icon-    folder.gif   Conference Reports for NATAP  
 
   
Back grey_arrow_rt.gif
 
 
 
Safety, Tolerability, and Pharmacokinetic Interactions of the Antituberculous Agent TMC207 (Bedaquiline) With Efavirenz in Healthy Volunteers: AIDS Clinical Trials Group Study A5267
 
 
  Download the PDF here

JAIDS Journal of Acquired Immune Deficiency Syndromes: 15 April 2012

In conclusion, bedaquiline concentrations were only modestly reduced when bedaquiline was given together with efavirenz, a decrease that is unlikely to be clinically significant. Peak concentrations of the M2 metabolite were increased in this single-dose study, but once formed, M2 seems to be more rapidly cleared in the setting of efavirenz, and this will likely result in steady-state concentrations of M2 that do not significantly differ when bedaquiline is given with or without efavirenz. The present study indicates that efavirenz should not compromise the therapeutic efficacy of bedaquiline against TB. Further analyses will be needed to understand the implications, if any, of the effects of efavirenz on M2 metabolite PK.

The novel diarylquinoline bedaquiline (formerly TMC207) is active against both drug-sensitive and drug-resistant TB and has the potential to shorten treatment duration.7,8 Bedaquiline inhibits the proton pump of M. tuberculosis, ATP synthase, a novel mechanism of action.7,9Bedaquiline has demonstrated bactericidal and sterilizing activity in mouse models, and it is synergistic with first-line and second-line TB drugs and TB compounds in development.10-13 In a phase 2 placebo-controlled trial among patients with multidrug-resistant (MDR) TB, 48% of those randomized to receive bedaquiline together with a 5-drug MDR regimen became sputum culture negative by 8 weeks of treatment compared with 9% who received just the 5-drug MDR regimen.14 By 6 months, 79% of patients who received bedaquiline plus standard MDR treatment became culture negative compared with 58% who received standard MDR treatment.15 Clinical trials are ongoing to identify optimal bedaquiline-containing regimens for drug-sensitive TB and MDR-TB.

Bedaquiline is largely metabolized by cytochrome P450 (CYP) 3A to a less-active N-monodesmethyl metabolite (M2). Bedaquiline undergoes triphasic elimination, and although the average half-life over a dosing interval is about 24 hours, its terminal half-life is approximately 5 months.7,16,17 Administration with food increases bioavailability by about 95%. Efavirenz, one of the most widely prescribed ARVs globally, may induce CYP3A, raising the possibility that efavirenz might adversely lower bedaquiline concentrations. In addition, efavirenz is metabolized primarily by CYP2B6, and genetic variants in CYP2B6 confer marked interindividual variability in efavirenz pharmacokinetics (PK).18,19


We conducted a phase 1 open-label clinical trial in healthy volunteers to evaluate the effects of efavirenz on the PKs of bedaquiline and its M2 metabolite and to describe the safety and tolerability of single doses of bedaquiline coadministered with steady-state efavirenz. We also compared the PKs of efavirenz in this study with historical control values, taking into account CYP2B6 genotype.

Dooley, Kelly E. MD, PhD*; Park, Jeong-Gun PhD; Swindells, Susan MBBS; Allen, Reena MA; Haas, David W. MD; Cramer, Yoninah MS; Aweeka, Francesca PharmD#; Wiggins, Ilene RN*; Gupta, Amita MD*; Lizak, Patricia BS#; Qasba, Sonia MD, MPH; van Heeswijk, Rolf PharmD, PhD**; Flexner, Charles MD*; the ACTG 5267 Study Team

*Johns Hopkins University School of Medicine, Baltimore, MD Harvard School of Public Health, Boston, MA University of Nebraska Medical Center, Omaha, NE AIDS Clinical Trials Group Operations, Center, Silver Spring, MD Division of AIDS, National Institutes of Health, Bethesda, MD Vanderbilt University School of Medicine, Nashville, TN #University of California, San Francisco, CA **Tibotec, BVBA, Mechelen, Belgium Montgomery County Department of Health and Human Services, Silver Spring, MD

Supported by the AIDS Clinical Trials Group sponsored by the National Institute of Allergy and Infectious Diseases

Abstract

Background: Drug-drug interactions complicate management of coinfection with HIV-1 and Mycobacterium tuberculosis. Bedaquiline (formerly TMC207), an investigational agent for the treatment of tuberculosis, is metabolized by cytochrome P450 (CYP) 3A which may be induced by the antiretroviral drug efavirenz.

