 integrase inhibitor resistance . Curr Opin Infect Dis 2012 ; 25 : 677 – 86 . Google Scholar CrossRef Search ADS PubMed 8 Abram ME , Hluhanich RM , Goodman DD et al. Impact of primary elvitegravir resistance-associated mutations in HIV-1 integrase on drug susceptibility and viral replication fitness . Antimicrob Agents Chemother 2013 ; 57 : 2654 – 63 . Google Scholar CrossRef Search ADS PubMed 9 Canducci F , Ceresola ER , Boeri E et al. Cross-resistance profile of the novel integrase inhibitor Dolutegravir (S/GSK1349572) using clonal viral variants selected in patients failing raltegravir . J Infect Dis 2011 ; 204 : 1811 – 5 . Google Scholar CrossRef Search ADS PubMed 10 Van Wesenbeeck L , Rondelez E , Feyaerts M et al. Cross-resistance profile determination of two second-generation HIV-1 integrase inhibitors using a panel of recombinant viruses derived from raltegravir-treated clinical isolates . Antimicrob Agents Chemother 2011 ; 55 : 321 – 5 . Google Scholar CrossRef Search ADS PubMed 11 Kobayashi M , Yoshinaga T , Seki T et al. In vitro antiretroviral properties of S/GSK1349572, a next-generation HIV integrase inhibitor . Antimicrob Agents Chemother 2011 ; 55 : 813 – 21 . Google Scholar CrossRef Search ADS PubMed 12 Garrido C , de Mendoza C , Alvarez E et al. Plasma raltegravir exposure influences the antiviral activity and selection of resistance mutations . AIDS Res Hum Retroviruses 2012 ; 28 : 156 – 64 . Google Scholar CrossRef Search ADS PubMed 13 Fourati S , Charpentier C , Amiel C et al. Cross-resistance to elvitegravir and dolutegravir in 502 patients failing on raltegravir: a French national study of raltegravir-experienced HIV-1-infected patients . J Antimicrob Chemother 2015 ; 70 : 1507 – 12 . Google Scholar CrossRef Search ADS PubMed 14 Mesplède T , Wainberg MA. Is resistance to dolutegravir possible when this drug is used in first-line therapy? Viruses 2014 ; 6 : 3377 – 85 . Google Scholar CrossRef Search ADS PubMed 15 Oldenbuettel C , Wolf E , Ritter A et al. Dolutegravir monotherapy as treatment de-escalation in HIV-infected adults with virological control: DoluMono cohort results . Antivir Ther 2017 ; 22 : 169 – 72 . Google Scholar CrossRef Search ADS PubMed 16 Cozzi-Lepri A , Noguera-Julian M , Di Giallonardo F et al. Low-frequency drug-resistant HIV-1 and risk of virological failure to first-line NNRTI-based ART: a multicohort European case-control study using centralized ultrasensitive 454 pyrosequencing . J Antimicrob Chemother 2015 ; 70 : 930 – 40 . Google Scholar CrossRef Search ADS PubMed 17 Li JZ , Paredes R , Ribaudo HJ et al. Impact of minority nonnucleoside reverse transcriptase inhibitor resistance mutations on resistance genotype after virologic failure . J Infect Dis 2013 ; 207 : 893 – 7 . Google Scholar CrossRef Search ADS PubMed 18 Charpentier C , Lee GQ , Rodriguez C et al. Highly frequent HIV-1 minority resistant variants at baseline of the ANRS 139 TRIO trial had a limited impact on virological response . J Antimicrob Chemother 2015 ; 70 : 2090 – 6 . Google Scholar PubMed 19 Lataillade M , Chiarella J , Yang R et al. Virologic failures on initial boosted-PI regimen infrequently possess low-level variants with major PI resistance mutations by ultra-deep sequencing . PLoS One 2012 ; 7 : e30118. Google Scholar CrossRef Search ADS PubMed 20 Perrier M , Visseaux B , Landman R et al. No impact of HIV-1 protease minority resistant variants on the virological response to a first-line PI-based regimen containing darunavir or atazanavir . J Antimicrob Chemother 2018 ; 73 : 173 – 6 . Google Scholar CrossRef Search ADS PubMed 21 Li JZ , Chapman B , Charlebois P et al. Comparison of Illumina and 454 deep sequencing in participants failing raltegravir-based antiretroviral therapy . PLoS One 2014 ; 9 : e90485. Google Scholar CrossRef Search ADS PubMed 22 Armenia D , Vandenbroucke I , Fabeni L et al. Study of genotypic and phenotypic HIV-1 dynamics of integrase mutations during raltegravir treatment: a refined analysis by ultra-deep 454 pyrosequencing . J Infect Dis 2012 ; 205 : 557 – 67 . Google Scholar CrossRef Search ADS PubMed 23 Jung BH , Rezk NL , Bridges AS et al. Simultaneous determination of 17 antiretroviral drugs in human plasma for quantitative analysis with liquid chromatography-tandem mass spectrometry . Biomed Chromatogr 2007 ; 21 : 1095 – 104 . Google Scholar CrossRef Search ADS PubMed 24 Markowitz M , Morales-Ramirez JO , Nguyen B-Y et al. Antiretroviral activity, pharmacokinetics, and tolerability of MK-0518, a novel inhibitor of HIV-1 integrase, dosed as monotherapy for 10 days in treatment-naive HIV-1-infected individuals . J Acquir Immune Defic Syndr 2006 ; 43 : 509 – 15 . Google Scholar CrossRef Search ADS PubMed 25 Elliot E , Amara A , Jackson A et al. Dolutegravir and elvitegravir plasma concentrations following cessation of drug intake . J Antimicrob Chemother 2016 ; 71 : 1031 – 6 . Google Scholar CrossRef Search ADS PubMed 26 Song I , Chen S , Piscitelli S et al. Pharmacokinetics (PK) and PK-pharmacodynamic (PD) relationship of dolutegravir (DTG) in integrase inhibitor-naïve subjects. In: Abstracts of the Fifty-third Interscience Conference on Antimicrobial Agents and Chemotherapy, Denver, CO, USA, 2013. Abstract A-1573. American Society for Microbiology, Washington, DC, USA. 27 Malet I , Subra F , Charpentier C et al. Mutations located outside the integrase gene can confer resistance to HIV-1 integrase strand transfer inhibitors . MBio 2017 ; 8 : e00922-17. 