An apparent paradox: resistance mutations in HIV-1 DNA predict improved virological responses to antiretroviral therapy

An apparent paradox: resistance mutations in HIV-1 DNA predict improved virological responses to... Abstract Background In sub-Saharan Africa, detecting resistance-associated mutations (RAMs) at failure of first-line ART with two NRTIs plus an NNRTI predicts improved virological responses to second-line therapy with two NRTIs plus a ritonavir-boosted PI (PI/r). This indicates residual NRTI activity in the presence of RAMs, although additional factors may contribute to the effect. Objectives The aim of this study was to investigate the influence of pre-existing RAMs on the outcomes of maintenance monotherapy with ritonavir-boosted darunavir within a randomized trial in Cameroon. Methods RAMs were detected in HIV-1 DNA using PBMCs collected at initiation of darunavir/ritonavir monotherapy. Adherence was assessed by pill count and visual analogue scale (VAS). Predictors of virological failure (confirmed or last available viral load >400 copies/mL) were explored by logistic regression analysis. Trial name = MANET (NCT02155101). Results After NNRTI-based therapy, participants (n = 81) had received PI/r-based therapy for a median of 3.2 years and had a confirmed viral load <60 copies/mL and a median CD4 count of 466 cells/mm3. NRTI and NNRTI RAMs were detected in 39/60 (65.0%) and 41/60 (68.3%) HIV-1 DNA sequences, respectively. Over 48 weeks of monotherapy, 16/81 (19.8%) patients experienced virological failure. After adjusting for age, HIV-1 DNA load, adherence by VAS and RAM status, virological failure was less likely with higher VAS-measured adherence (adjusted OR 0.04, 95% CI 0.01–0.37; P = 0.004) and detectable HIV-1 DNA RAMs (adjusted OR 0.15, 95% CI 0.03–0.82; P = 0.028). Conclusions Pre-existing NRTI and NNRTI RAMs are associated with improved virological responses to NRTI-sparing ART in sub-Saharan Africa, indicating a predictive effect that is independent of residual NRTI activity. Introduction The WHO recommends standardized regimens for first- and second-line ART in sub-Saharan Africa, comprising two NRTIs plus an NNRTI, a ritonavir-boosted PI (PI/r) or, more recently, an integrase inhibitor.1 Pooled estimates indicate adherence rates of 67%–77% in the region,2,3 and rates of virological suppression of 65% and 62% after 24 months of NNRTI-based first-line ART4 and PI/r-based second-line5 ART, respectively. Sparse virological monitoring and delayed introduction of second-line ART in viraemic patients favour the emergence of drug resistance, and resistance-associated mutations (RAMs) affecting both the NRTIs and the NNRTIs are common at failure of first-line NNRTI-based ART.6–10 The significance of NRTI RAMs in terms of predicting reduced responses to second-line NRTI-containing ART has been called into question.5,11 In a meta-analysis of pooled data from sub-Saharan Africa, we observed that detection of RAMs at failure of NNRTI-based ART, most commonly M184V and thymidine analogue mutations (TAMs), predicted higher (rather than lower) odds of virological suppression after starting two NRTIs plus a PI/r.5 Thus, it has been inferred that resistance testing underestimates the continued activity of NRTIs such as lamivudine and tenofovir in the presence of M184V and TAMs. Here we report the same predictive effect in patients receiving NRTI-sparing ART. Patients and methods Study population MANET (Monotherapy in Africa, New Evaluations of Treatment) was a randomized, open-label trial based at Hôpital Central Yaoundé (NCT02155101). Between August 2014 and July 2015, 120 virologically suppressed patients on two NRTIs + PI/r either switched to darunavir/ritonavir monotherapy (800/100 mg once daily) for 48 weeks (n = 81) or continued two NRTIs + PI/r for 24 weeks (n = 39). Eligibility criteria comprised receiving two NRTIs + PI/r for ≥12 weeks, CD4 count >100 cells/mm3, viral load <60 copies/mL at two measurements 4–12 weeks apart (median 7 weeks) and hepatitis B surface antigen (HBsAg) negative. This analysis focused on patients receiving monotherapy. Ethics Approval was granted by the University of Liverpool Ethics Committee (RETH000605) and the Cameroon National Ethics Committee (2013/07/347). Participants gave written informed consent. Adherence Participants collected a prescription from the pharmacy every 4 weeks and attended study visits at weeks 4, 12, 24, 36 and 48. At each pharmacy visit, the number of left over pills was counted to infer the percentage of pills taken. At each study visit, adherence was self-reported via a visual analogue scale (VAS) graded from 0% (complete non-adherence) to 100% (complete adherence) in 10% increments.10 Laboratory testing Safety parameters, CD4 cell counts and viral load (Biocentric, Bandol, France; lower limit of quantification 60 copies/mL) were measured at the Centre Pasteur of Cameroon in Yaoundé. In the UK, total HIV-1 DNA was quantified in PBMCs as previously described.12 RAMs were detected by Sanger sequencing of protease (amino acids 1–99) and reverse transcriptase (amino acids 1–335) as previously described,13 and defined by the Stanford HIV drug resistance algorithm (v8.4). Darunavir RAMs comprised V11I, V32I, L33F, I47V, I50V, I54L/M, T74P, L76V, I84V and L89V. Analyses The characteristics of patients with or without HIV-1 DNA sequencing results were compared using Wilcoxon rank-sum test, t-test or χ2 test. Pill counts and VAS percentages were averaged in each patient over the entire study period and compared by Spearman’s rank correlation. Patients with virological rebound were recalled for adherence counselling, followed by repeat testing 4 weeks later. Mean adherence levels in the period before versus the period after the first counselling session were compared by t-test. Treatment failure comprised virological failure (either confirmed or last available viral load >400 copies/mL) or discontinuation of monotherapy prior to week 48 due to a significant adverse event, death or loss to follow-up. Predictors of virological failure were explored by logistic regression analysis; all available variables were considered for inclusion and those with P ≤ 0.2 in the univariate analysis were retained in the multivariable model. Results HIV-1 DNA