A novel probe drug interaction study to investigate the effect of selected antiretroviral combinations on the pharmacokinetics of a single oral dose of maraviroc in HIV-positive subjects

Abstract

Aims

Maraviroc (UK-427 857), an antagonist of the CCR5 receptor with potent anti-HIV activity, was recently approved for use in treatment-experienced patients infected with CCR5-tropic HIV-1. The aim of this study was to evaluate the effect of selected commonly used antiretroviral therapy (ART) combinations on the pharmacokinetics of a single oral dose of maraviroc 300 mg in HIV-positive subjects compared with historical controls.

Methods

In this study, four cohorts of HIV-positive patients (n = 8 each) receiving one of the following combination therapies were recruited: cohort 1 – efavirenz + Combivir® (lamivudine/zidovudine); cohort 2 – efavirenz + didanosine + tenofovir; cohort 3 – nevirapine + lamivudine + tenofovir; cohort 4 – Kaletra® (lopinavir/ritonavir) + stavudine + lamivudine. Subjects continued on their prescribed ART and also received a single oral dose of maraviroc 300 mg. Serial blood samples and urine for determination of maraviroc pharmacokinetics were collected over 12 h postdose. Plasma pharmacokinetic parameters from this study were compared with historical data generated in HIV-positive subjects receiving maraviroc monotherapy in a Phase IIa study.

Results

A total of 29 subjects were recruited (eight each in cohorts 1–3, and five in cohort 4). The geometric mean ratios for AUC₁₂ and C_max for each treatment group compared with maraviroc monotherapy were: 47% and 67% (cohort 1); 48% and 76% (cohort 2); 101% and 154% (cohort 3); and 265% and 180% (cohort 4), respectively. T_max was similar in all treatment groups. Mean values for renal clearance ranged from 8.2 l h⁻¹ (cohort 1) to 13.2 l h⁻¹ (cohort 4). There were no renal clearance data collected in the comparator study.

Conclusions

The results of this study support those previously seen in healthy volunteer studies that showed that efavirenz reduces maraviroc exposure, whereas lopinavir/ritonavir increases maraviroc exposure. These data also suggest that nevirapine does not lead to a clinically significant effect on maraviroc pharmacokinetics.

Introduction

Maraviroc is a CCR5 antagonist with potent anti-human immunodeficiency virus (HIV) activity in vitro[1], which was recently approved for use along with other antiretroviral agents for treatment-experienced patients infected with CCR5-tropic HIV-1.

Infection with HIV-1 leads, in the vast majority of cases, to progressive disease and ultimately, acquired immunodeficiency syndrome (AIDS) and death, with highly active antiretroviral therapy (HAART) as the only means of slowing progression. Although HAART can be successful in controlling virus replication, the side-effect profiles for several of these drugs make long-term adherence difficult and the consequent emergence of drug resistance a serious problem [2–5]. Currently most patients will eventually fail combination therapies of drugs of the existing classes, either due to intolerance or resistance (or a combination of both), and therefore there is a high medical need for better tolerated and conveniently administered agents to treat HIV-1/AIDS. One new approach is via host cell entry inhibition [6], an example of which is through blockade of the CCR5 coreceptor.

HIV-1 must bind to the cluster of differentiation 4 (CD4) receptor and a coreceptor such as CCR5 or CXCR4, located on the CD4 cell surface, to invade the human cell. Binding to the CD4 receptor alone is not sufficient to render cells susceptible to infection by HIV-1. Macrophage-tropic HIV isolates, irrespective of subtype, predominately use the β-chemokine receptor CCR5, and are referred to as R5 viruses. R5 viruses are preferentially transmitted and predominate in early asymptomatic disease. Standard treatment of HIV in treatment-experienced patients typically involves at least three antiretroviral agents, usually consisting of two or more nucleoside/nucleotide reverse transcriptase inhibitors (N/NtRTIs) plus one or more boosted protease inhibitors (PIs) and/or enfuvirtide. Therapy could also contain a non-nucleoside reverse transcriptase inhibitor (NNRTI) instead of, or in combination with a boosted PI. PIs are typically inhibitors of cytochrome P450 (CYP) 3A4 and P-glycoprotein (Pgp) activity, whereas the NNRTIs efavirenz and nevirapine are CYP inducers. N/NtRTIs are generally renally cleared or metabolized by non-CYP mechanisms [7–11].

