Moxifloxacin Population Pharmacokinetics in Patients with Pulmonary Tuberculosis and the Effect of Intermittent High-Dose Rifapentine

Abstract

We described the population pharmacokinetics of moxifloxacin and the effect of high-dose intermittent rifapentine in patients with pulmonary tuberculosis who were randomized to a continuation-phase regimen of 400 mg moxifloxacin and 900 mg rifapentine twice weekly or 400 mg moxifloxacin and 1,200 mg rifapentine once weekly. A two-compartment model with transit absorption best described moxifloxacin pharmacokinetics. Although rifapentine increased the clearance of moxifloxacin by 8% during antituberculosis treatment compared to that after treatment completion without rifapentine, it did not result in a clinically significant change in moxifloxacin exposure.

TEXT

The duration of standard therapy for pulmonary tuberculosis is at least 6 months, and direct observation of treatment doses is promoted to support treatment adherence. Strategies aiming to shorten the duration of treatment or reduce the frequency of treatment doses are desirable. Rifapentine has a long half-life (7) and has superior in vitro activity against Mycobacterium tuberculosis compared to rifampin (9). Moxifloxacin has demonstrated potent bactericidal and sterilizing activity against M. tuberculosis in vitro and in mouse models (11, 12). Like rifapentine, moxifloxacin has a long half-life (9 to 12 h) (8, 18), making it an attractive companion drug to prevent selection of rifapentine-resistant strains when the drugs are administered intermittently. However, moxifloxacin is a substrate of p glycoprotein (6), sulfotransferases (17), and glucuronosyltransferases (19), which may be induced by rifapentine, thus reducing systemic concentrations of moxifloxacin.

To investigate the effects of rifapentine on moxifloxacin pharmacokinetics, we enrolled 28 adults diagnosed with pulmonary tuberculosis who were participating in the RIFAQUIN study (ISRCTN 44153044; http://ipc.nxgenomics.org/intertb/download/RIFAQUIN_Protocol_v_1.8_15_April_2011_FINAL.pdf) at the study site in Worcester, South Africa. Separate written informed consent for the pharmacokinetic study was obtained from RIFAQUIN study participants who were randomized to the two investigational arms with continuation-phase regimens of 1,200 mg rifapentine and 400 mg moxifloxacin once weekly or 900 mg rifapentine and 400 mg moxifloxacin twice weekly. The study protocol was reviewed and approved by the Research Ethics Committee of the University of Cape Town and the Medicines Control Council of South Africa.

Plasma concentration data were collected during coadministration of moxifloxacin and rifapentine in the fourth month of tuberculosis treatment and again after a single 400-mg moxifloxacin dose 4 to 8 weeks after completion of tuberculosis treatment. At each occasion, blood samples were collected immediately before and at 1, 3, 5, 7, 10, 12, 26, and 50 h after the dose.

Plasma moxifloxacin concentrations were determined using a validated liquid chromatography-tandem mass spectrometry assay method developed in the Division of Clinical Pharmacology, University of Cape Town. The samples were processed with a protein precipitation extraction method using 20 μl plasma with 200 μl precipitation solution (acetonitrile) containing the internal standard gatifloxacin at a concentration of 500 ng/ml. Gradient chromatographic separation was achieved on a Phenomenex, Gemini-NX 5-μm C₁₈ (110-A, 50-mm by 2-mm) analytical column using acetonitrile and 0.1% formic acid as the mobile phase and was delivered at a flow rate of 400 μl/min. An AB Sciex API 3200 Q-trap mass spectrometer was operated at unit resolution in the multiple-reaction-monitoring mode, monitoring the transition of the protonated molecular ions at m/z 402.1 to the product ions at m/z 358.3 for moxifloxacin and the protonated molecular ions at m/z 376.1 to the product ions at m/z 332.4 for the internal standard. The assay was validated over the concentration range of 0.063 μg/ml to 16 μg/ml. The accuracies for the moxifloxacin assay were 106.3%, 100.7%, and 102% at the low, medium, and high quality control (QC) levels, respectively, during interbatch validation. The precision (expressed as the percent coefficient of variation) was less than 5.5% at the low, medium, and high QC levels.

Moxifloxacin plasma concentration-time data were analyzed using a nonlinear mixed-effects approach. Estimation of standard population pharmacokinetic parameters, along with their interindividual (IIV) and interoccasional (IOV) variabilities, was performed in NONMEM 7 using the first-order conditional estimation method with ε-η interaction (3, 4). Graphical diagnostics of the models were made using Xpose 4 (13), and model evaluation included a visual predictive check (10). Allometric scaling was explored using either total body weight (WT), fat-free mass (FFM), or normal fat mass (NFM) (calculated separately for each sampling period) on the clearance (CL/F), intercompartmental clearance (Q/F), volume of distribution in plasma (V_c/F), and peripheral volume (V_p/F) according to Anderson and Holford (1, 2). The covariate relationships were tested in the model by stepwise addition using a change in objective function value (ΔOFV) of ≥3.84 (P ≤ 0.05) as the cutoff for inclusion, followed by stepwise deletion using an ΔOFV of ≥6.83 (P ≤ 0.01) for covariate retention.

