Pharmacokinetics of 2 Formulations of Transdermal Fentanyl in Cynomolgus Macaques (Macaca fascicularis)

Abstract

Fentanyl is a μ-opioid agonist that often is used as the analgesic component for balanced anesthesia in both human and veterinary patients. Minimal information has been published regarding appropriate dosing, and the pharmacokinetics of fentanyl are unknown in NHP. The pharmacokinetic properties of 2 transdermal fentanyl delivery methods, a solution (2.6 and 1.95 mg/kg) and a patch (25 µg/h), were determined when applied topically to the dorsal scapular area of cynomolgus macaques (Macaca fascicularis). Serum fentanyl concentrations were analyzed by using liquid chromatography–mass spectrometry. Compared with the patch, the transdermal fentanyl solution generated higher drug concentrations over longer time. Adverse reactions occurred in the macaques that received the transdermal fentanyl solution at 2.6 mg/kg. Both preparations showed significant interanimal variability in the maximal serum drug levels, time to achieve maximal fentanyl levels, elimination half-life, and AUC values. Both the maximal concentration and the time at which this concentration occurred were increased in macaques compared with most other species after application of the transdermal fentanyl patch and compared with dogs after application of the transdermal fentanyl solution. The pharmacokinetic properties of transdermal fentanyl in macaques are markedly different from those in other veterinary species and preclude its use as a long-acting analgesic drug in NHP.

An integral component of veterinary medical care is the prevention or alleviation of pain associated with procedural and surgical protocols. The administration of analgesics to manage pain in research animals is a cornerstone of refinement. The American College of Veterinary Anesthesiologists endorses a philosophy that promotes the prevention and alleviation of animal pain and suffering as an important and tenable therapeutic goal. The College's position statement on the treatment of pain in animals states “it is preferable to empirically administer analgesics preemptively if there is any question that a procedure will induce pain in an animal.”¹ The College's position is also in accordance with recommendations of the National Research Council as published in the Guide for the Care and Use of Laboratory Animals.¹⁵ TheGuide states that “Preemptive analgesia (the administration of preoperative and intraoperative analgesia) enhances intraoperative patient stability and optimizes postoperative care and wellbeing by reducing postoperative pain.”¹⁵ A commonly used method in the management of postoperative and postprocedural pain in veterinary medicine is systemic administration of opioid agonists (for example, fentanyl, morphine) or mixed agonists–antagonists (for example, buprenorphine). Extensive case reports, research studies, and veterinary pharmaceutical formularies support the clinical use of fentanyl in pain-management protocols for traditional small-animal veterinary species (for example, dogs, cats) and multiple large-animal veterinary species (for example, horses, sheep, goats, and swine).^{3,4,6,7,10-13,22} Currently little veterinary literature documents the pharmacokinetic properties of transdermal fentanyl, specifically the transdermal fentanyl solution, in NHP. It is difficult, if not impossible, to assess whether appropriate drug levels for analgesia are achieved in NHP species after the administration of doses extrapolated from domestic species.^20,23,25 Pharmacokinetic properties have been studied for other drugs used in pain management protocols in NHP, including buprenorphine, hydromorphone, and meloxicam.^{2,9,19, 21, 23} This point is supported by the cefovecin study conducted in cynomolgus macaques, in which the drug's pharmacokinetic properties were vastly different from those seen in dogs and cats and thus precluded its use as a long-acting antibiotic in NHP.²³ Principles of pain management are relatively easy to apply in clinically familiar species, such as dogs and cats, because the ranges of doses and drug combinations are well known. The application of these principles to other laboratory animal species, specifically NHP, is a matter of trial and error until sufficient scientific information is available to establish evidence-based guidelines, including the feasibility of various routes.

