Antiviral Therapy 2011; 16: (doi: /IMP1719)

Transcription

1 Antiviral Therapy 2011; 16: (doi: /IMP1719) Original article Virological characterization of patients failing darunavir/ritonavir or lopinavir/ritonavir treatment in the ARTEMIS study: 96-week analysis Erkki Lathouwers 1 *, Sandra De Meyer 1, Inge Dierynck 1, Tom Van de Casteele 1, Ludo Lavreys 1, Marie Pierre de Béthune 1, Gaston Picchio 2 1 Tibotec BVBA, Beerse, Belgium 2 Tibotec Inc., Titusville, NJ, USA *Corresponding author elathouw@its.jnj.com Background: In the Phase III ARTEMIS Trial, treatmentnaive patients received once-daily darunavir/ritonavir (DRV/r) 800/100 mg (n=343) or lopinavir/ritonavir (LPV/r) 800/200 mg (total daily dose; n=346) plus fixed-dose tenofovir disoproxil fumarate/emtricitabine. The primary outcome measure was non-inferiority of DRV/r versus LPV/r (HIV type-1 [HIV-1] RNA<50 copies/ml). Here, a detailed 96-week resistance analysis is presented. Methods: Virological failures (VFs) were defined as patients who had lost (rebounders) or who had never achieved (never suppressed) HIV-1 RNA<50 copies/ml after week 12. Genotypic and phenotypic determinations were performed on plasma samples with HIV-1 RNA 50 copies/ml. The end point was defined as the last on-treatment visit with available genotype and/or phenotype. Results: The VF rate was significantly lower in DRV/r (12%, n=40) versus LPV/r patients (17%, n=59; P=0.0437). Among DRV/r patients, 24 rebounded and 16 were never suppressed, whereas among LPV/r patients, 33 rebounded and 26 were never suppressed. Transient HIV-1 RNA increases ( 50 copies/ml) occurred in 50% (n=12) DRV/r and 48% (n=16) LPV/r rebounders; these viral levels returned to undetectable by end point without any changes to the study regimen. No major (primary) protease inhibitor (PI) resistance-associated mutations (RAMs) developed in VFs with an available genotype at baseline and end point, and almost all developing minor PI RAMs were polymorphic. At end point, all VFs with available phenotypes at baseline and end point remained susceptible to all PIs, including study PIs. Conclusions: The VF rate was lower with DRV/r than LPV/r. The findings of this resistance analysis confirmed the lack of development of major PI RAMs and the preservation of phenotypic susceptibility to all PIs in patients with VF. Introduction The emergence of resistance to highly active antiretroviral (ARV) therapy (HAART) presents a significant challenge to managing HIV type-1 (HIV-1) infection. The choice of first-line therapy is crucial as the regimen needs to durably suppress HIV replication, but ultimately should prevent the emergence of drugresistant strains, which can lead to treatment failure and therefore limit options for future therapy [1]. Thus, the effect of emergent viral resistance should be considered when initiating ARV therapy and the characterization of the development of resistance to new ARVs is important in helping clinicians improve treatment decisions. Although less common than emergent viral resistance, transmitted HIV-1 drug resistance can also affect the success of an initial ARV therapy, generally leading to a delay in virological suppression and an increased risk of earlier virological failure (VF) [2]. Adherence to an ARV drug regimen has been strongly correlated with HIV viral suppression and reduced rates of resistance [3]. Because of the high replication and mutation rates of HIV-1, a high level of adherence is required to achieve durable suppression of HIV-1 replication and to reduce the risk of resistance development [4]. Protease inhibitors (PIs) remain a widely used class for HIV therapy, with proven efficacy in treatmentnaive and treatment-experienced patients [3]. PIs have the advantage over other classes of ARV drugs in that resistance to most ritonavir-boosted PIs requires 2011 International Medical Press (print) (online) 99

2 E Lathouwers et al. multiple mutations; thus, PI resistance rarely develops after first-line VF when ritonavir boosting is used [3]. The PI darunavir (DRV; TMC114), has in vitro activity against both wild-type virus and multidrug-resistant HIV-1 strains [5]. In vitro, DRV has a high genetic barrier to the development of resistance [6], which allows the retention of antiviral activity despite the occurrence of mutations within the target viral protein. DRV boosted with low-dose ritonavir (DRV/r) 800/100 mg once daily is approved in the USA [7], Europe [8] and other countries for the treatment for HIV-1-infected treatment-naive adults as part of combination therapy. DRV/r at a dose of 600/100 mg twice daily is approved