28 Seki T , Suyama-Kagitani A , Kawauchi-Miki S et al. Effects of raltegravir or elvitegravir resistance signature mutations on the barrier to dolutegravir resistance in vitro . Antimicrob Agents Chemother 2015 ; 59 : 2596 – 606 . Google Scholar CrossRef Search ADS PubMed 29 Margot NA , Hluhanich RM , Jones GS et al. In vitro resistance selections using elvitegravir, raltegravir, and two metabolites of elvitegravir M1 and M4 . Antiviral Res 2012 ; 93 : 288 – 96 . Google Scholar CrossRef Search ADS PubMed 30 Fonager J , Larsson JT , Hussing C et al. Identification of minority resistance mutations in the HIV-1 integrase coding region using next generation sequencing . J Clin Virol 2015 ; 73 : 95 – 100 . Google Scholar CrossRef Search ADS PubMed 31 Mohamed S , Penaranda G , Gonzalez D et al. Comparison of ultra-deep versus Sanger sequencing detection of minority mutations on the HIV-1 drug resistance interpretations after virological failure . AIDS 2014 ; 28 : 1315 – 24 . Google Scholar CrossRef Search ADS PubMed 32 Codoñer FM , Pou C , Thielen A et al. Dynamic escape of pre-existing raltegravir-resistant HIV-1 from raltegravir selection pressure . Antiviral Res 2010 ; 88 : 281 – 6 . Google Scholar CrossRef Search ADS PubMed 33 Liu J , Miller MD , Danovich RM et al. Analysis of low-frequency mutations associated with drug resistance to raltegravir before antiretroviral treatment . Antimicrob Agents Chemother 2011 ; 55 : 1114 – 9 . Google Scholar CrossRef Search ADS PubMed 34 Charpentier C , Montes B , Perrier M et al. HIV-1 DNA ultra-deep sequencing analysis at initiation of the dual therapy dolutegravir + lamivudine in the maintenance DOLULAM pilot study . J Antimicrob Chemother 2017 ; 72 : 2831 – 6 . Google Scholar CrossRef Search ADS PubMed 35 Zanini F , Brodin J , Albert J et al. Error rates, PCR recombination, and sampling depth in HIV-1 whole genome deep sequencing . Virus Res 2017 ; 239 : 106 – 14 . Google Scholar CrossRef Search ADS PubMed 36 Raymond S , Nicot F , Jeanne N et al. Performance comparison of next-generation sequencing platforms for determining HIV-1 coreceptor use . Sci Rep 2017 ; 7 : 42215. Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

Prevalence and clinical impact of minority resistant variants in patients failing an integrase inhibitor-based regimen by ultra-deep sequencing

Loading next page...
 
/lp/ou_press/prevalence-and-clinical-impact-of-minority-resistant-variants-in-aKHPhDZ01C
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com.
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dky198
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Integrase strand transfer inhibitors (INSTIs) are recommended by international guidelines as first-line therapy in antiretroviral-naive and -experienced HIV-1-infected patients. Objectives This study aimed at evaluating the prevalence at failure of INSTI-resistant variants and the impact of baseline minority resistant variants (MiRVs) on the virological response to an INSTI-based regimen. Methods Samples at failure of 134 patients failing a raltegravir-containing (n = 65), an elvitegravir-containing (n = 20) or a dolutegravir-containing (n = 49) regimen were sequenced by Sanger sequencing and ultra-deep sequencing (UDS). Baseline samples of patients with virological failure (VF) (n = 34) and of those with virological success (VS) (n = 31) under INSTI treatment were sequenced by UDS. Data were analysed using the SmartGene platform, and resistance was interpreted according to the ANRS algorithm version 27. Results At failure, the prevalence of at least one INSTI-resistant variant was 39.6% by Sanger sequencing and 57.5% by UDS, changing the interpretation of resistance in 17/134 (13%) patients. Among 53 patients harbouring at least one resistance mutation detected by both techniques, the most dominant INSTI resistance mutations were N155H (45%), Q148H/K/R (23%), T97A (19%) and Y143C (11%). There was no difference in prevalence of baseline MiRVs between patients with VF and those with VS. MiRVs found at baseline in patients with VF were not detected at failure either in majority or minority mutations. Conclusions UDS is more sensitive than Sanger sequencing at detecting INSTI MiRVs at treatment failure. The presence of MiRVs at failure could be important to the decision to switch to other INSTIs. However, there was no association between the presence of baseline MiRVs and the response to INSTI-based therapies in our study. Introduction Integrase strand transfer inhibitors (INSTIs), which act by inhibiting the HIV integrase enzyme from inserting viral DNA genome into the host cell’s chromatin, are the most recent class of antiretroviral (ARV) drugs approved for treatment of HIV-infected individuals. Current practice guidelines (US Department of Health and Human Services, European AIDS Clinical Society and French guidelines) recommend the use of this class as first-line therapy in ARV-naive and -experienced patients.1–3 However, some studies have shown selection of raltegravir- and elvitegravir-resistant strains in patients experiencing virological failure (VF).4–6 The most frequent primary resistance pathways observed in vitro, in vivo and in clinical trials are Y143R, Q148H/R/K and N155H for raltegravir, and T66I, E92Q, Q148H/K/R and N155H for elvitegravir.7,8 Dolutegravir is known to create a higher genetic barrier to resistance, and thus an accumulation of multiple mutations, possibly selected under raltegravir or elvitegravir pressure, is required to reduce susceptibility to dolutegravir.9,10 In phenotypic studies, susceptibility to dolutegravir is reduced in the presence of Q148H/K/R mutations especially when combined with L74I, E138A/K/T, G140A/C/S or N155H.7,11 However, several studies have shown that raltegravir failures were also observed in the absence of detectable resistance. For example, a Spanish cohort treated with raltegravir showed that the viruses of 50/89 (56%) patients lacked emerging mutations on the integrase gene at failure.12 In a clinical setting, a study by Fourati et al.13 on 502 patients failing a raltegravir-based regimen showed a selection at failure of resistance mutations to the INSTI class in only 39% of cases. In clinical practice, VF and the emergence of resistance mutations have rarely been reported in patients receiving dolutegravir-based therapy.14,15 It should be noted that studies such as these are often performed by traditional Sanger sequencing and thus may underestimate the prevalence of drug-resistant variants. Therefore, our first objective was to investigate the prevalence at failure of resistant variants in patients failing an INSTI-based regimen by ultra-deep sequencing (UDS). Furthermore, to better document the VF under INSTI treatment, pharmacological determinations were performed concomitantly