RAMs at study entry The population is described in Table 1. Of 76 patients with PBMC samples, 60 (78.9%) yielded a HIV-1 DNA sequence. Relative to patients with sequencing results, those without had a lower HIV-1 DNA load (P = 0.0004) and a higher nadir CD4 cell count (P = 0.05) (Table S1, available as Supplementary data at JAC Online). NRTI and NNRTI RAMs occurred in 39/60 (65.0%) and 41/60 (68.3%) samples, respectively (Table 1); 37/60 (61.7%) samples had dual-class mutations. Table 1. Characteristics of the study population at study entry (N = 81) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) GSS, genotypic susceptibility score. a GSSs were determined from HIV-1 DNA sequencing results using the Stanford HIV drug resistance algorithm (v8.4) and assigning to each drug in the regimen 0 for high-level resistance, 0.25 for intermediate resistance, 0.5 for low-level resistance and 1 for potential low-level resistance/full susceptibility. b Five patients did not have a PBMC sample for sequencing; 16 samples did not yield an HIV-1 DNA sequence in ≥2 attempts and no further attempts were possible due to limited sample volume. c HIV-1 subtypes were determined phylogenetically from HIV-1 DNA sequences; other subtypes comprised CRF11_cpx (n = 3), and subtypes D, H, F1, CRF01_AE, CRF37_cpx, CRF06_cpx and CRF18_cpx (one of each). d NRTI and NNRTI RAMs of 60 patients with HIV-1 DNA sequencing results are detailed; 3 patients had D30N in protease but no record of exposure to a PI/r other than lopinavir/ritonavir and including nelfinavir; 11 reverse transcriptase sequences showed stop codons (positions 24, 42, 48, 71, 88, 120, 153, 212, 219, 229 and 239) suggestive of defective proviruses; among 29 patients with ≥1 TAM, the median number of TAMs was 2 (range 1–5); TAM-1 profile comprised ≥1 of (M41L, L210W, T215Y); TAM-2 profile comprised ≥1 of [D67N/G, K70R, T215F (or T215I/V occurring with other TAM-2 RAMs), K219 Q/E/R]; mixed/other TAM profile comprised mixtures of TAM-1 and TAM-2 RAMs or occurrence of T215S as the sole TAM. Of the 35 M184V/I, 30 were M184V (5 defective/hypermutated) and 5 were M184I (all defective/hypermutated). View Large Table 1. Characteristics of the study population at study entry (N = 81) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) GSS, genotypic susceptibility score. a GSSs were determined from HIV-1 DNA sequencing results using the Stanford HIV drug resistance algorithm (v8.4) and assigning to each drug in the regimen 0 for high-level resistance, 0.25 for intermediate resistance, 0.5 for low-level resistance and 1 for potential low-level resistance/full susceptibility. b Five patients did not have a PBMC sample for sequencing; 16 samples did not yield an HIV-1 DNA sequence in ≥2 attempts and no further attempts were possible due to limited sample volume. c HIV-1 subtypes were determined phylogenetically from HIV-1 DNA sequences; other subtypes comprised CRF11_cpx (n = 3), and subtypes D, H, F1, CRF01_AE, CRF37_cpx, CRF06_cpx and CRF18_cpx (one of each). d NRTI and NNRTI RAMs of 60 patients with HIV-1 DNA sequencing results are detailed; 3 patients had D30N in protease but no record of exposure to a PI/r other than lopinavir/ritonavir and including nelfinavir; 11 reverse transcriptase sequences showed stop codons (positions 24, 42, 48, 71, 88, 120, 153, 212, 219, 229 and 239) suggestive of defective proviruses; among 29 patients with ≥1 TAM, the median number of TAMs was 2 (range 1–5); TAM-1 profile comprised ≥1 of (M41L, L210W, T215Y); TAM-2 profile comprised ≥1 of [D67N/G, K70R, T215F (or T215I/V occurring with other TAM-2 RAMs), K219 Q/E/R]; mixed/other TAM profile comprised mixtures of TAM-1 and TAM-2 RAMs or occurrence of T215S as the sole TAM. Of the 35 M184V/I, 30 were M184V (5 defective/hypermutated) and 5 were M184I (all defective/hypermutated). View Large Outcomes of darunavir/ritonavir monotherapy Between week 12 and week 48 (median week 24), 16/81 (19.8%) patients experienced virological failure. A further 6/81 (7.4%) prematurely discontinued monotherapy owing to significant adverse events (n = 5) or loss to follow-up (n = 1) (Table S2). A further 12/81 (14.8%) patients did not meet the definition of virological failure but experienced one or more viraemic episodes (median viral load 2.2 log10 copies/mL; range 1.9–2.9), with 9/12 experiencing a single episode. Between week 12 and week 48 (median week 36), 21/28 viraemic patients (median viral load 3.0 log10 copies/mL; range 2.0–4.1) underwent resistance testing. Largely reflecting the profiles of HIV-1 DNA, NRTI and NNRTI RAMs occurred in 7/21 (33.3%) and 8/21 (38.1%) samples, respectively; 6/21 (28.6%) samples had dual-class mutations. No sample had darunavir RAMs. Adherence Mean adherence was 97.3% (±2.9) by pill count and 97.3% (±3.5) by VAS overall, and 97.3% (±2.5) and 94.9% (±5.2), respectively, in the virological failure population. The two adherence measures were weakly correlated (Spearman’s r = 0.26; P = 0.021) (Figure S1). Among viraemic patients, mean adherence levels in the period preceding versus that following adherence counselling were 97.6% versus 97.4% (P = 0.677) by pill count and 96.5% versus 95.7% (P = 0.264) by VAS. Predictors of virological failure By univariate analysis, the odds of virological failure were lower in older patients and those with a lower HIV-1 DNA load, higher adherence by VAS and either detected HIV-1 DNA RAMs or unknown HIV-1 DNA RAM status (due to missing data) (Table 2). After adjustment, higher VAS-measured adherence and detection of HIV-1 DNA RAMs remained independently predictive of reduced odds of virological failure. Table 2. Factors associated with virological failure (confirmed or last available viral load >400 copies/mL) Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 AOR, adjusted OR. View Large Table 2. Factors associated with virological failure (confirmed or last available viral load >400 copies/mL) Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 AOR, adjusted OR. View Large Discussion Among patients who switched from suppressive second-line ART with two NRTIs + PI/r to darunavir/ritonavir monotherapy, 19.8% experienced virological failure over 48 weeks. Detection of HIV-1 DNA RAMs at study entry and higher VAS-measured adherence during the study