for the treatment of ARV-experienced adult patients in the USA [7], Europe [8] and other countries. In the ongoing Phase III ARTEMIS Trial (AntiRetroviral Therapy with TMC114 ExaMined In Naive Subjects), results of the 96-week analysis demonstrated that once-daily DRV/r 800/100 mg achieved both primary (non-inferiority) and secondary (superiority) virological objectives (HIV-1 RNA<50 copies/ml) versus lopinavir (LPV) boosted with low-dose ritonavir (LPV/r) 800/200 mg total daily dose in treatmentnaive patients [9]. At week 96, 79% of DRV/r patients achieved HIV-1 RNA<50 copies/ml compared with 71% of LPV/r patients (estimated difference =8.3%, 95% confidence interval [CI] ; intent-to-treat/ time-to-loss of virological response [ITT-TLOVR] P=0.012 for superiority) [9]. The present week 96 analysis investigates the genotypic and phenotypic changes observed in VFs and the role of adherence on virological response in the ARTEMIS Trial. Methods Study design ARTEMIS is an ongoing, randomized, controlled, Phase III trial comparing the efficacy, safety, tolerability, resistance characteristics and pharmacokinetics of DRV/r versus LPV/r in treatment-naive HIV-1-infected patients (ClinicalTrials.gov identifier: NCT ). The methodology of this trial has been reported elsewhere [10]. Analyses were performed on the 96-week dataset, at which time all patients had reached at least week 96 of treatment or discontinued earlier. All VFs in the dataset were included in the analyses, including those occurring prior to or after the week 96 time point. The χ 2 test was used to determine between-group significance. Virological analyses Plasma samples were collected at screening, baseline and at regular intervals thereafter. Viral phenotypic (Antivirogram ) and genotypic determinations were performed by Virco BVBA (Beerse, Belgium). Samples taken at screening, baseline, at weeks 24, 48, 72, 96 or at withdrawal visit were analysed when received. Further samples taken from VFs at other time points were analysed. Initially, phenotypic and genotypic determinations were only performed on plasma samples with HIV-1 RNA 1,000 copies/ml [11]. In order to better assess the relationship between VF and resistance, additional testing was performed on all end point samples from VFs with low viraemia (HIV-1 RNA 50 to <1,000 copies/ml). This resulted in more genotypic and phenotypic data at end point (defined as the last on-treatment visit with an available genotype and/or phenotype; Figure 1). The same protocols were used for samples with low HIV-1 RNA concentration Sample HIV-1 subtype was derived as the best match between the determined protease reverse transcriptase nucleotide sequence (determined by Virco BVBA) and the corresponding sequences from the Los Alamos National Laboratory subtype reference subset [12]. Plasma HIV-1 RNA levels were measured using the Roche Amplicor HIV-1 Monitor Assay version 1.5 or the Roche Ultrasensitive method version 1.5 (Roche Molecular Systems, Branchburg, NJ, USA). For resistance determinations, VF was identified using the TLOVR (non-vf censored) imputation method, meaning that no viral load imputations were performed at time points after the last on-treatment viral load determination for patients who discontinued for reasons other than VF (non-vf). Patients who discontinued before week 12 were not taken into account to determine VF. Patients with VF were defined as never suppressed (never achieved a confirmed virological response, that is, HIV-1 RNA<50 copies/ml) or rebounders (achieved two consecutive HIV-1 RNA values of <50 copies/ml, but subsequently either reported HIV-1 RNA 50 copies/ml at two consecutive time points or discontinued with a last observed HIV-1 RNA value of 50 copies/ml). Rebounders were considered as VFs at all visits following confirmed rebound, even if confirmed resuppression (plasma HIV-1 RNA<50 copies/ml) was observed. The proportions of patients with VF were compared between treatments using the Pearson χ 2 test in a post hoc analysis. Phenotypic resistance was defined as having a fold change (FC) in 50% effective concentration above the biological/clinical cutoff, as determined by the Antivirogram. The clinical cutoff of 10 was used for both DRV [13] and LPV [14]; a clinical cutoff of 3 was used for tipranavir [15] and 1.4 was used for tenofovir disoproxil fumarate (TDF) [16]. The biological cutoffs of 2.2, 2.1, 2.3, 2.2, 1.8 and 3.1 were used for amprenavir, atazanavir (ATV), indinavir, nelfinavir, saquinavir (SQV) and emtricitabine (FTC), respectively International Medical Press