Fentanyl is a common opioid used as the analgesic component for balanced anesthesia in NHP. Many studies have been published describing the pharmacokinetic properties of fentanyl, effect of application site, and route of administration in multiple species, including dogs, cats, horses, goats, and sheep.^{3,4,6-13,16,17,25} The use of a transdermal fentanyl solution instead of the standard transdermal patch may prove to be beneficial, safer, and more advantageous compared with parenteral, oral, injectable, and patch-delivered opioids. The many disadvantages associated with the use of a transdermal patch for drug delivery include a slow onset of action, problems with maintaining appropriate patch contact with the skin, drug delivery variability, and ingestion of the patch. Reports of fatal overdose in 2 NHP after their ingestion of a transdermal fentanyl patch have driven the use of jacketing with this delivery route.⁵

Advantages of the transdermal fentanyl solution include ease of application, no risk of patch removal and its potential ingestion by NHP, no need for additional anesthetic events for removal of the patch, continuous delivery of fentanyl, and no need for frequent dosing of fentanyl. According to the manufacturer, the transdermal fentanyl solution (Recuvyra, Elanco, Greenfield, IN) is a novel, long-acting formulation recently approved for use in dogs. This product is the first transdermal fentanyl solution to be licensed for the control of postoperative pain associated with major orthopedic and soft tissue surgery. A single topical application provides active analgesic levels in 2 to 4 h after application and ongoing postoperative pain control for 4 d. It is delivered through a patented delivery transdermal technology, providing sustained plasma active concentration for 4 d.²⁴ Once applied to the skin, the transdermal solution dries rapidly, resulting in full absorption of fentanyl. However, in the event of an overdose, the transdermal fentanyl solution, unlike the patch, cannot be removed, and veterinarians must rely on the availability and use of a reversal agent, such as naloxone.

Naloxone, a pure opiate antagonist used in veterinary medicine almost exclusively for its opiate reversal effects. The exact mechanism for its activity is not understood, but it is believed that the drug acts as a competitive antagonist by binding to the μ-, κ-, and σ-opioid receptor sites. The drug has its highest affinity for the μ receptor. In humans, naloxone has a very rapid onset of action (usually 1 to 2 min), and the drug's activity usually persists for 45 to 90 min but potentially for as long as 3 h.²²

The current study aimed to determine the pharmacokinetic properties of transdermal fentanyl in cynomolgus macaques and compare differences between patch and solution formulations. The primary goal was to determine whether both formulations achieved quantifiable plasma concentrations and to assess how long these concentrations remained detectable. Our hypothesis was that the transdermal fentanyl solution would achieve concentrations faster and that lasted longer than those from the transdermal fentanyl patch. We postulated that the pharmacokinetic properties of transdermal fentanyl (patch or solution) in macaques would be similar to those observed in dogs. These data would facilitate the assessment of transdermal fentanyl for incorporation into pain management protocols for NHP. The pharmacokinetic properties of a transdermal fentanyl solution (doses, 2.6 and 1.95 mg/kg) and a transdermal fentanyl patch (25 µg/h) were determined when the formulations were applied topically to the dorsal scapular area of cynomolgus macaques. Secondary objectives of this study included 1) assessment of the transdermal fentanyl solution as an alternative to the transdermal fentanyl patch as a method of pain management in NHP and 2) establishment of an appropriate dosing strategy for the transdermal fentanyl solution and transdermal fentanyl patch. The establishment of an appropriate dosing strategy would assist in determining appropriate analgesic levels of transdermal fentanyl in NHP.

Materials and Methods

Animals.

Adult female cynomolgus macaques (Macaca fascicularis; n = 20; age, 6.1 ± 0.6 y; weight, 4.3 ± 1.4 kg) were used in this study (pilot experiment, n = 4; experiments 1 and 2, n = 8 each). Animals were identified with unique tattoo numbers. Each macaque was seronegative for Macacine herpesvirus 1, simian retrovirus type D, SIV, and simian T-lymphotropic leukemia virus. Macaques were tested twice annually for tuberculosis, and all remained negative. All procedures were conducted under an approved protocol from the United States Army Medical Research Institute of Infectious Disease IACUC. All macaques were housed in accordance with the Guide, Public Health Service Policy, and Animal Welfare Act and Regulations in an AAALAC-accredited facility. Macaques were fed a commercial diet (2050 Teklad Global 20% Protein Primate Diet, Harlan Laboratories, Frederick, MD). Water was provided without restriction. Fresh produce was offered daily as an edible enrichment item. Macaques also were provided with manipulable enrichment (balls, kongs, toys) daily. The environment was maintained at 69 to 75 °F (22.6 to 23.9 °C), at a relative humidity of 30 to 70%, and on a 12:12-h light:dark cycle. Macaques received baseline health assessments and behavioral assessments, conducted by a qualified veterinarian, 1 wk prior to study initiation. The baseline health assessment included a physical exam, CBC analysis, and plasma chemistry analysis. Behavioral assessments included screening for preexisting abnormal behavior (for example, flipping, spinning, self-injurious behavior) and were performed prior to jacket placement, during the jacket acclimation period, and daily throughout the course of the study. All macaques were weighed prior to each round of dosing to achieve accurate drug dosage.

