No Effect of a Single Supratherapeutic Dose of Lersivirine, a Next-Generation Nonnucleoside Reverse Transcriptase Inhibitor, on Corrected QT Interval in Healthy Subjects

Abstract

The objective of this study was to investigate the effect of a supratherapeutic dose of lersivirine (LRV) on corrected QT (QTc) interval using Fridericia's equation (QTcF) in healthy subjects. In this randomized, single-dose, placebo- and active-controlled 3-way crossover study, healthy adult males (n = 48) were randomized to receive LRV (2,400 mg), moxifloxacin (400 mg), or placebo for each treatment period. Triplicate 12-lead electrocardiogram measurements were performed, PK samples were collected, and vital signs were measured. Adverse event monitoring and safety laboratory testing were performed. All subjects were white (mean age, 39 years; body mass index [BMI], 25.6 kg/m²) and completed the study. Following LRV administration, the upper bound of the 90% confidence interval (CI) for time-matched adjusted mean differences to placebo QTcF at each time point postdose was below the regulatory threshold of 10 ms, satisfying the criteria for a negative thorough QT/QTc study. The highest upper bound of QTcF 90% CI occurred at 6 h for LRV (3.32 ms; 90% CI, 1.47 to 5.17 ms). The study was deemed adequately sensitive as the lower bound of the 90% CI for the adjusted mean QTcF differences between moxifloxacin and placebo at the moxifloxacin historical T_max of 3 h was >5 ms (15.29 ms; 90% CI, 13.44 to 17.14 ms). There was no statistically significant relationship between LRV exposure and placebo-adjusted change from baseline QTcF or clinically significant changes in QRS complex, pulse rate (PR) interval, heart rate, or blood pressure. LRV (2,400 mg) did not prolong the QTcF interval, and no clinically relevant electrocardiogram or vital sign changes were observed in healthy subjects.

INTRODUCTION

Lersivirine (UK-453,061) is a next-generation nonnucleoside reverse transcriptase inhibitor (NNRTI) under investigation for the treatment of HIV-1 in adults. Lersivirine exhibits a unique resistance profile due to a unique binding interaction in the NNRTI binding pocket (21) and demonstrates potent antiretroviral activity against wild-type virus and a number of clinically relevant NNRTI-resistant strains. X-ray crystallography (10, 19) has demonstrated that lersivirine binds to the HIV reverse transcriptase (RT) enzyme in a novel way which is thought to contribute to antiretroviral activity (10, 21).

Lersivirine is rapidly absorbed, with a median time to maximum plasma concentration (T_max) of between 1 and 3 h after dosing (13). Mean terminal half-life estimates with lersivirine are approximately 6 to 7 h (12). Lersivirine displays linear kinetics at single and multiple doses ranging from 1 to 1,800 mg and 250 to 1,000 mg, respectively, as systemic plasma was shown to increase proportionally with increasing dose (12).

QT prolongation has been reported with a prevalence as high as 65% in patients with HIV and autonomic neuropathy (17, 20, 26). In addition, some antiretroviral agents have been associated with corrected QT interval (QTc) prolongation (1, 2, 22, 23). Therefore, it is important to establish a lack of effect of lersivirine on QT interval in patients with HIV who may also develop other comorbidities. In vitro analyses suggest that lersivirine is a weak delayed rectifier potassium (I_kr) channel and calcium channel blocker. However, no clinically significant effects on QTc or any other electrocardiogram (ECG) parameter have been observed in any of the phase I or phase IIa studies completed thus far (12, 14).

Lersivirine is under investigation in two phase IIb studies at doses of 500 to 1,000 mg once a day (QD) (8, 9). This study was performed to investigate the effect of a single supratherapeutic dose of lersivirine on the corrected QT (QTc) interval using Fridericia's equation (QTcF = QT/³√RR, where RR [respiratory rate] is the interval from the onset of one QRS complex to the onset of the next) in healthy subjects relative to a matched placebo and active control. A 2,400-mg dose was chosen as it was expected to match the predicted exposure (maximum plasma lersivirine concentration [C_max] and area under the plasma concentration-time curve [AUC]) as a result of known drug interactions with the highest dose currently being studied in phase IIb trials (750 mg QD [treatment naive] or 1,000 mg QD dosed with darunavir-ritonavir [treatment experienced]). QT intervals were corrected for influences of heart rate using Fridericia's equation. The objective of this study was primarily to demonstrate a lack of effect of lersivirine on QTcF interval. The study also aimed to characterize the exposure-response relationship between lersivirine and changes in QTcF and to compare the magnitude of any effect to the effect of moxifloxacin on QTcF. Moxifloxacin is the most commonly used active control in thorough QT (TQT) studies and is used to establish the assay sensitivity (28). The safety and tolerability of lersivirine were also assessed.

