Comparative bioequivalence study between a novel matrix transdermal delivery system of fentanyl and a commercially available reservoir formulation

Abstract

Aim

To determine the pharmacokinetics, safety and performance of a novel matrix formulation of fentanyl.

Methods

Transdermal fentanyl was administered as the novel matrix and the Durogesic® reservoir formulations (24 subjects, 100 µg h⁻¹) in a randomized, fully replicate, four-way crossover study. Serum concentrations of fentanyl were assayed by LC/MS/MS. Pharmacokinetic parameters of fentanyl and performance (adherence and skin irritability) were evaluated.

Results

Test/reference ratio (90% confidence intervals) for AUC_0–t, AUC_inf and C_max were 105.5% (99.4, 112.0), 105.3% (99.3, 111.6) and 111.4% (100.4, 123.6), respectively. Adherence and skin irritability results of the two formulations were similar.

Conclusion

The two formulations are expected to result in similar efficacy for the management of severe pain.

Introduction

Fentanyl is indicated for the management of chronic pain in patients who require opioid analgesia for pain that cannot be managed by lesser means such as paracetamol–opioid combinations, nonsteroid analgesics, or as-needed dosing with short-acting opioids [1, 2]. Durogesic® is a transdermal system that provides continuous systemic delivery of fentanyl for 72 h at a nearly constant amount per unit of time. While there is important variation in dose delivered among patients, the nominal flux of the Durogesic® system (25, 50, 75 and 100 µg of fentanyl per hour) is sufficiently accurate to allow individual titration of dosage for a given patient. The risk of drug leakage and the ease of drug extraction appear to be a limitation of the reservoir technology, since various changes in product labelling were recently provided to include new safety information and warnings [3]. Furthermore, it was recently reported that cancer patients experienced significantly lower adhesion of the reservoir formulation compared with a matrix formulation of fentanyl [4] and that some patients may have discontinued their treatments due to adherence problems of the reservoir formulation [5].

Various prototypes of transdermal systems were developed by LTS (Andernach, Germany) in collaboration with the Nycomed Group (Roskilde, Denmark) based on the matrix technology. Silicone matrices consisting of fentanyl-containing dipropylene glycol droplets dispersed in a silicone matrix formulation were developed with and without a controlling membrane. Based on preliminary studies using these various prototypes, a modified membrane patch was developed in order to improve the release of drug. The chosen transdermal system is a system of fentanyl-containing dipropylene glycol droplets distributed in a silicone adhesive oil matrix and the patch contains a rate-controlling membrane (copolymer of ethylene and vinyl acetate) with 19% vinyl acetate. The special formulation of the matrix along with the controlling membrane improves the control of drug release to the extent that the drug load in the patch is reduced by almost 35% compared with other fentanyl prototypes based on the matrix technology. In addition, the membrane increases the thickness of the patch, which helps to stabilize it and makes it easy to handle. The skin adhesive layer is a silicone adhesive with known skin compatibility [6].

The objectives of the present study were to determine the pharmacokinetics, safety and performance of the novel matrix transdermal system of fentanyl relative to the commercialized reservoir formulation, after a single application in a four-way fully replicate study design in healthy subjects.

Methods

This was an open-label, randomized, fully replicate four-way crossover study of transdermal fentanyl administered as a single 72-h application of the novel matrix formulation of fentanyl (Formulation A: 100 µg h⁻¹, 11.0 mg fentanyl patch, total patch size =33.6 cm²; Nycomed Group) or as the commercially available reservoir formulation (Formulation B: Durogesic®, 100 µg h⁻¹, 10.0 mg fentanyl patch, reservoir size =40 cm², total patch size =57.0 cm²; Janssen-Cilag, Birkerod, Denmark) in 24 healthy male subjects. Prior to study start, all male volunteers gave informed consent according to local requirements. Subjects were medically healthy with clinically normal ECGs, had normal laboratory tests with no history of alcohol or drug abuse. The protocol and Informed Consent Form were submitted and accepted by the Institutional Review Board (IRB). The trial was conducted in accordance with The Declaration of Helsinki, Good Clinical Practice guidelines, relevant EU guidelines and any regulations for the protection of personal data.