Methods: This was a phase 1 pharmacokinetic drug interaction trial. Each healthy volunteer received two 400 mg doses of bedaquiline, the first alone and the second with concomitant steady-state efavirenz. Plasma pharmacokinetic sampling for bedaquiline and its N-monodesmethyl metabolite was performed over 14 days after each bedaquiline dose. Steady-state efavirenz pharmacokinetics were also determined. Efavirenz metabolizer status was based on CYP2B6 composite 516/983 genotype.

Results: Thirty-three of 37 enrolled subjects completed the study. Geometric mean of ratios for bedaquiline with efavirenz versus bedaquiline alone were 0.82 [90% confidence interval (CI): 0.75 to 0.89] for the 14-day area under the concentration-time curve (AUC0-336 h) and 1.00 (90% CI: 0.88 to 1.13) for the maximum concentration (Cmax). For N-monodesmethyl metabolite, the geometric mean of ratios was 1.07 (90% CI: 0.97 to 1.19) for AUC0-336 h and 1.89 (90% CI: 1.66 to 2.15) for Cmax. There were no grade 3 or 4 clinical adverse events. One subject developed asymptomatic grade 3 serum transaminase elevation, prompting study drug discontinuation. Efavirenz concentrations stratified by CYP2B6 genotype were similar to historical data.

Conclusions: Single-dose bedaquiline was well tolerated alone and with steady-state efavirenz. The effect of efavirenz on bedaquiline concentrations is unlikely to be clinically significant.

INTRODUCTION

Of the 9.4 million patients with incident tuberculosis (TB) annually, 1.4 million are coinfected with HIV-1, and globally, TB is the leading cause of death among HIV-infected individuals.1 Recent studies have demonstrated that survival among coinfected individuals is improved by initiating antiretroviral therapy early during TB treatment rather than waiting to complete TB therapy.2-5 Such antiretroviral therapy is particularly beneficial in coinfected individuals with lower CD4 counts. Concomitant treatment for HIV and TB is, thus, common and is becoming the standard of care in many settings. However, concurrent treatment of TB and HIV is complicated by drug-drug interactions and overlapping drug toxicities.6 There is an urgent need for new anti-TB medications that can be given safely with antiretrovirals (ARVs).

The novel diarylquinoline bedaquiline (formerly TMC207) is active against both drug-sensitive and drug-resistant TB and has the potential to shorten treatment duration.7,8 Bedaquiline inhibits the proton pump of M. tuberculosis, ATP synthase, a novel mechanism of action.7,9 Bedaquiline has demonstrated bactericidal and sterilizing activity in mouse models, and it is synergistic with first-line and second-line TB drugs and TB compounds in development.10-13 In a phase 2 placebo-controlled trial among patients with multidrug-resistant (MDR) TB, 48% of those randomized to receive bedaquiline together with a 5-drug MDR regimen became sputum culture negative by 8 weeks of treatment compared with 9% who received just the 5-drug MDR regimen.14 By 6 months, 79% of patients who received bedaquiline plus standard MDR treatment became culture negative compared with 58% who received standard MDR treatment.15 Clinical trials are ongoing to identify optimal bedaquiline-containing regimens for drug-sensitive TB and MDR-TB.

Bedaquiline is largely metabolized by cytochrome P450 (CYP) 3A to a less-active N-monodesmethyl metabolite (M2). Bedaquiline undergoes triphasic elimination, and although the average half-life over a dosing interval is about 24 hours, its terminal half-life is approximately 5 months.7,16,17 Administration with food increases bioavailability by about 95%. Efavirenz, one of the most widely prescribed ARVs globally, may induce CYP3A, raising the possibility that efavirenz might adversely lower bedaquiline concentrations. In addition, efavirenz is metabolized primarily by CYP2B6, and genetic variants in CYP2B6 confer marked interindividual variability in efavirenz pharmacokinetics (PK).18,19


We conducted a phase 1 open-label clinical trial in healthy volunteers to evaluate the effects of efavirenz on the PKs of bedaquiline and its M2 metabolite and to describe the safety and tolerability of single doses of bedaquiline coadministered with steady-state efavirenz. We also compared the PKs of efavirenz in this study with historical control values, taking into account CYP2B6 genotype.

