Rilpivirine (RPV, TMC278, Edurant) is a new non-nucleoside reverse transcriptase inhibitor (NNRTI) recently approved in the US, Canada and Europe for use in combination with other antiretrovirals (ARVs) in HIV-1-infected treatment-naive adult patients and also as a single-tablet regimen with tenofovir disoproxil fumarate (TDF) and emtricitabine (FTC) (Complera [US]; Eviplera [EU]).1–4 These approvals were based on the results of ECHO (TMC278-C209, NCT00540449)5 and THRIVE (TMC278-C215, NCT00543725).6 These were 96-week global, phase 3, randomized, double-blind, double-dummy, active-controlled trials with nearly identical design. In the week-48 primary analysis of each trial, RPV 25 mg once daily showed noninferior efficacy to efavirenz (EFV) 600 mg once daily (primary objective). RPV had a higher rate of virologic failure but lower rates of adverse events (AEs) leading to discontinuation, grade 2–4 overall AEs at least possibly related to treatment, rash, dizziness, abnormal dreams/nightmares, and grade 3 or 4 lipid abnormalities than EFV.5,6
As the ECHO and THRIVE trials have a nearly identical design, a preplanned pooled week-48 analysis of the data was conducted. The greater statistical power of this larger data set compared with the individual trials also allowed for: (1) subgroup analyses to be performed; and (2) a more in depth analysis of predictors of response and the higher rate of virologic failure with RPV. A detailed pooled virology analysis was described separately.7
ECHO and THRIVE Patient Populations and Study Designs
The trial design and methods have been reported in detail for the individual trials.5,6 Main inclusion criteria included treatment-naive, HIV-1-infected adults with baseline viral load ≥5000 copies per milliliter and confirmed viral sensitivity to the background nucleoside/nucleotide reverse transcriptase inhibitors (N[t]RTIs) (assessed using the vircoTYPE HIV-1 assay). Patients were excluded if they had documented presence of any NNRTI resistance-associated mutation (RAM) from a list of 39 (The list includes A98G, L100I, K101E/P/Q, K103H/N/S/T, V106A/M, V108I, E138A/G/K/Q/R, V179D/E, Y181C/I/V, Y188C/H/L, G190A/C/E/Q/S/T, P225H, F227C, M230I/L, P236L, K238N/T, and Y318F).5,6,8
In both trials, patients were randomized 1:1 to RPV 25 mg with EFV placebo once daily or EFV 600 mg with RPV placebo once daily, both in addition to (i) a fixed background N[t]RTI regimen of TDF and FTC in ECHO, or (ii) an N[t]RTI regimen based on the investigator's choice of TDF/FTC, zidovudine/lamivudine (3TC), or abacavir/3TC in THRIVE.5,6 Randomization was stratified by background regimen (THRIVE) and screening viral load (≤100,000 >100,000 to ≤500,000, and >500,000 copies per milliliter).
Although the NNRTIs are dosed once daily, to maintain the double-blind design, RPV/RPV placebo was taken with food, whereas EFV/EFV placebo was taken on an empty stomach at bedtime. These global trials were conducted at 112 sites across 21 countries for ECHO and at 98 sites across 21 countries for THRIVE. The site locations differed between trials with some overlap of countries.