predicted reduced odds of virological failure in this population. Previous randomized studies from Western Europe investigated darunavir/ritonavir monotherapy in patients without a history of treatment failure, and reported frequent viraemia but a low risk of darunavir RAMs.14 A study from Burkina Faso, Cameroon and Senegal evaluated patients who after failure of first-line ART with two NRTIs + NNRTI had achieved virological suppression on two NRTIs + PI/r,15 thus resembling the population of this study. Of 133 patients randomized to either darunavir/ritonavir (n = 56) or lopinavir/ritonavir monotherapy, 21% experienced virological failure (viral load >500 copies/mL) over 48 weeks. Similarly high rates of virological failure were observed in our study. Together, the data indicate that darunavir/ritonavir monotherapy is not indicated for treatment-experienced patients in sub-Saharan Africa due to the high risk of viraemia.16 Adding a further dimension, we previously observed a risk of both incident and reactivated HBV infection after discontinuation of NRTIs that may exert prophylactic activity in a high HBV endemicity setting such as Cameroon.17 VAS-measured adherence was an independent predictor of virological outcomes, and performed better than directly measured pill counts. This is promising given that a VAS is an easy tool to implement in resource-limited settings. Although there exists conflicting evidence about the relative performance of self-reported versus directly assessed adherence measures in sub-Saharan Africa,18,19 in the study from Burkina Faso, Cameroon and Senegal, pill counts were also less likely to detect adherence problems relative to the use of questionnaires.15 It is possible that the personal interaction required to administer a VAS (or a questionnaire) may bring benefit when assessing adherence in these settings. The study provides novel insights about the role of RAMs as predictors of virological outcomes in treatment-experienced populations in sub-Saharan Africa. At study entry, the mutational patterns in HIV-1 DNA were overall reflective of prolonged viraemia on first-line NNRTI-based ART. Mutations selected by lamivudine (M184V) and zidovudine or stavudine (TAMs) were common, as were the nevirapine and efavirenz RAMs K103N and Y181C. Further emergence of NRTI RAMs may have occurred during undocumented periods of viraemia on two NRTIs + PI/r. Our previous pooled analysis indicated that detecting RAMs at failure of first-line ART with two NRTIs + NNRTI predicted improved responses after starting second-line ART with two NRTIs + PI/r.5 Here, detecting NRTI and NNRTI RAMs in HIV-1 DNA of virologically suppressed patients receiving two NRTIs + PI/r predicted a reduced risk of virological failure after switching to darunavir/ritonavir monotherapy. The effect persisted after considering other proposed predictors of responses to PI/r monotherapy, including adherence, HIV-1 DNA load and nadir CD4 cell counts.20 While NRTIs are expected to retain partial antiviral activity despite the presence of RAMs, the findings imply a predictive role for pre-existing NRTI RAMs that is partially independent of residual NRTI activity. RAMs may be a proxy for unknown co-variables. It may also be proposed that patients who develop resistance while receiving ART have relatively higher levels of adherence (hence higher drug selective pressure) than patients who experience failure without resistance, and maintain higher adherence levels during subsequent treatment lines. Thus, RAMs may act as an additional, sensitive indicator of overall compliance with treatment in these populations. The predictive role of harbouring defective or hypermutated HIV-1 genomes remains to be explored. Acknowledgements We are grateful to the patients and staff of Hôpital Central Yaoundé, and to the staff of the University of Liverpool and the Centre Pasteur of Cameroon. Funding The study was funded jointly by Janssen, the University of Liverpool and Chantal Biya International Reference Centre for Research on HIV/AIDS Prevention & Management (CIRCB), and was sponsored by the University of Liverpool. The funders had no role in the conduct of the study, the analysis of the data and the preparation and submission of the manuscript. Transparency declarations A. M. G. reports honoraria from Cepheid, Gilead, Janssen and ViiV, outside the submitted work, and employment as an expert scientist by Roche Pharma Research and Early Development, outside the submitted work. The University of Liverpool is the recipient of research grants from BMS, Gilead, Roche Pharma and ViiV, for which A. M. G. is the principal investigator. All other authors: none to declare. Author contributions A. M. G., study design, data collection and analysis, and preparation of the manuscript. A. A., sequencing and data analysis. O. M. F., study coordination, and sample and data acquisition. L. B., statistical analysis. V. F. D., S. M. and C. K., clinical supervision and data acquisition. J. F., laboratory data acquisition. J. T., laboratory supervision. All authors contributed to the preparation of the manuscript. References 1 WHO . Updated Recommendations on First-Line and Second-Line Antiretroviral Regimens and Post-Exposure Prophylaxis and Recommendations on Early Infant Diagnosis of HIV: Interim Guidance, December 2018. https://www.who.int/hiv/pub/guidelines/ARV2018update/en/. 