WO2003097008A2 - Transdermal delivery device for the administration of fentanyl - Google Patents

Abstract

A simple, non-irritant and chemically stable transdermal delivery device for the administration of fentanyl comprises a pressure sensitive acrylic adhesive matrix formed by a non-functional copolymer containing 2-5 % by weight of fentanyl, and lauryl alcohol as a permeation enhancer.

Description

TRANSDERMAL DELIVERY DEVICE FOR THE ADMINISTRATION OF FENTANYL

Field of the Invention This invention relates to transdermal administration devices, and more particularly, to transdermal administration devices for the systemic administration of fentanyl.

Background of the Invention The effects of opium on the human mind had been known since ancient times, although it was Sydeham, in the seventeenth century who recognised the valuable medical use of opium extracts to treat painful conditions and alleviate suffering ofthe sick. The main active principle in opium is the alkaloid morphine. In the late 1970s the presence of morphine-like compounds in animal and human brains was described. These substances are peptides and belong to three families: enkephalins, endorphins and dynorphins. Nowadays we know that there exist several cellular receptors that mediate the psychotomimetics and pain relieving effects of opioids and the physiological role ofthe mentioned endogenous peptides. Among these receptors the so-called Mu receptor mediates most ofthe medical useful effects of opioids.

The effectiveness of morphine to alleviate pain drove industry interest to obtain synthetic opioids that could present some advantage over the natural alkaloid. Levorphanol, heroin, oxymorphone, meperidine, buprenorphine, pentazocine and fentanyl can be named as examples. From this group fentanyl can be selected as one ofthe most potent agents with a relative short half-life of elimination and a rapid onset of action.

Fentanyl is chemically different to the natural morphine. Its structural formula is related to the phenylpiperidines and it has a simpler structure than that of morphine which belongs to the morphinan group of molecules. Its potency is almost 100 times greater than that of morphine and produces shorter respiratory depression than another synthetic analogue, meperidine. The use of fentanyl has gained wide acceptance in its use by anaesthesiologists because it can be safely combined with neuroleptic agents such as droperidol. This combination is used as an intravenous anaesthetic without potentiation ofthe side effects. The low dose of this compound makes it suitable for its use in a transdermal device and the medical profession is currently using such a device for the management of severe pain. The pharmacokinetics of this compound is widely known and the therapeutic plasma levels usually accepted as safe and effective are within the range of 1 to 3 nanograms per ml as quoted in the "Goodmand & Gillman's The Pharmacological Basis of Therapeutics" ninth edition, MacGraw Hill 1996, Table AII-1, page 1742. Transdermal administration of fentanyl offers advantages over its administration by injection or oral administration. The safety margin of opioids is narrow. Doses in the therapeutic range can produce nausea and heavy vomiting whereas small increases in dosage will produce a decline in the respiratory function. Amounts near the maximum therapeutic dose will slow the respiratory rate whereas higher doses will further decrease the sensitivity ofthe medullar centres that control respiration and can produce respiratory arrest and finally death. Although opioids antagonists such as naloxone or naltrexone can reverse these toxic effects, if the drug has been injected or administered per os, absorption will continue maintaining dangerous levels ofthe opioids till the elimination rate surpass the absorption rate and the drug is eliminated through metabolism. In contrast, with a transdermal device the drug to be absorbed is maintained outside the body and if undesirable side effects occur, removal ofthe system will terminate the absorption phase and the decline of plasma levels will depend only on the metabolism rate. Thus, the transdermal route is more convenient in some cases than the more classical routes of administration for opioids. Gels or ointments, on the other hand are devoid ofthe advantages of transdermal devices since dosification is less precise (absorption will be extremely dependent on the size ofthe application area and the thickness ofthe layer applied). Moreover in the case that it could be helpful to remove the applied dosage either rubbing or alcohol application will rather increase the rate of absorption.

