Efficacy and safety of etravirine in treatment-experienced, HIV-1 patients: pooled 48 week analysis of two randomized, controlled trials
13 November 2009
Pharmacokinetics of Once-Daily Etravirine (ETR) Without and With Once-Daily Darunavir/ Ritonavir (DRV/r) in Antiretroviral-Naïve HIV-1 Infected Adults - (12/09/08)
Will Etravirine Work Once Daily--With Once-Daily Darunavir?
Etravirine given once daily to treatment-naive people attained equivalent levels with or without once-daily darunavir . ...
"Aside from rash, which generally occurred early and infrequently led to etravirine discontinuation, tolerability in the etravirine group was generally comparable with that in the placebo group, particularly in terms of hepatic, gastrointestinal and neuropsychiatric events. No new safety concerns were identified."
"Data from the current analysis illustrate the activity of etravirine against NNRTI-resistant virus. This is in line with in-vitro data that demonstrated a low propensity for the development of resistance to etravirine, as well as potent activity of this agent against HIV-1 virus with NNRTI RAMs [9,23]. The lowest virologic response rate with etravirine was observed in the subgroup of patients with an etravirine-weighted genotypic score of at least 4 at baseline. However, responses were higher with etravirine than with placebo in patients with weighted genotypic scores less than 4, and the majority of patients had a baseline etravirine-weighted genotypic score less than 4 (81%). Moreover, K103N, the most prevalent NNRTI mutation in the DUET studies at baseline in accordance with the overall resistance profile in NNRTI-experienced patients in the clinics, did not affect response to etravirine."
Katlama, Christine; Haubrich, Richard; Lalezari, Jacob; Lazzarin, Adriano; Madruga, Jose V; Molina, Jean-Michel; Schechter, Mauro; Peeters, Monika; Picchio, Gaston; Vingerhoets, Johan; Woodfall, Brian; De Smedt, Goedele; on behalf of the DUET-1, DUET-2 study groups
aUniversite Pierre et Marie Curie, Paris VI and H™pital Pitie-Salpetri¸re, Paris, France
bUniversity of California San Diego, San Diego, California, USA
cQuest Clinical Research, San Francisco, California, USA
dVita-Salute, San Raffaele University, Milan, Italy
eCentro de Referencia e Treinamento DST/AIDS, Sao Paulo, Brazil
fDepartment of Infectious Diseases, Saint-Louis Hospital, University of Paris, Diderot Paris 7, France
gProjeto Praca Onze, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil
hTibotec BVBA, Mechelen, Belgium
iTibotec, Inc., Yardley, Pennsylvania, USA.
* DUET-1 and DUET-2 study groups detailed in Acknowledgment section.
Objective: To evaluate the efficacy, safety and virologic resistance profile of etravirine (TMC125), a next-generation nonnucleoside reverse transcriptase inhibitor, over 48 weeks in treatment-experienced adults infected with HIV-1 strains resistant to a nonnucleoside reverse transcriptase inhibitor and other antiretrovirals.
Design: DUET-1 (NCT00254046) and DUET-2 (NCT00255099) are two identically designed, randomized, double-blind phase III trials.
Methods: Patients received twice-daily etravirine 200 mg or placebo, each plus a background regimen of darunavir/ritonavir, investigator-selected nucleoside/nucleotide reverse transcriptase inhibitors and optional enfuvirtide. Eligible patients had documented nonnucleoside reverse transcriptase inhibitor resistance, at least three primary protease inhibitor mutations at screening and were on a stable but virologically failing regimen for at least 8 weeks, with plasma viral load more than 5000 copies/ml. Pooled 48-week data from the two trials are presented.