2 Mills EJ , Nachega JB , Buchan I et al. Adherence to antiretroviral therapy in sub-Saharan Africa and North America . JAMA 2006 ; 296 : 679 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Ortego C , Javier TBH , Warren MR et al. Adherence to highly active antiretroviral therapy (HAART): a meta-analysis . AIDS Behav 2011 ; 15 : 1381 – 96 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Boender TS , Sigaloff KCE , McMahon JH et al. Long-term virological outcomes of first-line antiretroviral therapy for HIV-1 in low- and middle-income countries: a systematic review and meta-analysis . Clin Infect Dis 2015 ; 61 : 1453 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Stockdale AJ , Saunders MJ , Boyd MA et al. Effectiveness of protease inhibitor/nucleos(t)ide reverse transcriptase inhibitor-based second-line antiretroviral therapy for the treatment of HIV-1 infection in sub-Saharan Africa: a systematic review and meta-analysis . Clin Infect Dis 2018 ; 66 : 1846 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Hamers RL , Sigaloff KCE , Wensing AM et al. Patterns of HIV-1 drug resistance after first-line antiretroviral therapy (ART) failure in 6 sub-Saharan African countries: implications for second-line ART strategies . Clin Infect Dis 2012 ; 54 : 1660 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Lam EP , Moore CL , Gotuzzo E et al. Antiretroviral resistance after first-line antiretroviral therapy failure in diverse HIV-1 subtypes in the SECOND-LINE study . AIDS Res Hum Retroviruses 2016 ; 32 : 841 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat 8 TenoRes Study Group . Global epidemiology of drug resistance after failure of WHO recommended first-line regimens for adult HIV-1 infection: a multicentre retrospective cohort study . Lancet Infect Dis 2016 ; 16 : 565 – 75 . Crossref Search ADS PubMed WorldCat 9 Guichet E , Aghokeng A , Serrano L et al. High viral load and multidrug resistance due to late switch to second-line regimens could be a major obstacle to reach the 90-90-90 UNAIDS . AIDS Res Hum Retroviruses 2016 ; 32 : 1159 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Villa G , Phillips RO , Smith C et al. Drug resistance outcomes of long-term ART with tenofovir disoproxil fumarate in the absence of virological monitoring . J Antimicrob Chemother 2018 ; 73 : 3148 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Villabona-Arenas CJ , Eymard-Duvernay S , Aghokeng A et al. Nucleoside reverse transcriptase inhibitors with reduced predicted activity do not impair second-line therapy with lopinavir/ritonavir or darunavir/ritonavir . AIDS Res Hum Retroviruses 2018 ; 34 : 477 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Ruggiero A , De Spiegelaere W , Cozzi-Lepri A et al. During stably suppressive antiretroviral therapy integrated HIV-1 DNA load in peripheral blood is associated with the frequency of CD8 cells expressing HLA-DR/DP/DQ . EBioMedicine 2015 ; 2 : 1153 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Geretti AM , Conibear T , Hill A et al. Sensitive testing of plasma HIV-1 RNA and Sanger sequencing of cellular HIV-1 DNA for the detection of drug resistance prior to starting first-line antiretroviral therapy with etravirine or efavirenz . J Antimicrob Chemother 2014 ; 69 : 1090 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Arribas JR , Girard PM , Paton N et al. Efficacy of protease inhibitor monotherapy vs. triple therapy: meta-analysis of data from 2303 patients in 13 randomized trials . HIV Med 2016 ; 17 : 358 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Ciaffi L , Koulla-Shiro S , Sawadogo AB et al. Boosted protease inhibitor monotherapy versus boosted protease inhibitor plus lamivudine dual therapy as second-line maintenance treatment for HIV-1-infected patients in sub-Saharan Africa (ANRS12 286/MOBIDIP): a multicentre, randomised, parallel, open-label, superiority trial . Lancet HIV 2017 ; 4 : e384 – 92 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Thompson JA , Kityo C , Dunn D et al. Evolution of protease inhibitor resistance in human immunodeficiency virus type 1 infected patients failing protease inhibitor monotherapy as second-line therapy in low-income countries: an observational analysis within the EARNEST randomized trial . Clin Infect Dis 2018 ; doi:10.1093/cid/ciy589. WorldCat 17 Abdullahi A , Fopoussi OM , Torimiro J et al. Hepatitis B virus (HBV) infection and re-activation during nucleos(t)ide reverse transcriptase inhibitor-sparing antiretroviral therapy in a high-HBV endemicity setting . Open Forum Infect Dis 2018 ; 5 : ofy251. Google Scholar Crossref Search ADS PubMed WorldCat 18 Orrell C , Cohen K , Leisegang R et al. Comparison of six methods to estimate adherence in an ART-naive cohort in a resource-poor setting: which best predicts virological and resistance outcomes? AIDS Res Ther 2017 ; 14 : 20 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Stöhr W , Dunn DT , Arenas-Pinto A et al. Factors associated with virological rebound in HIV-infected patients receiving protease inhibitor monotherapy . AIDS 2016 ; 30 : 2617 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Lambert-Niclot S , Flandre P , Valantin MA et al. Factors associated with virological failure in HIV-1-infected patients receiving darunavir/ritonavir monotherapy . J Infect Dis 2011 ; 204 : 1211 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. 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/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

An apparent paradox: resistance mutations in HIV-1 DNA predict improved virological responses to antiretroviral therapy

Loading next page...
 
/lp/oxford-university-press/an-apparent-paradox-resistance-mutations-in-hiv-1-dna-predict-improved-iBbvzGwuW3
Publisher
Oxford University Press
Copyright
© The Author(s) 2019. 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/dkz264
Publisher site
See Article on Publisher Site

Abstract

Abstract Background In sub-Saharan Africa, detecting resistance-associated mutations (RAMs) at failure of first-line ART with two NRTIs plus an NNRTI predicts improved virological responses to second-line therapy with two NRTIs plus a ritonavir-boosted PI (PI/r). This indicates residual NRTI activity in the presence of RAMs, although additional factors may contribute to the effect. Objectives The aim of this study was to investigate the influence of pre-existing RAMs on the outcomes of maintenance monotherapy with ritonavir-boosted darunavir within a randomized trial in Cameroon. Methods RAMs were detected in HIV-1 DNA using PBMCs collected at initiation of darunavir/ritonavir monotherapy. Adherence was assessed by pill count and visual analogue scale (VAS). Predictors of virological failure (confirmed or last available viral load >400 copies/mL) were explored by logistic regression analysis. Trial name = MANET (NCT02155101). Results After NNRTI-based therapy, participants (n = 81) had received PI/r-based therapy for a median of 3.2 years and had a confirmed viral load <60 copies/mL and a median CD4 count of 466 cells/mm3. NRTI and NNRTI RAMs were detected in 39/60 (65.0%) and 41/60 (68.3%) HIV-1 DNA sequences, respectively. Over 48 