DISCUSSION

Coinfection with HIV-1 and M. tuberculosis is common, and concurrent treatment is now recommended.24,25 The widely prescribed ARV efavirenz can induce CYP3A isoforms, and the promising investigational agent for TB, bedaquiline, is a CYP3A substrate. The most important finding from the present study is that steady-state efavirenz reduced the AUC of bedaquiline by only a median of about 20%. Although the Cmax of the M2 metabolite was increased almost 90%, the AUC of M2 was unchanged. That efavirenz coadministration results in only a modest decrease in bedaquiline exposure is reassuring, given that drug-drug interactions can be problematic during treatment of TB/HIV coinfection and limit therapeutic options for HIV-infected patients with TB. Safety implications of higher peak concentrations of the M2 metabolite, however short-lived, are unclear.

The clinical consequences of diminished bedaquiline concentrations for patients receiving bedaquiline-containing multidrug therapy for the treatment of TB are not known, but the change associated with efavirenz coadministration is modest and unlikely to be clinically relevant. Pharmacodynamic targets for bedaquiline have not been well established, but preclinical and clinical data do suggest dose-response relationships. In dose-ranging monotherapy studies in the mouse acute infection model, bedaquiline at 6.5 mg/kg delivered 5 times per week was bacteriostatic, and a dose increase from 12.5 mg/kg to 25 mg/kg yielded a one-log decline in lung colony-forming units after 4 weeks.7,26 More recent dose-fractionation studies in the mouse acute infection model have shown a dose-response relationship between bedaquiline dose and lung colony-forming unit counts when doses from 8 to 64 mg/kg were tested, and the effects appeared to be concentration-dependent (personal communication, Dr Koen Andries, August 31, 2011). The optimal effect of bedaquiline in the mouse model was at plasma concentrations of approximately 0.3 mcg/mL, achieved with about 100 mg/kg per week.7 However, in mice, 80% of bedaquiline is converted to the M2 metabolite, whereas in humans, the circulating M2 to bedaquiline ratio is 1:4. Given that M2 is 3-fold to 6-fold less active against M. tuberculosis than bedaquiline, PK/pharmacodynamic targets may differ in mice and humans. In a 7-day early bactericidal activity study of bedaquiline monotherapy in humans, doses up to 200 mg daily showed no demonstrable activity, but 400 mg daily resulted in late declines in daily sputum colony counts.27 When given in multidrug regimens for MDR-TB, a dose of 400 mg once daily for 2 weeks followed by 200 mg thrice weekly achieved plasma concentrations above 600 ng/mL throughout the dosing interval, a target that may achieve therapeutic efficacy similar to 25 mg/kg in mice.14 No correlation between average bedaquiline concentration and outcomes, however, was seen in this phase 2 study, so target concentrations in humans remain uncertain.

Bedaquiline is a cationic amphiphilic drug, as defined by a hydrophobic ring structure plus a hydrophilic side chain with a charged cationic amine group. Cationic amphiphilic drugs can cause phospholipidosis, characterized by the accumulation of phospholipids in cells.28 Phospholipidosis is thought to be an adaptive response rather than a manifestation of direct cellular toxicity, but phospholipidosis can have clinical consequences, including prolonged QT interval, myopathy, hepatotoxicity, nephrotoxicity, or pulmonary dysfunction.29 These effects are generally reversible with drug discontinuation. There is no reliable biomarker predictive of drug-related phospholipidosis, and animal models do not seem to predict human responses well, posing regulatory challenges.30 In vitro, M2 was a stronger inducer of phospholipidosis than amiodarone, and M2 induced phospholipidosis at lower concentrations than bedaquiline (1.2 μM vs. 7.3 μM, respectively).31 Concentrations of M2 in our study were far lower than those that produced phospholipidosis in in vitro models (Cmax of 83.3 ng/mL = 0.154 μM), but these were not steady-state concentrations. In the current study, ECGs were performed 4-6 hours after each single dose of bedaquiline. In a post hoc analysis in this small dataset, there was no correlation between change in QTcF from baseline and either AUC or Cmax for bedaquiline or M2. Although QT interval corrected for heart rate was slightly prolonged with both doses of bedaquiline, all but 2 subjects had ECGs with intervals within the normal range (<450 milliseconds).