Maraviroc is a substrate for both CYP3A4 and Pgp [12, 13]; hence, drug interactions are expected with drugs that modulate the function of these enzymes/transporters. Specific drug interaction studies showing the effects of CYP3A4/Pgp inhibitors and inducers on maraviroc have been conducted in healthy volunteers [14, 15]. This probe drug interaction study was conducted to investigate the effect of combination antiretroviral therapy (ART) on the pharmacokinetics of maraviroc in HIV-positive subjects. The ART regimens were chosen to cover some of the most commonly prescribed agents from the three major classes of ART. Four ART combinations containing either efavirenz, nevirapine, or Kaletra® (lopinavir/ritonavir), plus two N/NtRTIs (didanosine, stavudine, zidovudine, lamivudine and tenofovir), were co-administered with maraviroc in this study. A single oral dose of 300 mg maraviroc was chosen for this study because it was within the expected therapeutic dose range.

Methods

Subjects

HIV-positive subjects (eight per cohort), who had been stable for at least 3 months on one of four ART regimens, were to be recruited into four cohorts (see Table 1 for details) and enrolled into this open, single-period, single-centre study.

Table 1.

Study cohorts

Cohort (n)	Antiretroviral therapy
1 (8)	Efavirenz 600 mg q.d.
	Combivir® (lamivudine 150 mg + zidovudine 300 mg) b.i.d.
2 (8)	Efavirenz 600 mg q.d.
	Didanosine enteric coated 250 mg q.d.
	Tenofovir 300 mg q.d.
3 (8)	Nevirapine 200 mg b.i.d.
	Lamivudine 150 mg b.i.d.
	Tenofovir 300 mg q.d.
4 (5)	Kaletra® (lopinavir 400 mg + ritonavir 100 mg) b.i.d.
	Stavudine 40 mg b.i.d.
	Lamivudine 150 mg b.i.d.

Subjects were HIV-1-infected men, aged 18–55 years, weight between 60 and 100 kg and within the permitted range for their height using Quetelet's index [weight (kg)/height² (m) between 18 and 28]. Subjects were excluded if they had a CD4 count <250 cells mm⁻³, had a history of severe drug hypersensitivity reactions, or a clinically significant abnormality in laboratory data (other than would be expected as part of HIV infection) or full physical examination, had received an experimental drug within the 4 months prior to their first dose of study medication or who drank >28 units of alcohol per week. Subjects with a social condition, psychological or addictive disorder that would preclude compliance with the protocol were also excluded. All concomitant medications had to be approved by the sponsor to exclude any other agents that may have affected the pharmacokinetics of maraviroc. The study was conducted in compliance with the ethical principles as outlined in the Declaration of Helsinki, and the study protocol was approved by an Institutional Review Board or Independent Ethics Committee at the study centre. All subjects provided written informed consent before participating.

Study design

This was an open, parallel-group study, involving three study visits: screening (up to 21 days before dosing), one study day (until 12 h postdose), and a follow-up visit (7–10 days after study drug administration).

Subjects were asked to fast from all food and drink, except water, from midnight the night before the study day until 1 h post maraviroc dose. Water intake was also restricted from 1 h before dosing until 1 h after dosing. On the study day, all subjects received a single oral dose of 300 mg maraviroc between 08.00 h and 10.00 h with 250 ml of water. For at least 1 h after dosing, subjects were not allowed to lie down, but were allowed to be semirecumbent. Subjects were given breakfast exactly 1 h postdose. A standard lunch and dinner were served after collection of the 4- and 8-h blood samples, respectively. Subjects supplied their own ART medication. Subjects were advised to take their ART concomitantly with maraviroc with the exceptions of efavirenz, which was administered the evening before maraviroc dosing, and tenofovir and Kaletra®, which were administered 1 h after maraviroc with food.

At the discretion of the investigator, subjects were allowed to leave the unit after the collection of the 12-h pharmacokinetic samples. Subjects continued to take their ART medication after leaving the unit.

Pharmacokinetic sampling and analysis

Blood samples (4 ml) for the analysis of maraviroc were collected at intervals up to 12 h postdose. All urine was collected for the time interval from 0 to 12 h postdose. Samples were prepared using solid-phase extraction (plasma) or protein precipitation (urine) and assayed for maraviroc using liquid chromatography and tandem mass spectrometry (Tandem Laboratories, West Trenton, NJ, USA). For plasma the overall method imprecision values for the analysis of plasma quality control (QC) samples were 4.0, 3.9 and 1.2% at maraviroc target concentrations of 1.5, 150 and 400 ng ml⁻¹, respectively. The mean inaccuracy (bias) of the assay ranged from +0.7 to +3.0% over the QC range. The calibration range was 0.5–500 ng ml⁻¹. For urine the overall method imprecision values for the analysis of urine QC samples were 4.4, 5.8 and 4.6% at maraviroc target concentrations of 15, 1500 and 3750 ng ml⁻¹, respectively. The mean inaccuracy (bias) of the assay ranged from −5.2 to +8.7% over the QC range. The calibration range was 5.0–5000 ng ml⁻¹.