Nowadays, there are different types of transdermal devices. These devices could be mainly grouped in two broad categories: reservoir-type devices and matrix- type devices. In the literature other types of devices have been described, but all of them could be considered as extension of one of these types or as a combination thereof.

The transdermal reservoir-type devices are characterized by a reservoir that contains the active pharmaceutical ingredient (API), a rate controlling membrane, a contact adhesive, a protective liner and a backing layer. From the reservoir, the API diffuses through the controlling membrane to the site of absorption. The controlling membrane may be microporous or continuous. The main advantage of this type of device is that the rate of drug delivery is maintained practically constant for long periods of time. Nevertheless, these devices are usually bulky, they have a total surface area that is bigger than the active surface and, besides, they have the disadvantage that damage ofthe controlling membrane or the reservoir sachet could produce a release higher than desired.

On the other hand, the matrix-type transdermal delivery devices comprise, generally, a nonpermeable backing liner, a polymeric adhesive matrix in which the pharmaceutical ingredient or ingredients are dissolved or dispersed and a release liner. They have a total surface area that is the same as that ofthe active surface. Generally, this type of device has had greater acceptance by patients than the reservoir-type patches.

The current trend in the design of transdermal delivery devices is directed to matrix-type devices. This is not only because the production costs are lower, but because it is possible to obtain devices with greater versatility than the reservoir-type ones.

One disadvantage ofthe matrix-type device is that, with some active substances, it is difficult to maintain a constant dose for extended periods of time. Generally, in this type of device, the delivery rate decreases during the wearing period as a consequence ofthe depletion ofthe API in the matrix.

The addition of polymeric layers or a semipermeable film acting as a controlling rate membrane for the drug has not been completely successful because they are usually less comfortable the use ofthe device by the patient, mainly because their mechanical properties become worse with the mentioned addition. Another problem to be solved in this type of device is related to physical stability because, generally, to assure a constant delivery, the active substance needs to be present in a saturation or supersaturated concentration. In the particular case of fentanyl, which is an expensive drug, high loading ofthe matrix yields in an expensive final product. Besides, it is important for the comfort ofthe user that the size ofthe device should be as small as possible.

Finally, it is very important that the transdermal delivery device has enough adhesiveness to be able to remain on the application site assuring the suitable drug delivery during the necessary period of time, while having, at the same time, a painless removal, which is not easy to obtain.

For the transdermal delivery of fentanyl there is only one device in the market. It is a reservoir-type device sold under the trademark Duragesic® and is described in US 4588580.

Several documents are related to the development of fentanyl matrix-type devices. In one of them, three chemical classes of polymers: polyisobutylene, acrylate and silicone were tested (ref. 1). The authors concluded that the silicone with 2% drug loading provided the highest drug flux and that this adhesive appeared to be a very promising candidate for designing a transdermal device. Nevertheless, this study is quite incomplete because, as we have found, not only the type of polymer is responsible for the release profile but the presence of different functional groups modify the release rate. Our experimental results have showed significant differences in the fentanyl delivery rate with different acrylic adhesives having different functional groups. In the mentioned study, the only acrylic adhesive used contained the -OH functional group.

Patents of Cygnus (see for example US 4906463, US 4911916, US 4915950 and US 5186939) disclose fentanyl transdermal devices with one of more adhesive layers in which rate controlling membranes are avoided. Nevertheless, the results obtained in the clinical trials described in ref. 2 and 3 led the authors to conclude that the Cygnus transdermal fentanyl device shows great variability in the rate of fentanyl absorption, resulting in highly variable plasma fentanyl concentrations. Since fentanyl has a narrow therapeutic range, this variability can cause in some subjects respiratory depression and other undesirable and dangerous side effects.

US 6174545 discloses a heat equilibration process for the manufacture of a multilaminated drug delivery device that includes fentanyl. The described end products have drug in excess of saturation in layers other than the drug reservoir, such as the contact adhesive. The disadvantages ofthe described devices are the amount of steps in the manufacture procedure because ofthe presence of different layers and for the heat equilibration process, and also the possibility of physical instability ofthe devices in excess of saturation.