MATERIALS AND METHODS

The study was conducted in compliance with the ethical principles originating in or derived from the Declaration of Helsinki (27) and in compliance with International Conference on Harmonisation (ICH) E14 guidelines (15). All subjects provided written informed consent.

Study design.

This was a randomized, single-dose, placebo- and active-controlled 3-way crossover study conducted at the Pfizer Clinical Research Unit in Brussels, Belgium. All subjects recruited were healthy males between 18 and 55 years old with a body mass index (BMI) of 17.5 to 30.5 kg/m² and a total body weight of >50 kg. “Healthy” was defined as having no clinically relevant abnormalities identified by a detailed medical history, full physical examination, including blood pressure and pulse rate measurement, 12-lead ECG, and clinical laboratory tests. Subjects with a QTcF interval of >450 ms at screening and any condition possibly affecting drug absorption or a history of risk factors for QT prolongation or torsades de pointes were excluded.

Subjects that entered the study (n = 48) were randomized to receive a single dose of (i) lersivirine (2,400 mg), (ii) moxifloxacin (400 mg), or (iii) placebo on day 1 (following an 8-h fast) of each treatment period on one of three occasions, according to one of six possible sequences. There was a minimum washout period of 7 days between each treatment period. This study was subject and investigator blinded, except for unblinded moxifloxacin administration, and sponsor unblinded.

A 2,400-mg dose was chosen as it was expected to match the predicted exposure (maximum plasma lersivirine concentration [C_max] and area under the plasma concentration-time curve [AUC]) as a result of known drug interactions with the highest dose currently being studied in phase IIb trials (750 mg QD [treatment naive] or 1,000 mg QD dosed with darunavir-ritonavir [treatment experienced]). The expected exposure was based on linear regression of C_max and AUC following single escalating lersivirine doses from 10 mg to 1,800 mg for AUC and from 800 mg to 1,800 mg for C_max (12). As a single 1,800-mg dose was the highest previously administered dose at the time of protocol development, a small cohort of subjects received a single dose of 2,400 mg prior to the start of this study to confirm that this dose could be tolerated.

Blood samples (4 ml) were collected predose (0 h) and 1, 2, 3, 4, 5, 6, 9, 12, and 24 h postdose in each study period. Triplicate 12-lead ECG measurements were performed predose (−1, −0.5, and 0 h) and at 1, 2, 3, 4, 5, 6, 9, 12, and 24 h postdose. QT intervals were corrected for the possible effects of heart rate (HR) using QTcF (ms) = QT (ms)/(RR)^1/3, where RR (s) = 60/HR (if not provided) for the primary analysis.

Statistics.

The sample size of 42 completers was selected to provide 99% power to exclude a 10-ms mean difference from the placebo at each time point if the expected mean difference between lersivirine and placebo was no greater than 5 ms at each time point. The minimum overall study power was at least 92% for 8 postdose time points (1, 2, 3, 4, 5, 6, 9, and 12 h). With the intersection-union test, no adjustment to the level of significance was required for multiple comparisons since each test was at the 5% level. These calculations were based on nQuery Advisor 6.0, using an equivalence limit of 10 ms and a common intrasubject standard deviation of 5.52 ms for QTcF intervals. To allow for any subjects who might not complete the study, 48 subjects were recruited.

The postdose QTcF intervals were analyzed using a linear mixed-effects model with sequence, period, treatment, time, and treatment-by-time interaction as fixed effects, subject within sequence as a random effect, and baseline QTcF as a covariate. The averages of the triplicate 12-lead ECGs collected at each time point were calculated; the average of the triplicate ECG measurements collected predose on day 1 during each of the study periods served as each subject's baseline QTcF value. The mixed-effects model was implemented using SAS Proc mixed with the REML estimation method and Kenward-Roger degrees of freedom algorithm. Compound symmetry was assumed for the variance-covariance structure of the ith subject between observations in different periods, as well as in the same period at different times. Estimates of the adjusted mean differences (test-reference) and the two-sided 90% confidence intervals (CIs) for each treatment and time were obtained from the model. To demonstrate the lack of effect of lersivirine on QTcF intervals at nine postdose time points, the raw postdose QTcF of the single supratherapeutic doses of lersivirine was the “test,” while that of placebo was the “reference.” To demonstrate the study sensitivity at 3 h postdose, the raw postdose QTcF of moxifloxacin 400 mg was the “test,” while that of placebo was the “reference.” The 24-h postdose time point was added in the ad hoc analysis in order to include all of the QTcF data being collected.

