Reassessment of tigecycline bone concentrations in volunteers undergoing elective orthopedic procedures
Article first published online: 24 OCT 2013
© 2013, The American College of Clinical Pharmacology
The Journal of Clinical Pharmacology
Volume 54, Issue 1, pages 70–74, January 2014
How to Cite
Bhattacharya, I., Gotfried, M. H., Ji, A. J., Saunders, J. P., Gourley, I., Diehl, A. and Korth-Bradley, J. M. (2014), Reassessment of tigecycline bone concentrations in volunteers undergoing elective orthopedic procedures. Journal of Clinical Pharma, 54: 70–74. doi: 10.1002/jcph.201
- Issue published online: 2 JAN 2014
- Article first published online: 24 OCT 2013
- Accepted manuscript online: 8 OCT 2013 05:25AM EST
- Manuscript Accepted: 1 OCT 2013
- Manuscript Received: 15 MAY 2013
- Wyeth Pharmaceuticals, Inc.
- bone assay;
- diabetic foot infection;
The goal of the this study was to re-evaluate tigecycline bone concentrations in subjects undergoing elective orthopedic surgery, using multiple doses and a more robust bone assay than was used in a previous study. Each subject received three intravenous doses of tigecycline (one 100-mg infusion followed by two 50-mg infusions, each administered over 30 minutes). A single bone sample was collected from each subject at one of the following times: 1, 2, 4, 6, 8, or 12 hours after the third dose. Four blood samples were collected from each subject: before the first dose, before and after the third dose, and within 15 minutes of the collection time of the bone sample. Noncompartmental pharmacokinetic analysis serum and bone area under the curve for the given dose interval (AUCτ) values were 2,402 ng h/mL and 11,465 ng h/g, and maximum concentration (Cmax) values were 974 ng/mL and 2,262 ng/g, respectively. The bone to serum ratio calculated using the AUCτ values was 4.77, confirming tigecycline penetration into bone.
Tigecycline is approved for the treatment of complicated skin and skin structure infections, complicated intra-abdominal infections, and community-acquired pneumonia using an initial dose of 100 mg followed by 50 mg every 12 hours.
The pharmacokinetics of tigecycline has been extensively studied and is reasonably well characterized.[1, 2] Following single intravenous (IV) infusions of 12.5–300 mg, or multiple day infusions of 25–100 mg, tigecycline demonstrated linear pharmacokinetics with a low systemic clearance (CL; 0.2–0.3 L/h/kg), a large volume of distribution (7–10 L/kg), and a half-life of 37–67 hours. After 10 days of every 12 hours multiple-dose administration, serum tigecycline concentrations (measured 12 hours after the start of the 1-hour infusion), showed an accumulation of approximately 2.5- to 3.4-fold, suggesting an effective half-life of 16–24 hours. Further, tigecycline does not have active metabolites and there are no age- or gender-related differences in its pharmacokinetics.
Structural similarities to tetracycline,[5, 6] along with data from preclinical studies using [14C]tigecycline in rats, suggested that tigecycline could readily penetrate into bone.[6, 7] However, a previously reported study in otherwise healthy subjects undergoing elective orthopedic surgery suggested the contrary. The analytical method used to determine tigecycline bone concentrations had a low extraction recovery and was perhaps the reason for the apparently low concentrations. Ji et al. later developed a new assay increasing the assay extraction recovery of tigecycline from bone.[9, 10] The previous study used a single dose of tigecycline, which may have provided inadequate time for penetration of the drug into bone. The goal of the present multiple-dose study was to re-evaluate tigecycline bone concentrations using multiple doses (to better access steady-state concentrations) and a more robust assay. The data from this study would be used to interpret the penetration of tigecycline into bone.
This study used an open-label, multiple-dose design to estimate tigecycline penetration into uninfected human bone. Subjects participating in this study were scheduled to undergo a planned bone or joint surgery or procedure. Male and female subjects older than 18 years of age were eligible for the study. Women of nonchildbearing potential needed to be either surgically sterile or postmenopausal for more than 1 year. Nonpregnant women of childbearing potential needed to be on acceptable birth control measures over a 1-month period, both pre- and post-dose of tigecycline. All subjects included were healthy as determined on the basis of medical history, physical examination, vital signs, and 12-lead electrocardiogram.