Several documents describe the use of lauryl alcohol as enhancer for the delivery and permeation of narcotics. Nevertheless, none of them disclose systems that could be adapted to a matrix type transdermal device for the administration of fentanyl according to the present invention. Additionally, none of them shows data of a device capable of delivering a therapeutic dose of fentanyl evaluated by human plasma levels with low matrix concentration ofthe active substance. US 5932227 discloses a hydro-alcoholic percutaneously administrable base composition in which the absorption promoter comprises lauryl alcohol or myristyl alcohol. The type of composition described is suitable for lotions, creams, gels, nose drops or reservoir- type patches, but would be incompatible with a monolithic matrix type patch. WO

0187276 refers to an hydrogel composition for transdermal drug delivery comprising a hydrophilic polymer base. Even though lauryl alcohol is mentioned as one ofthe useful skin penetration enhancers, there is no example of its specific combination with fentanyl. In US 5866157, a matrix type patch formulation which comprises an adhesive layer containing a physiologically active substance, an organic acid, a hydrophobic high molecular material, a tackifying resin, a plasticizer and an absorption enhancer, in which the active drug forms ion-pair with the organic acid, is disclosed. In the described device lauryl alcohol is one ofthe mentioned enhancers. Although fentanyl is mentioned as one ofthe suitable active drugs, the formation of the ion-pair described would not be possible using fentanyl base because it is not stable in acidic media. Neither an example containing this drug in its base form, nor the specific combination of any form ofthe fentanyl molecule with lauryl alcohol are present in the examples. Finally, in US 4626539, liquid or gelled compositions for topical administration of opioids are described. Several enhancers are mentioned, most of which were found incompatible (or chemically unstable) with fentanyl in our experiments. Moreover, there are no examples including fentanyl and there are no descriptions of matrix type devices disclosed in this document. Finally, WO 02/26217 and WO 02/074286 disclose monolithic, subsaturated patches or compositions that can be incorporated in a patch for transdermal administration of fentanyl. In the first, the content of fentanyl in the device is very high, 8-30%, which makes the finished product very expensive. The amount of enhancer used is also very high, 14.9-35 %, which is not recommendable for the irritation ofthe skin and the performance ofthe adhesive properties. Moreover, the thickness ofthe devices described makes necessary, in some cases, a manufacturing process of at least two coating/drying steps to obtain the desired device. In the second, the concentration of fentanyl in all examples is also high (2.8% or higher in the adhesive solution that yields about 5.5% or more by weight of total content excluding adhesive solvents). Also, in this document, the adhesives used in all examples have amonomeric composition identical to one ofthe adhesives described below (GELNA 788), with an -OH functionality. We have found that the chemical stability of fentanyl is diminished when adhesives with -OH functionality are used.

Summary of the Invention

We have now found that fentanyl can be administered transdermally using a simple transdermal device that overcomes or reduces the disadvantages mentioned above. The present invention has as its object to provide a simple, non-irritant and chemically stable transdermal delivery device for the systemic administration of fentanyl.

As noted in the discussion ofthe background ofthe invention, the state of the art indicates that to manufacture a medically useful transdermal device for the delivery of fentanyl, complex formulation and cumbersome procedures have been proposed. Surprisingly, we can obtain by the present invention a transdermal device with a simple formulation, low percentage of drug and few manufacturing steps that provides doses of fentanyl apt for detriment of pain.

According to the invention there is provided a matrix-type transdermal delivery device having a) a pressure sensitive acrylic adhesive matrix formed by one or more nonfunctional copolymers in which the fentanyl is incorporated, b) a backing layer, and c) a release liner wherein said matrix includes lauryl alcohol as enhancer of percutaneous permeation. The device is ofthe matrix type, i.e. having a body of adhesive which contains the fentanyl to be delivered and is applied directly to the skin in use.