The lack of effect of lersivirine on QTcF intervals was to be concluded if the upper bound of the two-sided 90% CIs for all time-matched QTcF differences between lersivirine and placebo was less than 10 ms. Study sensitivity was to be demonstrated if the lower bound of the two-sided 90% CI for the difference between moxifloxacin and placebo was greater than 5 ms at the historical moxifloxacin T_max of 3 h.

Nonlinear mixed-effects modeling (NONMEM 6.2; ICON Development Solutions, Ellicott City, MD) was used to describe the relationship between lersivirine plasma concentration and placebo-adjusted change from baseline QTcF. A total of 48 individuals provided 476 pairs of concentration-QT data points for the analysis. A linear slope model with additive random effects on intercept, slope, and residual error was used to describe the relationship. The first-order conditional estimation method was performed.

Changes from baseline for QT, QTcF, pulse rate (PR) interval, QRS interval, and heart rate were calculated and summarized descriptively. Descriptive statistics were used to analyze PK parameters: AUC_last, C_max, and T_max. AUC_last was calculated using the trapezoidal rule.

Safety.

The safety analysis set included all subjects who received at least one dose of study treatment. Adverse events (AEs) and serious AEs were recorded along with the severity and the investigator's opinion of the relationship to study treatment. Vital signs were measured predose, 3 h postdose, and at discharge. Safety laboratory testing was performed throughout. Any clinical laboratory, ECG, blood pressure (BP), or pulse rate abnormality of potential clinical concern was required to be reported as an AE.

RESULTS

All 48 subjects enrolled were white males with a mean age of 39.1 years (range, 26 to 55 years) and mean body mass index (BMI) of 25.6 kg/m² (range, 19.5 to 29.8 kg/m²), and all completed the study.

Lersivirine effects on TQT interval.

The highest upper bound of 90% CI for time-matched adjusted mean difference between the lersivirine 2,400-mg group and placebo for QTcF occurred at 6 h (3.32 ms; 90% CI, 1.47 to 5.17 ms) (Table 1). The upper bound of the 90% CI was below the 10-ms regulatory threshold, thus satisfying the criteria for a negative TQT/QTc study. In addition, the estimated adjusted mean difference was below 5 ms at all postdose time points (Fig. 1). The study was deemed adequately sensitive as the lower bound of the 90% CI for the adjusted time-matched mean differences between moxifloxacin and placebo in QTcF at the moxifloxacin historical T_max of 3 h was >5 ms (15.29 ms; 90% CI, 13.44 to 17.14 ms), meeting the prespecified criterion for sensitivity discussed earlier (Table 1).

Table 1.

Summary of statistical analysis of 2,400 mg lersivirine compared with placebo for QTcF^a

Time postdose (h)	Adjusted mean QTcF (ms)		Difference between lersivirine and placebo QTcFs (ms)	90% CI (ms)
	Lersivirine	Placebo
1	410.23	411.51	−1.28	−3.14–0.58
2	410.25	410.57	−0.31	−2.17–1.55
3	410.46	408.89	1.57	−0.28–3.43
4	412.27	409.61	2.67	0.81–4.52
5	413.17	410.17	3.00	1.15–4.84
6	414.20	410.88	3.32	1.47–5.17
9	406.69	404.85	1.84	−0.01–3.69
12	407.81	406.83	0.98	−0.87–2.83
24	408.48	409.63	−1.15	−3.00–0.70

^aCI, confidence interval; QTcF, QT interval corrected using Fredericia's equation.

Fig 1.

Lersivirine exposure and placebo-adjusted change from baseline QTcF. Dotted line, 10 ms regulatory threshold for a positive TQT/QTc study. DD, double delta.

