Pharmacokinetic and Pharmacodynamic Study of High-Dose Rifapentine and Moxifloxacin for Treatment of Tuberculosis

RIFAPENTINE Rifapentine is approved by the United States Food and Drug Administration for treatment of drug-susceptible pulmonary TB based on an open-label prospective randomized phase 3 study of 722 participants with pulmonary TB. Rifapentine has a longer half-life than rifampin, but both drugs inhibit bacterial RNA synthesis by binding to the β-subunit of DNA-dependent RNA polymerase. Rifapentine Pharmacokinetics With once-weekly treatment in participants with pulmonary TB, mean plasma AUC0-inf was 296, 410, and 477 μg\*h/mL after 600, 900, and 1200 mg doses administered without food, and Cmax was 12, 15, and 19 μg/mL. The bioavailability of rifapentine was 70% in healthy adult volunteers who were given a single 600-mg dose. Food (total, 850 calories 33 g protein, 55 g fat, and 58 g carbohydrate) increases AUC0-inf by 43% and Cmax by 44% compared with fasting conditions. Rifapentine and its 25-desacetyl metabolite are 98% and 93% protein-bound in serum, primarily to albumin. Rifapentine is metabolized to 25-desacetyl rifapentine by an esterase enzyme present in the liver and blood. The half-life of rifapentine is 14 to 17 hours and the half-life of 25-desacetyl rifapentine is 13 hours. Rifapentine is excreted in bile and eliminated in feces, and \< 10% rifapentine is excreted unchanged in urine. The microbiologically active metabolite contributes 40% of the overall activity of the drug. The MIC of rifapentine against M. tuberculosis is 0.05 μg/mL (0.25 μg/mL for 25-desacetyl rifapentine). Single-dose rifapentine PK parameters are similar in healthy females and males. Clinical Trials of Rifapentine The TBTC performed phase 2 studies to assess the antimicrobial safety and efficacy of daily rifapentine administered with isoniazid, pyrazinamide, and ethambutol for pulmonary TB treatment. In Study 29, adults (n = 531) who had sputum-smear-positive pulmonary TB were randomized to receive rifapentine 10 mg/kg/dose or rifampin 10 mg/kg/dose without food, 5 days per week for 8 weeks, with isoniazid, pyrazinamide, and ethambutol. This study showed no significant difference in antimicrobial activity between regimens, based on the surrogate marker of culture status at completion of the intensive phase (culture conversion on solid media: rifampin group, 83%; rifapentine group, 86%) (culture conversion in liquid media: rifampin group, 65%; rifapentine group, 68%). The rifapentine regimen was well tolerated, and similar proportions of participants between groups discontinued the assigned treatment overall (rifampin group, 16%; rifapentine group, 15%) or because of toxicity (rifampin group, 1.2%; rifapentine group, 1.5%). There were no differences in proportions of participants between treatment groups who had a serious adverse event related to study treatment (rifampin group, 0.4%; rifapentine group, 1.1%) or type or severity of adverse events between groups. The investigators concluded that the rifapentine regimen given on an empty stomach 5 days/week for 8 weeks was safe and well tolerated but not significantly more active than rifampin-based therapy. A subsequent dose-ranging study to determine the optimal dose of daily rifapentine during the first 8 weeks of TB treatment was performed by the TBTC. In Study 29X, adults (n = 331) with sputum-smear-positive pulmonary TB were randomized and received or rifampin (10 mg/kg/dose) or rifapentine (10, 15, or 20 mg/kg/dose; maximum dose, 1500 mg) given with a high fat meal daily (7 days per week) for 8 weeks, in addition to isoniazid, pyrazinamide, and ethambutol. At completion of the intensive phase, negative cultures on solid media were noted in 52 of 64 subjects in the rifampin group (81%); 62 of 67 subjects in the rifapentine 10 mg/kg group (93%; not significant); 59 of 66 subjects in the rifapentine 15 mg/kg group (89%; not significant); and 54 of 57 subjects in the rifapentine 20 mg/kg group (95%; P ≤ .05). Negative cultures in liquid media in the modified intention-to-treat analysis group occurred in 36 of 64 subjects in the rifampin group (56%); 50 of 67 subjects in the rifapentine 10 mg/kg group (75%; P ≤ .04); 46 of 66 subjects in the rifapentine 15 mg/kg group (70%; not significant); and 47 of 57 subjects in the rifapentine 20 mg/kg group (83%; P ≤ .01). High-dose daily rifapentine in combination with standard anti-TB drugs was well tolerated and highly active in surrogate bacteriologic tests. In the treatment trials 29 and 29X, the population PK properties of rifapentine were characterized in 415 participants. Rifapentine oral clearance did not significantly change with weight. Rifapentine bioavailability decreased with greater rifapentine doses and the increase in exposure was less than dose-proportional. Median exposures at steady state of rifapentine AUC0-24 were 292, 474, and 579 μg\*h/mL, respectively, for 600, 900, and 1200 mg daily doses. Because rifapentine clearance was not weight-dependent, a level rifapentine treatment dose (that is, a mg rather than a