Comparison of the Drug-Drug Interaction Potential of Daptomycin in Combination with Rifampin in Healthy Adult Volunteers

We evaluated the effects of rifampin coadministration and MDR1 single nucleotide polymorphisms on the disposition of daptomycin in twelve healthy adults. There were no significant changes from baseline in the clearance (0.53 versus 0.55 liters/h, P = 1.00), volume of distribution (7.0 versus 7.2 liter, P = 0.62), or half-life (9.7 versus 9.6 h, P = 0.89) of daptomycin after exposure to rifampin.

ABSTRACT

We evaluated the effects of rifampin coadministration and MDR1 single nucleotide polymorphisms on the disposition of daptomycin in twelve healthy adults. There were no significant changes from baseline in the clearance (0.53 versus 0.55 liters/h, P = 1.00), volume of distribution (7.0 versus 7.2 liter, P = 0.62), or half-life (9.7 versus 9.6 h, P = 0.89) of daptomycin after exposure to rifampin. The tested MDR1 polymorphisms were not associated with significant differences in daptomycin disposition.

TEXT

Although not a substrate of traditional metabolic pathways like cytochrome P450, daptomycin has been reported to be a substrate of the efflux transporter p-glycoprotein (PGP) (1). Rifampin is a potent activator of certain transcription factors that increase the production of PGP (2–4), creating the potential for a drug-drug interaction between rifampin and daptomycin that may lead to increased clearance (CL) of daptomycin. The primary aim of this study was to determine the effect of rifampin coadministration on the pharmacokinetic profile of daptomycin. We also assessed whether single nucleotide polymorphisms (SNPs) in the multidrug resistance gene MDR1, which encodes PGP, influenced the disposition of daptomycin.

This study was reviewed and approved by the University of Utah Health Sciences Center institutional review board. This was an open-label, sequential pharmacokinetic study in 12 healthy volunteers. After providing written informed consent, the subjects meeting the inclusion criteria received daptomycin 6 mg/kg intravenously over 2 min, and blood and urine samples were collected over a 24-h period after infusion. After the 24-h study visit, the subjects were given a 13-day supply of rifampin capsules (600 mg) to be taken daily at home. After the completion of the rifampin course, a second pharmacokinetic sampling was performed. At the second study visit, the subjects were administered rifampin 600 mg orally, followed by daptomycin 6 mg/kg intravenously over 2 min, and blood and urine samples were collected over a second 24-h period.

At each infusion, a peripheral venous catheter was placed in each arm: one for daptomycin administration and one for sampling. Six milliliters of blood was drawn immediately before the dose of daptomycin and at 0.5, 2, 4, 8, 12, and 24 h after the start of the infusion for the determination of daptomycin plasma concentrations. Additionally, 10 ml of blood was collected at 4 h for plasma protein binding studies. A 24-h urine collection starting at the time of daptomycin infusion was performed to determine creatinine clearance (mCL_CR) and renal clearance (CL_R) of daptomycin. Ten milliliters of blood was drawn at the first study visit prior to the administration of daptomycin for MDR1 genotyping studies.

Daptomycin protein binding was performed in triplicate using Centrifree centrifugal filters (Millipore) according to the manufacturer’s instructions. Daptomycin concentrations in plasma, ultrafiltrate, and urine samples were determined with a previously described high-performance liquid chromatography assay at the Center for Anti-Infective Research and Development (5). Genomic DNA was extracted from whole blood, and MDR1 genotypes were determined in duplicate by a real-time PCR assay (6). Haplotypes were determined from genotypic data using the method of Clark (7). Plasma concentration-time data were analyzed by a noncompartmental analysis with WinNonlin Pro (Phoenix 64 build 6.3.0.395, WinNonlin version 6.3; Certara USA, Inc., Princeton, NJ).

A statistical analysis of pharmacokinetic parameters was performed using JMP Pro, version 13.0.0. The a priori alpha value for all statistical analyses was set at ≤0.05; no adjustment for multiplicity was considered. Differences in the distribution of daptomycin pharmacokinetic parameters before and after rifampin were compared using Wilcoxon’s signed-rank test. Linear mixed-effects models were constructed for each pharmacokinetic parameter as the dependent variable separately, including fixed effects for MDR1 polymorphism or diplotype and the presence/absence of rifampin and also a subject-level random effect. Each genotype, haplotype, or diplotype was considered a fixed effect, with separate means assumed for each level and no interactions between levels. The individual covariates of age, total body weight, race, sex, and mCL_CR were also evaluated in the models. Model terms were first selected by an automated forward/backward stepwise method based on Akaike’s information criterion and then followed by backward selection based on individual tests using an alpha level of 0.05 for significance. The Kenward-Roger approximation was used to obtain P values from mixed-effects models (8). The null hypothesis for all comparisons was that there was no difference in any of the observed daptomycin pharmacokinetic parameters across the levels of the tested variables. By assuming a mean baseline CL of 0.8 liters/h and a standard deviation of 0.1 liters/h, 10 subjects were necessary to detect a 20% change in CL after rifampin administration with 80% power and a two-sided alpha of 0.05 (9).