Results: Patients (1203) were randomized and treated (n = 599, etravirine; n = 604, placebo). Significantly more patients in the etravirine than in the placebo group achieved viral load less than 50 copies/ml at week 48 (61 vs. 40%, respectively; P < 0.0001). Significantly fewer patients in the etravirine group experienced at least one confirmed or probable AIDS-defining illness/death (6 vs. 10%; P = 0.0408). Safety and tolerability in the etravirine group was comparable to the placebo group. Rash was the only adverse event to occur at a significantly higher incidence in the etravirine group (19 vs. 11%, respectively, P < 0.0001), occurring primarily in the second week of treatment.
Conclusion: At 48 weeks, treatment-experienced patients receiving etravirine plus background regimen had statistically superior and durable virologic responses (viral load less than 50 copies/ml) than those receiving placebo plus background regimen, with comparable tolerability and no new safety signals reported since week 24.
The first-generation nonnucleoside reverse transcriptase inhibitors (NNRTIs), efavirenz and nevirapine, have a low genetic barrier to the development of resistance. Extensive cross-resistance between these agents effectively limits NNRTI use to one failed regimen [1,2]. Efavirenz and nevirapine are associated with neuropsychiatric and hepatic events, respectively, which limit their utility among patient populations with existing comorbidities [3-8].
Etravirine (TMC125) is a next-generation NNRTI active against wild-type and NNRTI-resistant virus , which broadens the possible use of NNRTIs in treatment-experienced patients. The DUET-1 and DUET-2 trials were designed to investigate the efficacy and safety of etravirine in combination with other antiretroviral agents in clinically advanced, three antiretroviral class-experienced, human immunodeficiency virus type 1 (HIV-1)-infected adult patients who had evidence of viral replication and HIV-1 strains resistant to an NNRTI and other antiretroviral agents. Both trials were of identical design, allowing a robust, prespecified, pooled analysis. Results from 24-week analyses of the individual DUET (TMC125 to Demonstrate Undetectable viral load in patients Experienced with ARV Therapy) trials demonstrated better efficacy and comparable tolerability (except for rash) with etravirine versus the placebo group [10,11].
The primary aim of this pooled 48-week analysis was to evaluate whether etravirine, as part of combination therapy, could maintain the efficacy and safety findings over the longer term in treatment-experienced, HIV-1-infected patients.
Patient disposition and baseline characteristics
The disposition of patients during screening and treatment is shown in Fig. 1. Baseline characteristics and demographic data were well balanced between treatment groups (Table 1). During the screening period, 11.6% of patients were taking an NNRTI. Eight percent of patients in each treatment group had no documented previous NNRTI use. However, in line with the inclusion criteria for each trial, these patients had at least one NNRTI RAM documented at screening or from genotyping prior to trial entry. Median (range) exposure to study drugs for patients in the etravirine and placebo groups was 52.3 (1.6-85) and 51.0 (3.4-80) weeks, respectively.
The investigator-selected background regimen was similar between groups. Overall, 554 (46%) patients used enfuvirtide in the background regimen, with 312 (26%) using enfuvirtide de novo. Of note, a higher proportion of patients in the placebo group than in the etravirine group had darunavir fold-change less than 2 at baseline. A higher proportion of patients in the etravirine group than in the placebo group had darunavir fold-change more than 40. In total, 318 (53%) patients in the etravirine group and 328 (55%) in the placebo group were receiving less than two active drugs in the background regimen.
Virologic and immunologic responses
The primary and secondary virologic and immunologic endpoints, for both the overall population and according to enfuvirtide use (primary prespecified analysis), are shown in Table 2. Significantly more patients receiving etravirine than placebo achieved viral load less than 50 copies/ml (61 vs. 40%, respectively; P < 0.0001) (Fig. 2a), regardless of the total number of active antiretrovirals (determined by phenotype) in the background regimen (Fig. 2b). Virologic response rates increased in both groups with an increasing number of active background agents.