Study design.

A randomized crossover design was implemented for experiment 1 by using 8 adult female cynomolgus macaques. Animals were randomly assigned to 1 of 2 groups (n = 4 each), one of which received the transdermal fentanyl patch first, and the other received the transdermal fentanyl solution (dose, 2.6 mg/kg) first. After a 14-d washout period (defined according to pilot study results), the application method was to have been switched between groups but was not pursued due to persistent fentanyl levels and the inability to achieve sufficient drug washout in the time available.

Another set of adult female cynomolgus macaques (n = 8) were used in experiment 2. For all animals, the transdermal fentanyl patch was applied first, followed by a 14-d washout period and then application of the transdermal fentanyl solution at a lower dose (that is, 1.95 mg/kg) than used in experiment 1. We chose the lower dose for experiment 2 in light of the adverse effects observed in the NHP that received the transdermal fentanyl solution at the published dose for dogs (2.6 mg/kg).

Pilot study.

As a reduction strategy, each macaque was intended to act as its own control, thus requiring establishment of an appropriate washout period between the applications of 2 formulations in each macaque. A pilot study was therefore conducted to determine an appropriate washout period and to validate the accuracy of the laboratory's analysis method. We selected 4 adult female cynomolgus macaques (age, 5.2 ± 0.3 y; weight, 6.0 ± 1.5 kg) for the pilot study, placed a transdermal fentanyl patch in the dorsal scapular area of 2 animals, and applied transdermal fentanyl solution (2.6 mg/kg) to the dorsal scapular area of the other 2 macaques. Blood samples were collected through femoral venipuncture at 0 h (predose), 48 h (2 d), and 240 h (10 d) after fentanyl application. Collection time points were established by using published pharmacokinetic data for both the transdermal fentanyl patch and solution in dogs. These time points provided a negative control (0 h), positive control (48 h), and a time point likely beyond when the drug was eliminated from the body (240 h). Transdermal fentanyl patches were removed 48 h after application. According to the results from this pilot study, a washout period of 14 d was deemed acceptable but proved unsuccessful in experiment 1, necessitating the need for experiment 2. Because plasma fentanyl was detectable at 360 and 720 h after the application of transdermal fentanyl, experiment 1 was not continued.

Central venous catheter (CVC) surgery.

All macaques in the study underwent surgical implantation of a jugular CVC (7 French, Bard Access Systems, Salt Lake City, UT) to facilitate blood collection. Prior to surgery, macaques were acclimated to custom primate jackets (Lomir Biomedical, Malone, NY) for a minimum of 6 d prior to CVC surgery and to the tether–swivel system (Lomir Biomedical) for a minimum of 3 d. All animals underwent a standard postoperative recovery period (10 d) after CVC surgery. To minimize any potential source of variability, no opioids (specifically buprenorphine) were used during the postoperative period, because these drugs can antagonize the fentanyl mode of action¹⁴; instead a NSAID was used for postoperative analgesia in all macaques. After the postoperative recovery period, macaques were sedated with tiletamine–zolazepam (5 mg/kg IM; 100 mg/mL, Zoetis, Florham Park, NJ) and connected to the tether–swivel system. Due to restrictions on blood volumes, a 2.5-mL sample was taken from the femoral vein on the day of CVC placement to serve as time 0 (predose). All subsequent blood collections were performed by using the CVC. In the event of catheter failure (that is, failure to withdraw blood from the catheter), macaques were sedated with ketamine (3 to 10 mg/kg IM; KetaVed 100 mg/mL, Vedco, St Joseph, MO) and dexmedetomidine (0.02 to 0.05 mg/kg IM; Dexdomitor 0.5 mg/mL, Zoetis) followed by reversal with atipamezole (0.2 to 0.5 mg/kg IM; Antisedan 5 mg/mL, Zoetis) to perform blood collection through femoral venipuncture.