3 Resistance analysis of ARV-naive patients on DRV/r or LPV/r in ARTEMIS Figure 1. Overview of available genotype results at end point according to HIV-1 RNA levels DRV/r VFs (n=40) a Viral load 1,000 copies/ml (n=13) 12 VFs: genotype available 1 VF: no genotype available Viral load 50 <1,000 copies/ml (n=25) 19 VFs: genotype available 6 VFs: no genotype available LPV/r VFs (n=59) b Viral load 1,000 copies/ml (n=26) 26 VFs: genotype available Viral load 50 <1,000 copies/ml (n=32) 20 VFs: genotype available 12 VFs: no genotype available a Two missing genotypes for virological failures (VFs) in the darunavir/ritonavir (DRV/r) arm. b One missing genotype for a VF in the lopinavir/ritonavir (LPV/r) arm. HIV-1, HIV type-1. PI resistance-associated mutations (RAMs) were based on the International AIDS Society (IAS) USA list [17] and were classified as major (also known as primary) PI RAMs and minor PI RAMs. In order to provide data on developing mutations that were comparable to that presented in another study of boosted PIs in treatment-naive patients (CASTLE) [18], naturally occurring polymorphisms were excluded from the IAS USA list of PI RAMs. Accordingly, mutations were further classified as non-polymorphic major (D30N, V32I, M46I/L, I47A/V, G48V, I50L/V, I54A/L/M/S/T/V, L76V, V82A/F/L/S/T, IV, N88D/S and L90M) or nonpolymorphic minor (L10F/I/R/V, V11I, L24I, L33F, E35G, K43T, F53L/Y, Q58E, A71I/T/V, G73A/C/S/T, T74P, N83D and L89V). DRV RAMs and LPV RAMs were based on the IAS USA list [19]. The development of resistance at end point was studied in patients who experienced VF. The development of a mutation was defined as the detection of a mutation by population sequencing at end point, but not at baseline or pre-baseline. Adherence Adherence to the trial medication was assessed using patient self-reported adherence (Modified Medication Adherence Self-Report Inventory [M-MASRI]) [20] and pharmacokinetic sampling. Suboptimally adherent patients were defined as those with week 4 through to week 96 average adherence 95% [21] or those with at least one measurement below the lower limit of quantification (5.0 ng/ml and 10.0 ng/ml for DRV and LPV, respectively) at any time point. All patients with available data, including non-vfs, were included in the adherence assessments. Results Baseline characteristics In the ARTEMIS Trial, a total of 689 patients were randomized to receive once-daily DRV/r 800/100 mg (n=343) or LPV/r 800/200 mg total daily dose (n=346) plus fixed-dose TDF/FTC once daily. Baseline demographics and disease characteristics were well-balanced between treatment groups (Table 1). Mean baseline HIV-1 RNA was 4.85 log 10 copies/ml and the median CD4 + T-cell count was 225 cells/µl. Overall, most patients (64%) had Centers for Disease Control and Prevention (CDC) category A HIV infection (Table 1). The majority of patients (61%) harboured HIV-1 subtype B, 13% harboured HIV-1 subtype C, 17% harboured HIV-1 subtype CRF01_AE and 9% harboured another HIV-1 subtype. There were no relevant differences between the treatment groups with respect to any baseline protease mutations. Overall, at baseline, the median number of PI RAMs was 3 (range 0 11) and the median number of major (primary) PI RAMs was 0 (range 0 3). At baseline and/or pre-baseline, a low level of transmitted drug resistance was observed in the study population, with 2% of patients harbouring at least one PI RAM and 4% harbouring at least one nucleoside Antiviral Therapy

4 E Lathouwers et al. Table 1. Main baseline demographics, disease characteristics and resistance Demographic/characteristic DRV/r (n=343) LPV/r (n=346) Baseline demographics Female gender, n (%) 104 (30) 105 (30) Mean age, years (±sd) 36 (9) 35 (9) Ethnicity Caucasian, n (%) 137 (40) 153 (44) Black, n (%) 80 (23) 71 (21) Hispanic, n (%) 77 (22) 77 (22) Asian, n (%) 44 (13) 38 (11) Baseline disease characteristics Mean HIV-1 RNA, log 10 copies/ml (±sd) 4.86 (0.64) 4. (0.60) Median CD4 + T-cell count, cells/µl (range) 228 (4 750) 218 (2 714) HBV HCV-coinfected, n (%) 43 (13) 48 (14) HIV-1 subtype B, n (%) 210 (61) 208 (60) C, n (%) 39 (11) 51 (15) CRF01_AE, n (%) 62 (18) 55 (16) Other, n (%) 31 (9) 32 (9) CDC clinical stage of HIV infection A, n (%) 226 (66) 217 (63) B, n (%) 91 (27) 95 (27) C, n (%) 26 (8) 34 (10) Baseline resistance a PI RAMs, median (range) 3 (0 11) 3 (0 8) Major (primary) PI RAMs, median (range) 0 (0 3) 0 (0 3) NRTI RAMs, median (range) 0 (0 5) 0 (0 2) a Data from [19]. CDC, Centers for Disease Control and Prevention; DRV/r, darunavir/ritonavir; HIV-1, HIV type-1; LPV/r, lopinavir/ritonavir; NRTI, nucleoside reverse transciptase inhibitor; PI, protease