The primary objective of both studies was to demonstrate noninferiority in confirmed overall response with RPV versus EFV, with a margin of 12% at week 48 according to the intent-to-treat time-to-loss-of-virologic-response (ITT-TLOVR) imputation algorithm.9 Secondary endpoints included durability of antiviral activity, treatment adherence as measured by the Modified Medication Adherence Self-Report Inventory (M-MASRI), CD4+ cell count changes from baseline, safety, tolerability, and HIV genotypic characteristics in virologic failures in the resistance analysis (VFres [VFres is defined as any patient in the intent-to-treat population experiencing treatment failure regardless of time of failure, treatment status, or reason for discontinuation providing the following criteria were met: never achieved 2 consecutive viral load values <50 copies per milliliter and had an increase in viral load ≥0.5 log10 copies per milliliter above the nadir (never suppressed) or first achieved 2 consecutive viral load values <50 copies per milliliter followed by 2 consecutive (or single, when last available) viral load values ≥50 copies per milliliter (rebounder)]). An outcome analysis using the ITT-snapshot approach at week 48 (ie, viral load <50 copies per milliliter observed and patients with missing viral load at week 48 classified as failures) was also performed. For the ITT-snapshot analysis, the last available viral load value in the week-48 time-point window (weeks 44–54) was used to determine response. The non–VFres-censored population (which excludes patients discontinuing for reasons other than virologic failure) was used for a sensitivity analysis of the primary endpoint.
AEs were monitored from screening onwards and at each visit throughout the trial and were coded using the Medical Dictionary for Regulatory Activities (MedDRA, version 11.0), with severity determined according to the DAIDS grading scale.10 Based on preclinical data, which showed effects of RPV on adrenal steroidogenesis in mice, rats, and dogs but not in juvenile female cynomolgus monkeys,11 endocrine parameters were assessed. An electrocardiogram was conducted at screening and at weeks 2, 12, 24, and 48.
ECHO and THRIVE Pooled Analysis
In this report, preplanned pooled analyses of the primary and secondary endpoints described above were conducted, enabling a more comprehensive assessment of the efficacy and safety of RPV in a larger sample size. As mentioned above, the greater statistical power of this data set allowed several subgroup analyses to be performed on the pooled population as follows: baseline viral load, background regimen, gender, race, CD4+ cell count, and clade. Analysis of response by self-reported adherence (assessed by M-MASRI by ≤95% vs. >95%) was also performed in the pooled population.
All presented statistical analyses were preplanned, unless otherwise stated. Fisher exact test (5% significance level) was used to compare preplanned AEs, for which a significant difference had been recorded in the phase 2b trial.12 Lipid changes between treatment groups were compared using the nonparametric Wilcoxon rank-sum test.
Baseline Characteristics and Patient Disposition
Baseline characteristics were well balanced between the treatment groups (see Table, Supplemental Digital Content 1,http://links.lww.com/QAI/A285). Twenty-four percent of patients were female, and 61% were white. N[t]RTI background regimens were similar between groups, with 80% of patients receiving TDF/FTC.
At week 48, patient dispositions between groups across the 2 trials were similar (see Figure, Supplemental Digital Content 2,http://links.lww.com/QAI/A286). Similar proportions of patients in the RPV and EFV groups were still on treatment at week 48. The most common reasons for discontinuation as reported by the trial investigators were AEs and reaching of virologic endpoint. The most common major protocol violation in both trials was use of a disallowed drug (usually a proton-pump inhibitor) during the treatment period.
Overall treatment response (confirmed viral load <50 copies per milliliter; ITT-TLOVR) at week 48 was similar with RPV 84% (578 of 686) and EFV 82% (561 of 682) (Table 1). The difference in response rates [95% confidence interval (CI)] was 2.0% (–2.0% to 6.0%). Virologic failure rates in the efficacy analysis (VFeff [VFeff included rebounders: confirmed response before week 48 with confirmed rebound at or before week 48 or never suppressed: patients with no confirmed response before week 48]) were 9% for RPV and 5% for EFV. There were no major differences between the RPV and EFV groups in percentages of responders over time (Fig. 1A).
The model adjusted responses (using the covariate baseline log10 plasma viral load, background N(t)RTI regimen and study as factors), the TLOVR per-protocol result, and the ITT-snapshot results were all similar to ITT-TLOVR responses (Table 1). In a preplanned analysis focused on the virologic failures, using the TLOVR non–VFres-censored population, response rates were lower with RPV than with EFV (Table 1).