Seventeen subjects were screened and twelve were enrolled in the study. All 12 subjects who were enrolled completed the trial. The demographic characteristics of the subjects who completed the study are included in Table 1.

TABLE 1.

Demographic characteristics of study subjects

Subject	Age (yr)	Weight (kg)	Sex	Race	Baseline eCLCRa (ml/min)	Baseline mCLCR (ml/min)	SNPs			Haplotypes		Daptomycin CL (liters/h)		AUC0-∞ (mg·h/liter)		C30min (mg/liter)		C24h (mg/liter)		VSS (liters)		t1/2 (h)
							C1236T	G2677T/A	C3435T	1	2	Visit 1	Visit 2	Visit 1	Visit 2	Visit 1	Visit 2	Visit 1	Visit 2	Visit 1	Visit 2	Visit 1	Visit 2
1	26	71.8	Male	White	117.2	116.5	CC	GG	TT	CGT	CGT	0.52	0.56	824.2	752.9	73.3	70.4	10.5	9.6	7.5	7.6	10.7	9.6
2	23	71.9	Male	White	125.6	123.5	TT	TT	CT	TTC	TTT	0.55	0.56	782.8	805.4	72.9	80.9	9.6	9.9	8.0	6.9	11.3	8.8
3	20	62.1	Female	White	107.3	100.2	TT	TT	CT	TTC	TTT	0.45	0.45	825.8	826.8	83.7	83.6	9.5	9.1	5.4	5.0	8.7	7.9
4	22	83.7	Male	White	130.6	174.2	CC	GG	CC	CGC	CGC	0.74	0.73	678.2	705.6	61.7	92.1	8.5	7.6	10.9	8.5	11.1	9.1
5	40	64.9	Female	White	83.3	93.4	CT	GG	CT			0.57	0.56	679.8	695.8	64.4	60.4	8.0	9.3	7.2	9.1	9.1	11.9
6	25	49.4	Female	White	88.2	107.9	TT	TT	CT	TTC	TTT	0.44	0.39	681.3	776.0	73.2	79.4	8.4	9.0	4.9	5.1	7.6	10.0
7	26	68.4	Male	White	99.4	125.0	CC	GG	CC	CGC	CGC	0.47	0.44	873.0	924.2	64.6	77.3	11.2	12.3	6.7	6.7	10.2	11.1
8	26	70.1	Female	Hispanic	104.8	113.6	CT	GT	CT			0.57	0.54	739.9	780.1	75.4	77.5	8.7	10.3	7.7	7.9	10.4	11.0
9	46	56.5	Female	White	110.0	120.9	TT	GG	CT	TGC	TGT	0.43	0.45	789.7	728.3	61.9	74.1	10.1	8.9	6.5	6.0	10.9	10.1
10	23	69.3	Male	White	97.1	102.4	CT	GT	TT	CGT	TTT	0.53	0.49	772.4	846.0	76.1	82.6	5.6	8.8	5.0	5.6	6.4	8.1
11	22	71.4	Male	White	105.4	141.9	CC	GG	Undeterminedb	CG	CG	0.72	0.66	598.4	641.6	67.1	77.4	5.8	6.8	7.4	7.4	7.4	8.3
12	31	74	Female	Other	97.2	107.6	CC	GG	CC	CGC	CGC	0.53	0.60	837.6	724.6	78.0	66.3	9.2	8.5	6.4	7.8	9.2	9.7
Median (IQR)	26 (22–30)	69.7 (62.8–71.9)			105.1 (97.1–115.4)	115.0 (103.7–124.6)						0.53 (0.46–0.57)	0.55 (0.45–0.59)	777.6 (680.2–825.4)	764.4 (710.3–821.5)	73.0 (64.5–76.0)	77.4 (71.3–82.2)	8.9 (8.1–9.9)	9.1 (8.6–9.8)	7.0 (5.6–7.7)	7.2 (5.7–7.9)	9.7 (7.8–10.8)	9.6 (8.4–10.8)
P valuec												1.00		0.89		0.08		0.71		0.62		0.89

^aeCL_CR, estimated creatinine clearance.

^bMost likely due to a previously undescribed SNP at position 3435 (i.e., C3435G or C3435A).

^cFor comparison of differences in pharmacokinetic parameters on visit 1 versus visit 2.

The pharmacokinetic parameters of daptomycin before and after exposure to rifampin are summarized in Table 1. Rifampin had no significant impact on the CL, area under the plasma concentration-time curve from time zero to infinity (AUC_0-∞), volume of distribution at steady-state (V_SS), daptomycin plasma concentrations at 30 min (C_30min) and 24 h (C_24h), terminal elimination half-life (t_1/2), or protein binding. For 7 of 24 subject visits, the calculated amount of daptomycin excreted in urine was larger than the dose administered. Due to concerns for a systemic error in our methodology, the urinary concentration data set was discarded, and planned determinations of CL_R and nonrenal clearance were not performed.