to investigate treatment adherence in patients failing INSTI-based regimen. Our second objective was to investigate the clinical impact of baseline minority resistant variants (MiRVs) on an INSTI-based regimen, one of the major questions still under discussion. Indeed, the detection of pre-existing MiRVs was correlated with a higher risk of VF for the first generation of NNRTIs,16,17 while few other studies have failed to establish this association for NRTIs and PIs.18–20 For INSTIs, several small-scale studies have shown that MiRVs have little impact on the virological response to a raltegravir-based regimen.21,22 Methods Study population and design Sanger sequencing and UDS were performed on plasma samples at failure of 134 patients failing an INSTI-based regimen (65 failed under raltegravir, 20 under elvitegravir and 49 under dolutegravir), which were collected between January 2014 and March 2017 at Pitié-Salpêtrière, Saint-Antoine and Bichat hospitals, Paris, France. Patients were defined as having failure under an INSTI-based regimen if two consecutive viral loads were >50 copies/mL during treatment. To evaluate the clinical impact of baseline MiRVs on INSTI response, we sequenced the integrase gene by UDS and compared the MiRV presence at baseline of patients having VF with those having virological success (VS) under INSTI treatment in the same period. For the VF group, of 134 patients failing INSTIs who were enrolled, we selected 34 patients whose samples were available prior to INSTI initiation. For the VS group, we selected 31 patients initiating a first-line INSTI-based regimen, with samples available prior to INSTI initiation and undetectable viral load (<50 copies/mL) for at least 6 consecutive months. All patients signed an informed consent for the anonymous use of their clinical and biological data. RNA extraction, PCR and Sanger sequencing Samples had been previously sequenced in our laboratory during the clinical monitoring of patients. Briefly, 80 μL of HIV RNA was extracted from 1 mL of plasma (NucliSENS® easyMAG®, bioMérieux Clinical Diagnostics). The extracted RNA was reverse transcribed into cDNA and a fragment of the integrase gene (corresponding to amino acids 49–286) was amplified by PCR in a two-round process. The first round was performed with forward primer (4339–4359) 5′-TAG TAG CCA GCT GTG ATA AAT GTC-3′ and reverse primer (5082–5102) 5′-TTC CAT GTT CTA ATC CTC ATC CTG-3′ using a Transcriptor One Step RT-PCR kit (Roche Life Science, Germany) following the manufacturer’s protocol. PCR products were diluted to 1/10 and subjected to a nested round with forward primer (4374–4389) 5′-GAA GCC ATG CAT GGA CAA G-3′ and reverse primer (5072–5090) 5′-ATC CTC ATC CTG TCT ACT TGC C-3′ using the Q5® High-Fidelity PCR Kit (New England Biolabs, USA) following the manufacturer’s protocol. The PCR products were purified by Sephadex gel and sequenced using the Sanger method (BigDye Terminator, Applied Biosystems, Foster City, CA, USA). UDS and computational method Integrase amplicons were deep-sequenced using the Illumina MiSeq platform. The nested PCR products were purified by SPRIselect beads (Beckman Coulter, France), ‘tagmentated’ (fragmented and tagged) and prepared for libraries using Nextera® DNA Sample Preparation and Index Kit (Illumina, USA) according to the manufacturer’s protocol. Resulting libraries were quantified on a 2100 Bioanalyzer (Agilent Technologies), normalized and pooled equimolarly. Pooling libraries were subjected to standard Illumina paired-end sequencing at 2 × 150 bp on the MiSeq platform. UDS analyses were performed using a fully automated analysis pipeline commercially available via SmartGene (www.smartgene.com; SmartGene, Zug, Switzerland). Briefly, paired-end reads are merged and quality-filtered to remove noise. Alignment is performed using a target-specific profile and a consensus is produced based on a user-selected ambiguity threshold. Mutations are called by frame-aware alignment with reference sequence HXB2 (Los Alamos, accession number AF033819) above a user-selected threshold (range = 0.5%–30%) at a predetermined minimum coverage of reads. Resistance interpretation Interpretation of resistance was in accordance with the ANRS resistance algorithm version 27 (updated September 2017, www.hivfrenchresistance.org). Briefly, integrase gene mutations A49G, T66I/A/K, L74M/I, E92Q, T97A, G118R, F121Y, E138A/K/T, G140A/C/S, Y143A/C/G/H/R/S, P145S, S147G, Q148E/G/H/K/R, V151L, S153Y/F, N155H/S/T, E157Q, S230G/R and R263K were analysed. Mutational load was calculated based on the frequency of mutations detected by UDS and plasma viral load. Mutations were defined as majority if they accounted for ≥20% of viral population and were detected by both techniques, and as minority if they accounted for <20% of viral population and were detected only by UDS. The susceptibility of virus at baseline to ARVs received including INSTIs was evaluated by calculating the genotypic susceptibility score (GSS) based on Sanger genotypic resistance test results at baseline. Results of ‘susceptible’, ‘possible resistance’ and ‘resistance’ to an ARV were interpreted into scores of 1, 0.5 and 0, respectively. We classified patients into two groups: those with baseline GSS equal to 1 or 1.5 (considered as functional INSTI monotherapy) and those with GSS of ≥2. ARV plasma concentration measurement To evaluate each patient’s adherence to ART, trough plasma concentrations of INSTI (Ctrough at 12 or 24 h) were determined using ultra-performance LC coupled with tandem MS (Acquity UPLC/TQ24H, Waters Corp., Milford, MA, USA) as described previously.23 INSTI plasma concentrations were interpreted according to the respective in vitro protein-adjusted IC95 for WT HIV-1 (raltegravir, 15 ng/mL;24 elvitegravir, 45 ng/mL25) and according to the pharmacokinetic/pharmacodynamic relationship of the SAILING trial26 (dolutegravir, 1000 ng/mL). Results Patient characteristics Of 134 patients failing an INSTI-based regimen, 65, 20 and 49 patients failed raltegravir-, elvitegravir- and dolutegravir-based regimens, respectively. Patient characteristics are described