weeks of monotherapy, 16/81 (19.8%) patients experienced virological failure. After adjusting for age, HIV-1 DNA load, adherence by VAS and RAM status, virological failure was less likely with higher VAS-measured adherence (adjusted OR 0.04, 95% CI 0.01–0.37; P = 0.004) and detectable HIV-1 DNA RAMs (adjusted OR 0.15, 95% CI 0.03–0.82; P = 0.028). Conclusions Pre-existing NRTI and NNRTI RAMs are associated with improved virological responses to NRTI-sparing ART in sub-Saharan Africa, indicating a predictive effect that is independent of residual NRTI activity. Introduction The WHO recommends standardized regimens for first- and second-line ART in sub-Saharan Africa, comprising two NRTIs plus an NNRTI, a ritonavir-boosted PI (PI/r) or, more recently, an integrase inhibitor.1 Pooled estimates indicate adherence rates of 67%–77% in the region,2,3 and rates of virological suppression of 65% and 62% after 24 months of NNRTI-based first-line ART4 and PI/r-based second-line5 ART, respectively. Sparse virological monitoring and delayed introduction of second-line ART in viraemic patients favour the emergence of drug resistance, and resistance-associated mutations (RAMs) affecting both the NRTIs and the NNRTIs are common at failure of first-line NNRTI-based ART.6–10 The significance of NRTI RAMs in terms of predicting reduced responses to second-line NRTI-containing ART has been called into question.5,11 In a meta-analysis of pooled data from sub-Saharan Africa, we observed that detection of RAMs at failure of NNRTI-based ART, most commonly M184V and thymidine analogue mutations (TAMs), predicted higher (rather than lower) odds of virological suppression after starting two NRTIs plus a PI/r.5 Thus, it has been inferred that resistance testing underestimates the continued activity of NRTIs such as lamivudine and tenofovir in the presence of M184V and TAMs. Here we report the same predictive effect in patients receiving NRTI-sparing ART. Patients and methods Study population MANET (Monotherapy in Africa, New Evaluations of Treatment) was a randomized, open-label trial based at Hôpital Central Yaoundé (NCT02155101). Between August 2014 and July 2015, 120 virologically suppressed patients on two NRTIs + PI/r either switched to darunavir/ritonavir monotherapy (800/100 mg once daily) for 48 weeks (n = 81) or continued two NRTIs + PI/r for 24 weeks (n = 39). Eligibility criteria comprised receiving two NRTIs + PI/r for ≥12 weeks, CD4 count >100 cells/mm3, viral load <60 copies/mL at two measurements 4–12 weeks apart (median 7 weeks) and hepatitis B surface antigen (HBsAg) negative. This analysis focused on patients receiving monotherapy. Ethics Approval was granted by the University of Liverpool Ethics Committee (RETH000605) and the Cameroon National Ethics Committee (2013/07/347). Participants gave written informed consent. Adherence Participants collected a prescription from the pharmacy every 4 weeks and attended study visits at weeks 4, 12, 24, 36 and 48. At each pharmacy visit, the number of left over pills was counted to infer the percentage of pills taken. At each study visit, adherence was self-reported via a visual analogue scale (VAS) graded from 0% (complete non-adherence) to 100% (complete adherence) in 10% increments.10 Laboratory testing Safety parameters, CD4 cell counts and viral load (Biocentric, Bandol, France; lower limit of quantification 60 copies/mL) were measured at the Centre Pasteur of Cameroon in Yaoundé. In the UK, total HIV-1 DNA was quantified in PBMCs as previously described.12 RAMs were detected by Sanger sequencing of protease (amino acids 1–99) and reverse transcriptase (amino acids 1–335) as previously described,13 and defined by the Stanford HIV drug resistance algorithm (v8.4). Darunavir RAMs comprised V11I, V32I, L33F, I47V, I50V, I54L/M, T74P, L76V, I84V and L89V. Analyses The characteristics of patients with or without HIV-1 DNA sequencing results were compared using Wilcoxon rank-sum test, t-test or χ2 test. Pill counts and VAS percentages were averaged in each patient over the entire study period and compared by Spearman’s rank correlation. Patients with virological rebound were recalled for adherence counselling, followed by repeat testing 4 weeks later. Mean adherence levels in the period before versus the period after the first counselling session were compared by t-test. Treatment failure comprised virological failure (either confirmed or last available viral load >400 copies/mL) or discontinuation of monotherapy prior to week 48 due to a significant adverse event, death or loss to follow-up. Predictors of virological failure were explored by logistic regression analysis; all available variables were considered for inclusion and those with P ≤ 0.2 in the univariate analysis were retained in the multivariable model. Results HIV-1 DNA RAMs at study entry The population is described in Table 1. Of 76 patients with PBMC samples, 60 (78.9%) yielded a HIV-1 DNA sequence. Relative to patients with sequencing results, those without had a lower HIV-1 DNA load (P = 0.0004) and a higher nadir CD4 cell count (P = 0.05) (Table S1, available as Supplementary data at JAC Online). NRTI and NNRTI RAMs occurred in 39/60 (65.0%) and 41/60 (68.3%) samples, respectively (Table 1); 37/60 (61.7%) samples had dual-class mutations. Table 1. Characteristics of the study population at study entry (N = 81) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) GSS, genotypic susceptibility score. a GSSs were determined from HIV-1 DNA sequencing results using the Stanford HIV drug resistance algorithm (v8.4) and assigning to each drug in the regimen 0 for high-level resistance, 0.25 for intermediate resistance, 0.5 for low-level resistance and 1 for potential low-level resistance/full susceptibility. b Five patients did not have a PBMC sample for sequencing; 16 samples did not yield an HIV-1 DNA sequence in ≥2 attempts and no further attempts were possible due to limited sample volume. c HIV-1 subtypes were determined phylogenetically from HIV-1 DNA sequences; other subtypes comprised CRF11_cpx (n = 3), and subtypes D, H, F1, CRF01_AE, CRF37_cpx, CRF06_cpx and CRF18_cpx (one of each). d NRTI and NNRTI RAMs of 60 patients with HIV-1 DNA sequencing results are detailed; 3 patients had D30N in protease but no