inhibitor; RAM, resistance-associated mutation. Figure 2. Number of VFs, including rebounders and patients who were never suppressed VFs, % Never suppressed 40/343 (12%) DRV/r P<0.05 a Rebounders 59/346 (17%) LPV/r a P<0.05, Pearson χ 2 test. DRV/r, darunavir/ritonavir; LPV/r, lopinavir/ritonavir; VF, virological failure. reverse transciptase inhibitor (NRTI) RAM according to the list of transmitted drug resistance mutations proposed by Shafer et al. [22]. The prevalence of PI RAMs in this trial is in agreement with the reported prevalence in PI-naive HIV-1- infected patients [23]. Virological failures In the 96-week analysis, the VF rate was significantly lower in the DRV/r group (12%, 40/343 patients) than in the LPV/r group (17%, 59/346 patients; P=0.0437), with fewer rebounders or never suppressed patients in the DRV/r than the LPV/r group (Figure 2). A similar proportion of rebounders in the DRV/r and LPV/r groups (50% [12/24] versus 48% [16/33], respectively) had transient HIV-1 RNA increases (confirmed HIV-1 RNA 50 copies/ml) and subsequently returned to undetectable by the time of the final HIV-1 RNA measurement without any changes to their study regimen. VF was not associated with transmitted drug resistance in this study. Only one out of 40 DRV/r VFs harboured PI resistance mutations [22] at baseline or at pre-baseline (IV and L90M), and one DRV/r VF and one LPV/r VF harboured NRTI resistance mutations [22] (DRV/r: M41L, L210W and T215D/S/Y; and International Medical Press

5 Resistance analysis of ARV-naive patients on DRV/r or LPV/r in ARTEMIS Table 2. Emergence of resistance in VFs Patients DRV/r (n=343) LPV/r (n=346) VFs a 40 (12) 59 (17) Paired genotypes Developing major (IAS USA) PI RAMs b 0 (0) 0 (0) Developing minor (IAS USA) PI RAMs b 4 (13) 7 (15) Developing major non-polymorphic PI RAMs c 0 (0) 0 (0) Developing minor non-polymorphic PI RAMs c 1 (3) 2 (4) Developing (IAS USA) NRTI RAMs b 2 (6) 5 (11) Paired phenotypes Loss of susceptibility to any PI d 0 (0) 0 (0) Loss of susceptibility to FTC 2 (7) 4 (9) Loss of susceptibility to TDF 1 (3) 0 (0) Data are presented as n (%). a Based on HIV type-1 (HIV-1) RNA 50 copies/ml. b Data from [19]. c Data from [18]. d Darunavir (DRV), lopinavir (LPV), amprenavir, atazanavir, indinavir, nelfinavir, saquinavir and tipranavir. DRV/r, darunavir/ritonavir; FTC, emtricitibine; IAS, International AIDS Society; LPV/r, lopinavir/ritonavir; PI, protease inhibitor; RAM, resistance-associated mutation; TDF, tenofovir disproxil fumarate; VF, virological failure. LPV/r: T69D). These two DRV/r VFs were rebounders that had low transient increases in HIV-1 RNA that subsequently returned to undetectable at week 96 (no end point genotype or phenotype data were available for these patients as a result of low viral loads). The LPV/r patient with VF withdrew from the trial at week 16 with an unconfirmed undetectable viral load. Development of mutations and changes in ARV drug susceptibility in VFs The development of mutations was studied in VF patients with available baseline and end point genotypes. Figure 1 presents an overview of end point genotypes obtained from all patients with VFs according to the HIV-1 RNA levels. In the DRV/r group, 31 out of 40 VFs had paired baseline/end point genotypes of which 19 had HIV-1 RNA 50 to <1,000 copies/ml (76% success rate) and 12 had HIV-1 RNA 1,000 copies/ml (92% success rate). In the LPV/r group, 46 out of 59 VFs had paired baseline/end point genotypes, of which 20 had HIV-1 RNA 50 to <1,000 copies/ml (63% success rate) and 26 had HIV-1 RNA 1,000 copies/ml (100% success rate). Among the VFs in either group with an available genotype at baseline and end point, no major (primary) IAS USA PI RAMs developed (Table 2). In the DRV/r group, four VFs developed one or two IAS USA minor PI RAMs at end point: L10V, I13V and G16E (each in one patient) and I13V+G16E (in one patient; Table 2 and Figure 3). After exclusion of naturally occurring mutations (polymorphisms), only one VF in the DRV/r group developed one minor nonpolymorphic IAS USA PI RAM (L10V). None of the developing mutations corresponded to DRV RAMs. In the LPV/r group, seven VFs developed one or two IAS USA minor PI RAMs at end point (Table 2 and Figure 4). These included I13V (in two patients), L33V, M36I, A71V, V77I (each in one patient) and A71T+V77I (in one patient). After exclusion of poly morphisms, two VFs developed one minor nonpolymorphic IAS USA PI RAM (A71V and A71T). Both mutations are described as LPV RAMs. Changes in susceptibility to study PIs or NRTIs at end point as compared with baseline were studied