Some patients did not have available M-MASRI data (missing from 59 RPV patients versus 95 EFV patients). Of those patients with data, the majority self-reported >95% adherence (assessed by M-MASRI) (Table 1). Most patients had baseline viral load ≤500,000 copies per milliliter or CD4+ cell count >50 cells per cubic millimeter (Table 1). Response rates were high and similar between treatment groups in these 3 subgroups. Suboptimal adherence and higher baseline viral load were associated with lower responses in both treatment groups. Overall responses decreased with lower baseline CD4+ cell counts in the RPV group, with little variation in the EFV group (Table 1). The response rate in those with both baseline viral load >500,000 copies per milliliter and CD4+ cell count <50 cells per cubic millimeter was similar in each treatment group [RPV 71% (10 of 14) vs. EFV 73% (11 of 15)]. The effect of suboptimal adherence, baseline viral load >500,000 copies per milliliter, or CD4+ cell count <50 cells per cubic millimeter on VFeff was more apparent in the respective RPV than EFV subgroups. However, numbers of patients in these 3 subgroups were low (Table 1).
Response rates were similar between the 2 treatments by background N[t]RTI regimen, gender, race, and clade (Table 2). In both treatment groups, response rates were lowest in black/African American patients and highest in Asian patients and in patients with clade CRF01_AE. Discontinuations for other reasons were greater in Black/African American patients (RPV and EFV: 10%) and lower in Asian patients (RPV: 1% vs. EFV: 0%) when compared with the overall population (Table 2).
CD4+ Cell Count Response
The mean (95% CI) change from baseline in CD4+ cell count to week 48 was 192 (181 to 203) cells per cubic millimeter for RPV versus 176 (165 to 188) cells per cubic millimeter for EFV (Fig. 1B). Corresponding mean (95% CI) changes from baseline in observed CD4+ cell counts were 217 (207 to 228) and 207 (195 to 219) cells per cubic millimeter, respectively.
Full resistance data for the pooled week-48 analysis have been presented previously7 and are presented in Supplemental Table 2 (see Supplemental Digital Content 2,http://links.lww.com/QAI/A287). The resistance analysis included data beyond week 48. In total, 72 patients (10%) in the RPV and 39 patients (6%) in the EFV group, met the definition for VFres. Of these, 62 RPV (86%) and 28 EFV (72%) VFres had resistance data at time of failure. Due to low viral load at failure, some VFres could not be genotyped or phenotyped. In the RPV group, 29% (18 of 62) and in the EFV group 43% (12 of 28) of VFres had no NNRTI or International AIDS Society N(t)RTI RAMs13; 63% (39 of 62) and 54% (15 of 28) had emergent NNRTI RAMs, and 68% (42 of 62) and 32% (9 of 28) had emergent IAS N(t)RTI RAMs, respectively. The most prevalent treatment-emergent NNRTI and IAS N(t)RTI RAMs in RPV VFres were, respectively, E138K (28 of 39; 72%) and M184I (29 of 42; 69%) and in EFV VFres were K103N (11 of 15; 73%) and M184V (6 of 9; 67%). Of the 31 VFres on RPV, phenotypically resistant to RPV [fold change (FC) > biological cut-off of 3.7], 14 of 31 (45%) were resistant to nevirapine (FC > 6.0), 27 of 31 (87%) were resistant to EFV (FC > 3.3), and 28 of 31 (90%) were resistant to etravirine (ETR) (FC > 3.2). Twelve of the 28 EFV VFres were phenotypically resistant to EFV. All the EFV VFres resistant to EFV were cross-resistant to nevirapine, and all remained susceptible to RPV and ETR.
The safety data are those available at the time of the week-48 analysis, and therefore include some data beyond week 48. The median treatment duration was 56 weeks in both groups.
The incidence of any grade 2–4 AE at least possibly related to treatment was significantly lower in the RPV group than in the EFV group (Table 3). The most frequent grade 2–4 AEs at least possibly related to treatment (excluding laboratory abnormalities) and seen in ≥2% of patients in either group were rash, dizziness, abnormal dreams/nightmares, headache, insomnia, and nausea (Table 3).