Drugs and materials.

A transdermal fentanyl patch (25 µg/h; Actavis, Corium International, Grand Rapids, MI) was placed on the dorsal scapular area of 16 macaques. The dorsal scapular area, where the patch was applied, was clipped in a manner that provided an adequate margin (4 to 5 in.) around the patch application site. According to the manufacturer, the application of the transdermal fentanyl solution in an anatomic site other than the dorsal scapular area is contraindicated because absorption characteristics may differ between the sites. Therefore, the transdermal fentanyl patch was applied to the same area as for the transdermal fentanyl solution to eliminate the introduction of a potential location variable in the study; in addition, the same person applied both the patch and the solution in all animals. The operator held the patch in place for approximately 2 to 3 min after application. The edges of the patch were then secured by using tissue adhesive (Vetbond, 3M, St Paul, MN), and a transparent dressing (Tegaderm, 3M) was placed over the patch (Figure 1 A and B). In addition, all macaques wore jackets (Lomir) which provided an additional level of security and ensured that macaques could not remove the patch.

Figure 1.

Methods of application of transdermal fentanyl to cynomolgus macaques. (A) Transdermal fentanyl patch (25 µg/h). (B) Fentanyl patch (25 µg/h) secured with tissue glue and membrane dressing. (C) Patented applicator tip used to apply transdermal fentanyl solution. (D) Location (dorsal scapular area) and angle of application for transdermal fentanyl solution.

A transdermal fentanyl solution (50 mg/mL, Recuvyra, Elanco, Indianapolis, IN) was applied to the dorsal scapular area at a dose of 2.6 mg/kg in experiment 1 (4 macaques) and 1.95 mg/kg in experiment 2 (8 macaques). In experiment 1, a single 2.6-mg/kg dose of transdermal fentanyl solution was applied according to manufacturer guidelines (Figure 1 C and D). The area was allowed to dry for 5 min before the jacket was replaced on the macaque. An alternate application method was used in experiment 2 due to constraints encountered with the decreased dose in this experiment. Specifically, because the syringe included with the transdermal fentanyl solution was too large to allow for accurate dosing, and a generic smaller syringe could not be used with the patented applicator tip due to O-ring incompatibility. Therefore, we used a calibrated 250-µL pipette (Gilson, Middleton, WI) to apply the transdermal fentanyl solution on the dorsal scapular area. A calibrated analytical balance (Adventurer Pro, Ohaus, Pine Brooks, NJ) was used to verify accuracy of the dose prior to application. The transdermal fentanyl solution was applied to 2 separate areas of the dorsal scapular area to mimic the applicator tip used in experiment 1. The area was allowed to dry for 5 min after application, and then the macaque was jacketed.

Observation of macaques.

The macaques were observed for adverse reactions immediately after dosing and at each scheduled blood collection time point (approximately every 12 h). The animals were monitored for reported side effects of fentanyl, including sedation, vomiting, diarrhea, constipation, hypersalivation, and anorexia or poor appetite.

Sample collection.

In both groups, blood samples were collected into additive-free glass blood collection tubes (5 mL, Monoject, Tyco Healthcare Group, Mansfield, MA) at each of the 10 designated time points to determine fentanyl plasma concentrations. Blood collection time points included predose (0 h) and 4, 8, 12, 24, 48, 60, 72, 84, and 96 h after dosing. Blood volumes were based on the individual body weight of each macaque and were limited to 7% of the total blood volume during the study period. The transdermal fentanyl patch was removed after the final blood collection at 96 h (4 d). In experiment 1, additional samples were collected at 360 h (15 d) and 720 h (30 d) to determine whether washout of fentanyl had been achieved, because macaques given the solution formulation showed detectable levels of fentanyl at 96 h (4 d) after application.

Sample analysis.