in VFs with available phenotypes at both time points. Baseline and end point phenotypes (Antivirogram ) were available for 30 of the 40 VFs in the DRV/r group and 43 of the 59 VFs in the LPV/r group (Table 2). All HIV isolates from VFs with available phenotypes at baseline and end point remained susceptible at end point to all PIs, including the regimen PIs. Two VFs in the DRV/r group developed an NRTI RAM: M1V and M1I/V (FTC RAMs; Figure 3). The development of both these mutations was associated with decreased phenotypic susceptibility to FTC, which was included in the background regimen. A further VF in the DRV/r group did not develop any NRTI RAMs, but showed decreased susceptibility at end point to TDF, although the TDF FC value was just above the biological cutoff at end point (TDF FC value changed from 0.7 to 2.7 at end point). Five VFs in the LPV/r group developed an NRTI RAM: K70E (TDF RAM), M1I and M1V (in three patients; Figure 4). Development of the FTC RAMs M1I and M1V was associated with decreased phenotypic susceptibility to FTC. A detailed overview of resistance data for all VFs that developed PI or NRTI RAMs in the DRV/r and LPV/r treatment groups is presented in Figures 3 and 4, respectively. Adherence in VFs Overall, adherence was high. There was no significant difference between treatment groups with respect to the percentage of adherent patients throughout the trial: Antiviral Therapy

6 E Lathouwers et al. Figure 3. Overview of resistance data of all VFs that developed PI or NRTI RAMs in the DRV/r group VF EP Developing PI RAMs at EP FC DRV FC LPV FC FTC FC TDF HIV-1 RNA Patient profile a HIV-1 subtype BL HIV-1 Time Type RNA, of VF of copies/ml (visit) VF Visit HIV-1 RNA, copies/ml Major IAS USA PI RAMs b Minor IAS USA PI RAMs b Major nonpolymorphipolymorphic IAS USA NRTI Minor non- Developing PI RAMs c PI RAMs c BL EP BL EP RAMs at EP b BL EP BL EP 1 CRF01_AE 1,520,000 NS 96 6,720 I13V, G16E M1I/V 0.4 > CRF02_AG 39, ,020 L10V L10V B 612,000 NS I13V B 450, G16E < B 140, M1V 0.8 > a X-axis labels show time of visit in weeks; Y-axis labels show log10 HIV type-1 (HIV-1) RNA copies/ml; the horizontal line on graphs represents HIV-1 RNA 50 copies/ml; and the arrow indicates time of the genotype end point (EP; last available genotype/phenotype on treatment) visit. b Data from [19]. c Data from [18]. For Antivirogram data, fold change (FC) in 50% effective concentration values above the biological/clinical cutoff (decreased susceptibility) are shown in italics, and baseline (BL) to EP changes in FC values from susceptible to decreased susceptibility are presented in shaded boxes. DRV, darunavir; DRV/r, darunavir/ritonavir; FTC, emtricitabine; IAS USA, International AIDS Society; LPV, lopinavir; NRTI, nucleoside reverse transcriptase inhibitor; NS, never suppressed patient; PI, protease inhibitor; RAM, resistance-associated mutation;, rebounder; TDF, tenofovir disoproxil fumarate; VF, virological failure;, week International Medical Press

7 Resistance analysis of ARV-naive patients on DRV/r or LPV/r in ARTEMIS Figure 4. Overview of resistance data of all VFs that developed PI or NRTI RAMs in the LPV/r group VF EP Developing PI RAMs at EP FC DRV FC LPV FC FTC FC TDF Patient HIV-1 RNA profile a HIV-1 subtype BL HIV-1 RNA, copies/ml Time of VF (visit) Type of VF Visit HIV-1 RNA, copies/ml LPV dose Major IAS USA PI RAMs b Minor IAS USA PI RAMs b Major nonpolymorphic PI RAMs c Minor nonpolymorphic c PI RAMs BL EP BL EP Developing IAS USA NRTI b RAMs at EP BL EP BL EP 1 B 205,000 NS 142 bid I13V B 13, ,400 bid I13V B 10,100 NS 12 1,170 bid L33V B 23, ,800 bid M36I B 8,800 NS qd A71V A71V B 517,000 NS ,000 bid A71T, V77I A71T B 356, bid V77I 0.8 NA 1.5 NA 0.9 NA 1.1 NA 8 B 494,000 NS 83,000 qd M1V 1.4 > C 171,000 NS 16 12,200 bid M1V 3.3 > CRF01 _AE 16, ,440 bid M1V 0.7 > B 349, bid + qd M1I 1.8 > C 86, ,270 bid K70E a X-axis labels show time of visit in weeks; Y-axis labels show log10 HIV type-1 (HIV-1) RNA copies/ml; the horizontal line on graphs represents HIV-1 RNA 50 copies/ml; and the arrow indicates time of the genotype end point (EP; last available genotype/phenotype on treatment) visit. b Data from [19]. c Data from [18]. For Antivirogram data, fold change (FC) in 50% effective concentration values above the biological/clinical cutoff (decreased susceptibility) are shown in italics, and baseline (BL) to EP changes in FC values from susceptible to decreased susceptibility are presented in shaded boxes. Lopinavir (LPV) resistance-associated mutations (RAMs; International AIDS Society [IAS] USA list) are shown in bold. BID, twice daily; DRV, darunavir; FTC, emtricitabine; LPV/r, lopinavir/ritonavir; NA, not available; NRTI, nucleoside reverse transcriptase inhibitor; NS, never suppressed patient; PI, protease inhibitor; qd, once daily;, rebounder; TDF, tenofovir disoproxil fumarate; VF, virological failure;, week. Antiviral Therapy