Subjects who were lactating, or those with any acute disease, hypersensitivity to tigecycline or tetracyclines, any disorder that might potentially affect pharmacokinetic assessment, or those taking other investigational drugs were excluded from the study. The study was conducted in accordance with the Declaration of Helsinki and its amendments as well as local laws and guidelines and was approved by the Western Investigational Review Board. Informed consent was obtained from all subjects who participated in the study prior to enrollment.
One or 2 days prior to their procedure, subjects were admitted to the hospital for tigecycline administration. Each subject received three IV doses; one 100-mg infusion followed by two 50-mg infusions. Each dose was administered over 30 minutes. The tigecycline dosing intervals were not less than 8 hours and not more than 12 hours.
Sampling and Analysis
A single bone sample was collected from each subject during the operation, at one of the target collection times of 1, 2, 4, 6, 8, or 12 hours after the third dose. Although the sample was collected from the bone or joint being operated on, no specific direction was given to the particular type of bone (e.g., cancellous or cortical bone) to collect. Four blood samples were collected from each subject: before the first dose, before and immediately after the third dose, and within 15 minutes of the collection time of the bone sample. All blood samples were collected from the arm that was not used for the infusion of tigecycline. The exact actual collection times of blood and bone samples were recorded and used in the pharmacokinetic calculations.
All blood samples were collected in tubes without any anticoagulant, allowed to clot, and then centrifuged under refrigerated conditions (4–6°C). The serum obtained was stored at −70°C for future analysis. For the bone samples, soft tissue and bone marrow were removed as much as possible, followed by rinsing to remove blood and air drying. Bone was cut into small pieces <2 cm3 and stored in airtight polypropylene containers at −70°C until the time of analysis, wherein the pieces were thawed and ground into particles of approximately 1 mm or less in diameter prior to tigecycline concentration determination.
Tigecycline serum and bone samples were quantified using two different liquid chromatography–mass spectrometry (LC/MS/MS) methods. The serum samples were analyzed by a method previously described.[8, 9] Briefly, serum samples (0.2 mL) were mixed with 0.6 mL of t-butyl-d9 tigecycline solution (internal standard, 150 ng/mL in acetonitrile), followed by centrifugation, separation of supernatant, and evaporation. The dried sample was reconstituted with mobile phase (200 µL) and then injected (10 µL) on to a C18 high-performance liquid chromatography (HPLC) column and detected with an API 3000 LC/MS/MS system. The mass-to-charge ratios of tigecycline precursor and product ion were monitored at 586.3–513.3 ion transition for tigecycline and at 595.4–514.3 ion transition for the internal standard (t-butyl-d9 tigecycline), respectively. The standard curve was linear from 10 to 2,000 ng/mL.
Tigecycline bone concentrations were quantified using a validated and sensitive human bone assay developed by Ji et al.[9, 10] This method was developed based on a rat bone assay. The bone method uses a special grinding procedure for bone, a special stabilizing agent, and a strong acidic extraction solvent to extract the analyte out of the bone. Briefly, t-butyl-d9 tigecycline (internal standard) and 0.1 M L-ascorbic acid solution (stabilizing agent) were added to the ground bone samples, followed by the addition of a strong acidic solvent mixture of perchloric acid and phosphoric acid in 50:50 methanol:water. This was followed by homogenization and centrifugation of the bone suspension. The supernatant from the above procedure was then injected on to a HPLC column for separation and detected using an API 3000 LC/MS/MS system for quantitation. The improved extraction efficiency was 79%, and the standard curve was linear from 50 to 20,000 ng/g using 0.1 g human bone. The chromatographic and mass spectrometry conditions for the human bone method were similar to the serum method.
Serum (or bone) concentrations were treated as composite data wherein all serum (or bone) concentrations were used collectively to obtain one serum (or bone) concentration–time profile. If multiple concentrations were present at precisely the same sample time, the mean of the concentrations, collected at that time, was used.