in Table 1. Table 1. Characteristics of patients failing an INSTI-based regimen (n = 134) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; RAL, raltegravir; /r, ritonavir. a Patients received and failed their first-line INSTI-based regimen. b Patients received INSTIs previously but developed failure on their last INSTI-based regimen. Table 1. Characteristics of patients failing an INSTI-based regimen (n = 134) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) Age (years), median (IQR) 52 (43–58) Male, n (%) 93 (69.4) Plasma HIV viral load (copies/mL) at failure, median (IQR) 459 (130–4687) CD4 cell count (cells/mm3) at failure, median (IQR) 458 (242–664) Nadir CD4 cell count (cells/mm3), median (IQR) 159 (53–255) INSTI treatment history, n (%)  naivea 90 (67.2)  experiencedb 44 (32.8) Viral subtype, n (%)  B 77 (57.5)  CRF02_AG 33 (24.6)  A1 6 (4.5)  others 18 (13.4) INSTI therapies, n (%)  RAL (400 mg q12h) 65 (48.5)  EVG/c (150/150 mg q24h) 20 (14.9)  DTG (50 mg q24h/q12h) 49 (36.6)   DTG q12h 11 (22.45)   DTG q24h 38 (77.55)   DTG in association with other ARVs 44 (89.79)   DTG monotherapy 5 (10.21) ARVs associated with INSTIs, n (%)  NRTIs 70 (52.24)  NNRTIs 12 (8.96)  PIs/r 13 (9.70)  NRTIs + PIs/r 8 (5.97)  NNRTIs + PIs 3 (2.24)  others 20 (14.93) Time to failure (months) from the beginning of treatment, median (IQR) 6 (3–15) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; RAL, raltegravir; /r, ritonavir. a Patients received and failed their first-line INSTI-based regimen. b Patients received INSTIs previously but developed failure on their last INSTI-based regimen. Viruses at baseline of 34 patients with VF were compared with those of 31 patients with VS under INSTI treatment. Significant differences between both groups were observed for gender, viral load at baseline, nadir CD4 and use of NRTI class in association with INSTIs. Patient characteristics are shown in Table 2. Table 2. Characteristics of patients with VS and VF under INSTI treatment Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — RAL, raltegravir; EVG/c, elvitegravir/cobicistat; DTG, dolutegravir; —, not applicable. Table 2. Characteristics of patients with VS and VF under INSTI treatment Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — Characteristic Total (N = 65) VS group (N = 31) VF group (N = 34) P Age (years), median (IQR) 45 (37–54) 42 (21–51) 47 (39.0–57) 0.069 Male, n (%) 45 (69.2) 30 (96.8) 15 (44.1) <0.001 HIV viral load (copies/mL) at baseline, median (IQR) 15 870 (767–95 580) 3800 (152–55 502) 715 (33–2994) <0.001 CD4 cell count (cells/mm3) at baseline, median (IQR) 350 (148–504) 350 (230–518) 360 (176–532) 0.511 Nadir CD4 cell count (cells/mm3), median (IQR) 201 (46–422) 350 (184–488) 74 (26–214) <0.001 INSTI treatment history, n (%)  naive 59 (90.8) 31 (100) 28 (82.4)  experienced 6 (9.2) 0 (0) 6 (17.6) Viral subtype, n (%) 0.275  B 30 (46.2) 17 (54.8) 13 (38.2)  CRF02_AG 18 (27.7) 8 (25.8) 10 (29.4)  A1 1 (1.5) 1 (3.2) 0 (0)  others 16 (24.6) 5 (16.2) 11 (32.4) INSTI therapies, n (%)  RAL 14 (21.5) 3 (9.7) 11 (32.4) 0.055  EVG/c 25 (38.5) 15 (48.4) 10 (29.4) 0.188  DTG 26 (40) 13 (41.9) 13 (38.2) 0.96 ARVs associated with INSTIs, n (%)  NRTIs 46 (70.8) 26 (87.1) 20 (58.8) 0.023  NNRTIs 5 (7.7) 3 (9.7) 2 (5.9) 0.914  PIs 2 (3.1) 0 (0) 2 (5.9) 0.514  NRTIs+PIs 5 (7.7) 0 (0) 5 (14.7) 0.079  others 7 (10.8) 2 (6.5) 5 (14.7) — Time (months) to failure from the beginning of treatment, median (IQR) — — 4 (2–11) — Time (months) with undetectable viral load under INSTI treatment, median (IQR) 5 (0–15) 15 (12–21) 0 (0–0) — RAL, raltegravir; EVG/c, elvitegravir/cobicistat; DTG, dolutegravir; —, not applicable. Read coverage and technical validation of UDS A median of 62 250 reads per amplicon (IQR = 55 147–100 483) was obtained. The cellular clone 8E5, harbouring a single HIV provirus, was also amplified and sequenced following the same procedure as a control for error rate. The mean ± SD of error rate per bp (%) was 0.004 ± 0.0766. The maximal error rate obtained at one nucleotide position was 1.6% but was only observed once in the 5′ extremity where the reads coverage diminished significantly, far fewer than 5000 reads. Few data are available regarding the threshold to interpret the impact of MiRVs in a clinical context. Therefore, thresholds of 1% and 5% with a minimum coverage of 5000 reads at one nucleotide position were used to retain MiRVs for resistance interpretation. Sanger sequences were submitted to GenBank with accession numbers MH143121–MH143242 and UDS sequences were submitted to GenBank with accession number SRP137063. INSTI plasma concentration results INSTI plasma concentrations at time of treatment failure were available for 102/134 patients. Eighty-five patients (79.4%) had an effective Ctrough of the INSTIs used, while 6 (5.6%) and 11 (10.2%) patients had low level and undetectable levels of the INSTIs used, respectively. The median (IQR) Ctrough values for raltegravir q12h, dolutegravir q24h, dolutegravir q12h and elvitegravir/cobicistat on available dosage results were 107 (53.5–500), 1868 (1310–2877), 2002 (787–3171) and 472 (258–826) ng/mL, respectively (Figure 1). Figure 1. View largeDownload slide Measurement of INSTI Cmin at time of failure to INSTI based on available dosage results. DTG BID, dolutegravir q12h; DTG QD, dolutegravir q24h; EVG, elvitegravir; RAL BID, raltegravir q12h. Figure 1. View largeDownload slide Measurement of INSTI Cmin at time of failure to INSTI based on available dosage results. DTG BID, dolutegravir q12h; DTG QD, dolutegravir q24h; EVG, elvitegravir; RAL BID, raltegravir q12h. GSS at baseline based on Sanger genotypic resistance test results Sanger genotypic resistance test results were available for 130/134 patients failing INSTI-based regimens. Thirteen patients (10%) had a total GSS at baseline of 1 or 1.5, while 117 patients (90%) had a GSS of 2 (44/130, 33.8%) or >2 (73/130, 56.2%). Prevalence of majority resistant variants (MaRVs) at failure detected by both Sanger sequencing and UDS techniques Overall, viruses harboured no MaRVs and were then considered as fully susceptible to all INSTIs in 60.4% cases (n = 81/134). INSTI Ctrough measurement was available for 60 of them. Fifty of these 60 patients (83.3%) had an effective Ctrough of the INSTIs used, while 4 (6.7%) and 6 (10.0%) patients had a low level and an undetectable level of the INSTIs used, respectively. Ctrough measurement was available for 42 of 51 patients whose virus was resistant genotypically to at least one INSTI. Effective, low-level and undetectable Ctrough values were determined in 34 (81.0%), 3 (7.1%) and 5 (11.9%) patients, respectively. Genotypically, viruses of 35.8% (n = 48), 38.8% (n = 52), 24.6% (n = 33) and 4.5% (n = 6) of patients were considered to be resistant to raltegravir, elvitegravir, dolutegravir q24h and dolutegravir q12h, respectively. Regarding INSTI mutation patterns, among 53 patients detected at failure with at least one MaRV, the most dominant pathways of resistance are N155H and Q148H/K/R, which were observed in 24 (45.2%) and 12 (22.6%) patients, respectively, whereas Y143C was detected in 6 (11.3%) patients. The other INSTI mutations detected were T66A/K/I (n = 3), L74M/I (n = 12), E92Q (n = 6), T97A (n = 10), E138A/K (n = 4), G140A/C/S (n = 7), S147G (n = 4), S153Y (n = 1) and E157Q (n = 3). Q148H/K/R was accompanied by G140A/C/S in 7 cases among the 12 patients harbouring this resistance profile. No patients had the R263K mutation. No statistical difference was observed in the prevalence of MaRVs at failure in INSTI-naive and -experienced patients (37.8% and 43.2%, respectively; P = 0.729). Prevalence of MiRVs at failure additionally detected by UDS UDS additionally detected MiRVs in viruses of 9% (12/134) of patients at the 5% threshold and in 18% (24/134) of patients at the 1% threshold. Among those 24 patients, 13, 3 and 8 failed a raltegravir-, a dolutegravir- and an elvitegravir-based regimen, respectively. Ten out of the 24 patients were previously treated with an INSTI-based regimen. Considering the 1% threshold, among the 13 patients failing a raltegravir-based regimen (6 were INSTI-experienced patients), we detected the presence of MiRVs at failure, including T66A/I (3/13), E92Q (3/13), T97A (3/13), E138A/K (4/13), G140S (1/13), Y143C (2/13), S147G (1/13), N155H (3/13) and S230G (1/13). In particular, the virus of one patient carried mutations E92Q at 21.5% and N155H at 25.7%, which were detected only by UDS. Among the three patients failing an elvitegravir-based regimen, R263K was detected at a non-negligible frequency (7.9% and 20.4%) in viruses of two patients who were previously treated with raltegravir. Furthermore, in the virus of another patient naive to INSTI treatment, S230R was additionally detected by UDS at 41%, which was in association with the major mutation N155H found at 100% by both techniques. Four out of eight patients failing a dolutegravir-based regimen were previously treated with a raltegravir-containing regimen. Only two among them harboured viruses carrying MaRVs detected by both techniques, such as E138K, G140S and Q148H. UDS additionally detected the T66A as MiRVs in these two patients. Six out of 8 patients’ viruses (75%) carried solely MiRVs detected by UDS at failure, such as T66A/I (2/6), Y143C/H (2/6), S147G (1/6), Q148R (2/6) and S153F (1/6). No MiRVs detected by UDS at frequencies of ≥20% were identifiable from the Sanger sequencing chromatograms even as background noise. Overall, the presence of MiRVs led to changes in resistance interpretation of the INSTI class in 13% (17/134) of patients when taking into account mutations at the 1% threshold and in 7% (10/134) of patients at the 5% threshold. No statistical difference was observed in the prevalence of MiRVs at failure in INSTI-naive and -experienced patients (55.6% and 61.3%, respectively; P = 0.527). To verify if these MiRVs existed prior to treatment or were selected under treatment pressure, we were able to sequence viruses at baseline of 10/24 patients who carried MiRVs at failure; this sequencing was not possible for the 14 other patients due to the absence of samples or low viral load prior to INSTI treatment. Finally, no MiRVs were detected in viruses at baseline of the 10 patients for whom viral sequences were obtained. MiRVs and MaRVs at failure are shown in Table S1 (available as Supplementary data at JAC Online). Presence of MiRVs at baseline in patients with VF and in those with VS To further evaluate the impact of MiRVs on response to INSTI treatment, samples at baseline of patients with VF (n = 34) and those with VS (n = 31) were sequenced by UDS. No MaRVs were detected at baseline in either group. Among the 34 patients failing INSTI treatment, 17 (50%) harboured at failure viruses carrying INSTI MaRVs. We detected MiRVs in 5/34 (14.7%) viruses at baseline, but none of these mutations was detected either as majority or minority mutations at failure. Regarding the 31 patients who achieved VS under INSTI treatment, MiRVs were found at baseline in viruses of 4 (12.9%) patients, with one in particular harbouring Q148H at 13.1%. Statistical analysis showed no difference between both groups in terms of presence of MiRVs at baseline (P = 0.817). Detected mutations and clinical data of patients are presented in Tables 3 and 4. Table 3. Prevalence of resistant variants at baseline in patients who achieved treatment success Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; RPV, rilpivirine; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Table 3. Prevalence of resistant variants at baseline in patients who achieved treatment success Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) Patient Current treatment Previous INSTI treatment Viral load at baseline (copies/mL) Duration with undetectable viral load (months) Resistant variants at baseline (frequency in %; mutational load in copies/mL) 011 EVG/c/FTC/TDF none 59 261 6 T97A (1.1; 652) 016 EVG/c/FTC/TDF none 1515 28 Q148H (13.1; 198) 024 DTG q24h/RPV none 11 278 17 R263K (3; 338) 034 EVG/c/FTC/TDF none 64 334 27 E157Q (1.3; 836) DTG, dolutegravir; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; RPV, rilpivirine; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Table 4. Prevalence of resistant