Bedaquiline is primarily metabolized via oxidative N-demethylation to M2, and this biotransformation is largely catalyzed by CYP3A. M2 can be further N-demethylated to M3 or, alternatively, oxidized to M4. In this study, when efavirenz was coadministered with bedaquiline, M2 concentrations were initially increased resulting in a higher Cmax, but clearance of M2 also seemed to be enhanced, leading to similar overall M2 exposures, or AUC0-336 h. This could be explained by increased velocity and capacity of enzymatic transformation of bedaquiline to M2 in the setting of efavirenz coupled with induction of further biotransformation of M2 to a secondary metabolite. M3 or M4 concentration-time data would be needed to confirm this. If M2 is both formed and cleared more rapidly with concomitant efavirenz, steady-state M2 Cmax is unlikely to be higher when bedaquiline is given with efavirenz than when bedaquiline is given alone. To test this assumption, nonlinear mixed effects modeling is underway to estimate steady-state M2 parameters using these single-dose data and taking into account triphasic bedaquiline elimination and M2 metabolite kinetics.

Efavirenz is a substrate of CYP2B6 and CYP3A and also induces these metabolizing enzymes, resulting in autoinduction of metabolism and diminished efavirenz concentrations with multiple dosing.32 The CYP2B6 polymorphisms 516G->T and 983T->C predict decreased plasma clearance and increased steady-state efavirenz exposure.18 In the present study, as expected, plasma efavirenz exposure was strongly associated with CYP2B6 metabolizer genotype. The PKs of efavirenz should not be affected by bedaquiline. It is therefore reassuring that, within each CYP2B6 genotype group, efavirenz PK parameters were similar to previously reported values.22 In drug-drug interaction studies that involve efavirenz, it is important to determine CYP2B6 metabolizer genotypes of study participants so as to more accurately interpret efavirenz PK parameters. In addition, results from participants with CYP2B6 slow, intermediate, and extensive metabolizer genotypes suggest that our findings are generalizable regardless of CYP2B6 genotype. This is important because CYP2B6 slow metabolizer genotypes vary in frequency depending on geographic region of ancestry and are most frequent among populations of African descent.33 There is a correlation between efavirenz AUC and hepatic CYP3A induction, but the dose or clinical exposure at which maximal induction is achieved is unknown.34

There are several limitations of this study. Because bedaquiline was given as 2 single doses, the full safety and tolerability profile of bedaquiline with efavirenz could not be assessed. Although M2 is eliminated more rapidly when bedaquiline is given with efavirenz, it is unknown whether this represents increased metabolism of M2 to a secondary metabolite or another process because we did not measure other metabolites. Rifampin reduces bedaquiline concentrations by 50%.35 Because the mechanism by which efavirenz induces CYP3A has not been fully elucidated,36 it is unclear whether efavirenz would further diminish bedaquiline concentrations among patients who are also receiving concomitant rifampin. Finally, although we found no clear association between efavirenz concentrations and change in bedaquiline exposures, we studied relatively few individuals with CYP2B6 slow metabolizer genotypes; trends seen in our analyses would need to be confirmed in a larger group of patients.

In conclusion, bedaquiline concentrations were only modestly reduced when bedaquiline was given together with efavirenz, a decrease that is unlikely to be clinically significant. Peak concentrations of the M2 metabolite were increased in this single-dose study, but once formed, M2 seems to be more rapidly cleared in the setting of efavirenz, and this will likely result in steady-state concentrations of M2 that do not significantly differ when bedaquiline is given with or without efavirenz. The present study indicates that efavirenz should not compromise the therapeutic efficacy of bedaquiline against TB. Further analyses will be needed to understand the implications, if any, of the effects of efavirenz on M2 metabolite PK.

RESULTS

Subjects


Thirty-seven subjects were enrolled and received the first dose of bedaquiline. The median age was 44 years (range 19-62), and 34 (92%) were male. The median weight and body mass index were 82.9 kg (range 57-119) and 26 (19-36), respectively. Twenty-six (70%) participants were white, non-Hispanic; 8 (22%) were black, non-Hispanic; 2 (5%) were Hispanic; and 1 (3%) was Asian. Of the 37, 1 was removed from the study for missing study visits, 1 withdrew for personal reasons (relocation), 1 was discontinued for study drug nonadherence, and 1 developed grade-3 serum transaminase elevations, prompting study drug discontinuation. Thirty-three participants were, thus, eligible for analysis of PK endpoints.