We have found that for different adhesive matrixes containing fentanyl and formulated with different acrylic polymers, the delivery ofthe drug varies according to the functional group or groups present in the adhesive polymer. Moreover, for -COOH functional groups, the delivery of fentanyl from the adhesive matrix is too low to provide an adequate permeation rate.

Besides, the chemical stability of fentanyl in different acrylic adhesives proved to be better when no -OH groups are present in the copolymer.

Surprisingly, we have also found that the chemical stability of mixtures of fentanyl with different permeation enhancers was optimal with the saturated fatty alcohol lauryl alcohol.

Experiments in vitro (permeation through human cadaveric skin) as well as in vivo (plasma levels in healthy volunteers) showed that the simple device ofthe invention with a fentanyl load of 2-5% and an enhancer load lower than 15% has a performance comparable to Duragesic®. Preferably, the load concentration for fentanyl is between 2-4.5%, more preferably 3.5-4.5% and for lauryl alcohol is between 5-12%, more preferably 5-10 %. The device ofthe present invention also has good adhesive properties and it is non-irritant.

The specific combination of a particular pressure sensitive acrylic adhesive without functional groups and lauryl alcohol as permeation enhancer with fentanyl can provide a device for the transdermal administration of fentanyl that delivers a sustained and controlled amount of fentanyl capable to offer suitable therapeutic plasma levels ofthe active drug for at least three days, using lower loading of fentanyl than devices previously described. Additionally, the device with this specific combination has an excellent chemical and physical stability as well as good adhesive performance without irritation ofthe application site.

The performance mentioned can be achieved with a final matrix thickness less than 100 μm. Preferably, it should be less than 80 μm. This thickness allows manufacture ofthe device through a simple coating/drying process.

Brief Description of the Drawings The above and other objects and advantages of our invention will be readily apparent from the following description with reference to the accompanying drawings, wherein:

Figure 1 is a schematic view of a transdermal delivery device according the present invention. Figure 2 shows the dissolution profile of fentanyl from several adhesive matrixes in Example 1.

Figure 3 shows the in vitro permeation of fentanyl from different adhesive matrixes in Example 1.

Figure 4 shows the in vitro permeation of fentanyl from matrixes with lauryl alcohol or 1-decanol as permeation enhancer in Example 4.

Figure 5 shows the in vitro permeation of fentanyl from matrixes containing lauryl alcohol and methyl laurate as permeation enhancers in Example 5.

Figure 6 shows the in vitro permeation of fentanyl from matrixes with different drug load in Example 6. Figure 7 shows the in vitro permeation of fentanyl from matrixes with lauryl alcohol and/or myristyl alcohol as permeation enhancers in Example 7. Figure 8 shows fentanyl plasma levels in rabbits in Example 9. Figure 9 shows fentanyl plasma levels in healthy volunteers in Example 10.

Detailed Description of the Invention

In the transdermal device described in the present invention fentanyl is the active drug, which is dissolved and incorporated homogeneously in the selected adhesive polymeric matrix. Adhesive matrixes were prepared with different content of fentanyl using acrylic adhesives as well as polyisobutylene. We have found that its physical stability is greater in acrylic adhesives than in polyisobutylene. Within the acrylic adhesives, copolymers with -OH functional groups, -COOH functional groups and nonfunctional groups were tested. Fentanyl showed good physical stability in all of them, but we have found that the chemical stability is optimal if copolymers with -OH functional groups are avoided. Regarding the drug delivery rate, dissolution experiments were performed with the mentioned acrylic adhesives, showing that the drug delivery rate from acrylic copolymer with -COOH functional group was too low to provide an adequate permeation rate.

We have found that the best pressure sensitive adhesive for a transdermal delivery device of fentanyl, taking into account both physical and chemical stability as well as drug delivery rate, is an acrylic adhesive substantially without functional groups in the polymeric molecules present.