All samples were centrifuged (Allegra X-15R Centrifuge, Beckman Coulter, Pasadena, CA) at 2000 × g for 15 min at 4 °C. Serum was transferred into a 2-mL Nalgene cryogenic vial (Sigma–Aldrich, St Louis, MO) and stored at –20 °C until shipment on dry ice for analysis (NMS Labs, Willow Grove, PA). Samples were analyzed by using liquid chromatography and tandem mass spectrometry. After the addition of internal standards, the biologic matrix was pH-adjusted and the analytes extracted by using a liquid–liquid extraction technique. Analysis was achieved by using HPLC separation with positive-ion electrospray tandem mass spectrometry for detection and quantitation. The lower limit of quantitation of the assay for fentanyl was 0.1 ng/mL. If the sample volume was insufficient for analysis, the value was as ‘QNS’ (quantity not sufficient). During experiment 2, one macaque that received the patch first had fentanyl levels that were more than 10 fold higher than those of other macaques; the outlier's samples were reanalyzed where possible, but insufficient sample volume precluded reanalysis for some time points.

Pharmacokinetic and statistical analyses.

The serum concentrations over time were summarized by mean, standard deviation, and select quantiles by using SAS version 9.4 (SAS Institute, Cary, NC). Simple pharmacokinetic parameters were derived from a noncompartmental model. T_max was defined as the latest sampled time point at which the maximal plasma concentration was observed, and C_max was defined as the concentration observed at T_max. The AUC was estimated by application of a trapezoid rule extending from time 0 to time 96 h after dose. No extrapolation was made past 96 h. The elimination half-life was defined here as (log₂/λ),where the rate constant λ is the slope of the least-squares line relating log plasma concentration to time over all time points greater than or equal to T_max. This half-life is almost surely an overestimate of the true elimination half-life, because it does not consider the effect of continued absorption past T_max, but it nonetheless serves to illustrate the nature of the drug-elimination phase.

Results

Experiment 1.

Pharmacokinetic parameters for the transdermal fentanyl patch and transdermal fentanyl solution (dose extrapolated from that used in dogs) are presented in Table 1.

Table 1.

Experiment 1: Pharmacokinetic parameters of transdermal fentanyl in cynomolgus macaques (n = 4 per group)

	Patch (25 μg/h)	Solution (2.6 mg/kg)	Difference (Patch – Solution)
AUC (ng×h/mL)	8.5 ± 2.0	773.4 ± 165.0	−764.9 ± 165.0
Cmax (ng/mL)	2.4 ± 0.6	159.8 ± 35.0	−157.4 ± 35.0
Elimination half-life (h)	45.2 ± 14.6	165.2 ± 72.7	−120.0 ± 74.2
Tmax (h)	36.0 ± 12.0	56.0 ± 18.1	−20.0 ± 21.7

Data are given as mean ± 1 SD.

Transdermal fentanyl patch.

Two of the macaques that received the patch formulation first had detectable levels 8 h after patch application, with peak levels occurring at 24 h (1 d). The remaining 2 NHP had detectable levels 12 h after patch application, with peak levels occurring at 72 and 24 h respectively. Serum levels in all 4 macaques were still detectable at 96 h (4 d). Serum fentanyl levels for these animals ranged from 0.14 to 3.4 ng/mL over the 96-h study period (Figure 2). The T_max for the transdermal fentanyl patch ranged from 24 to 72 h, and C_max ranged from 0.76 to 3.40 ng/mL. The median C_max for the patch group was 2.4 ng/mL, with a median T_max of 36 h. The median elimination half-life was 45.2 h, whereas the median AUC was 8.5 ng×h/mL for the patch group. The 2-wk washout period was sufficient in the macaques that received the transdermal fentanyl patch.

Figure 2.

Serum fentanyl concentrations after the application of the transdermal fentanyl patch (25 µg/h) to all macaques in experiments 1 and 2 combined (n = 12). Data are shown as the mean ± 1 SD (error bars) at each time point.

Transdermal fentanyl solution.