All pharmacokinetic parameters were calculated by noncompartmental analysis using WinNonLin™ Version 4.1 (Pharsight Corp., Mountain View, CA, USA). The pharmacokinetic parameters calculated for maraviroc were: maximum observed plasma concentration (C_max), time to reach maximum plasma concentration (T_max), area under the plasma concentration–time curve from time zero to 12 h postdose (AUC₁₂), and the renal clearance (CL_R), calculated as total urinary recovery of unchanged drug from time zero to 12 h postdose (Ae₁₂)/AUC₁₂. The pharmacokinetic data (C_max and AUC₁₂) from this study were compared with historical data (Study A4001007) generated in HIV-positive subjects who received 300 mg b.i.d. maraviroc as monotherapy [16]. The data generated up to 12 h postdose on day 1 were used as the historical control data.

Safety

Laboratory safety tests, physical examination, vital sign measurements, and 12-lead electrocardiograms (ECGs) were performed at screening and at the follow-up visit. In addition, laboratory safety tests and breath alcohol and urine drug screen were performed before dosing. The nature and severity of all adverse events (AEs) were recorded throughout the study period. The investigator obtained all AEs using nonleading questions and recorded their opinion of the relationship to study treatment.

Statistical analysis

The sample size of eight subjects in each group was estimated to provide a 90% confidence interval (CI) of ±0.432 on the natural log scale for AUC₁₂ with 80% coverage possibility. Assuming no change in AUC₁₂, it was estimated that the 90% CI would be (64.9%, 154%). Assuming a 50% decrease in AUC₁₂ it was estimated that the 90% CI would be (32.5%, 77.0%). Assuming a 400% increase in AUC₁₂ it was estimated that the 90% CI would be (325%, 770%). The sample size calculation assumed all data came from the same study. The CIs may have been wider due to increased variability, because the data from this study were compared with historical data.

Log-transformed AUC₁₂ and C_max and untransformed T_max were subject to analysis of variance including a term for treatment group. There were four comparisons of interest, each including one of the four ART regimens: maraviroc 300 mg plus ART vs. maraviroc 300 mg alone (historical control).

Differences between adjusted treatment means, associated standard errors and 90% CIs for the differences were presented on the log scale for AUC₁₂ and C_max and on the nominal scale for T_max. Log-transformed data were back transformed to provide the ratio between the geometric means and the associated 90% CI.

Results

Subjects

This study enrolled 29 HIV-positive male patients (eight in cohorts 1–3 and five in cohort 4) with a mean age of 39.5 years and a mean weight of 79.3 kg. There were 27 White subjects, one Black subject and one described as ‘other race’. The demographics of subjects in the control group were similar to this (eight White males, mean age 33.1 years, mean weight 71.8 kg). All 29 subjects were included in the pharmacokinetic and safety analyses. No subjects discontinued the study.

Pharmacokinetics

Summary statistics for the maraviroc pharmacokinetic parameters by treatment group are presented in Table 2. A summary of the statistical analysis of the maraviroc pharmacokinetic parameters is presented in Table 3. The geometric mean ratios for AUC₁₂ and C_max for each treatment group compared with maraviroc monotherapy were: 47% and 67% (cohort 1); 48% and 76% (cohort 2); 101% and 154% (cohort 3); and 265% and 180% (cohort 4), respectively. T_max was similar in all treatment groups (approximately 2 h). Mean values for CL_R ranged from 8.2 l h⁻¹ (cohort 1) to 13.2 l h⁻¹ (cohort 4). There were no CL_R data available for comparison, because none were collected in the historical control study. The mean plasma concentration profiles of maraviroc in the presence of each of the combination therapies and alone is shown in Figure 1.

Table 2.

Summary statistics for maraviroc pharmacokinetic parameters by treatment group

Treatment	Tmax (h)*	Cmax (ng ml−1)†	AUC12(ng h ml−1)†	CLR (l h−1)*
Cohort 1: maraviroc + efavirenz/lamivudine/zidovudine	2.1 (0.8)	389 (45%)	1060 (39%)	12.0 (3.6)
Cohort 2: maraviroc + efavirenz/didanosine/tenofovir	1.9 (0.4)	447 (89%)	1090 (85%)	8.2 (2.5)
Cohort 3: maraviroc + nevirapine/lamivudine/tenofovir	2.0 (0.0)	900 (68%)	2270 (51%)	9.7 (3.2)
Cohort 4: maraviroc + lopinavir/ritonavir/stavudine/lamivudine	2.2 (1.1)	1050 (48%)	5990 (30%)	13.2 (4.7)
Historical control: maraviroc 300 mg	2.9 (1.2)	585 (23%)	2260 (31%)	NA

^*Arithmetic mean (SD).