Particularly, acrylic adhesives with -OH and -COOH functional groups are excluded from use in our invention for the reasons explained above. Also, amine (primary, secondary and tertiary), keto and silanol groups should preferably be excluded because they can, potentially, interact with the active drug. The presence of a small amount of acrylic pressure sensitive adhesives containing a functional group can be allowed only if the total number of monomeric units carrying a functional group is less than 1% by weight of total content of adhesive, excluding adhesive solvents, in the final product.

There is to be understood by "acrylic pressure sensitive adhesives without functional groups" an acrylic based polymer which has no or substantially no functional free reactive moieties present in its polymeric chain. These non-functional acrylate copolymer adhesives are generally copolymers consisting of or containing acrylic acid esters or methacrylic acid esters as monomers. Typical alkyl acrylate and alkyl methacrylate monomers in such adhesives are those having 4 to 12 carbon atoms in the alkyl group. Such a monomer or monomers may be copolymerised with another monomer or monomers which do ^"not have functional groups, such as vinyl acetate (for example DuroTak 87-4098 obtainable from National Starch and Chemical Co.). A preferred matrix is a copolymer of 2-ethylhexyl acrylate and vinyl acetate. A mixture ofthe non-functional copolymers may be employed to form the matrix. Surprisingly we have found that, even though the chemical stability of fentanyl in acrylic adhesive matrixes is optimal in the absence of -OH functional groups, the permeation enhancers that show the least chemical interaction with the drug are saturated fatty alcohols. Particularly 1-decanol, lauryl alcohol and myristyl alcohol are the least aggressive to the pharmaceutical moiety. Fentanyl showed better chemical stability in mixtures with those saturated fatty alcohols than in mixtures with compounds having acidic groups, oleyl groups, polyols and unsaturated fatty acid esters. So, enhancers such as fatty acids, fatty acid esters and glyceryl fatty acid esters should be avoided.

Within the mentioned saturated fatty alcohols, lauryl alcohol has the best performance in drug delivery rates and adhesive properties. Also, the use of lauryl alcohol allows us to obtain a device in which the drug load is between 2-5% and the enhancer load is between 5-15% with a performance similar to Duragesic®. This low concentration for both components yields a device that is non-irritant, has good adhesive properties and can be cheaper than other proposed matrix type fentanyl devices. Lauryl alcohol is thus preferably the sole permeation enhancer present.

However, other permeation enhancer or enhancers may be present, e.g. up to 10% by weight of total permeation enhancers.

As is conventional, antioxidants and stabilising agents as butylhydroxytoluene, butylhydroxyanisol, tocopherols, lecitin, polyvinyl pyrrolidone, gum guar or carboxymethylcellulose, carbomers, ethylcelluloses and other well known excipients may be included.

The addition of fillers, such as bentonite, titanium dioxide, talc or silicon dioxide may be useful, but the concentration of filler is preferably lower than 7.5%. A schematic view of a matrix-type transdermal delivery device for the administration of fentanyl that contains a single adhesive layer embodying the present invention is shown in Figure 1. The device comprises a backing layer 1 that acts as a protective layer for the adhesive matrix 2 composed of a pressure sensitive adhesive containing the permeation enhancer and the active drug dissolved in it and a release liner 3 that is detached before the application ofthe device onto the skin. Materials suitable for the backing layer are cellulose xanthate film, polyvinylidene chloride film, polyvinyl chloride, polyethylene, polypropylene, polyurethane, polyesters such as polyethylene terephthalate including binary structures such as aluminium-polyethylene coatings, etc.

For the release liner, any ofthe above mentioned materials can be used, preferably polyesters, such as polyethylene terephthalate, etc. covered with a silicone to prevent sticking ofthe adhesive.

The devices of this invention are preferably applied on those regions ofthe body far from mucosa or excessively keratinizated sites (such as palm of hand or sole ofthe foot).

This invention is further illustrated by the following examples. These examples should not be taken as limiting the scope of this invention in any manner.

EXAMPLE 1: Fentanyl performance in different adhesive matrixes.