In patients who achieved viral load less than 50 copies/ml at week 24, suppression was sustained to week 48 in 92% (334/363) and 89% (218/246) of patients in the etravirine and placebo groups, respectively. Approximately one-third of those patients with a viral load between 50 and 400 copies/ml at week 24 [35% (29/83) in the etravirine group; 31% (21/67) in the placebo group] went on to achieve a viral load less than 50 copies/ml at week 48. Patients taking etravirine achieved a virologic response significantly more quickly than with placebo (median 15.7 weeks and 32.7 weeks for etravirine and placebo, respectively; P < 0.0001).
Factors predicting virologic response
Factors found to predict virologic response (proportion of patients with viral load less than 50 copies/ml) at 48 weeks in both treatment groups were baseline viral load (76, 61 and 49% responders for less than 30 000, 30 000-100 000 and more than 100 000 copies/ml, respectively, for the etravirine group, P < 0.0001), baseline CD4 cell count (45, 65, 74 and 72% responders for less than 50, 50-200, 200-300 and at least 350 cells/µl, respectively for the etravirine group, P = 0.0059), adherence (16, 59 and 63 responders for less than 95, at least 95-less than or equal to 97.5 and more than 97.5% adherence, respectively for the etravirine group, P < 0.0001), number of active agents in the background regimen (Fig. 2b; P < 0.0001), number of previously used NNRTIs (43, 66 and 58% responders for 0, 1 and at least 2 NNRTIs for the etravirine group, P = 0.0052) and enfuvirtide use (59, 52 and 71% responders for enfuvirtide not used, reused and used de novo for the etravirine group, P = 0.0018). At week 48, a consistently higher proportion of patients in the etravirine group achieved viral load less than 50 copies/ml than those in the placebo group, even after accounting for these factors in a multivariate analysis (data not shown). Race, sex, HIV-1 clade, use of an NNRTI during screening and previous experience with efavirenz or nevirapine were not predictive of response.
Baseline etravirine fold-change was also a significant predictor of virologic response at 48 weeks in the etravirine-treated patient arm (P < 0.0001). The greatest virologic response rate with etravirine was observed in patients with minimal resistance to etravirine (baseline etravirine fold-change 3 or less and weighted genotypic score of 0-2) (Table 3). The majority of patients in DUET had an etravirine baseline fold-change of 3 or less (65%; 779/1190) and weighted genotypic score of 0-2 (56%; 227/406). Responses were higher with etravirine than in the overall placebo group in patients with weighted genotypic scores less than 4 (72 vs. 44%, respectively; P < 0.0001). A reduced response to etravirine was seen in patients with baseline fold-change more than 13 (overall population: 48%; 35/73) or weighted genotypic score of at least 4 (46%; 46/100). Results showing greater virologic response between etravirine genotypic and phenotypic subgroups and overall placebo were generally similar in the not de-novo subgroup and overall group.
Among the 17 etravirine RAMs, the presence at baseline of V179D, V179F, V179T, Y181V or G190S were associated with the largest impact on virologic response to etravirine. In the DUET studies, these mutations were each present in less than 5% of patients at baseline. The most prevalent NNRTI substitution at baseline, K103N, was not associated with a reduced response to etravirine and thus not classified as an etravirine RAM.
At week 48, a significantly lower proportion of patients in the etravirine group experienced at least one confirmed or probable AIDS-defining illness or death (independently adjudicated) than in the placebo group (Table 2). Mean time to AIDS-defining illness or death was longer in the etravirine group than in the placebo group (59 vs. 45 weeks; P = 0.0108). The most commonly reported confirmed or probable AIDS-defining illnesses (in 6 or more patients) in the etravirine and placebo groups, respectively, were Candida esophagitis (one vs. nine patients), Pneumocystis jiroveci pneumonia (three vs. six patients), herpes simplex (four vs. four patients), Mycobacterium avium complex infection (two vs. five patients), cytomegalovirus retinitis (one vs. five patients) and Kaposi's sarcoma (two vs. four patients). During the treatment period, deaths occurred in 20 placebo-treated and 12 etravirine-treated patients. Of these, death was the first clinical event to occur in seven out of 20 patients receiving placebo and eight out of 12 patients receiving etravirine (1.2 vs. 1.3% overall, respectively).