Pharmacokinetic parameters were determined with a noncompartmental approach using WinNonlin Pro version 5.1.1 (Pharsight Corp., Cary, NC), making the assumption that steady-state had been reached. Data from all subjects were used to estimate a single parameter value for the entire study population. Serum and bone area under curve for the given dose interval (AUCτ) were calculated using the log-linear trapezoidal method with a weighting scheme of 1/Y. The maximum concentration (Cmax) and the time to reach Cmax (tmax) were obtained directly from the concentration–time plots. Total serum CL was calculated as the ratio of dose to AUCτ.
Because the sampling times for bone were more restricted than serum, to estimate an AUCτ over the same time interval as for serum, the mean of all bone concentrations observed was used as an estimate of C0 and C12 for bone. A single AUC bone-to-AUC serum ratio was determined. Individual bone:serum ratios were calculated for all subjects using the individual bone concentration reported for each subject and the corresponding serum sample collected closest in time.
The 33 subjects in this study ranged in age from 39 to 86 years, with a median age of 66 years. The study population included 19 (58%) women and 14 (42%) men, and 97% of the subjects were white. All subjects were healthy and did not have any underlying medical condition(s) that might have interfered with the metabolism or excretion of tigecycline or with the interpretation of the study results.
All subjects completed the study and no serious adverse reactions related to drug treatment were reported. Unfortunately, most experienced nausea (76%) or vomiting (27%) of mild or moderate severity, which are known adverse reactions of tigecycline administration.
The individual serum and bone concentrations of patients, all of which were collected after dose 3, are shown in Figure 1. As may be seen, there was a distribution of sample collection times, permitting a robust characterization of the AUCτ. Although several serum concentrations were collected at the same time, no bone concentrations were collected at the same time. The range of individual bone-to-serum ratios observed was 1.11–14.31 with a median of 4.46 (Figure 2).
Serum pharmacokinetic parameters AUCτ, CL, Cmax and tmax values observed were 2402 ng h/mL, 20.8 L/h, 974 ng/mL, and 0.5 hours, respectively. The mean of all the bone concentrations reported in this study (range, 259–2,262 ng/g) was 898 ng/g. The bone AUCτ value was 11,465 ng h/g. The bone-to-serum ratio calculated using the AUCτ values was 4.77.
The goal of this study was to document the penetration of tigecycline into bone, rather than a definitive pharmacokinetic assessment. Serum concentrations were collected before the first dose was given to document lack of interference in the assay and then before and after the third dose to document that the concentrations were consistent with what would be expected in patients receiving the recommended therapeutic dose. The fourth serum sample, as noted previously, was used to calculate an individual bone:serum ratio, which would be expected to vary with time post-dose because of the variability of the serum concentration. The composite tigecycline serum concentration–time profile, not shown, was consistent with what has been previously reported after single- and multiple-doses. It declined rapidly after the infusion, suggesting a fast distribution phase, and then entered a more shallow elimination phase, a typical characteristic of tigecycline pharmacokinetics. The pharmacokinetic analysis was limited by the number of samples collected (four serum and one bone per individual) and the duration of sampling time (0–11 hours post-dose for serum and 1.5–11 hours post-dose for bone) after only 16–24 hours of tigecycline treatment, but provides an exposure estimate that is consistent with what has been reported for preclinical studies of tigecycline and in related drugs.
The extent of penetration into tissues is important to understanding the efficacy and safety of antibiotics. Single time points of assessment are difficult to interpret because of the dynamic nature of concentrations; thus, comparisons of steady-state exposures are preferred. In this present study, tigecycline was administered to otherwise healthy subjects undergoing an elective surgical procedure, and the bone and serum concentrations were determined to interpret the bone penetration of tigecycline after the administration of a total of three doses over 16–24 hours. The conditions attempted to mimic steady-state concentrations with repeated doses, but it should be recognized that true steady-state of serum concentrations would not be expected to be achieved for at least a week in the absence of a loading dose, given the long half-life. The high incidence of nausea and vomiting limit the acceptability of prolonged administration in the absence of infections.