variants at baseline in patients failing an INSTI-based regimen Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective ATV, atazanavir; DRV/r, darunavir/ritonavir; DTG, dolutegravir; ETR, etravirine; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; MVC, maraviroc; RAL, raltegravir; T20, enfuvirtide; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Table 4. Prevalence of resistant variants at baseline in patients failing an INSTI-based regimen Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective Patient Treatment at failure Previous INSTI treatment Working viral load at baseline (copies/mL) Viral load at failure (copies/mL) Resistant variants at baseline (frequency in %; mutational load) Resistant variants at failure (frequency in %; mutational load) INSTI Ctrough at failure (ng/mL); interpretation 038 DTG q24h EVG/c/FTC/TDF 62 173 L74I (100; 62), S230G (28.3; 17) L74I (100; 173), E92Q (11; 19) 1814; effective 072 EVG/c/FTC/TDF ETR/RAL/TDF 155 105 7496 N155S (6.6; 10236) E92Q (100; 155 105) <10; no use of treatment 087 ATV/DRV/r/DTG q12h/T20/TDF DRV/r/FTC/ RAL/TDF 4181 1403 A49G (1.4; 59), T66I (2.3; 96), E157Q (1.2; 50) T97A (100; 4181), G140S (100; 4181), Q148H (100; 4181) 2736; effective 089 ETR/RAL q12h none 775 1000 S147G (1.8; 14) N155H (100; 775) 91; effective 110 DRV/r/DTG q24h/ MVC none 2599 69 S230G (11.5; 299) none 1178; effective ATV, atazanavir; DRV/r, darunavir/ritonavir; DTG, dolutegravir; ETR, etravirine; EVG/c, elvitegravir/cobicistat; FTC, emtricitabine; MVC, maraviroc; RAL, raltegravir; T20, enfuvirtide; TDF, tenofovir disoproxil fumarate. Mutations in bold: mutations detected only by UDS. Discussion In this study, we analysed the prevalence and impact of INSTI-resistant variants on treatment outcome by a sensitive technique, UDS, compared with Sanger sequencing. In accordance with results from other studies, we detected by both techniques the presence of INSTI MaRVs at failure in only 39.6% of INSTI non-responders. Indeed, Fourati et al.13 found the same prevalence (39%) of resistant variants in their study of 502 patients failing a raltegravir-based regimen. Moreover, both studies concluded that, when resistant variants were detected, the most dominant resistant pathways evidenced were N155H and Q148H/K/R.13 In this study, the INSTI plasma concentration measurement results available (102/134 patients) showed good adherence to ART in 79% of patients. No difference in patient adherence was observed between patients whose viruses carried at least one INSTI MaRV and those without an MaRV. Effective INSTI Ctrough values in the former and the latter were 83% and 81%, respectively. To explore other factors, such as mutations in the reverse transcriptase and protease at baseline inducing resistance to associated drugs, we evaluated the susceptibility of virus at baseline to ARVs received by calculating the GSS. We found that only 10% of patients had a total GSS of 1–1.5, while 90% of patients had a total GSS of 2 (33.8%) or >2 (56.2%). This suggests that patients were treated with a solid backbone of ARVs and therefore resistance to other drug classes at baseline could not explain failure of the INSTI-based regimen. Therefore, in addition to drug resistance testing and adherence follow-up, other factors might be considered should an INSTI-based regimen fail. A recent in vitro study by Malet et al.27 has interestingly described a new mechanism of resistance to INSTI with the emergence of mutations located in the nef region of the virus. To understand and better explain therapeutic failure to INSTI-based regimen, we investigated by UDS the prevalence at failure of MiRVs in 134 INSTI non-responders. We additionally detected the presence of MiRVs in 24/134 viruses (18%) at the 1% threshold, leading to changes in INSTI resistance interpretation in 17 patients. The resistance mutations detected most frequently only by UDS were T66A/I (7/24, 29.2%), E92Q and E138A/K (both at 4/24, 16.7%), T97A, Y143C and N155H (all at 3/24, 12.5%), and S230G and R263K (both at 2/24, 8.3%). Interestingly, R263K, which was reported to be selected in vitro by elvitegravir,28,29 was detected by UDS at a non-negligible frequency (7.9% and 20.4%) in viruses of two patients failing an elvitegravir-based regimen. It should be noted that several mutations present at frequencies >20% but only detected by UDS have been described previously30,31 but the contrary is rarely reported. In this study, one mutation (E92Q) was detected by UDS at 11% but was also detected by Sanger sequencing. This may be explained either by sequencing technique or PCR bias. In our study, some low viral loads (<1000 copies/mL) may have led to possible PCR bias and hence inaccurate quantification by UDS. However, a more important question here is whether MiRVs at any frequency could, given INSTI selection pressure, turn into MaRVs and therefore contribute to INSTI VF. In fact, several studies have tried to answer this question but the clinical significance of INSTI MiRVs is still debated.18,32–34 To the best of our knowledge, few studies have conclusively established the relationship between the presence of integrase gene MiRVs and the risk of failure under INSTI treatment. A study by Charpentier et al.18 using UDS at a detection threshold of 1% reported a prevalence of 9% of raltegravir-resistant variants at baseline but no association with VF was established. Another study by Charpentier et al.34 using UDS on 27 patients receiving dolutegravir plus lamivudine treatment showed that 26% of patients harboured INSTI MiRVs in baseline DNA genotypes, although all maintained virological suppression. To better characterize this association, we used UDS to search for the presence at baseline of integrase gene MiRVs in patients with VF and in those with VS receiving INSTI treatment. We found that MiRVs at baseline were present in both groups with similar prevalence (14.7% versus 12.9% in treatment failure and treatment success groups, respectively). Moreover, MiRVs detected at baseline in 5/34 patients failing an INSTI-based regimen were not found at failure as either majority or minority mutations