All subjects received a single oral dose of 50 mg naltrexone (Revia®, 1 × 50 mg tablet; Duramed Pharmaceuticals, Cincinnati, OH, USA) prior to each patch application and 100 mg (2 × 50 mg tablet of Revia®) once daily for 5 days in each test period to prevent opioid adverse events. All subjects were randomized to receive the matrix (A) and reservoir (B) formulations twice in a four-way crossover study according to the following sequences: ABAB or BABA. Between each treatment, a wash-out period of 11 days was used. Vital signs were recorded before the first patch application and over 144 h. ECGs, readings of oxygen saturation and respiratory rate were obtained before patch applications and approximately every 4–6 h each day. All adverse events were evaluated by a physician. Blood samples (7 ml each) were collected in Vacutainers tubes with no additives before patch application (time 0) and at 6, 12, 24, 36, 48, 60, 72, 78, 84, 96, 108, 120 and 144 h after patch application. Adhesive properties of each patch were graded every 12 h (within approximately 10 min). The area of the patch was divided into squares of equal size, forming a grid. Assessment of skin irritation was performed every 24 h. Tolerability (skin irritation) of the two delivery systems was evaluated using the Skin Irritation Scoring Systems [7].

Fentanyl concentrations in serum samples were determined by a sensitive and specific method using liquid chromatography with mass spectrometry detection (LC/MS/MS) developed at MDS Pharma Services [8]. Human serum lots, free of significant interference, were pooled and used to prepare calibration standard and quality control (QC) samples. Blank samples were free of significant interference at the retention times of fentanyl and the internal standard (fentanyl-D₅). The retention times for fentanyl and fentanyl-D₅ were 1.93 min and 1.92 min, respectively [8]. Residual fentanyl patch content was determined using a validated high-performance liquid chromatography with ultraviolet detection based on a method previously published [9].

The following pharmacokinetic parameters were calculated using noncompartmental methods: the area under the curve from time 0 to the last measurable concentration (AUC_0–t) using the linear trapezoidal rule, the area under the curve extrapolated to infinity (AUC_inf = AUC_0–t + C_last/k_el, where C_last is the last measurable concentration), the maximum serum concentration (C_max), the time of maximum serum concentration (T_max), the apparent first-order terminal elimination rate constant (K_el) and the apparent first-order terminal elimination half-life (T_1/2, calculated as 0.693/K_el). Analysis of variance (anova) was performed on natural log-transformed parameters (AUC_0–t, AUC_inf, C_max and residual fentanyl patch content) and untransformed parameters (T_max, T_1/2 and patch adherence) to compare the least-squares mean (LSM) values between the two formulations (SAS® MIXED procedure; SAS Inc., Cary, NC, USA). Ratios of LSM between the two formulations (Matrix/Reservoir) were estimated using 90% confidence intervals (CI) for pharmacokinetic parameters and 95% CI for patch performance parameters. Based on the intrasubject CV% of AUC_0–t for fentanyl, a sample size of 48 subjects was initially proposed to meet the 80.0–125.0% criteria for the 90% CI, assuming an expected difference in AUC_0–t between the two treatments of 5% and a power of at least 80%. Since a four-way fully replicate study design was used, the sample size could be reduced by half while respecting the same assumptions on the expected difference in AUC_0–t and power of the analysis.

Results

A total of 24 healthy male subjects (all White) completed all four periods of the study. Mean age of subjects was 35 years (range 23–45 years), mean height was 176 cm (range 168–183 cm) and mean weight was 75.1 kg (range 63.5–87.2 kg). Of the 281 adverse events included in the safety analysis, all were nonserious and 139 were judged to be possibly or probably related to the drug (70 for the matrix and 69 for the reservoir formulation). Adverse events possibly or probably related to the drug were well known adverse reactions to transdermal fentanyl such as application site reactions, disturbance in attention, dizziness and somnolence. A total of seven subjects reported nausea (two for the matrix and five for the reservoir formulation).