8 E Lathouwers et al. the percentage of adherent patients (M-MASRI, self-reported week 4 96 mean adherence >95%) was 83% for DRV/r patients and 78% for LPV/r patients [24]. A higher frequency of suboptimal adherence (M-MASRI) was reported in patients with VF compared with patients without VF (defined as >50 copies/ml, using TLOVR non-vf censored imputation algorithm); this difference was greater in the LPV/r than the DRV/r group. In this analysis, 24% of the DRV/r VFs versus 17% of patients in the DRV/r group excluding the VFs were suboptimally adherent. In the LPV/r group, 40% of VFs versus 20% of patients in the LPV/r group excluding the VFs were suboptimally adherent. According to the pharmacokinetic sampling, 20% of the VFs in the DRV/r group had at least one DRV plasma concentration measurement below the lower limit of quantification and were classed as suboptimally adherent, as compared with 4% of patients in the DRV/r group excluding the VFs. Similarly, in the LPV/r group more VFs than the subgroup excluding the VFs were suboptimally adherent according to pharmacokinetic sampling (32% versus 9%, respectively). Discussion In addition to durable efficacy, favourable tolerability and convenience, the potential for the development of resistance is an important factor to consider when choosing a first-line ARV regimen [25]. The sustained efficacy and favourable tolerability of oncedaily DRV/r 800/100 mg in treatment-naive patients in ARTEMIS have been previously reported [9]. The present analyses showed that once-daily DRV/r 800/100 mg was associated with a low rate of VF, lower than that with LPV/r, and no major IAS USA PI RAMs developed over 96 weeks. As expected in this treatment-naive patient population, there was a low frequency of patients harbouring transmitted resistance mutations [22] at screening and/ or at baseline. The prevalence of PI RAMs at baseline in this trial is consistent with the reported prevalence in PI-naive HIV-1-infected patients [23]. Importantly, transmitted resistance was not associated with an increased risk of VF in this study. The percentage of patients with VF was low in both the DRV/r and LPV/r groups. The lower percentage of VFs on DRV/r compared with LPV/r might be a result of the longer plasma half-life of DRV (15 h) versus LPV (5 6 h) and of the tight binding of DRV to wild type HIV-1 protease. This binding affinity is >100-fold higher compared with other PIs, including LPV, because of the very high dissociative half-life of DRV, indicating that the duration of DRV drug action is longer than LPV [26]. It should be noted that the current study used the COBAS HIV-1 Ampliprep Amplicor Monitor ultrasensitive assay version 1.5 (Roche Diagnostics) to quantify viraemia. When other assays such as the COBAS Ampliprep TaqMan HIV-1 assay are employed to monitor viral load, a greater frequency of patients with viraemia might be recorded. However, the clinical relevance of this increased HIV RNA detection has not been clearly established [27]. The same resistance testing protocols were used for samples with low HIV RNA concentration. It has been observed that for HIV-1 RNA levels ranging from 0 1,000 copies/ml, the success rate increases from 28% (0 100 copies/ml) to 80% (901 1,000 copies/ml). Genotyping can be reliably performed on samples with HIV-1 RNA between 500 and 1,000 copies/ml [28]. In this study, samples from VFs with HIV-1 RNA<500 copies/ml were also analysed. The risk of these results from plasma samples with a very low viral load being less representative of the circulating viral population is higher than with results from samples with a higher viral load. However, in this study, all attempts were made to detect the presence of resistance mutations at end point independently of the viral load, recognizing that when a mutation is detected it is present although possibly on less abundant quasispecies. All PI and NRTI RAMs detected by population sequencing at end point, but not at baseline or at pre-baseline, were presented. Consistent with previous first-line boosted PI-based studies [29], no major PI RAMs developed in VFs with an available genotype at baseline and end point in either the DRV/r or LPV/r group and almost all minor IAS USA PI RAMs that developed in VFs were polymorphic. Furthermore, in both DRV/r and LPV/r