There was no statistically significant relationship between lersivirine exposure (i.e., concentration) and placebo-adjusted change from baseline QTcF. The model estimated slope was 8.46E⁻⁰⁴ (90% CI, −3.01E⁻⁰⁴ to 19.93E⁻⁰⁴) with 90% CI including zero (Fig. 2). No subjects in the lersivirine 2,400-mg treatment group had an absolute maximum QTcF interval of ≥450 ms or a change from baseline in QTcF of ≥30 ms.

Fig 2.

Time-matched placebo adjusted mean differences (90% CI) for single doses of lersivirine (2,400 mg) and moxifloxacin (400 mg) for QTcF. Dotted line, 10 ms regulatory threshold for a positive TQT/QTc study.

The relationship between lersivirine 2,400-mg plasma concentration and the difference from placebo in lersivirine QTcF change from baseline over time was examined (Fig. 3). The QTcF difference from baseline followed a similar pattern to the increase in lersivirine concentration. The maximum QTcF difference from baseline occurred at approximately 6 h, which was 3 h after plasma concentrations of lersivirine peaked.

Fig 3.

Relationship between lersivirine plasma concentration following a single 2,400-mg supratherapeutic dose and the time-matched placebo-adjusted mean difference (90% CI) in lersivirine QTcF.

QTcF, QT (uncorrected) PR interval, QRS interval, and heart rate were summarized descriptively. There were no differences between treatments in mean baseline and mean change from baseline values. No clinically significant differences in ECG parameters (changes from baseline for QTcF, QT [uncorrected], PR interval, or QRS interval), heart rate, or blood pressure were observed in healthy subjects receiving a single dose of lersivirine (2,400 mg).

Lersivirine pharmacokinetics.

The pharmacokinetic parameters for lersivirine were geometric mean AUC_last of 17,750 ng · h/ml (28% coefficient of variation [CV]), geometric mean C_max of 1,727 ng/ml (33% CV), and median T_max of 3.00 h.

Safety.

There were no deaths, serious AEs, or permanent withdrawals due to AEs; one subject temporarily discontinued the study due to a moderate AE of blood creatinine phosphokinase increase 6 days after receiving a single-dose moxifloxacin treatment of 400 mg in period 2; this AE was considered to be unrelated to treatment. A total of 71 AEs were reported by 46 subjects: 26 subjects receiving 2,400 mg lersivirine, 13 subjects receiving 400 mg moxifloxacin, and 12 subjects receiving placebo. The most commonly reported all-causality AE across all treatments was headache, which was reported by a total of 22 subjects. Nausea and vomiting were the second and third most common AEs reported across all treatment groups, and were more commonly reported in the 2,400-mg lersivirine group than in the placebo group (Table 2). All AEs reported were of mild to moderate intensity. There were no clinically significant laboratory abnormalities or changes in vital sign values of clinical concern.

Table 2.

Summary of the most commonly reported all-causality adverse events that occurred in more than two subjects in any treatment group

No. of subjects with adverse eventa	No. of subjects (n = 48) with adverse event by treatment group:
	Lersivirine (2,400 mg)	Moxifloxacin (400 mg)	Placebo
Headache	13	4	5
Nausea	11	0	0
Vomiting	5	0	0
Diarrhea	3	3	1
Abdominal discomfort	2	0	2

^aShown are the Medical Dictionary for Regulatory Activities v12.1 preferred terms (19a).

DISCUSSION

Lersivirine, administered at a single supratherapeutic dose of 2,400 mg, had no clinically relevant effect on the QTcF interval or other ECG parameters in healthy subjects. There was no association observed between lersivirine pharmacokinetics and QTcF interval. ECG results from this study confirm previous observations that QT and QTcF do not appear to be prolonged over the lersivirine dose concentrations investigated to date (12, 14). Single-dose administration of 2,400 mg lersivirine did not lead to any unexpected AEs or appear to compromise subject safety.