The incidence of AEs leading to permanent discontinuation was lower in the RPV group than in the EFV group. The most common AEs leading to discontinuation were any rash (0.1% of RPV vs. 1.8% of EFV patients) and depression (0.3% vs. 0.6% of patients, respectively). Pregnancy led to discontinuation in 0.4% of patients in each group.
There were 5 deaths [1 in the RPV group (grade 3 bronchopneumonia) and 4 in the EFV group (grade 3 Burkitt's lymphoma, grade 4 cerebral toxoplasmosis/respiratory failure, grade 4 dysentery, grade 4 cerebrovascular accident)], none of which was considered related to treatment.
The incidence of any rash (grouped term) at least possibly related to treatment was significantly lower in the RPV group than in the EFV group (Table 3). The incidence of rash was highest in the first 4 weeks of treatment. The majority of rashes were of grade 1 or 2 severity, and there were no grade 4 rashes.
Neurologic AEs, including dizziness, and psychiatric disorders, including abnormal dreams/nightmares at least possibly related to treatment (any grade), were observed significantly less often with RPV than with EFV (Table 3). Most neurologic and psychiatric events of interest were grade 1 or 2. Table 3 also shows that the incidence of any grade 2−4 treatment-emergent laboratory abnormalities was lower with RPV than with EFV.
RPV was associated with significantly smaller mean changes from baseline in total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein (HDL) cholesterol and triglyceride levels than EFV (Fig. 2). Both mean low-density lipoprotein cholesterol and triglyceride levels did not increase above baseline with RPV, but did with EFV. There was no statistical difference in the change from baseline at week 48 in the total/HDL-cholesterol ratio between groups (Fig. 2).
There was a minimal change from baseline in mean serum creatinine with RPV at week 2 (first on-treatment assessment), remaining stable to week 48 (RPV 0.1 mg/dL vs. EFV 0 mg/dL). This seems to arise from an effect on the tubular secretion of creatinine and not from a change in glomerular filtration rate.6
There was a small decrease from baseline at week 48 in basal cortisol of 13.1 nmol/L for RPV and a small increase of 9.0 nmol/L for EFV. The proportions of patients with at least 2 consecutive (treatment emergent) abnormal cortisol responses to an adrenocorticotropic hormone test (≤500 nmol/L) during the trial were 1.7% (11 of 643) in the RPV group and 0% in the EFV group. No patients had signs or symptoms of adrenal insufficiency or discontinued the study secondary to abnormal adrenocorticotropic hormone test results. There were no clinically relevant changes from baseline at week 48 in androstenedione, dehydroepiandrosterone sulfate, luteinizing hormone, total testosterone, or progesterone.
The QT interval of the electrocardiogram, corrected using Frederica formula (QTcF interval) increased over time in both groups; mean (95% CI) increases were +11 (10 to 13) milliseconds and 13 (12 to 14) milliseconds, respectively. There were no QTcF intervals >500 milliseconds.
Because the phase 3 studies were identical in design except for background N[t]RTI regimen and geographic distribution, a pooled analysis was performed as predefined by the statistical analysis plans to increase the robustness of the analyses of the data. The 84% treatment response (ITT-TLOVR) with RPV and 82% with EFV were at least as high as has been reported in previous trials in treatment-naive patients.14–18 Mean increases in CD4+ cell count occurred in both groups throughout the 48-week period.
Response rates were similar with EFV and RPV irrespective of background regimen and baseline characteristics such as gender, race, and HIV clade. Response rates were lowest in black/African American patients in both treatment groups and highest in Asian patients, consistent with findings from other trials.19–21 Discontinuations for other reasons were highest in black/African American patients and lowest in Asian patients. Highest responses were observed in patients with clade CRF01_AE, which also had the highest incidence in Asian patients. As expected,22–27 suboptimal adherence (≤95% by M-MASRI) and higher baseline viral load were correlated with a lower rate of responses. In patients with suboptimal adherence, response rates were similar between treatment groups. There was a greater influence of higher baseline viral load reducing response in the RPV group compared with in the EFV group. Lower baseline CD4+ cell count was also a predictor of response in the RPV group. However, the response rates of patients with both higher baseline viral load and low CD4+ cell count were similar in each treatment group. The effect of CD4+ cell count on treatment response was largely, but not completely explained by the effect of baseline viral load.