Effect of Efavirenz on Bedaquiline and M2 PKs

Figure 2 shows the mean (SE) single-dose bedaquiline and M2 log-transformed plasma concentration-time curves for single doses of bedaquiline 400 mg taken alone [pharmacokinetic period 1, (PK1)] or with steady-state efavirenz 600 mg daily [pharmacokinetic period 2, (PK2)]. The median bedaquiline AUC0-336 h was 58,155 ng·hr·mL-1 for bedaquiline alone and 52,135 ng·hr·mL-1 for bedaquiline coadministered with efavirenz (P < 0.001) (Table 1). The AUC0-336 h PK2:AUC0-336 h PK1 GMR was 0.82 (90% CI: 0.75 to 0.89). Oral clearance of bedaquiline was increased with concomitant efavirenz, whereas bedaquiline Cmax was not significantly affected. The median M2 Cmax was 42.4 ng/mL for bedaquiline alone and 83.3 ng/mL for bedaquiline coadministered with efavirenz (P < 0.001) (Table 1). The ratio of CmaxPK2 to CmaxPK1 GMR was 1.89 (90% CI: 1.65 to 2.15). The AUC0-336 h of M2, however, did not differ significantly between the 2 PK periods. Our analyses suggested a nonsignificant trend toward inverse correlations between efavirenz AUC0-24 h values and change from period 1 to period 2 in AUC0-336 h values for bedaquiline (r = -0.34, P = 0.053) and M2 (r = -0.3, P = 0.089; see Figure, Supplemental Digital Content 1, http://links.lww.com/QAI/A238).

Efavirenz Pks

Median (IQR) AUC0-24 h for efavirenz was 56.87 μg·hr·mL-1 (42.27-74.99), Cmax was 3.98 μg/mL (3.32-5.49), and Cmin was 1.58 μg/mL (1.04-2.06). These values are similar to mean steady-state parameters reported among HIV-infected adults (58.09 μg·hr·mL-1, 4.07 μg/mL, and 1.77 μg/mL, respectively).22 Among the 33 study participants eligible for PK analysis, 3 (9%) had CYP2B6 slow metabolizer genotypes (2 with 516TT, 1 with 516GT/983TC), 12 (36%) intermediate metabolizer genotypes (10 with 516GT, 2 with 983TC), and 18 (55%) extensive metabolizer genotypes. Median efavirenz concentrations stratified by CYP2B6 genotype are shown in Table 2 and Figure 3. Efavirenz C24 h values among CYP2B6 extensive, intermediate, and slow metabolizers in the present study were similar to previously described steady-state C24 h values stratified by CYP2B6 metabolizer genotype among HIV-infected adults.23

Safety and Tolerability of Single-dose TMC With Steady-State Efavirenz

Grade 2 or higher clinical or laboratory adverse events are reported in the supplemental Table (see Supplemental Digital Content 2, http://links.lww.com/QAI/A239). There were no serious adverse events and no clinical adverse events of grade 3 or higher. One subject developed grade-2 serum transaminase elevations 14 days after the first dose of bedaquiline that increased to grade 3 on the first day of efavirenz dosing, prompting discontinuation of study drug. Another developed grade-3 hypoglycemia, while taking efavirenz, which resolved without specific intervention. Mean increase in QT interval corrected for heart rate using Fridericia's formula (QTcF) from baseline to 4-6 hours after the first and second bedaquiline doses was 12.3 milliseconds (95% CI: 6.92 to 17.6) and 12.8 milliseconds (95% CI: 7.49 to 18.1), respectively. One subject had a QTcF that was 450 milliseconds after the second bedaquiline dose, and another had a QTcF that was 452 after the first dose and 450 after the second dose (all were grade-1 toxicities), but all other postdose QTcF were below the upper limit of normal (450 milliseconds by Fredericia). There were no statistically significant associations between bedaquiline or M2 Cmax and changes in QTcF for either bedaquiline dose administration period.