groups, all VFs with available phenotypes at baseline and end point remained phenotypically susceptible at end point to all PIs, including the study PIs used in the regimen. The reasons for the absence of major protease RAMs in treatment-naive patients failing a boosted PI-based regimen are not fully understood, but might result from a complex interaction between the favourable pharmacokinetics and cell/compartment penetration and the high genetic barrier to the development of resistance [29]. Overall adherence in ARTEMIS was high. The correlation between adherence and treatment success has been demonstrated across a number of ARV therapies, including PIs [30,31]. Consistent with these findings, in ARTEMIS a higher frequency of suboptimal adherence was observed in VFs than in patients not failing virologically in both treatment groups. The difference in adherence level between VFs and patients not failing virologically was greater for LPV/r than for DRV/r patients International Medical Press

9 Resistance analysis of ARV-naive patients on DRV/r or LPV/r in ARTEMIS The findings from this trial are consistent with those from other clinical trials (CASTLE [ATV/r], GEMINI [SQV/r] and KLEAN [fosamprenavir/r]), which examined the resistance profiles of ritonavir-boosted PIs versus LPV/r in treatment-naive patients [18,32,33]. Although comparison across trials is difficult because of multiple confounders, patients in these trials demonstrated low rates of VF and a low frequency of development of major PI RAMs [18,32,33]. In conclusion, the findings of these resistance analyses in treatment-naive patients from ARTEMIS show a low rate of VF with once-daily DRV/r 800/100 mg, which was lower than with LPV/r. Furthermore, the results of these analyses also confirm the lack of development of any major PI RAM with once-daily DRV/r 800/100 mg or LPV/r 800/200 mg total daily dose and demonstrate that phenotypic susceptibility to all PIs studied was preserved in patients with VF in ARTEMIS. Overall, these analyses could provide a better understanding of the resistance profile of once-daily DRV/r in treatment-naive HIV-1-infected patients, which might assist clinicians when making treatment decisions. Acknowledgements The authors wish to thank Gilead for the kind donation of Truvada (TDF and FTC) for this study. The authors would also like to thank Eric Lefebvre and Bjorn Daems from Tibotec for their important contributions to the manuscript. In addition, the authors would like to acknowledge Catherine Elliott (Medical Writer, Gardiner Caldwell Communications) for assistance in drafting the manuscript and collating author contributions. Financial assistance to support this service was provided by Tibotec BVBA. The ARTEMIS study was funded by Tibotec BVBA. Disclosure statement All authors are current employees of Tibotec. References 1. Perno CF, Moyle G, Tsoukas C, Ratanasuwan W, Gatell J, Schechter M. Overcoming resistance to existing therapies in HIV-infected patients: the role of new antiretroviral drugs. J Med Virol 2008; 80: Hirsch MS, Günthard HF, Schapiro JM, et al. Antiretroviral drug resistance testing in adult HIV-1 infection: 2008 recommendations of an International AIDS Society USA panel. Clin Infect Dis 2008; 47: US Department of Health and Human Services panel on antiretroviral guidelines for adults and adolescents. Guidelines for the use of antiretroviral agents in HIV- 1-infected adults and adolescents (Updated 1 December Accessed 09 February 2010.) Available from AdultandAdolescentGL.pdf 4. World Health Organization. Adherence to long-term therapies: evidence for action (2003). (Accessed 09 February 2010.) Available from s4883e/s4883e.pdf 5. De Meyer S, Azijn H, Surleraux D, et al. TMC114, a novel human immunodeficiency virus type 1 protease inhibitor active against protease inhibitor-resistant viruses, including a broad range of clinical isolates. Antimicrob Agents Chemother 2005; 49: de Béthune MP, Hertogs K. Screening and selecting for optimized antiretroviral drugs: rising to the challenge of drug resistance. Curr Med Res Opin 2006; 22: Tibotec, Inc. PREZISTA (darunavir) prescribing information. (Updated January Accessed 09 February 2010.) Available from 8. European Medicines Agency. PREZISTA (darunavir) summary of product characteristics (Updated October Accessed 09 February 2010.) Available from emea-combined-h707en.pdf 9. Mills AM, Nelson M, Jayaweera D, et al. Once-daily darunavir/ritonavir vs. lopinavir/ritonavir in treatmentnaive, HIV-1-infected patients: 96-week analysis. AIDS 2009; 23: Ortiz R, DeJesus E, Khanlou H, et al. Efficacy and safety of once-daily darunavir/ritonavir versus lopinavir/ritonavir in treatment-naive HIV-1-infected patients at week 48. AIDS 2008; 22: Cohen C. The ARTEMIS trial: once-daily darunavir/ ritonavir in the management of treatment-naive, HIVinfected patients. HIV Therapy 2009; 3: Los Alamos National Laboratory. HIV-1 subtype and circulating recombinant form reference sequences (Accessed 09 February 2010.) Available from De Meyer S, Vangeneugden T, Lefebvre E, et al. Phenotypic and genotypic determinants of resistance to TMC114: pooled analysis of POWER 1, 2 and 3. 