In experiment 1, 3 of the 4 macaques that received the transdermal fentanyl solution first (dose, 2.6 mg/kg) had adverse reactions to the drug, characterized in all 3 animals by severe respiratory depression, hypothermia, bradycardia, and unresponsiveness. All 3 macaques received epinephrine (1 mg/mL IV, IMS Limited, El Monte, CA), atropine (0.3 mg/mL IV, Henry Schein Animal Health, Dublin, OH), intravenous fluid therapy (Lactated Ringers Solution, Hospira, Lake Forest, IL), and thermal support. In addition, the macaques underwent pharmacologic reversal with 0.02 mg/kg IV naloxone (0.4 mg/mL, Hospira), and vital signs returned to normal (temperature, 98.6 to 103.1 °F; heart rate, 115 to 243 bpm; respiratory rate, 30 to 54 rpm)⁹ within 2 h of administering the reversal agent and providing supportive care. A single dose of naloxone was adequate to manage the side effects observed in the macaques that received the transdermal fentanyl solution.

All 4 macaques that received the transdermal fentanyl solution had detectable serum levels of fentanyl at 4 h after application. Serum fentanyl levels ranged from 17 to 230 ng/mL over the 96-h (4 d) study period (Figure 3). The T_max for transdermal fentanyl solution ranged from 8 to 84 h, with C_max ranging from 99 to 230 ng/mL. These parameters did not differ significantly between the 3 macaques that experienced adverse reactions compared with the macaque with no adverse reaction. The median C_max for the transdermal fentanyl solution group was 159.8 ng/mL, with median T_max of 56 h. The median elimination half-life was 165.2 h, whereas the median AUC was 773.4 ng×h/mL for the solution group. Fentanyl was detected at both 360 h (15 d) and 720 h (30 d) in the serum of all 4 macaques that received the solution formulation, thus prompting experiment 2.

Figure 3.

Serum fentanyl concentrations (n = 4) in cynomolgus macaques after application of the transdermal fentanyl solution (2.6 mg/kg). Data are shown as the mean ± 1 SD (error bars) at each time point.

Experiment 2.

Pharmacokinetic parameters for the transdermal fentanyl patch and transdermal fentanyl solution (at the reduced dose) are presented in Table 2.

Table 2.

Experiment 2: Pharmacokinetic parameters of transdermal fentanyl in cynomolgus macaques (n = 8 per group)

	Patch (25 μg/h)	Solution (1.95 mg/kg)	Difference (Patch – Solution)a
AUC96 h (ng·hr/mL)	2.8 ± 1.0 (8)a	646.8 ± 600.8 (8)	−643.9 ± 601.0
Cmax (ng/mL)	0.9 ± 0.2 (6)	177.1 ± 160.6 (8)	−176.2 ± 187.4
Elimination half-life (h)	47.6 ± 8.9 (5)	32.8 ± 16.3 (4)	+9.7 ± 18.5
Tmax (h)	44.0 ± 11.5 (6)	56.0 ± 17.6 (7)	−12.7 ± 26.5

Data are given as mean ± 1 SD (n)

^aEstimated by using a repeated-measures mixed model.

Transdermal fentanyl patch.

Two macaques had no detectable fentanyl levels throughout the entire 96-h (4 d) study period. Serum fentanyl levels were detectable in the 6 remaining macaques from 4 to 48 h. Serum fentanyl levels for all 8 animals that received the patch ranged from 0 to 1.6 ng/mL over the 96-h study period (Figure 2). The T_max for the patch ranged from 24 to 96 h, with C_max ranging from 0.17 to 1.60 ng/mL. The median C_max for the transdermal fentanyl patch group was 0.9 ng/mL, with median T_max of 44 h. The median elimination half-life was 47.6 h, whereas the median AUC was 2.8 ng×h/mL for the transdermal fentanyl patch group.

Transdermal fentanyl solution.

This group of macaques received a lower fentanyl dose (1.95 mg/kg) than did the solution group from experiment 1(2.6 mg/kg), due to the presence of adverse reactions in experiment 1. In experiment 2, 3 macaques had detectable levels of fentanyl at time of application (0 h); these levels were higher than those measured at the time of patch removal, which was 17 d prior. The remaining 5 macaques had detectable levels of fentanyl at 4 h after application of the solution. Serum fentanyl levels for all 8 animals that received the solution formulation ranged from 0.49 to 1300 ng/mL over the 96-h (4 d) study period (Figure 4). The T_max for the transdermal fentanyl solution ranged from 0 to 96 h, with C_max ranging from 1.2 to 1300 ng/mL. The median C_max for the transdermal fentanyl solution group was 177.1 ng/mL, with median T_max of 56 h. The median elimination half-life was 32.8 h, whereas the median AUC was 646.8 ng×h/mL for the transdermal fentanyl solution group.