Treatment success in patients with HIV-1 infection is achieved through a combination of virologic suppression and the patient's ability to tolerate treatment. Overall, RPV had a more favorable tolerability profile than EFV with fewer treatment discontinuations. The rate of virologic failures was low in both treatment groups while numerically higher for RPV, but still within the ranges described in antiretroviral trials performed recently in treatment-naive HIV-infected patients.18,28,29
Consistent with the higher VFres rate for RPV, the non–VFres-censored analysis showed a lower response rate with RPV than with EFV. This analysis excludes the beneficial impact of better tolerability of the RPV regimen. Lower adherence to a HIV-1 treatment regimen has been shown to be a significant predictor of virologic failure,23,24 sometimes the strongest predictor.25 This was an important factor here as well, but the difference between treatment groups in VFeff cannot be explained by suboptimal adherence alone. In both treatment groups, in addition to suboptimal adherence, higher baseline viral load was associated with increased VFeff rates. The effect of suboptimal adherence or of higher baseline viral load on VFeff was more apparent in the RPV than the EFV respective subgroups. These observations contribute to understanding the reasons for the higher virologic failure rate with RPV than with EFV. As described previously for RPV, high baseline viral load was a predictor of a higher rate of VFres, treatment-emerging RAMs in VFres and cross-resistance to the NNRTI class compared with EFV.7 However, the proportion of discontinuations for AEs/deaths and, in most cases, the proportion of discontinuations for other reasons were lower for RPV than for EFV irrespective of adherence or baseline viral load. The United States prescribing information provides information about virologic failure with RPV, stating that more RPV patients with baseline viral load >100,000 copies per milliliter experienced virologic failure compared with patients with baseline viral load ≤100,000 copies per milliliter.1 In Europe, rilpivirine in combination with other ARVs, is indicated for the treatment of HIV-1 infection in ARV treatment-naive adults with a viral load ≤100,000 copies per milliliter.2
There are additional factors which may have contributed to the virologic outcomes observed. Due to the double-dummy design, patients had to take trial medication twice a day rather than once daily, which may have negatively impacted adherence. Furthermore, the recommendation for patients to take RPV (or matching placebo) with a meal may not have been followed strictly, resulting in some patients taking RPV on an empty stomach, potentially resulting in suboptimal RPV absorption in some cases. The roles of adherence, drug exposure, N(t)RTI combination, baseline viral load, and CD4+ cell count in virologic failure are being analyzed further.
Among patients with VFres, the rate of resistance was high for both NNRTIs and N(t)RTIs, as reported with other NNRTI-based regimens.30 The proportion of VFres with ≥1 emergent NNRTI RAM was similar in each group, whereas the proportion of VFres with ≥1 emergent N[t]RTI RAM was higher in the RPV group than in the EFV group. NNRTI RAMs seen in RPV VFres were typically not found in the EFV VFres, and vice versa. More RPV patients developed 3TC/FTC-associated resistance compared with EFV. In VFres phenotypically resistant to RPV, there was 90% phenotypic cross resistance with ETR. An analysis of phenotypic sensitivity to NNRTIs is presented separately for the pooled data.7 At low baseline viral load, ETR cross-resistance was less common than at high baseline viral load.7 The clinical implications of all these findings are not yet elucidated.