Mean serum concentration profiles of fentanyl for the novel matrix and reservoir delivery systems are presented in Figure 1. Serum concentration values of fentanyl at 6 and 36 h for one particular subject were unrealistically high after one matrix patch (period 2) and one reservoir patch (period 3). These periods were removed from the statistical analysis since the unexpected concentrations were not observed in the other periods of either the matrix or the reservoir formulation. Summary statistics of pharmacokinetic parameters of fentanyl and patch performance are presented in Table 1. The ratio of LSM and 90% CI for natural log-transformed AUC_0–t, AUC_inf and C_max were within 80–125%, confirming that both formulations displayed similar rates and extents of fentanyl absorption. Residual amounts of fentanyl from the novel matrix formulation were higher than those measured for the reservoir formulation. This corresponded to the difference in the nominal amounts of fentanyl in the matrix formulation compared with the reservoir formulation (11.0 vs. 10.0 mg, respectively). Mean patch adherence and skin irritation scores for the matrix delivery system were not statistically different from those observed for the reservoir delivery system since 0 is included within the 95% CI.

Figure 1.

Novel Matrix Delivery System (•), Reservoir Delivery System (○)

Table 1.

Summary of pharmacokinetic parameters of fentanyl in serum and performance of the novel matrix and reservoir transdermal delivery system of fentanyl (100 µg h⁻¹) in healthy subjects (n = 24)

	Descriptive statistics
Parameters	Novel matrix delivery system	Reservoir delivery system	anovaRatio of LSM(90% CI)
Pharmacokinetics
AUC0–t (pg h ml−1)*	142 682 (25.0%)	135 877 (22.1%)	105.5 (99.4, 112.0)
AUCinf (pg h ml−1)*	145 126 (25.2%)	138 822 (22.6%)	105.3 (99.3, 111.6)
Cmax (pg ml−1)*	2 352 (34.8%)	2 117 (27.2%)	111.4 (100.4, 123.6)
Tmax (h)†	31.5 (35.7%)	33.1 (40.3%)	95.2 (84.0, 106.4)
T1/2 (h)†	20.3 (22.4%)	20.9 (27.3%)	97.7 (89.8, 105.5)
Performance
Residual amount (mg)*	2.08 (49.6%)	1.69 (36.9%)	123.9 (108.8, 141.1)
Adherence (%)†	63.5 (17.5%)	63.7 (21.3%)	−0.400 (−4.90, 4.10)‡
Irritation score†	0.895 (99.3%)	0.678 (72.5%)	0.178 (−0.060, 0.416)‡

AUC_0–t, Area under the concentration–time curve from time 0 to the last measurable concentration; AUC_inf, area under the concentration–time curve extrapolated to infinity; C_max, maximum plasma concentration; T_max, time of maximum measured plasma concentration; T_1/2, terminal elimination half-life.

^*Geometric mean (geometric CV%).

^†Arithmetic mean (arithmetic CV%).

^‡Difference in least-squares mean (LSM) (95% CI).

Discussion

The novel matrix and reservoir transdermal delivery systems of fentanyl were safe and equally well tolerated. A fully replicate crossover design was used to evaluate pharmacokinetic differences between the two treatments. The two delivery systems were considered bioequivalent since they resulted in similar rates and extents of exposure of fentanyl. Overall, the novel matrix formulation with rate-controlling membrane has a number of advantages over the conventional reservoir formulation. Since fentanyl is completely dissolved in a semisolid state in the matrix formulation, the risk of incidental drug leakage is reduced and extraction of drug for use and abuse is more complex. Furthermore, the matrix formulation is more convenient to use than the reservoir formulation because of its smaller size (33.6 vs. 57.0 cm², respectively). The novel matrix formulation displayed similar patch adherence and skin irritability profiles to the marketed formulation. Caution should be used for disposal of the novel matrix patch as well as for the reservoir patch [10].

The safety, pharmacokinetics and performance profiles of the novel matrix transdermal formulation of fentanyl were evaluated and compared with those of the commercialized reservoir formulation. Since the novel matrix delivery system resulted in similar safety and pharmacokinetic profiles to the reservoir formulation after a single application, the two formulations are expected to result in similar efficacy for the management of severe pain.