record of exposure to a PI/r other than lopinavir/ritonavir and including nelfinavir; 11 reverse transcriptase sequences showed stop codons (positions 24, 42, 48, 71, 88, 120, 153, 212, 219, 229 and 239) suggestive of defective proviruses; among 29 patients with ≥1 TAM, the median number of TAMs was 2 (range 1–5); TAM-1 profile comprised ≥1 of (M41L, L210W, T215Y); TAM-2 profile comprised ≥1 of [D67N/G, K70R, T215F (or T215I/V occurring with other TAM-2 RAMs), K219 Q/E/R]; mixed/other TAM profile comprised mixtures of TAM-1 and TAM-2 RAMs or occurrence of T215S as the sole TAM. Of the 35 M184V/I, 30 were M184V (5 defective/hypermutated) and 5 were M184I (all defective/hypermutated). View Large Table 1. Characteristics of the study population at study entry (N = 81) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) Characteristic Value Female, n (%) 61 (75.3) Age (years), median (IQR) 45 (38–51) Time since HIV diagnosis (years), median (IQR) 8.8 (5.9–11.1) CD4 count (cells/mm3), median (IQR) 466 (341–615) Nadir CD4 count (cells/mm3), median (IQR) 90 (37–167) History of AIDS, n (%) 12 (14.8) HIV-1 DNA (log10 copies/106 PBMCs), median (IQR) 2.9 (2.5–3.3) Antiretroviral drugs ever experienced, n (%)  lamivudine 81 (100)  tenofovir 73 (90.1)  zidovudine 72 (88.9)  stavudine 41 (50.6)  didanosine 13 (16.0)  abacavir 9 (11.1)  nevirapine 50 (61.7)  efavirenz 43 (53.1)  lopinavir/ritonavir 79 (97.5)  indinavir/ritonavir 9 (11.1)  atazanavir/ritonavir 5 (6.2) Duration of exposure (years), median (IQR)  total ART 7.6 (5.3–9.8)  tenofovir 2.9 (1.3–4.6)  zidovudine 2.4 (1.5–4.0)  stavudine 2.6 (1.3–4.0)  NNRTI 3.1 (1.6–5.5)  PI/r 3.2 (1.3–5.8) ART regimen at study entry, n (%)  tenofovir/lamivudine + lopinavir/ritonavir 69 (85.2)  tenofovir/lamivudine + atazanavir/ritonavir 2 (2.5)  tenofovir + abacavir + lopinavir/ritonavir 1 (1.2)  abacavir + didanosine + lopinavir/ritonavir 5 (6.2)  zidovudine/lamivudine + lopinavir/ritonavir 4 (4.9) GSS of ART regimen at study entry, n (%)a  3 22 (27.2)  2 22 (27.2)  1.5 1 (1.2)  1.25 10 (12.3)  1 5 (6.2)  not availableb 21 (25.9) HIV-1 subtype, n (%)c  CRF02_AG 35 (43.2)  A1 10 (12.3)  G 5 (6.2)  other 10 (12.3)  not availableb 21 (25.9) RAMs in HIV-1 DNA, n (%)d  any 44 (54.3)  none 16 (19.8)  not available 21 (25.9)  any NRTI 39 (48.1)  1 TAM 13 (16.0)  2 TAMs 5 (6.2)  ≥3 TAMs 11 (13.6)  TAM-1 profile 10 (12.3)  TAM-2 profile 15 (18.5)  mixed/other TAM profile 4 (4.9)  K65R 2 (2.5)  L74V/I 5 (6.2)  M184I/V 35 (43.2)  Q151M/L 3 (3.7)  T69ins 1 (1.2)  T69D/N 1 (1.2)  any NNRTI 41 (50.6)  A98G 9 (11.1)  L100I 1 (1.2)  K101E/H/P/T 6 (7.4)  K103N 20 (24.7)  V106A 3 (3.7)  V108I 6 (7.4)  E138G/K 3 (3.7)  Y181C 14 (17.3)  Y188L/C/F/H/L 4 (4.9)  G190A 9 (11.1)  H221Y 5 (6.2)  P225H 2 (2.5)  F227L 3 (3.7)  M230I/L 4 (4.9)  K238T 4 (4.9)  Y318F 1 (1.2) GSS, genotypic susceptibility score. a GSSs were determined from HIV-1 DNA sequencing results using the Stanford HIV drug resistance algorithm (v8.4) and assigning to each drug in the regimen 0 for high-level resistance, 0.25 for intermediate resistance, 0.5 for low-level resistance and 1 for potential low-level resistance/full susceptibility. b Five patients did not have a PBMC sample for sequencing; 16 samples did not yield an HIV-1 DNA sequence in ≥2 attempts and no further attempts were possible due to limited sample volume. c HIV-1 subtypes were determined phylogenetically from HIV-1 DNA sequences; other subtypes comprised CRF11_cpx (n = 3), and subtypes D, H, F1, CRF01_AE, CRF37_cpx, CRF06_cpx and CRF18_cpx (one of each). d NRTI and NNRTI RAMs of 60 patients with HIV-1 DNA sequencing results are detailed; 3 patients had D30N in protease but no record of exposure to a PI/r other than lopinavir/ritonavir and including nelfinavir; 11 reverse transcriptase sequences showed stop codons (positions 24, 42, 48, 71, 88, 120, 153, 212, 219, 229 and 239) suggestive of defective proviruses; among 29 patients with ≥1 TAM, the median number of TAMs was 2 (range 1–5); TAM-1 profile comprised ≥1 of (M41L, L210W, T215Y); TAM-2 profile comprised ≥1 of [D67N/G, K70R, T215F (or T215I/V occurring with other TAM-2 RAMs), K219 Q/E/R]; mixed/other TAM profile comprised mixtures of TAM-1 and TAM-2 RAMs or occurrence of T215S as the sole TAM. Of the 35 M184V/I, 30 were M184V (5 defective/hypermutated) and 5 were M184I (all defective/hypermutated). View Large Outcomes of darunavir/ritonavir monotherapy Between week 12 and week 48 (median week 24), 16/81 (19.8%) patients experienced virological failure. A further 6/81 (7.4%) prematurely discontinued monotherapy owing to significant adverse events (n = 5) or loss to follow-up (n = 1) (Table S2). A further 12/81 (14.8%) patients did not meet the definition of virological failure but experienced one or more viraemic episodes (median viral load 2.2 log10 copies/mL; range 1.9–2.9), with 9/12 experiencing a single episode. Between week 12 and week 48 (median week 36), 21/28 viraemic patients (median viral load 3.0 log10 copies/mL; range 2.0–4.1) underwent resistance testing. Largely reflecting the profiles of HIV-1 DNA, NRTI and NNRTI RAMs occurred in 7/21 (33.3%) and 8/21 (38.1%) samples, respectively; 6/21 (28.6%) samples had dual-class mutations. No sample had darunavir RAMs. Adherence Mean adherence was 97.3% (±2.9) by pill count and 97.3% (±3.5) by VAS overall, and 97.3% (±2.5) and 94.9% (±5.2), respectively, in the virological failure population. The two adherence measures were weakly correlated (Spearman’s r = 0.26; P = 0.021) (Figure S1). Among viraemic patients, mean adherence levels in the period preceding versus that following adherence counselling were 97.6% versus 97.4% (P = 0.677) by pill count and 96.5% versus 95.7% (P = 0.264) by VAS. Predictors of virological failure By univariate analysis, the odds of virological failure were lower in older patients and those with a lower HIV-1 DNA load, higher adherence by VAS and either detected HIV-1 DNA RAMs or unknown HIV-1 DNA RAM status (due to missing data) (Table 2). After adjustment, higher VAS-measured adherence and detection of HIV-1 DNA RAMs remained independently predictive of reduced odds of virological failure. Table 2. Factors associated with virological failure (confirmed or last available viral load >400 copies/mL) Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 