Figure 4.

Serum fentanyl concentrations per time point in cynomolgus macaques after application of transdermal fentanyl solution (1.95 mg/kg). Data are shown as the mean ± 1 SD (error bars) at each time point. Note that the data at the top of the graph (different units) depict serum fentanyl concentrations for the individual macaque that yielded increased concentrations of 0.49 to 1300 ng/mL; these data were not included when calculating the mean or SD values.

Adverse effects.

In experiment 1, adverse effects were seen in 3 of the 4 macaques that received the transdermal fentanyl solution and developed within 30 min of application. The reaction was characterized by severe respiratory depression, unresponsiveness, hypothermia, and bradycardia. All 3 macaques received epinephrine, atropine, intravenous fluid therapy, thermal support, and pharmacologic reversal. In all 3 animals, vital signs returned to normal within 2 h of administering reversal and providing supportive care. The dose for experiment 2 was decreased by 25% (that is, 1.95 mg/kg), and no adverse reactions were seen with the decreased dose of transdermal fentanyl solution used in experiment 2. Application of the patch was tolerated well by all macaques, and none developed any adverse responses to the patch in either experiment. In addition, no adverse reactions to the decreased dose of the transdermal fentanyl solution (experiment 2) occurred.

Discussion

This study is the first to document the pharmacokinetic properties of transdermal fentanyl delivered by patch or topical solution in cynomolgus macaques. For both experiments, the time points we chose were based on previously published literature in other species and took into account limitations regarding the blood volume for the NHP species used in this study. This study reveals a marked difference in the pharmacokinetic properties of the transdermal fentanyl solution between individual macaques.

Why 2 of the macaques in the transdermal fentanyl patch group of experiment 2 had undetectable fentanyl levels during the entire 96-h (4 d) study period is unclear. All patches were applied by the same person, and all patches were undamaged and did not contain visible drug residue within the patch after removal. In addition the transdermal dressing applied over the patch was undamaged and secure at removal time, and the skin of both macaques was grossly normal after removal of the patches. All transdermal fentanyl patches used on all macaques in both experiments were from the same lot and manufacturer. These findings bring into question whether levels observed with the transdermal fentanyl patch are sufficiently consistent to be an effective component of a pain management protocol for NHP.

Significant interanimal variability was observed among all macaques in both experiments when time to achieve fentanyl levels, maximal serum levels, elimination half -life, and AUC values were assessed. A variety of potential factors might account for these findings in cynomolgus macaques. After the application of a fentanyl patch, the underlying skin absorbs the drug, and a depot of fentanyl concentrates in the upper skin layers, where it then becomes available to the systemic circulation. The pharmacokinetic properties were differentiated by a more rapid initial absorption of fentanyl and higher plasma levels for the solution formulation compared with the patch, such that a depot effect might be responsible for the variability observed in the fentanyl levels. In addition, the rate of absorption is dependent on a number of factors, including body temperature, skin type, skin thickness, hydration of the stratum corneum, amount of body fat, and location of patch placement. These factors might also be potential contributors to the variability observed in the study. Furthermore, fentanyl is metabolized by the liver, with the primary route of excretion for the metabolites occurring through the urine¹⁷. The accumulates in the skeletal muscle and fat and is released slowly into the blood. More rapid release into the blood, with subsequent higher fentanyl blood levels might be seen in an animal that has a low body-fat percentage. The rate-limiting membrane of the transdermal fentanyl system in the patch controls the rate of drug delivery to the surface of the skin. Fentanyl moves from the patch reservoir to the patch–skin interface, where it passes through the stratum corneum, epidermis, and dermis. Eventually, fentanyl is taken up by the cutaneous circulation. An interaction between fentanyl and the skin can occur at any step in this process and might contribute to the variability of absorption.²⁶