No statistically significant associations were observed between the tested MDR1 SNPs or diplotypes and daptomycin pharmacokinetic parameters. Total body weight was independently associated with daptomycin CL (0.009 liter/h/kg; 95% confidence interval [CI], 0.004 to 0.014; between-subject covariance, 0.003; 95% CI, <0.001 to 0.006) and V_SS (0.12 liter/kg; 95% CI, 0.05 to 0.19; between-subject covariance, 0.63; 95% CI, −0.18 to 1.44), and age was associated with C_30min (−0.56 mg/liter/year; 95% CI, −0.86 to −0.25; between-subject covariance, −21.3; 95% CI, −51.1 to 8.4). The median daptomycin CL was lower in subjects homozygous TT at position 1236 than in subjects homozygous CC or heterozygous CT (CC, 0.58 liters/h; CT, 0.55 liters/h; TT, 0.44 liters/h; P = 0.12) and lower in subjects homozygous TT at position 2677 than in subjects homozygous GG or heterozygous GT/A (GG, 0.56 liters/h; GT/A, 0.55 liters/h; T/A T/A, 0.44 liters/h; P = 0.32). Daptomycin CL was similar across known C3435T polymorphisms (CC, 0.56 liters/h; CT, 0.53 liters/h; TT, 0.52 liters/h; P = 0.46).

The observed daptomycin CL values for each of the tested genotypes before and after exposure to rifampin are shown in Fig. 1. None of the tested SNPs or the associated haplotypes or diplotypes was associated with a difference in any daptomycin pharmacokinetic parameter after exposure to rifampin. Daptomycin CL was higher after rifampin exposure in subjects homozygous for the wild-type alleles at each of the MDR1 SNPs than in subjects that were heterozygous or homozygous recessive, but these observations were not statistically significant.

FIG 1.

Daptomycin clearance based on MDR1 genotype and study visit. Observed daptomycin CL values at each study visit according to MDR1 genotype are presented. Open circle, observations from visit 1 (daptomycin alone); open triangle, observations from visit 2 (rifampin coadministered with daptomycin); **, subject with undeterminable genotype. Horizontal lines denote mean CL values observed for each of the categories.

Despite a plausible theoretical mechanism, rifampin had a minimal effect on the pharmacokinetic disposition of daptomycin. The observed disposition of daptomycin in our study cohort was consistent with results from previous studies in healthy volunteers, with similar values observed for CL, V_SS, and t_1/2 (10–14). Although daptomycin has been reported to be a PGP substrate in in vitro models (1) and C3435T SNPs have been associated with altered daptomycin CL (15), the dependence on PGP in the overall elimination of daptomycin is unclear. The failure to observe a difference in daptomycin CL or other pharmacokinetic parameters after exposure to rifampin suggests that pathways other than PGP-mediated transport may play a larger role in the elimination of daptomycin. Notably, a post hoc power analysis found this study was 80% powered to detect a difference in daptomycin CL of 0.18 liters/h. In the absence of another indication, combination treatment with rifampin should not be the basis for more aggressive daptomycin dosing schemes. This is relevant clinically, since rifampin is added to daptomycin with increasing frequency in clinical practice, particularly for prosthetic device-associated infections (16, 17). For such infections, doses greater than the standard daptomycin dose of 6 mg/kg/day are often employed, creating a scenario in which doses adjusted further to account for a potential induction effect from rifampin might lead to daptomycin toxicity.

Unlike previous reports, we did not observe a statistically significant difference in daptomycin CL in subjects with different C3435T genotypes. This may be explained by sampling variability associated with a limited sample size, particularly regarding the races and ethnicities of the subjects enrolled in each study. Alternatively, the focus on individual SNPs may be too narrow, and subject haplotype and diplotype may better explain the observations across studies. Although not statistically significant, subjects with the wild-type CGC-CGC diplotype were observed to have the highest daptomycin CL. This is consistent with previous observations with other PGP substrates, such as digoxin and cloxacillin (18, 19). The CGC haplotype is most frequently found in subjects of African descent and to a lesser extent in Caucasian subjects, suggesting that daptomycin CL may be higher in these populations than in subjects of Indian or eastern Asian ethnicity for whom the TTT haplotype is most common (20). The degree to which daptomycin CL varies between various racial/ethnic groups and whether it is associated with differences in safety or effectiveness outcomes between populations of various genetic ancestries requires further study.

This study is primarily limited by the small sample size. Although we were successful in including subjects with each of the respective SNPs in C1236T, G2677T/A, and C3435T, our statistical power to detect differences between patients with various SNPs was limited. The sample size was an even greater limitation to our haplotype- and diplotype-based analyses, because we were only able to include a small fraction of the possible observations in our sample. Consequently, our results should be viewed as hypothesis generating. Additional studies with a much larger sample would be needed to characterize the role of various MDR1 diplotypes in the pharmacokinetics of daptomycin.

ACKNOWLEDGMENTS

This work was funded by an investigator-initiated research grant from Cubist Pharmaceuticals (now Merck & Co). This study was performed at the University of Utah Center for Clinical and Translational Services, which is funded by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number UL1TR001067.

The content is solely the responsibility of the authors and does not necessarily represent the official views of Cubist, Merck, or the National Institutes of Health.