15th International HIV Drug Resistance Workshop June 2006, Sitges, Spain. Abstract Abbott Laboratories. KALETRA (lopinavir/ritonavir) prescribing information. (Updated October Accessed 09 February 2010.) Available from Boehringer Ingelheim. APTIVUS (tipranavir) prescribing information. Boehringer Ingelheim (Updated June Accessed 09 February 2010.) Available from aptivus.com 16. Miller MD, Margot N, Lu B, et al. Genotypic and phenotypic predictors of the magnitude of response to tenofovir disoproxil fumarate treatment in antiretroviralexperienced patients. J Infect Dis 2004; 189: Johnson VA, Brun-Vezinet F, Clotet B, et al. Update of the drug resistance mutations in HIV-1: Top HIV Med 2007; 15: Molina JM, Andrade-Villanueva J, Echevarria J, et al. Atazanavir/ritonavir vs lopinavir/ritonavir in antiretroviralnaive HIV-1-infected patients: CASTLE 96 week efficacy and safety. 48th Annual ICAAC/IDSA October 2008, Washington, DC, USA. Abstract H-1250d. 19. Johnson VA, Brun-Vezinet F, Clotet B, et al. Update of the drug resistance mutations in HIV-1: December 2009 Top HIV Med 2009; 17: Walsh JC, Mandalia S, Gazzard BG. Responses to a 1 month self-report on adherence to antiretroviral therapy are consistent with electronic data and virological treatment outcome. AIDS 2002; 16: Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med 2000; 133: Shafer RW, Rhee SY, Pillay D, et al. HIV-1 protease and reverse transcriptase mutations for drug resistance surveillance. AIDS 2007; 21: Antiviral Therapy

10 E Lathouwers et al. 23. Stanford University HIV Drug Resistance Database. Mutation prevalence according to subtype and treatment (Updated January Accessed 09 February 2010.) Available from MutPrevBySubtypeRx.cgi 24. Nelson M, Girard PM, DeMasi R, et al. Adherence to darunavir/ritonavir and lopinavir/ritonavir in treatmentnaive, HIV-infected patients in ARTEMIS: 96-week data. 15th Annual Conference of the British HIV Association (BHIVA). 1 3 April 2009, Liverpool, UK. Abstract Hammer SM, Eron JJ, Jr., Reiss P, et al. Antiretroviral treatment of adult HIV infection: 2008 recommendations of the International AIDS Society USA panel. JAMA 2008; 300: De Wit M, Keuleers I, Gustin E, et al. Binding kinetics of protease inhibitors to wild-type and multi-drug resistant HIV-1 proteases: a mechanistic study of the genetic barrier to resistance of darunavir. 14th Conference on Retroviruses and Opportunistic Infections February 2007, Los Angeles, CA, USA. Abstract M Lima V, Harrigan R, Montaner JS. Increased reporting of detectable plasma HIV-1 RNA levels at the critical threshold of 50 copies per milliliter with the TaqMan assay in comparison to the Amplicor assay. J Acquir Immune Defic Syndr 2009; 51:3 6. Accepted 25 June 2010; published online 17 January Pattery T, De Wolf H, Verlinden H, et al. Efficiency of HIV-1 PR-RT genotyping on clinical isolates with viral load less than 1000 copies/ml: a 12 year analysis. Reviews in Antiviral Therapy and Infectious Diseases. 2010; 1: Zazzi M. Lack of major resistance mutations in HIV-1 protease at failure of boosted protease inhibitor-based first-line antiretroviral treatment. Rev Antivir Ther 2007; 2: Wood E, Hogg RS, Yip B, Harrigan PR, O Shaughnessy MV, Montaner JS. The impact of adherence on CD4 cell count responses among HIV-infected patients. J Acquir Immune Defic Syndr 2004; 35: Moore DM, Hogg RS, Yip B, et al. Discordant immunologic and virologic responses to highly active antiretroviral therapy are associated with increased mortality and poor adherence to therapy. J Acquir Immune Defic Syndr 2005; 40: Walmsley S, Avihingsanon A, Slim J, et al. Gemini: a noninferiority study of saquinavir/ritonavir versus lopinavir/ ritonavir as initial HIV-1 therapy in adults. J Acquir Immune Defic Syndr 2009; 50: Eron J, Jr., Yeni P, Gathe J, Jr., et al. The KLEAN study of fosamprenavir-ritonavir versus lopinavir-ritonavir, each in combination with abacavir-lamivudine, for initial treatment of HIV infection over 48 weeks: a randomised noninferiority trial. Lancet 2006; 368: International Medical Press