AOR, adjusted OR. View Large Table 2. Factors associated with virological failure (confirmed or last available viral load >400 copies/mL) Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 Univariate analysis Multivariable analysis Characteristic Alteration OR (95% CI) P value AOR (95% CI) P value Gender female versus male 0.66 (0.20–2.20) 0.500 Age per 5 year older 0.75 (0.54–1.04) 0.087 0.94 (0.87–1.01) 0.119 Time since HIV diagnosis per year longer 0.92 (0.78–1.09) 0.348 CD4 count per 50 cells/mm3 higher 1.04 (0.94–1.14) 0.496 Nadir CD4 count per 50 cells/mm3 higher 0.92 (0.64–1.31) 0.867 History of AIDS yes versus no 1.44 (0.34–6.06) 0.622 HIV-1 DNA per log10 copies/106 PBMCs higher 1.80 (0.72–4.50) 0.211 2.80 (0.80–9.99) 0.109 ART duration per year longer 0.91 (0.76–1.09) 0.325 PI/r duration per year longer 0.96 (0.79–1.16) 0.667 ≥1 RAM in HIV-1 DNA yes versus no 0.31 (0.09–1.09) 0.069 0.15 (0.03–0.82) 0.028 unknown versus no 0.17 (0.03–0.94) 0.042 0.29 (0.04–2.06) 0.218 Pill count per 10% increment 0.55 (0.15–2.03) 0.367 VAS per 10% increment 0.21 (0.06–0.69) 0.011 0.04 (0.01–0.37) 0.004 AOR, adjusted OR. View Large Discussion Among patients who switched from suppressive second-line ART with two NRTIs + PI/r to darunavir/ritonavir monotherapy, 19.8% experienced virological failure over 48 weeks. Detection of HIV-1 DNA RAMs at study entry and higher VAS-measured adherence during the study predicted reduced odds of virological failure in this population. Previous randomized studies from Western Europe investigated darunavir/ritonavir monotherapy in patients without a history of treatment failure, and reported frequent viraemia but a low risk of darunavir RAMs.14 A study from Burkina Faso, Cameroon and Senegal evaluated patients who after failure of first-line ART with two NRTIs + NNRTI had achieved virological suppression on two NRTIs + PI/r,15 thus resembling the population of this study. Of 133 patients randomized to either darunavir/ritonavir (n = 56) or lopinavir/ritonavir monotherapy, 21% experienced virological failure (viral load >500 copies/mL) over 48 weeks. Similarly high rates of virological failure were observed in our study. Together, the data indicate that darunavir/ritonavir monotherapy is not indicated for treatment-experienced patients in sub-Saharan Africa due to the high risk of viraemia.16 Adding a further dimension, we previously observed a risk of both incident and reactivated HBV infection after discontinuation of NRTIs that may exert prophylactic activity in a high HBV endemicity setting such as Cameroon.17 VAS-measured adherence was an independent predictor of virological outcomes, and performed better than directly measured pill counts. This is promising given that a VAS is an easy tool to implement in resource-limited settings. Although there exists conflicting evidence about the relative performance of self-reported versus directly assessed adherence measures in sub-Saharan Africa,18,19 in the study from Burkina Faso, Cameroon and Senegal, pill counts were also less likely to detect adherence problems relative to the use of questionnaires.15 It is possible that the personal interaction required to administer a VAS (or a questionnaire) may bring benefit when assessing adherence in these settings. The study provides novel insights about the role of RAMs as predictors of virological outcomes in treatment-experienced populations in sub-Saharan Africa. At study entry, the mutational patterns in HIV-1 DNA were overall reflective of prolonged viraemia on first-line NNRTI-based ART. Mutations selected by lamivudine (M184V) and zidovudine or stavudine (TAMs) were common, as were the nevirapine and efavirenz RAMs K103N and Y181C. Further emergence of NRTI RAMs may have occurred during undocumented periods of viraemia on two NRTIs + PI/r. Our previous pooled analysis indicated that detecting RAMs at failure of first-line ART with two NRTIs + NNRTI predicted improved responses after starting second-line ART with two NRTIs + PI/r.5 Here, detecting NRTI and NNRTI RAMs in HIV-1 DNA of virologically suppressed patients receiving two NRTIs + PI/r predicted a reduced risk of virological failure after switching to darunavir/ritonavir monotherapy. The effect persisted after considering other proposed predictors of responses to PI/r monotherapy, including adherence, HIV-1 DNA load and nadir CD4 cell counts.20 While NRTIs are expected to retain partial antiviral activity despite the presence of RAMs, the findings imply a predictive role for pre-existing NRTI RAMs that is partially independent of residual NRTI activity. RAMs may be a proxy for unknown co-variables. It may also be proposed that patients who develop resistance while receiving ART have relatively higher levels of adherence (hence higher drug selective pressure) than patients who experience failure without resistance, and maintain higher adherence levels during subsequent treatment lines. Thus, RAMs may act as an additional, sensitive indicator of overall compliance with treatment in these populations. The predictive role of harbouring defective or hypermutated HIV-1 genomes remains to be explored. Acknowledgements We are grateful to the patients and staff of Hôpital Central Yaoundé, and to the staff of the University of Liverpool and the Centre Pasteur of Cameroon. Funding The study was funded jointly by Janssen, the University of Liverpool and Chantal Biya International Reference Centre for Research on HIV/AIDS Prevention & Management (CIRCB), and was sponsored by the University of Liverpool. The funders had no role in the conduct of the study, the analysis of the data and the preparation and submission of the manuscript. Transparency declarations A. M. G. reports honoraria from Cepheid, Gilead, Janssen and ViiV, outside the submitted work, and employment as an expert scientist by Roche Pharma Research and Early Development, outside the submitted work. The University of Liverpool is the recipient of research grants from BMS, Gilead, Roche Pharma and ViiV, for which A. M. G. is the principal investigator. All other authors: none to declare. Author contributions A. M. G., study design, data collection and analysis, and preparation of the manuscript. A. A., sequencing and data analysis. O. M. F., study coordination, and sample and data acquisition. L. B., statistical analysis. V. F. D., S. M. and C. K., clinical supervision and data acquisition. J. F., laboratory data acquisition. J. T., laboratory supervision. All authors contributed to the preparation of the manuscript. References 1 WHO . Updated Recommendations on First-Line and Second-Line Antiretroviral Regimens and Post-Exposure Prophylaxis and Recommendations on Early Infant Diagnosis of HIV: Interim Guidance, December 2018. https://www.who.int/hiv/pub/guidelines/ARV2018update/en/. 