The most frequent treatment-emergent NNRTI RAM in RPV VFres was E138K, a RAM for which the clinical implications are not yet fully understood. Currently, E138K has a low prevalence in routine clinical resistance testing (<1%).31 In the RPV VFres from these phase 3 studies, E138K never emerged in isolation and always occurred with other NNRTI RAMs and/or N(t)RTI RAMs, most often M184I/V.7 The high proportion of M184I in RPV VFres could be in part explained by the effect that this RAM has in combination with E138K on RPV susceptibility.7,32,33 An alternative explanation, although not confirmed in a recent study,33 could be that M184I improves the fitness of E138K-containing isolates.34,35 Although clinical data are currently lacking, as discussed previously, 7 the likelihood that ETR retains full activity after RPV virologic failure, with treatment-emergent RAMs, seems low. In vitro data support that after an EFV failure, RPV would be active, but there are no clinical data available to assess this.
As in the open-label phase 2b TMC278-C204 trial,12,36 RPV was associated with significantly lower incidences than EFV of grades 2–4 AEs including rash, dizziness, and abnormal dreams/nightmares. Incidences of depression were low and similar in both groups. Changes in HDL-cholesterol and proatherogenic lipid profiles were less pronounced with RPV than with EFV, but the changes in total cholesterol/HDL-cholesterol ratios were similar between groups. A small but higher proportion of patients on RPV than on EFV had abnormal cortisol responses, but this was not considered clinically relevant. There was a very small but significant increase in serum creatinine in the RPV group; this is most likely related to changes in tubular secretion of creatinine and not to direct effects on glomerular filtration as described previously.6 The change in QTcF interval from baseline was similar for RPV and EFV and is not considered clinically relevant given that there were no patients with a QTcF >500 milliseconds.
Some limitations of these trials have been described previously,5,6 including the double-dummy design (oral doses taken twice daily, rather than once daily as in clinical practice). One advantage of pooling the data is that it overcomes the limitation of subgroup analyses in the individual trials, which are not powered for this. The results of the subgroup analyses should still be interpreted with caution, however, as even in this pooled analysis, some of the subgroups contained only small numbers of patients.
RPV and EFV showed high responses at week 48 in this pooled analysis. Response rates were similar between the 2 treatment groups by background N[t]RTI regimen, gender, race, and clade and in patients with baseline viral load <500,000 copies per milliliter. RPV displayed a more favorable tolerability profile than EFV, with a lower rate of discontinuations due to AEs. The rate of virologic failure was higher in the RPV group than in the EFV group, and this was noted primarily in those with suboptimal adherence (M-MASRI) and high baseline viral load. These results highlight the importance of good adherence to any antiretroviral treatment. The proportion of VFres with treatment-emergent N[t]RTI RAMs, particularly leading to 3TC/FTC-associated resistance, was higher in the RPV group than in the EFV group, whereas a similar proportion of VFres in each group had treatment-emergent NNRTI RAMs.
In sum, these data support once-daily RPV or a once-daily single-tablet regimen of RPV coformulated with TDF/FTC3,4 as valuable treatment options for the majority of ARV-naive HIV-1–infected patients.
The authors thank the patients and their families for their participation and support during the trial and the investigators, trial centre staff and trial coordinators from each centre, and Janssen trial personnel. Both trials were designed and conducted by Janssen, the trials' sponsor and developer of RPV. The authors received medical writing support and assistance in coordinating and collating author contributions from Ian Woolveridge and Karen Pilgram (Gardiner-Caldwell Communications Ltd, Macclesfield, United Kingdom), funded by Janssen. Finally, the authors would like to thank the following Janssen people for their input into this article: Guy De La Rosa, Dave Anderson, Bryan Baugh, Steven Nijs, Deborah Schaible, and Kati Vandermeulen. All authors involved in the studies had full access to all of the data and took full responsibility for the accuracy of the data analysis.
3. Prescribing information for Complera® (emtricitabine/rilpivirine/tenofovir disoproxil fumarate) tablets. Gilead Sciences, August 2011. Available at: http://www.gilead.com/pdf/complera_pi.pdf
. Accessed August 11, 2011.