Safety and tolerability
The safety assessment was based on all data from 1203 patients. No new safety concerns were identified in the 48-week analysis compared with week 24 (Table 4). The majority of adverse events were grade 1 or 2 in severity. The incidence of grade 3 or 4 adverse events was comparable between treatment groups. Any deaths that occurred in the etravirine group were considered not or doubtfully related to trial medication and were mostly associated with HIV-1 disease progression.
Rash was the only adverse event to occur significantly more frequently with etravirine than with placebo (19.2 vs. 10.9%, P = 0.0001 prespecified Fisher's exact test). Within the etravirine group, rash was mild-to-moderate in severity (17.9% grade 1 or 2, 1.3% grade 3 and no grade 4 rashes reported), occurred primarily in the second week of therapy (median time-to-onset was 14 days; range 1-473 days) and was infrequent after week 4. Rash generally resolved within 1-2 weeks on continued therapy (median duration of 15 days; range 1-402 days). A total of 2.2% of HIV-1-infected patients receiving etravirine discontinued due to rash; there were no grade 3 rashes or discontinuations due to rash in the placebo group. One placebo patient developed Stevens-Johnson Syndrome after approximately 3 months of treatment, most likely related to a nonantiretroviral medication. The incidence of rash was higher in women (30.0%) than in men (18.0%; P = 0.0365) in the etravirine group. Patients with a history of NNRTI-related rash did not appear to be at increased risk for the development of etravirine-related rash compared with patients with no history of NNRTI-related rash (22 vs. 19%, respectively). There was no association between baseline CD4 cell count and development of rash for etravirine, regardless of sex. No relationship was apparent between the incidence and severity of rash and exposure to etravirine.
The overall incidence of nervous system and psychiatric-related disorders (including but not limited to headache, dizziness, insomnia, depression and abnormal dreams/nightmares) was low. Events were mostly grade 1 or 2 in severity and comparable between the etravirine and placebo groups (Table 4). The incidence of hepatic adverse events and profile of laboratory abnormalities, including hepatic and lipid parameters, was generally comparable between groups and mostly grade 1 or 2 in severity. There were no consistent or clinically relevant changes in laboratory, vital signs or electrocardiogram data, including QTc interval.
The pooled 48-week results from the DUET trials demonstrate that the superior virologic efficacy observed with etravirine versus placebo (both with a background regimen) is sustained to 48 weeks. At 48 weeks, 61% of patients receiving etravirine compared with 40% in the placebo plus background regimen group responded with a confirmed undetectable viral load of less than 50 copies/ml, which is the revised treatment goal for all treatment-experienced HIV-infected patients [3,17-20]. As all patients from both groups received darunavir, the relatively high efficacy rate in the placebo group is in accordance with the efficacy rate observed in the POWER trials (45%) . The difference between the two groups reflects the added efficacy with the addition of etravirine.
This article is the first to describe the independently adjudicated data on clinical events in the DUET trials. A significant benefit in favor of the etravirine group versus the placebo group was seen in terms of the proportion of patients experiencing clinical events (6 vs. 10%; P = 0.0408) and the mean time to clinical events (59 vs. 45 weeks; P = 0.0108). As recently highlighted by Hirschel and Perneger , morbidity and mortality, not just surrogate laboratory endpoints, are of substantial interest in the management of patients with highly resistant virus.
A statistically significantly greater response in the etravirine group than in the placebo group was also observed regardless of the number of active background agents. However, in line with current treatment guidelines [3,17-19], the use of an increasing number of other active antiretrovirals with etravirine was associated with an increased likelihood of treatment response.