^†Geometric mean (CV%).

Table 3.

Summary of statistical analysis of maraviroc pharmacokinetics parameters

	AUC12 (ng h ml−1)*	Cmax (ng ml−1)*	Tmax (h)†
Maraviroc + efavirenz/lamivudine/ zidovudine vs. maraviroc alone	46.9 (30.3, 72.4)	66.5 (40.8, 109)	−0.75 (−1.44, −0.06)
Maraviroc + efavirenz/didanosine/ tenofovir vs. maraviroc alone	48.3 (31.3, 74.6)	76.4 (46.8, 125)	−1.00 (−1.69, −0.31)
Maraviroc + nevirapine/lamivudine/ tenofovir vs. maraviroc alone	101 (65.1, 155)	154 (94.3, 251)	−0.88 (−1.57, −0.18)
Maraviroc + lopinavir/ritonavir/stavudine/ lamivudine vs. maraviroc alone	265 (161, 435)	180 (103, 314)	0.68 (−1.47, 0.12)

^*Ratio of geometric means (90% CI), log-transformed parameters.

^†Difference between means (90% CI), untransformed parameters.

Figure 1.

Mean plasma concentrations of maraviroc in cohorts 1–4 compared with maraviroc historical control data (A4001007). 300 mg maraviroc + efavirenz/Combivir™ (▪); 300 mg maraviroc + efavirenz/didanosine/tenofovir (▴); 300 mg maraviroc + nevirapine/lamivudine/tenofovir (▾); 300 mg maraviroc + Kaletra®/lamivudine/stavudine (♦); Historical control (A4001007) ()

The geometric mean ratios for AUC₁₂ of 300 mg maraviroc was approximately 50% lower, and C_max was approximately 25–40% lower in subjects who received efavirenz (cohorts 1 and 2) compared with those who received maraviroc alone (A4001007). The geometric mean ratios for AUC₁₂ of 300 mg maraviroc were similar, and C_max was 1.5-fold higher in subjects who received nevirapine (cohort 3) compared with those who received maraviroc alone. The geometric mean ratios for AUC₁₂ of 300 mg maraviroc were 2.6-fold higher, and C_max was 1.8-fold higher in subjects who received lopinavir/ritonavir (cohort 4) compared with those who received maraviroc alone.

Safety and tolerability

A total of seven treatment-emergent AEs were reported by seven of the 29 subjects dosed on the study. All AEs were considered as study drug related by the investigator. The AEs reported on the study were headache (n = 5), herpes simplex (n = 1) and sweating (n = 1). All AEs were of mild to moderate intensity and all resolved on completion of the study. The incidence of AEs did not appear to correlate with the observed maraviroc plasma concentrations, and there were no obvious trends in the type or incidence of AEs between cohorts. No severe or serious AEs were reported. There were no clinically significant changes in laboratory safety data, vital signs, 12-lead ECG or physical examination findings.

Discussion

Co-administration of the lopinavir/ritonavir-containing ART regimen resulted in a more than doubling of maraviroc exposure (2.6-fold higher AUC₁₂), whereas co-administration of the efavirenz-containing ART regimens resulted in approximately 50% reduction in systemic exposure of maraviroc (AUC₁₂). These data are consistent with drug interaction studies in healthy volunteers [14, 15], where maraviroc was co-administered with lopinavir/ritonavir b.i.d. and/or efavirenz for 7–21 days. Maraviroc is predominately metabolized by CYP3A4, with <25% of parent drug renally cleared [17], and hence the plasma concentrations of maraviroc are expected to be affected by modulators of CYP3A4 activity. The data from this study are therefore consistent with the known CYP3A4 modulatory activity of these agents; the PI ritonavir is a known CYP3A4 inhibitor, whereas efavirenz is a CYP3A4 inducer.

The nevirapine-containing regimen resulted in a small (1.5-fold) increase in C_max, but no effect on overall systemic exposure (AUC₁₂). This lack of interaction with nevirapine is perhaps unexpected, as nevirapine has been shown to act as a CYP3A4 inducer [18, 19]; however, the data on this are conflicting with other studies showing no effect of nevirapine on CYP3A4 activity [20].

A single oral dose of maraviroc 300 mg was well tolerated in HIV-positive subjects when co-administered with four different ART regimens.

This was a novel study design that has allowed assessment of potential drug interactions due to multiple combination therapies. This study design has the further advantage that it facilitated investigation of the effects of these combination therapies in a clinically relevant population (i.e. HIV+ patients) and hence removed the need to expose healthy volunteers to compounds such as nevirapine, which are associated with potentially significant AEs (e.g. hepatotoxicity and rash).