5. Molina JM, Cahn P, Grinsztejn B, et al.. Rilpivirine versus efavirenz with tenofovir and emtricitabine in treatment-naive adults infected with HIV-1 (ECHO): a phase 3 randomised double-blind active-controlled trial. Lancet. 2011;378:238–246.
6. Cohen CJ, Andrade-Villanueva J, Clotet B, et al.. Rilpivirine versus efavirenz with two background nucleoside or nucleotide reverse transcriptase inhibitors in treatment-naive adults infected with HIV-1 (THRIVE): a phase 3, randomised, non-inferiority trial. Lancet. 2011;378:229–237.
7. Rimsky L, Vingerhoets J, Van Eygen V, et al.. Genotypic and phenotypic characterization of HIV-1 isolates obtained from patients on rilpivirine therapy experiencing virologic failure in the phase 3 ECHO and THRIVE studies: 48-week analysis. J Acquir Immune Defic Syndr. 2012;59:39–46.
8. Vingerhoets J, Rimsky L, Van Eygen V, et al.. Screening and baseline mutations in the TMC278 phase III trials ECHO and THRIVE: prevalence and impact on virological response [abstract 41]. Antivir Ther. 2011;16(suppl 1):A51.
11. van Velsen F, Sternberg J, Lachau-Durand S, et al. Study of the endocrine effects of TMC278, a novel HIV-NNRTI, in juvenile female cynomolgus monkeys [abstract 622]. Presented at: 50th Annual Meeting of the Society of Toxicology; March 6–10, 2011; Washington DC.
12. Pozniak AL, Morales-Ramirez J, Katabira E, et al.. Efficacy and safety of TMC278 in antiretroviral-naive HIV-1 patients: week 96 results of a phase IIb randomized trial. AIDS. 2010;24:55–65.
13. Johnson VA, Brun-Vézinet F, Clotet B, et al.. Update of the drug resistance mutations in HIV-1: December 2009. Top HIV Med. 2009;17:138–145.
14. DeJesus E, McCarty D, Farthing CF, et al.. Once-daily versus twice-daily lamivudine, in combination with zidovudine and efavirenz, for the treatment of antiretroviral-naive adults with HIV infection: a randomized equivalence trial. Clin Infect Dis. 2004;39:411–418.
15. DeJesus E, Herrera G, Teofilo E, et al.. Abacavir versus zidovudine combined with lamivudine and efavirenz, for the treatment of antiretroviral-naive HIV-infected adults. Clin Infect Dis. 2004;39:1038–1046.
16. Moyle GJ, DeJesus E, Cahn P, et al.. Abacavir once or twice daily combined with once-daily lamivudine and efavirenz for the treatment of antiretroviral-naive HIV-infected adults: results of the Ziagen Once Daily in Antiretroviral Combination Study. J Acquir Immune Defic Syndr. 2005;38:417–425.
17. Pozniak AL, Gallant JE, DeJesus E, et al.. Tenofovir disoproxil fumarate, emtricitabine, and efavirenz versus fixed-dose zidovudine/lamivudine and efavirenz in antiretroviral-naïve patients: virologic, immunologic, and morphologic changes—a 96-week analysis. J Acquir Immune Defic Syndr. 2006;43:535–540.
18. Lennox JL, DeJesus E, Lazzarin A, et al.. Safety and efficacy of raltegravir-based versus efavirenz-based combination therapy in treatment-naive patients with HIV-1 infection: a multicentre, double-blind randomised controlled trial. Lancet. 2009;374:796–806.
19. Smith KY, Kumar PN, Patel P, et al. Differences in virologic response among African-Americans and females regardless of therapy in the HEAT study [abstract MOPEB033]. Presented at: 5th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 19–22, 2009; Cape Town, South Africa.
20. Kumar P, Currier J, Squires K, et al. Predictors of response in GRACE (Gender, Race And Clinical Experience) [poster 335]. Presented at: 47th Annual Meeting of the Infectious Diseases Society of America; October 29–November 1, 2009; Philadelphia, PA.