mg/kg dose) is recommended for adults. Rifapentine 1200 mg daily is incorporated into the Study 31 treatment trial protocol. Because food taken with rifapentine study doses significantly increased rifapentine exposure, it is recommended in Study 31 that a meal be taken prior to rifapentine study doses. An important knowledge gap in TB clinical trials is how to efficiently and economically assess new drug efficacy, as biomarkers now commonly used in phase 2 trials have had varying success in predicting drug efficacy in phase 3 trials. However, optimal use of a drug can be achieved through population pharmacokinetic-pharmacodynamic (PK/PD) studies that combine pharmacokinetic properties of a drug and efficacy outcomes of treatment. In TBTC Study 29-29X PK/PD, the relation between rifapentine exposure and response by time to conversion in sputum cultures was established. In these analyses, the efficacy of rifapentine was exposure-dependent. Using maximum inhibitory effect (Emax) in a time-to-event model, a highly significant rifapentine exposure-response relationship was observed. With rifapentine AUC0-24 \> 350 μg\*h/mL, the percentage of participants with culture conversion of sputa from positive to negative in liquid media at 2 months of rifapentine treatment was modeled to be 74% and with rifapentine AUC0-24 \< 300 μg\*h/mL ("low rifapentine AUC) to be 55%. In summary, adequately high rifapentine exposure resulted in better responses, and large cavity size in poorer responses. The TBTC Study 31 is a phase 3, multicenter, international, randomized, controlled, open-label, 3-arm, non-inferiority trial that will enroll newly diagnosed, previously untreated participants with pulmonary TB. Two investigational regimens will be evaluated to determine if the duration of treatment for drug-susceptible pulmonary TB can be reduced compared to a standard rifampin-containing six month treatment regimen. In one investigational arm, the efficacy of the single substitution of rifapentine for rifampin for the 17 week regimen will be assessed. The second investigational arm will evaluate a rifapentine-containing regimen that also substitutes moxifloxacin for ethambutol and continues moxifloxacin for 17 weeks. The Study 31 primary efficacy endpoint is TB disease-free survival at 12 months after study treatment assignment. In addition, time to culture conversion will be assessed. The primary safety endpoint is the proportion of participants with grade 3 or higher adverse events during study drug treatment. This PK/PD Study is a component of TBTC Study 31. MOXIFLOXACIN Moxifloxacin Pharmacokinetics Moxifloxacin is a fluoroquinolone that has potent activity against M. tuberculosis. Moxifloxacin is well absorbed and has 90% bioavailability. The PK parameters are linear from 50 to 800 mg (single dose) and ≤ 600 mg (once daily dosing) over 10 days). Steady state is reached within 3 days. The mean ± SD Cmax and AUC0-24 values at steady state with a 400-mg once-daily dosage regimen are 4.5 ± 0.5 μg/mL and 48 ± 3 μg\*h/mL. Trough plasma concentration at steady state (400 mg once daily) is 1.0 ± 0.1 μg/L. The time of Cmax (Tmax) is 1 to 3 hours. The mean plasma half-life is 12 ± 3 hours. Coadministration with food may slightly prolong Tmax and reduce Cmax by 16%, but these effects are not clinically important. Administration with yogurt or a high fat meal did not significantly affect the AUC. However, coadministered with aluminum-, magnesium-, or calcium-containing antacids markedly reduced oral bioavailability of fluoroquinolones. Moxifloxacin is 50% bound to plasma proteins. It is widely distributed, with some tissue concentrations in excess of plasma levels. Moxifloxacin is metabolized by glucuronide and sulfate conjugation. The sulfate conjugate (M1) accounts for 38% oral dose and is excreted in feces; 14% of an oral dose is converted to the glucuronide conjugate (M2) and is excreted in urine. Peak plasma levels of M1 are \< 10% and M2 are 40% parent drug levels. Much (45%) of an oral dose is excreted as the parent drug and 51% as known metabolites. Based on population PK modeling, moxifloxacin AUC0-24 in participants who have TB can be predicted with acceptable accuracy using limited sampling strategy. Rifampin induces the activity of the phase 2 enzymes glucuronosyltransferase and sulfotransferase. In a PK drug interaction between rifampin and moxifloxacin, the moxifloxacin AUC0-24 decreased 27% (ratio of geometric mean, 73 \[90% CI, 64 to 84\]) with coadministration of rifampin at a dose of 600 mg daily. In a PK interaction study of 19 participants who had tuberculosis, moxifloxacin AUC0-24 and Cmax decreased when moxifloxacin was administered daily with rifampin and isoniazid compared with moxifloxacin alone (ratio of geometric mean AUC0-24, 0.69 \[90% CI, 0.65 to 0.74\]; Cmax, 0.68 \[90% CI, 0.64 to 0.73\]). With rifapentine dosing 3 times per week in a PK interaction study in healthy volunteers, moxifloxacin AUC0-inf was decreased 17%. In the RIFAQUIN PK study with 28 adults who had pulmonary tuberculosis treated with a continuation-phase