Preliminary Pharmacokinetic Study of Repeated Doses of Rifampin and Rifapentine in Guinea Pigs

TEXT

Clinical studies suggest that antituberculous treatment can be improved with greater rifamycin exposures. We recently reported that substitution of rifapentine (RFP) for rifampin (RIF) in the standard antituberculous regimen did not reduce the time required to cure guinea pigs with established Mycobacterium tuberculosis infection (1). This result may be quite important, as it contradicts similar studies in mice (2), but appears to corroborate recent findings from a clinical trial (TBTC Study 29) investigating the substitution of daily RFP for RIF in the first-line regimen for treatment of drug-susceptible tuberculosis (TB) (3). Guinea pigs received RFP and RIF exposures in combination (together with isoniazid and pyrazinamide) regimens equivalent to those in human and mouse studies based on area under the serum concentration-time curve from 0 to 24 h (AUC_0-24) and the maximum concentration of drug in serum (C_max) (1). However, since the doses selected for this study were based on single-dose pharmacokinetic (PK) experiments, it is possible that they may not accurately represent steady-state human PK parameters of each drug. For example, RIF is known to induce its own metabolism in humans, with a significantly lower AUC and half-life observed after only several days (4). Whether the same degree of autoinduction occurs in guinea pigs has not been studied to our knowledge. If RIF clearance is more significantly induced in mice relative to guinea pigs, it is possible that the relative potency of RIF is underrepresented in mice. Conversely, RFP accumulates in humans with daily dosing (5, 6). Therefore, if RFP accumulation over time is greater in mice relative to that in guinea pigs and humans, its activity may be overrepresented relative to RIF. Thus, differential metabolism of either drug may account for the observed differences in potency when used in combination treatment in each species. We performed a steady-state PK study of healthy guinea pigs to evaluate the metabolism and autoinduction of RIF and RFP in this species and compared our results to corresponding values in humans.

Female outbred Hartley guinea pigs (300 to 350 g) with jugular vein vascular catheters (Charles River Labs, Wilmington, MA) were maintained under specific-pathogen-free conditions and fed water and chow ad libitum. All procedures followed protocols approved by the Institutional Animal Care and Use Committee at Johns Hopkins University. To generate the steady-state data, separate groups of four guinea pigs were treated once daily for 7 days with either 100 mg/kg RIF (R₁₀₀) (Bio-World, Dublin, OH) or 100 mg/kg RFP (RFP₁₀₀) (Priftin; Sanofi-Aventis). Doses were prepared in 40% sucrose in a final volume of 0.5 ml and delivered in the posterior oropharynx by an automatic pipette with a disposable tip, as previously described (1, 7). Blood (∼0.3 ml) was drawn serially from the intravenous catheter before drug administration on days 0 and day 8 and ½, 1, 2, 4, 6, 8, and 24 h after dosing. Samples were collected in microcentrifuge tubes, held for 30 min at room temperature, and centrifuged at 6,500 rpm for 10 min to obtain serum, which was frozen at −80°C until the time of analysis. Drug concentrations were determined using validated high-performance liquid chromatography (HPLC) assays and confirmed using a newly developed liquid chromatography-tandem mass spectrometry (LC-MS/MS) approach (8, 9). Analyses of PK parameters were performed using noncompartmental techniques (WinNonLin PK software, version 4; PharSight, Sunnyvale, CA). Total drug exposure was calculated using AUC_0-24.

Steady-state PK values of RIF and RFP (Table 1) differed from those of single-dose studies (1, 7). The C_max and AUC were significantly increased at the steady state for both RFP and RIF compared to single-dose data. In the present study, we determined metabolite concentrations from serum samples obtained in earlier single-dose studies (1) and found that desRFP was absent in guinea pigs that received a single dose of RFP of 10 or 50 mg/kg but present in 2 of 3 animals that received a single dose of 100 mg/kg. The guinea pig in which desRFP serum concentrations were undetectable following a single dose of RFP at 100 mg/kg was found to have significantly lower serum RFP concentrations than those of the other two animals receiving the same dose (data not shown). In the present steady-state data set, desRFP was present in most samples (Table 1) and showed a similar concentration-versus-time profile, but at much lower concentrations than the parent RFP. Most desRFP concentrations were below 1 μg/ml, with 2 of 4 animals achieving 2- to 3-μg/ml peak metabolite values. A pattern of low or absent desRIF concentrations was seen in guinea pigs treated with RIF, with only 1 of 4 animals achieving 2.8-μg/ml peak metabolite values (Table 1). Possible types of interference in the measurement of desRIF by HPLC were eliminated by retesting with LC-MS/MS. Comparisons of single-dose versus steady-state C_max and AUC data show no evidence of autoinduction of clearance for either RIF or RFP in guinea pigs (Table 1). Although the median RFP half-life (t_1/2) decreased with the 4 steady-state animals, the range for the single-dose data appeared to be affected by some outlying high values. With relatively sparse late samples, it can be difficult to calculate the t_1/2 confidently (Table 1).

To our knowledge, this is the first report to explore the steady-state PK of RIF and RFP in guinea pigs. While repetitive dosing of RFP leads to accumulation of the drug and increased toxicity in rats (Priftin package insert; Hoechst Marion Roussel, Kansas City, MO), data from mice and monkeys revealed autoinduction with repeated dosing, leading to reduced accumulation and good tolerability at relatively high weight-based doses. In guinea pigs, some accumulation of both RIF and RFP occurred in this limited data set. Our results provide further evidence that RFP is differentially metabolized and cleared in different species (see Table S1 in the supplemental material) and underscore the observation that steady-state concentrations cannot necessarily be predicted from single-dose studies. Although both RFP and, to a lesser degree, desRFP showed roughly similar half-lives to corresponding values in humans (10), the desRFP metabolite was detectable at almost 10-fold lower concentrations in guinea pigs than in humans (see Table S1). The AUC ratio of metabolite to parent drug in guinea pigs was 0.077, compared to 0.64 in humans (5, 6). The steady-state PK profiles of RFP also differ between guinea pigs and mice (see Table S1), as the desRFP metabolite was not detected in mouse sera (2, 11). Depending on test conditions, desRFP is 10 to 50% as active as RFP in vitro against M. tuberculosis. The relative ability of the metabolite versus the parent drug to reach the pathogen in lesions and to bind to its target is not known (5, 6). Thus, desRFP activity might be added to RFP activity if there is an excess of targets, or it might compete with (antagonize) RFP activity if the weaker metabolite occupies targets at the expense of the parent drug. At least in the paper by Rastogi, the desRFP metabolite was similar in potency to RIF, so one might anticipate that additive effects might be more likely (11). Comparing guinea pigs with mice, the high intracellular accumulation of RFP likely favored its activity against TB infection in murine lungs, in which the organisms are almost exclusively intracellular (1, 2). Furthermore, the high protein binding of RFP (97.1% in healthy volunteers) may have limited penetration of the drug into lung granulomas of chronically infected guinea pigs, and the activity of RFP-containing regimens may be more modest against the predominantly extracellular TB infection within the necrotic cores of guinea pig lung granulomas. The results of ongoing clinical studies (TBTC Study 29PK) determining if free (non-protein-bound) RFP and free RIF exposures are directly associated with antimycobacterial activity are awaited to determine the clinical relevance of our findings in the guinea pig model. Further studies of protein-binding properties and penetration of RIF and RFP in lungs of infected guinea pigs or those of larger species are warranted.