Preclinical data from rats administered a single IV dose of 300 mg/kg [14C]tigecycline suggested that, within the first 8 hours post-dose, tigecycline concentrations in bones were approximately 4.1- to 44.1-fold higher than those in serum. Preclinical efficacy seemed to confirm distribution to bone. Yin et al. evaluated subcutaneous doses of tigecycline administered with and without rifampicin in a rabbit model of methicillin-resistant Staphylococcus aureus experimental osteomyelitis and observed that the rabbits treated with tigecycline showed a 90% CL of their infections, which was the same as those who were treated with vancomycin and oral rifampicin and similar to rabbits treated with vancomycin alone who had a 82% CL. Untreated controls showed only a 26% CL.
However, Rodvold et al. found that in subjects receiving a single pre-operative 100-mg dose of tigecycline and undergoing elective surgical procedures, though tigecycline tissue-to-serum exposure ratios in certain tissues (e.g., gall bladder [23×], colon [2×], and lungs [2×]) were high, the bone-to-serum exposure ratio was low (0.41×). The discrepancy between the preclinical data and the results from Rodvold et al. was thought to be because of the low extraction recovery of tigecycline from human bone, whereas bone standards were prepared with blank bone spiked with tigecycline neat solutions, which resulted in 100% extraction recovery. Therefore, the concentration of tigecycline in bone was significantly underestimated. This study was conducted using multiple doses and a more robust assay in a similar group of subjects and reported the bone-to-serum exposure ratio of 4.77.
Tigecycline's use in diabetic foot infection was investigated in a phase 3 clinical trial and did not meet noninferiority criteria when compared with ertapenem. Knowledge of the true bone penetration, as observed using a robust assay, and activity of tigecycline, as demonstrated in the rabbit study, is important when interpreting the results of the study because a limitation to treating diabetic foot infections and osteomyelitis is poor antibiotic tissue penetration and activity. The findings of the diabetic foot infection study cannot be explained due to the lack of penetration of tigecycline in bone. The analyses conducted to understand the outcome of the study may be found in Sabol et al.
The study design limits expansive extrapolation of the results beyond simply that tigecycline does penetrate bone, similar to the tetracycline class of antibiotics, in contrast to the results reported in the study by Rodvold. The impracticality of collecting more than one bone sample from individual subjects meant that intersubject variability is included in the result. There is a single estimate of AUCτ for bone and for serum and thus no estimate of variability in the estimate. There are few data in the literature that describe whether or not penetration of drugs into bone is altered in patients with osteoporosis, nor whether there is a difference in bone penetration between men and women, or between young and old subjects.
In summary, in this study of subjects without infections undergoing elective orthopedic procedures, multiple doses of tigecycline were shown to result in bone-to-serum ratios of approximately 4.77, showing penetration into bone.
Editorial support in the preparation of this manuscript was provided by Annie Jones of Upside Endeavors LLC and Paul Hassan and Charlotte Kenreigh of Engage Scientific Solutions and funded by Pfizer Inc.
Indranil Bhattacharya, Allena J. Ji, James P. Saunders, Ian Gourley, Annette Diehl, and Joan M. Korth-Bradley are employees or former employees of Pfizer Inc. Mark H. Gotfried is an investigator participating in clinical trials for Pfizer Inc. and has been a member of Pfizer Inc. speakers boards.
This study was funded by Wyeth Pharmaceuticals, Inc., which was acquired by Pfizer Inc. in October 2009.
- 1Wyeth Pharmaceuticals Inc. Tygacil® [tigecycline] for injection for intravenous use. Full prescribing information. http://labeling.pfizer.com/showlabeling.aspx?id=491. Accessed March 18, 2013.
- 71999.Tissue distribution of GAR-936, a broad spectrum antibiotic, in male rats (abstr. 413). American Society of Microbiology WD, USA., trans.: 39th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco, CA, USA;
- 11WIRB. Western Investigational Review Board. www.wirb.com Accessed June 1, 2009.
- 132009., , , , , Phase 3 study comparing tigecycline (tgc) and ertapenem (ert) in patients (pts) with diabetic foot infections (dfi) with and without osteomyelitis, (abstr. LB-42). 47th Infectious Disease Society of America. Philadelphia, PA. USA;
- 14Therapeutic options for diabetic foot infections: a review with an emphasis on tissue penetration characteristics. J Am Podiatr Med Assoc. 2010; 100:52–63.,