The median (5th and 95th percentiles) age, weight, and height of the study patients were 40 (21, 51) years, 52 (42, 71) kg, and 163 (154, 176) cm, respectively. The pharmacokinetics of moxifloxacin was best described by a model with transit absorption compartments (16). The introduction of a second compartment to which distribution appears relatively late (Fig. 1) resulted in an ΔOFV of 99.8 points. The final structural model is represented in Fig. 1. Allometric scaling using FFM gave the best fit and was applied on CL/F, Q/F, V_c/F, and V_d/F. Inclusion of the effect of rifapentine on moxifloxacin CL/F further improved the fit (ΔOFV = −29 points) and reduced the IOV in bioavailability from 22.9% to17.5%. Parameter estimates of the final moxifloxacin model are shown in Table 1. A one-compartment model was previously used to describe moxifloxacin pharmacokinetics (15) and reported a somewhat lower half-life (6.53 h), CL/F (6.66 liters/h), and apparent volume of distribution (62.79 liters); these differences are probably largely due to a different study design and model structure.

Fig 1.

Illustration of the final moxifloxacin pharmacokinetic model (bottom). N₁ to N_n represents a series of hypothetical transit compartments used to model the delay in onset of absorption, and k_tr is the transit rate constant. k_a is the absorption rate constant from the hypothetical drug absorption site (depot) compartment to plasma. V_c, V_p, CL, and Q are the central and peripheral volumes of distribution and the oral and intercompartmental clearance, respectively. (Top) Visual predictive check of the final moxifloxacin pharmacokinetic model. The lower, middle, and upper solid lines are the 5th percentiles, medians, and 95th percentiles of the observed data, respectively. The shaded areas are the 95% confidence intervals for the 5th percentile, median, and 95th percentile of the simulated data. The open black circles are the observed concentrations.

Table 1.

Parameter estimates of the final moxifloxacin pharmacokinetic model^a

Parameter	Typical value (RSE [%])	IIVb (RSE [%])	IOVc (RSE [%])
TV[CL/F] (liters/h)	8.50 (4.9)	12.6 (28.1)
Vc/F (liters)	114 (4.0)
ka (h−1)	1.85 (22.3)		74.7 (31.0)
MTT (h)	0.483 (22.4)		70.4 (27.0)
NN	22.5 (44.9)
Q/F (liters/h)	2.90 (12.1)
Vp/F(liters)	41.6 (9.1)
F	1.00f		17.5 (16.3)
EFFRFP (%)d	8.03 (21.5)
t1/2 (h)e	9.23
Additive residual error (mg/liter)	0.0189 (19.5)
Proportional residual error (%)	8.94 (4.7)

^aRSE, relative standard error reported on the standard deviation scale (it is approximate for IIV and IOV). CL/F, oral clearance; V_c/F, apparent volume of distribution of the central compartment; k_a, first-order absorption rate constant; MTT, absorption mean transit time; NN, number of hypothetical transit compartments; Q/F, intercompartmental clearance; V/F_p, apparent volume of distribution of the peripheral compartment; F, oral bioavailability; C_max, maximum plasma concentration; AUC, area under the concentration-time curve.

^bIIV, interindividual variability, expressed as the approximate percent coefficient of variation (% CV).

^cIOV, interoccasional variability, expressed as the approximate % CV.

^dCL/F = TV[CL/F] · (1+EFF_RFP), where TV[CL/F] is the typical value of moxifloxacin CL/F when not coadministered with rifapentine and EFF_RFP is the fractional change in CL/F due to the coadministration of rifapentine.

^et_1/2, calculated population elimination half-life when moxifloxacin is not coadministered with rifapentine.

^fFixed value.

In our study, we found that high-dose intermittent rifapentine increased the CL/F of moxifloxacin by only 8%, corresponding to a reduction in the area under the concentration-time curve from 0 h to infinity (AUC_0-∞) of approximately the same magnitude (Table 2). A direct effect on bioavailability was tested but did not reach statistical significance. The extent of interaction in our study was lower than the 17.2% decrease in moxifloxacin exposure previously reported in healthy volunteers dosed three times a week (8). The discrepancy between these results may be due to differences in physiology between tuberculosis patients and healthy volunteers with respect to the activity of drug-metabolizing enzymes, genetic differences in the study population, or the different dosing strategies. Carry-over effects due to multiple dosing of moxifloxacin during rifapentine administration are unlikely to have affected the results, as predose moxifloxacin concentrations were below the validated range in all patients but two (both received twice-weekly doses, and they had predose concentrations of 0.088 and 0.066 μg/ml, respectively). The induction effect of rifapentine on moxifloxacin in our study was less than that of 450 mg of rifampin dosed three times weekly (14), possibly due to different dosing frequencies. Interestingly, Bliven-Sizemore et al. recently showed that rifapentine in doses of 10 to 20 mg/kg is as potent an inducer of CYP3A as rifampin (5), all dosed daily. We could not detect differences in moxifloxacin exposure or clearance between the regimens used in our study. In conclusion, moxifloxacin exposures were not affected by high doses of rifapentine given to tuberculosis patients in once- or twice-weekly doses, and the potential pharmacokinetic interaction is unlikely to affect treatment outcomes.

Table 2.

Pharmacokinetic values for different moxifloxacin dosing regimens

Moxifloxacin dose	Median Cmax (mg/liter) (range)	Median AUC (mg · h/liter) (range)	No. of participants
Alone (single dose)	3.8 (2.1–4.6)	50.8 (31.8–65.3)	27
Once weeklya	2.9 (2.0–3.8)	46.2 (29.5–60.6)	13
Twice weeklya	2.8 (2.0–4.5)	45.3 (26.8–65.3)	15

^aMoxifloxacin coadministered with rifapentine.