regimen of 400 mg moxifloxacin and 900 mg rifapentine twice weekly, or 400 mg moxifloxacin and 1200 mg rifapentine once weekly, median moxifloxacin AUC0-inf with rifapentine treatment once weekly was decreased 9% or twice weekly was decreased 11%. These PK drug interaction studies indicate that moxifloxacin AUC0-inf is decreased with increased frequency of rifamycin dosing (from once weekly up to daily dosing). The magnitude of reduction in moxifloxacin concentrations when it is given with rifapentine at a dose of 1200 mg daily is unknown. Moxifloxacin for Tuberculosis Treatment Two phase 2 TB clinical trials that have shown that substitution of moxifloxacin for ethambutol during the intensive phase of pulmonary TB treatment increases the bactericidal activity with multi-drug, intensive-phase regimen, as assessed by sputum culture conversions to negative after 2 month of treatment. In another study, treatment with a weekly regimen of rifapentine and moxifloxacin during the continuation phase of therapy (for a total TB treatment for 6 months) was not inferior to daily isoniazid plus rifampin; the efficacy of rifapentine plus moxifloxacin in this trial was important because once-weekly rifapentine with isoniazid (instead of moxifloxacin) was associated with a higher frequency of relapse and treatment failure. In the REMox phase 3 trial, randomized double-blind trial to test non-inferiority of a seventeen weeks of isoniazid, rifampicin and moxifloxacin supplemented by pyrazinamide for the first eight weeks (INH-arm) compared to standard 6 months of rifampin based regimen for tuberculosis, the number of participants classified as favorable in the control regimen per protocol was 467 (92%) and in the INH-arm 436 (85%) a difference of 6.1% (97.5% CI 1.7-10.5). In this trial, the moxifloxacin containing regimen was more bactericidal than the control regimen, but there was insufficient activity to permit a rifampin-moxifloxacin regimen to be shortened to 4 months. The relationship between moxifloxacin concentrations and TB treatment outcomes when moxifloxacin is given as part of high-dose rifapentine-based multidrug treatment for drug-sensitive TB is unknown. Further, the target exposure of moxifloxacin for treatment of TB has not been defined. Rationale for Intensive and "Late" PK Samplings for Rifapentine at Two Different Occasions Autoinduction has been described previously with daily dosing of rifamycins, including rifampin and rifapentine. With autoinduction of metabolizing enzymes or transporters, both bioavailability and clearance can theoretically be affected. In a population PK model from healthy volunteers, rifapentine clearance appeared to be time-dependent, whereas time had a less significant effect on bioavailability. In addition, there was no clear relationship between dose and clearance, or dose and the effects of time on clearance. Further in this prior study, the time to maximal autoinduction with daily dosing could not be estimated as increases in clearance were seen up until the final PK sampling day after 14 days of dosing. PK sampling over longer dosing intervals is required to determine the time course of autoinduction and the change in exposures over time with repeated dosing. Information from this PK study is needed to fill this knowledge gap, and to model the population PK model parameters without bias and with satisfactory precision (relative standard error \< 20%). Reiterated simulations and re-estimations ("sse") of rifapentine data in Study 29-29X were performed to assess the optimal Study 31 PK/PD design for: (1) between-subject variability (BSV) and precision of the population PK parameters, and (2) the longitudinal component of rifapentine clearance by autoinduction. Uncertainty in parameter estimates (RSE) of \<10% was considered acceptable for the population PK parameters and \< 20% for the parameter describing autoinduction. These simulations indicated that for the parameter describing rifapentine auto-induction, it is important that multiple PK samplings are obtained, and that PK sampling occasions be time-balanced. If this is not performed, the uncertainty of the longitudinal change in clearance substantially increases, i.e. a protocol design with only one initial PK sampling is not able to identify longitudinal change in clearance with satisfactory precision, and leads to high uncertainty in parameter estimates. A parsimonious design that efficiently assesses longitudinal changes in clearance with adequate certainty and limited costs, will include the following features: (i) a minimum of 60 participants; (ii) intensive PK sampling (6 optimal samples/participant (times = 0.5, 3, 6, 9, 12, 24 hours) on at least one occasion; and (iii) late sampling with 2 to 3 PK samples at another, later occasion separated by a minimum of 14 days. In summary, this PK/PD Study is valuable because, it will allow a robust development of population PK model parameters for rifapentine and for moxifloxacin in presence of high-dose rifapentine, and evaluate the PK/PD relationships of study drugs with efficacy outcomes.