Data from the current analysis illustrate the activity of etravirine against NNRTI-resistant virus. This is in line with in-vitro data that demonstrated a low propensity for the development of resistance to etravirine, as well as potent activity of this agent against HIV-1 virus with NNRTI RAMs [9,23]. The lowest virologic response rate with etravirine was observed in the subgroup of patients with an etravirine-weighted genotypic score of at least 4 at baseline. However, responses were higher with etravirine than with placebo in patients with weighted genotypic scores less than 4, and the majority of patients had a baseline etravirine-weighted genotypic score less than 4 (81%). Moreover, K103N, the most prevalent NNRTI mutation in the DUET studies at baseline in accordance with the overall resistance profile in NNRTI-experienced patients in the clinics, did not affect response to etravirine.
When data were analyzed by enfuvirtide use, it was found that the virologic response (less than 50 copies/ml) was significantly greater in the etravirine group versus the placebo group, regardless of enfuvirtide use. This is despite the imbalance between groups in baseline darunavir fold-change in patients who used enfuvirtide de novo and is in contrast with the pooled 24-week findings. At 24 weeks, there was no significant difference in response between the etravirine and placebo groups in patients who used de-novo enfuvirtide (67 vs. 61%, P = 0.276) . The greater between-group difference at week 48 (13%) than week 24 (6%) might be explained by the longer time-to-reach undetectable viral load among those with viral load of at least 100 000 copies/ml at baseline  (over a third of patients); the requirement for viral load to be confirmed as undetectable on two consecutive visits and a loss of response among some patients in the placebo group.
This 48-week pooled analysis of the DUET trials confirms the safety and tolerability profile of etravirine observed in the 24-week analysis [10,11]. Aside from rash, which generally occurred early and infrequently led to etravirine discontinuation, tolerability in the etravirine group was generally comparable with that in the placebo group, particularly in terms of hepatic, gastrointestinal and neuropsychiatric events. No new safety concerns were identified.
Among the limitations of the current study is that no information was generated on the use of etravirine with protease inhibitors other than darunavir. On the basis of inclusion criteria of at least three primary protease inhibitor mutations at screening, the boosted protease inhibitors that would be most likely to retain activity were tipranavir and darunavir. Tipranavir/ritonavir was found to substantially decrease etravirine exposure in healthy volunteers , so for the purposes of this study, coadministration was not allowed. Similarly, recently licensed agents that were experimental at the time of DUET enrolment, such as raltegravir and maraviroc, were not allowed because information on drug-drug interactions was lacking at trial onset. However, available findings from recent drug interaction studies demonstrate that the combination of etravirine with these agents is feasible [27,28], albeit dose adjustment of maraviroc may be necessary. Of late, the TRIO National agency for AIDS Research; Agence nationale de recherches sur la sida trial has shown excellent antiviral activity and good tolerability in a new three-drug combination of etravirine, darunavir and raltegravir .
In DUET, most patients were not on a failing NNRTI-containing regimen during the screening period. NNRTI RAMS were present in 88% of patients at baseline, confirming the persistence of some NNRTI mutations in the absence of drug pressure [30,31]. However, not all possible RAMS may have been detected in circulating viral strains at baseline, potentially resulting in an underestimation of the true impact of baseline resistance to etravirine on virologic outcome.
After 48 weeks, etravirine plus background regimen demonstrated durable virologic and immunologic responses, which were superior to the placebo group in treatment-experienced, HIV-1-infected adult patients. The incidence of clinical events was lower with etravirine plus background regimen than with placebo plus background regimen. No new safety concerns were identified at 48 weeks. With the exception of rash, the incidence of adverse events and laboratory abnormalities in the etravirine group was generally comparable to that in the placebo group. The DUET studies demonstrate the clinical utility of etravirine in treatment-experienced patients with resistance to NNRTI and other antiretroviral agents and highlight the clinical benefit that is now achievable from an NNRTI with activity in the face of multidrug resistance.