21. Fourie J, Flamm J, Rodriguez-French A, et al.. Effect of baseline characteristics on the efficacy and safety of once-daily darunavir/ritonavir in HIV-1-infected, treatment-navie ARTEMIS patients at week 96. HIV Clin Trials. 2011;12:313–322.
22. Nachega JB, Hislop M, Dowdy DW, et al.. Adherence to nonnucleoside reverse transcriptase inhibitor-based HIV therapy and virologic outcomes. Ann Intern Med. 2007;146:564–573.
23. Knobel H, Guelar A, Carmona A, et al.. Virologic outcome and predictors of virologic failure of highly active antiretroviral therapy containing protease inhibitors. AIDS Patient Care STDS. 2001;15:193–199.
24. Goldman JD, Cantrell RA, Mulenga LB, et al.. Simple adherence assessments to predict virologic failure among HIV-infected adults with discordant immunologic and clinical responses to antiretroviral therapy. AIDS Res Hum Retroviruses. 2008;24:1031–1035.
25. Back DJ, Khoo SH, Gibbons SE, et al.. Therapeutic drug monitoring of antiretrovirals in human immunodeficiency virus infection. Ther Drug Monit. 2000;22:122–126.
26. Mocroft A, Phillips AN, Friis-Møller N, et al.. Response to antiretroviral therapy among patients exposed to three classes of antiretrovirals: results from the EuroSIDA study. Antivir Ther. 2002;7:21–30.
27. Egger M, May M, Chêne G, et al.. Prognosis of HIV-1-infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet. 2002;360:119–129.
28. 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;22:1389–1397.
29. Molina JM, Andrade-Villanueva J, Echevarria J, et al.. Once-daily atazanavir/ritonavir compared with twice-daily lopinavir/ritonavir, each in combination with tenofovir and emtricitabine, for management of antiretroviral-naive HIV-1-infected patients: 96-week efficacy and safety results of the CASTLE study. J Acquir Immune Defic Syndr. 2010;53:323–332.
30. Adams J, Patel N, Mankaryous N, et al.. Nonnucleoside reverse transcriptase inhibitor resistance and the role of the second-generation agents. Ann Pharmacother. 2010;44:157–165.
31. Picchio G, Vingerhoets J, Tambuyzer L, et al. Prevalence of genotypic and phenotypic susceptibility to etravirine in US samples received for routine resistance testing [abstract MOPDB105]. Presented at: 18th International AIDS Conference; July 18–23, 2010; Vienna, Austria.
32. Azijn H, Tirry I, Vingerhoets J, et al.. TMC278, a next-generation nonnucleoside reverse transcriptase inhibitor (NNRTI), active against wild-type and NNRTI-resistant HIV-1. Antimicrob Agents Chemother. 2010;54:718–727.
33. Kulkarni R, Babaoglu K, Lansdon E, et al.. The HIV-1 reverse transcriptase M184I mutation enhances the E138K-associated resistance to rilpivirine and decreases viral fitness. J Acquir Immune Defic Syndr. 2012;59:48–55.
34. Xu HT, Asahchop EL, Oliveira M, et al.. Compensation by the E138K mutation in HIV-1 reverse transcriptase for deficits in viral replication capacity and enzyme processivity. J Virol.. 2011;85:11300–11308.
35. Hu ZX, Li J, Gallien S, et al.. Impact of the interactions of rilpivirine (E138K) and lamivudine/emtricitabine (M184V/I) resistance mutations on viral DNA synthesis and fitness of HIV-1 [abstract 12]. Antivir Ther. 2011;16(suppl 1):A20.
36. Wilkin A, Pozniak AL, Morales-Ramirez J, et al.. Long-term efficacy, safety, and tolerability of rilpivirine (RPV, TMC278) in HIV type 1-infected antiretroviral-naive patients: week 192 results from a phase llb randomized trial. AIDS Res Hum Retroviruses. 2011. Epub ahead of print.