The single-supratherapeutic-dose study design with thorough crossover used in this study has been considered appropriate, as previously reported in an investigation of the effects of raltegravir on QTcF interval (16). The Food and Drug Administration's (FDA's) E14 guidance for industry provides recommendations to sponsors concerning the design, conduct, analysis, and interpretation of clinical studies to assess the potential of a drug to delay cardiac repolarization. Several important recommendations were proposed, all of which were incorporated into this study. These included the use of a positive control (moxifloxacin) and the testing of a supratherapeutic dose of lersivirine, as well as several methods for reducing variability in QTcF interval measurement (triplicate ECGs, a single core laboratory to measure QTcF interval, and the use of the 10-ms threshold to demonstrate study negativity). The pharmacological profiles of the most abundant metabolites for lersivirine have been tested against a range of physiologic receptors, transporters, enzymes, and ion channels in which no pharmacological activities likely to be of biological relevance were observed. In addition, drug accumulation was taken into consideration in the dose selection. The 1,800-mg dose, the highest lersivirine single dose studied to date, would not provide a high enough C_max concentration of lersivirine to cover a worst case scenario interaction—for example, an interaction with a potent CYP3A4 inhibitor (11, 12). Therefore, based on a linear regression of C_max and AUC following a single dose of lersivirine in a previous study, subjects would need a single dose of approximately 2,400 mg in order to reach the targeted concentrations proposed for the study.

With current combination antiretroviral therapies, patients are experiencing close to normal quality of life (18, 24), and consequently extra care must be taken to safeguard against serious toxicities being introduced by some new drugs. Many antiretroviral drugs have been shown to have an effect on the QT interval. With patients using the FDA's voluntary AE reporting system, Anson and colleagues identified 14 cases of QT/QTc interval prolongation and/or arrhythmia in patients treated with protease inhibitors (PIs), such as atazanavir, indinavir, amprenavir, lopinavir, nelfinavir, ritonavir, and saquinavir (2). A prospective study (4) has shown atazanavir to have a nonsignificant QTc interval prolongation 2 h after administration and no prolongation at steady state, which suggests that further examination may be warranted. In another study, statistically significant changes in PR and QRS intervals were reported in 21 patients receiving atazanavir with or without ritonavir for 1 month (7.4 ms and 3.7 ms, respectively) (5). The mean changes in the PR intervals were similar to those reported elsewhere (3).

Within the NNRTI class, rilpivirine (TMC278) was originally found to have an effect on QT interval at the 75-mg QD therapeutic dose (25) and a higher supratherapeutic dose, leading to a reduction in the planned dose for phase III trials from 75 mg QD to 25 mg QD. A recent study carried out by Venveggel and colleagues demonstrated that treatment with the NNRTIs rilpivirine (dosed at 25 mg QD) and efavirenz did not result in QTc interval prolongation (25). However, in a case study of a patient with recurrent syncope and polymorphic ventricular tachycardia, efavirenz was found to mediate QTc prolongation, with the potential to produce life-threatening arrhythmia (6). Furthermore, a study by Chinello and colleagues reported an increased risk of prolonged QTc interval among patients taking efavirenz (7).

Both male and female subjects were permitted to be recruited to the study. However, participants were competitively enrolled, and as a result, only male subjects were enrolled in the study. Although we cannot definitively claim that a comparable effect on QTcF would be seen in women, who have higher baseline QTcF intervals, a post hoc analysis showed that QTcF intervals were comparable between male subjects with high baseline QTcF and male subjects with low baseline QTcF, suggesting that subjects with higher baseline QTcF may not be at greater risk for changes in QTcF when dosed with lersivirine.

Lersivirine was shown to display linear kinetics for AUC and C_max in the dose range of 10 to 1,800 mg and 800 to 1,800 mg, respectively (12). These data were generated from single-dose escalation studies carried out at the same clinical research unit (Pfizer, Inc., Brussels, Belgium) used in this study. The PK data from the small cohort that preceded the QT study did not confirm linear kinetics; however, subjects in the QT study were still dosed with 2,400 mg as the sample size in the small cohort was too small to definitively conclude that the PK data in the QT study would not confirm linearity. The AUC and C_max observed with the 2,400-mg dose in this study were 17,750 ng · h/ml and 1,727 ng/ml, respectively, which were 13% lower than the values observed with the 1,800-mg dose (AUC, 20,383 ng · h/ml; C_max, 1,983 ng/ml). The reason for this variation is unknown as the formulation and demographic characteristics were similar between both studies. The AUC and C_max achieved with the single 2,400-mg dose in the QT study were approximately 100% and 28% greater, respectively, than the exposures observed with the highest dose (750 mg QD) from the PK substudy in the ongoing phase 2b treatment-naive study of lersivirine (study A5271015) and approximately 196% and 64% greater than the exposures observed with the 500-mg QD dose.

In conclusion, this randomized, single-dose, placebo- and active-controlled, 3-way crossover study in healthy volunteers provides evidence that lersivirine, a next-generation NNRTI, can be safely administered without the need for ECG monitoring.