The mean maximal concentration, T_max, AUC, and elimination half-life measured in NHP that received transdermal fentanyl solution (2.6 mg/kg) were markedly different from those of dogs that received the same dose. Compared with the published study of transdermal fentanyl solution in dogs, ^11,12 all pharmacokinetic parameters in macaques at either dose (2.6 and 1.95 mg/kg) were markedly different from those of dogs, even those that received a higher dose (5.2 mg/kg) of transdermal fentanyl solution. All macaques had increased (28 to 300 fold higher) serum levels of fentanyl compared with levels measured in dogs (mean, 2.6 ng/mL). Fentanyl levels were still detectable in all macaques that received naloxone throughout the study period 96 h (4 d), and these levels waxed and waned throughout the study.

Most drugs used in NHP have not been studied in these species and consequently are used off-label; only the pharmacokinetic properties and drug dosing regimens for the drugs buprenorphine, hydromorphone, and meloxicam have been studied in NHP. ^2,16,21,23The dosing regimen for the transdermal fentanyl patch was based on the postoperative pain management protocol used at our institute and was extrapolated from dosing regimens used in dogs. The findings in the current study highlight the difficulty in establishing appropriate pain management protocols in NHP by using dosing regimens established in other species. In this study, the published canine dosing regimen (2.6 mg/kg) for the transdermal fentanyl solution was determined to be unsuitable for use in NHP. The decreased dose (1.95 mg/kg) used for macaques in experiment 2 did not induce the adverse reactions to the dose (2.6 mg/kg) used in experiment 1. Despite the decreased dose, blood fentanyl levels were still significantly higher in macaques than dogs and significant interanimal variability still occurred. The pronounced variability in fentanyl levels and the inability to maintain steady-state concentrations in groups that received the solution formulation emphasize the need for additional studies to determine an appropriate dose that achieves steady-state levels in NHP. The large intersubject variation in serum fentanyl concentrations observed in macaques given the transdermal fentanyl solution indicates unpredictable drug absorption of transdermal fentanyl when applied according to manufacturer's recommendation. Once an appropriate dose is found, further studies are warranted to determine therapeutic levels that would provide adequate levels of analgesia in NHP. The tail-withdrawal latency assay has been used in rhesus macaques to evaluate the thermal antinociception effect of fentanyl. One study showed dose-dependent antinociception when fentanyl was administered to rhesus macaques at doses ranging from 0.01 to 0.032 mg/kg SC.⁴ Another study showed that concentrations of fentanyl ranging from 3 to 40 mg/mL IV as sufficient to produce analgesia in rhesus macaques according to the tail-withdrawal latency assay.²² This assay is a useful tool to further characterize fentanyl-induced analgesia in future NHP studies.^4,27

Because the published canine dose (2.6 mg/kg) resulted in adverse reactions to fentanyl in macaques, it was necessary to use a lower dose (1.95 mg/kg). This dose could not be delivered reliably by using the recommended syringe and applicator tip and instead was applied by using an alternative method, which yielded detectable drug concentrations. Using the transdermal fentanyl solution in NHP is infeasible if an alternative method of application is unavailable, such as a syringe capable of providing smaller volume than that used for dogs. Another alternative is to use a more dilute drug solution in NHP. In addition, it would be helpful to assess alternate anatomic sites for the application of the transdermal fentanyl solution and patch in NHP. Differences in the absorption characteristics of different application sites have been documented in humans, dogs, and horses.^{10, 13,18} The variability in pharmacokinetic data may be due to the lipophilic nature of fentanyl and inconsistent adipose tissue of individual NHP, which might alter the absorption and metabolism of the drug. The current study shows that the pharmacokinetic properties of transdermal fentanyl, specifically the solution formulation, in macaques differ markedly from those observed in dogs.^8-11

The current study shows that dosing regimens can differ markedly between species, and dosing regimens for NHP cannot be extrapolated reliably from data in other species. According to this study, the use of transdermal fentanyl (patch or solution) as a component of pain management protocols in NHP is not recommended, due to the inconsistent and unreliable pharmacokinetic properties. Further studies are warranted before transdermal delivery of fentanyl can be recommended for postoperative treatment of pain in NHP.