2 Mills EJ , Nachega JB , Buchan I et al. Adherence to antiretroviral therapy in sub-Saharan Africa and North America . JAMA 2006 ; 296 : 679 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Ortego C , Javier TBH , Warren MR et al. Adherence to highly active antiretroviral therapy (HAART): a meta-analysis . AIDS Behav 2011 ; 15 : 1381 – 96 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Boender TS , Sigaloff KCE , McMahon JH et al. Long-term virological outcomes of first-line antiretroviral therapy for HIV-1 in low- and middle-income countries: a systematic review and meta-analysis . Clin Infect Dis 2015 ; 61 : 1453 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Stockdale AJ , Saunders MJ , Boyd MA et al. Effectiveness of protease inhibitor/nucleos(t)ide reverse transcriptase inhibitor-based second-line antiretroviral therapy for the treatment of HIV-1 infection in sub-Saharan Africa: a systematic review and meta-analysis . Clin Infect Dis 2018 ; 66 : 1846 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Hamers RL , Sigaloff KCE , Wensing AM et al. Patterns of HIV-1 drug resistance after first-line antiretroviral therapy (ART) failure in 6 sub-Saharan African countries: implications for second-line ART strategies . Clin Infect Dis 2012 ; 54 : 1660 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Lam EP , Moore CL , Gotuzzo E et al. Antiretroviral resistance after first-line antiretroviral therapy failure in diverse HIV-1 subtypes in the SECOND-LINE study . AIDS Res Hum Retroviruses 2016 ; 32 : 841 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat 8 TenoRes Study Group . Global epidemiology of drug resistance after failure of WHO recommended first-line regimens for adult HIV-1 infection: a multicentre retrospective cohort study . Lancet Infect Dis 2016 ; 16 : 565 – 75 . Crossref Search ADS PubMed WorldCat 9 Guichet E , Aghokeng A , Serrano L et al. High viral load and multidrug resistance due to late switch to second-line regimens could be a major obstacle to reach the 90-90-90 UNAIDS . AIDS Res Hum Retroviruses 2016 ; 32 : 1159 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Villa G , Phillips RO , Smith C et al. Drug resistance outcomes of long-term ART with tenofovir disoproxil fumarate in the absence of virological monitoring . J Antimicrob Chemother 2018 ; 73 : 3148 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Villabona-Arenas CJ , Eymard-Duvernay S , Aghokeng A et al. Nucleoside reverse transcriptase inhibitors with reduced predicted activity do not impair second-line therapy with lopinavir/ritonavir or darunavir/ritonavir . AIDS Res Hum Retroviruses 2018 ; 34 : 477 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Ruggiero A , De Spiegelaere W , Cozzi-Lepri A et al. During stably suppressive antiretroviral therapy integrated HIV-1 DNA load in peripheral blood is associated with the frequency of CD8 cells expressing HLA-DR/DP/DQ . EBioMedicine 2015 ; 2 : 1153 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Geretti AM , Conibear T , Hill A et al. Sensitive testing of plasma HIV-1 RNA and Sanger sequencing of cellular HIV-1 DNA for the detection of drug resistance prior to starting first-line antiretroviral therapy with etravirine or efavirenz . J Antimicrob Chemother 2014 ; 69 : 1090 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Arribas JR , Girard PM , Paton N et al. Efficacy of protease inhibitor monotherapy vs. triple therapy: meta-analysis of data from 2303 patients in 13 randomized trials . HIV Med 2016 ; 17 : 358 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Ciaffi L , Koulla-Shiro S , Sawadogo AB et al. Boosted protease inhibitor monotherapy versus boosted protease inhibitor plus lamivudine dual therapy as second-line maintenance treatment for HIV-1-infected patients in sub-Saharan Africa (ANRS12 286/MOBIDIP): a multicentre, randomised, parallel, open-label, superiority trial . Lancet HIV 2017 ; 4 : e384 – 92 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Thompson JA , Kityo C , Dunn D et al. Evolution of protease inhibitor resistance in human immunodeficiency virus type 1 infected patients failing protease inhibitor monotherapy as second-line therapy in low-income countries: an observational analysis within the EARNEST randomized trial . Clin Infect Dis 2018 ; doi:10.1093/cid/ciy589. WorldCat 17 Abdullahi A , Fopoussi OM , Torimiro J et al. Hepatitis B virus (HBV) infection and re-activation during nucleos(t)ide reverse transcriptase inhibitor-sparing antiretroviral therapy in a high-HBV endemicity setting . Open Forum Infect Dis 2018 ; 5 : ofy251. Google Scholar Crossref Search ADS PubMed WorldCat 18 Orrell C , Cohen K , Leisegang R et al. Comparison of six methods to estimate adherence in an ART-naive cohort in a resource-poor setting: which best predicts virological and resistance outcomes? AIDS Res Ther 2017 ; 14 : 20 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Stöhr W , Dunn DT , Arenas-Pinto A et al. Factors associated with virological rebound in HIV-infected patients receiving protease inhibitor monotherapy . AIDS 2016 ; 30 : 2617 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Lambert-Niclot S , Flandre P , Valantin MA et al. Factors associated with virological failure in HIV-1-infected patients receiving darunavir/ritonavir monotherapy . J Infect Dis 2011 ; 204 : 1211 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. 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/open_access/funder_policies/chorus/standard_publication_model)

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Jan 9, 18

References

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 folders to
organize your research

Export folders, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off