regardless of selection of other MaRVs at failure. These results suggest that the presence of INSTI MiRVs at baseline might not be associated with risk of VF in patients receiving INSTI treatment. However, the clinical significance of MiRVs at baseline is somehow still questionable because the presence of certain pre-existing MiRVs could probably facilitate selection of other MaRVs at failure, although the mechanism for this has not been demonstrated. Moreover, if switching to other INSTIs is considered after failure to an INSTI, the presence of MiRVs at failure could be important for choice of switch especially if patients are not treated with a fully active backbone or if their virus replicates for a long time under INSTI treatment. This study has some limitations. We determined the detection threshold of UDS based on an external control, the cellular clone 8E5 harbouring a provirus instead of a viral RNA template. The error rate could be higher than measured if the reverse transcription step is taken into account and if samples are of low viral load creating insufficient templates for sequencing. However, the two rounds of PCR were performed with highly effective and high-fidelity enzymes, which are expected to induce a very low error rate as described elsewhere (<1%).35,36 Nevertheless, based on this limitation, two thresholds (1% and 5%) of MiRV detection were presented in this study. Another limitation in this study was the low number of patients selected in the two groups of patients with VF and VS, making it difficult to draw a robust conclusion about the clinical impact of baseline MiRVs. Importantly, to ensure that MiRVs detected in patients with VS would not be affected by previous exposures to ARVs or INSTIs, we included only those initiating their first-line ART via an INSTI-based regimen. Several differences were observed in the two groups of patients, but these were coherent with our selection criteria. The same prevalence of baseline MiRVs detected in patients with VS compared with those with VF can therefore be presumed to enhance our conclusion about the small impact of baseline MiRVs on INSTI response. In conclusion, the present study showed a prevalence of 39.6% of INSTI-resistant variants by Sanger sequencing and 57.5% by UDS at the 1% threshold among 134 patients failing an INSTI-containing regimen, without significant impact on the virological outcome of an INSTI-based regimen containing raltegravir, elvitegravir or dolutegravir. Acknowledgements Orally presented in part at the Conference on Retroviruses and Opportunistic Infections, Boston, MA, USA, 2018 (Abstract 545).  We would like to thank Doris Loutan and Stefan Emler (SmartGene IDNS) for technical assistance with UDS data analysis. Funding This work was supported by the Agence Nationale de Recherches sur le SIDA et les Hépatites virales (ANRS). Transparency declarations None to declare. Supplementary data Table S1 is available as Supplementary data at JAC Online. References 1 Conseil national du sida et des hépatites virales . Prise en charge du VIH-Recommandations du groupe d’experts. Mis à jour mai 2017 . https://cns.sante.fr/actualites/prise-en-charge-du-vih-recommandations-du-groupe-dexperts/. 2 US Department of Health and Human Services . Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. 2016 . https://aidsinfo.nih.gov/guidelines/htmL/1/adult-and-adolescent-treatment-guidelines/0/. 3 European AIDS Clinical Society . European Guidelines for Treatment of HIV-Positive Adults in Europe. 2017 . http://www.eacsociety.org/guidelines/eacs-guidelines/eacs-guidelines.html. 4 Sax PE , DeJesus E , Mills A et al. Co-formulated elvitegravir, cobicistat, emtricitabine, and tenofovir versus co-formulated efavirenz, emtricitabine, and tenofovir for initial treatment of HIV-1 infection: a randomised, double-blind, phase 3 trial, analysis of results after 48 weeks . Lancet 2012 ; 379 : 2439 – 48 . Google Scholar CrossRef Search ADS PubMed 5 Marcelin A-G , Delaugerre C , Beaudoux C et al. A cohort study of treatment-experienced HIV-1-infected patients treated with raltegravir: factors associated with virological response and mutations selected at failure . Int J Antimicrob Agents 2013 ; 42 : 42 – 7 . Google Scholar CrossRef Search ADS PubMed 6 Messiaen P , Wensing AMJ , Fun A et al. Clinical use of HIV integrase inhibitors: a systematic review and meta-analysis . PLoS One 2013 ; 8 : e52562. Google Scholar CrossRef Search ADS PubMed 7 Geretti AM , Armenia D , Ceccherini-Silberstein F. Emerging patterns and implications of HIV-1 integrase inhibitor resistance . Curr Opin Infect Dis 2012 ; 25 : 677 – 86 . Google Scholar CrossRef Search ADS PubMed 8 Abram ME , Hluhanich RM , Goodman DD et al. Impact of primary elvitegravir resistance-associated mutations in HIV-1 integrase on drug susceptibility and viral replication fitness . Antimicrob Agents Chemother 2013 ; 57 : 2654 – 63 . Google Scholar CrossRef Search ADS PubMed 9 Canducci F , Ceresola ER , Boeri E et al. Cross-resistance profile of the novel integrase inhibitor Dolutegravir (S/GSK1349572) using clonal viral variants selected in patients failing raltegravir . J Infect Dis 2011 ; 204 : 1811 – 5 . Google Scholar CrossRef Search ADS PubMed 10 Van Wesenbeeck L , Rondelez E , Feyaerts M et al. Cross-resistance profile determination of two second-generation HIV-1 integrase inhibitors using a panel of recombinant viruses derived from raltegravir-treated clinical isolates . Antimicrob Agents Chemother 2011 ; 55 : 321 – 5 . Google Scholar CrossRef Search ADS PubMed 11 Kobayashi M , Yoshinaga T , Seki T et al. In vitro antiretroviral properties of S/GSK1349572, a next-generation HIV integrase inhibitor . Antimicrob Agents Chemother 2011 ; 55 : 813 – 21 . Google Scholar CrossRef Search ADS PubMed 12 Garrido C , de Mendoza C , Alvarez E et al. Plasma raltegravir exposure influences the antiviral activity and selection of resistance mutations . AIDS Res Hum Retroviruses 2012 ; 28 : 156 – 64 . Google Scholar CrossRef Search ADS PubMed 13 Fourati S , Charpentier C , Amiel C et al. Cross-resistance to elvitegravir and dolutegravir in 502 patients failing on raltegravir: a French national study of raltegravir-experienced HIV-1-infected patients . J Antimicrob Chemother 2015 ; 70 : 1507 – 12 . Google Scholar CrossRef Search ADS PubMed 14 Mesplède T , Wainberg MA. Is resistance to dolutegravir possible when this drug is used in first-line therapy? Viruses 2014 ; 6 : 3377 – 85 . Google Scholar CrossRef Search ADS PubMed 15 Oldenbuettel C , Wolf E , Ritter A et al. Dolutegravir monotherapy as treatment de-escalation in HIV-infected adults with virological control: DoluMono cohort results . Antivir Ther 2017 ; 22 : 169 – 72 . Google Scholar CrossRef Search ADS PubMed 16 Cozzi-Lepri A , Noguera-Julian M , Di Giallonardo F et al. Low-frequency drug-resistant HIV-1 and risk of virological failure to first-line NNRTI-based ART: a multicohort European case-control study using centralized ultrasensitive 454 pyrosequencing . J Antimicrob Chemother 2015 ; 70 : 930 – 40 . Google Scholar CrossRef Search ADS PubMed 17 Li JZ , Paredes R , Ribaudo HJ et al. Impact of minority nonnucleoside reverse transcriptase inhibitor resistance mutations on resistance genotype after virologic failure . J Infect Dis 2013 ; 207 : 893 – 7 . Google Scholar CrossRef Search ADS PubMed 18 Charpentier C , Lee GQ , Rodriguez C et al. Highly frequent HIV-1 minority resistant variants at baseline of the ANRS 139 TRIO trial had a limited impact on virological response . J Antimicrob Chemother 2015 ; 70 : 2090 – 6 . Google Scholar PubMed 19 Lataillade M , Chiarella J , Yang R et al. Virologic failures on initial boosted-PI regimen infrequently possess low-level variants with major PI resistance mutations by ultra-deep sequencing . PLoS One 2012 ; 7 : e30118. Google Scholar CrossRef Search ADS PubMed 20 Perrier M , Visseaux B , Landman R et al. No impact of HIV-1 protease minority resistant variants on the virological response to a first-line PI-based regimen containing darunavir or atazanavir . J Antimicrob Chemother 2018 ; 73 : 173 – 6 . Google Scholar CrossRef Search ADS PubMed 21 Li JZ , Chapman B , Charlebois P et al. Comparison of Illumina and 454 deep sequencing in participants failing raltegravir-based antiretroviral therapy . PLoS One 2014 ; 9 : e90485. Google Scholar CrossRef Search ADS PubMed 22 Armenia D , Vandenbroucke I , Fabeni L et al. Study of genotypic and phenotypic HIV-1 dynamics of integrase mutations during raltegravir treatment: a refined analysis by ultra-deep 454 pyrosequencing . J Infect Dis 2012 ; 205 : 557 – 67 . Google Scholar CrossRef Search ADS PubMed 23 Jung BH , Rezk NL , Bridges AS et al. Simultaneous determination of 17 antiretroviral drugs in human plasma for quantitative analysis with liquid chromatography-tandem mass spectrometry . Biomed Chromatogr 2007 ; 21 : 1095 – 104 . Google Scholar CrossRef Search ADS PubMed 24 Markowitz M , Morales-Ramirez JO , Nguyen B-Y et al. Antiretroviral activity, pharmacokinetics, and tolerability of MK-0518, a novel inhibitor of HIV-1 integrase, dosed as monotherapy for 10 days in treatment-naive HIV-1-infected individuals . J Acquir Immune Defic Syndr 2006 ; 43 : 509 – 15 . Google Scholar CrossRef Search ADS PubMed 25 Elliot E , Amara A , Jackson A et al. Dolutegravir and elvitegravir plasma concentrations following cessation of drug intake . J Antimicrob Chemother 2016 ; 71 : 1031 – 6 . Google Scholar CrossRef Search ADS PubMed 26 Song I , Chen S , Piscitelli S et al. Pharmacokinetics (PK) and PK-pharmacodynamic (PD) relationship of dolutegravir (DTG) in integrase inhibitor-naïve subjects. In: Abstracts of the Fifty-third Interscience Conference on Antimicrobial Agents and Chemotherapy, Denver, CO, USA, 2013. Abstract A-1573. American Society for Microbiology, Washington, DC, USA. 27 Malet I , Subra F , Charpentier C et al. Mutations located outside the integrase gene can confer resistance to HIV-1 integrase strand transfer inhibitors . MBio 2017 ; 8 : e00922-17. 28 Seki T , Suyama-Kagitani A , Kawauchi-Miki S et al. Effects of raltegravir or elvitegravir resistance signature mutations on the barrier to dolutegravir resistance in vitro . Antimicrob Agents Chemother 2015 ; 59 : 2596 – 606 . Google Scholar CrossRef Search ADS PubMed 29 Margot NA , Hluhanich RM , Jones GS et al. In vitro resistance selections using elvitegravir, raltegravir, and two metabolites of elvitegravir M1 and M4 . Antiviral Res 2012 ; 93 : 288 – 96 . Google Scholar CrossRef Search ADS PubMed 30 Fonager J , Larsson JT , Hussing C et al. Identification of minority resistance mutations in the HIV-1 integrase coding region using next generation sequencing . J Clin Virol 2015 ; 73 : 95 – 100 . Google Scholar CrossRef Search ADS PubMed 31 Mohamed S , Penaranda G , Gonzalez D et al. Comparison of ultra-deep versus Sanger sequencing detection of minority mutations on the HIV-1 drug resistance interpretations after virological failure . AIDS 2014 ; 28 : 1315 – 24 . Google Scholar CrossRef Search ADS PubMed 32 Codoñer FM , Pou C , Thielen A et al. Dynamic escape of pre-existing raltegravir-resistant HIV-1 from raltegravir selection pressure . Antiviral Res 2010 ; 88 : 281 – 6 . Google Scholar CrossRef Search ADS PubMed 33 Liu J , Miller MD , Danovich RM et al. Analysis of low-frequency mutations associated with drug resistance to raltegravir before antiretroviral treatment . Antimicrob Agents Chemother 2011 ; 55 : 1114 – 9 . Google Scholar CrossRef Search ADS PubMed 34 Charpentier C , Montes B , Perrier M et al. HIV-1 DNA ultra-deep sequencing analysis at initiation of the dual therapy dolutegravir + lamivudine in the maintenance DOLULAM pilot study . J Antimicrob Chemother 2017 ; 72 : 2831 – 6 . Google Scholar CrossRef Search ADS PubMed 35 Zanini F , Brodin J , Albert J et al. Error rates, PCR recombination, and sampling depth in HIV-1 whole genome deep sequencing . Virus Res 2017 ; 239 : 106 – 14 . Google Scholar CrossRef Search ADS PubMed 36 Raymond S , Nicot F , Jeanne N et al. Performance comparison of next-generation sequencing platforms for determining HIV-1 coreceptor use . Sci Rep 2017 ; 7 : 42215. Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Jun 4, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off