The performance of fentanyl in different adhesive matrixes was tested evaluating physical and chemical stability as well as dissolution and permeation profiles. Eleven different adhesive matrixes comprising fentanyl were prepared according to the following procedure:

1. Add the previously calculated amount of fentanyl base to adhesive mixture. Stir during 30 minutes at 750 RPM using impeller type ring.

2. Perform the coating of each coating mixture, using a standard procedure, setting the machine parameters as follows:

3. The coating is performed onto siliconised polyester (1-5 PESTR Clear 6200 (P2)), supplied by DCP Lohja, which acts as release liner. The dry adhesive matrix is laminated onto a polyester/EVA laminate (Scotchpack 1012), supplied by 3M, as backing layer.

4. The bulk laminate obtained is die cut to 4-cm length pieces to obtain an active surface of 20 cm² (approximately). 5. The resulting patches are stored in aluminium pouches under the following conditions:

R.H. = relative humidity.

The weight percent of fentanyl, identity ofthe adhesive and its functionality are given in table 1.

Table 1: Experimental lots of fentanyl in different adhesive matrixes

DT 87-4098 is available from National Starch and is a copolymer comprising 2-ethylhexyl acrylate and vinyl acetate.

DT 87-2353 is available from National Starch and is an acrylic copolymer. Gelva 788 is available from Solutia Inc. and is a copolymer comprising glycidyl methacrylate, 2-ethylhexyl acrylate, vinyl acetate and 2- hydroxyethylacrylate.

L100:4H (45:55) is a mixture of polyisobutylene L-100 (mol. weight 1,000,000), available from Exxon and polyisobutylene 4H (mol. weight 40,000), available from Rit-Chem Co. in a ratio of 45:55.

Part a: Physical stability

Physical stability ofthe prepared formulations was tested in three different conditions. The obtained results are listed on tables 2, 3 and 4 in which OK means "No crystals observed" and FAIL or F means "Crystals observed".

Table 2: Storage Condition:55 °C / Ambient Humidity

Table 3: Storage Condition:40 °C / 75 % R.H.

Fentanyl is physically stable in all the acrylic adhesives matrixes tested, while it is unstable in matrixes made of polyisobutylene for a load of fentanyl higher than 2%. Part b: Chemical stability

Chemical stability of lots with different adhesive materials were tested in two different conditions. The obtained results are listed on tables 5 and 6.

Table 5: Storage Condition: 55 °C / Ambient Humidity

Table 6: Storage Condition: 40°C / 75% R.H.

Optimal chemical stability for fentanyl is obtained in acrylic adhesives without -OH functional group.

Part c: Dissolution profile of fentanyl from different adhesive matrixes.

Efflux of fentanyl from different matrixes that support the maximal concentration of fentanyl in solution (i.e., without crystallisation) and the reference product Duragesic® was compared: Table 7: Matrixes used in efflux experiments

The experiment was performed using a "Hanson Research SR8" Dissolutor according to the following procedure: 1. Arrange the dissolutor devices placing one patch per vessel (8 vessels).

2. Set the following parameters: Temperature: 32°C Paddle speed: 100 RPM Vessel average volume: 500 ml. 3. Use as dissolution media phosphate buffer pH 6.

4. Obtain samples at the following times:

1, 4, 8, 12, 24, 32, 48, 56, 72, 80, 150, and 170 h.

5. Perform 4 assays for each lot (except for Duragesic® when only one was tested).

6. Analyse the fentanyl content under the following HPLC method: Mobile phase: ACN: Phosphate Buffer pH=6: THF (25:60:15)

Stationary phase: RP18 column 125 mm x 4 mm. 5 μ Particles.

Flow: 1 ml min

Temperature: 40°C

Wavelength: 225 nrn. Injection Volume: lOOμl

Results:

As is seen in figure 2, non-functional as well as -OH acrylic adhesive matrixes show dissolution profiles for fentanyl that are similar to the reference product Duragesic®. On the other hand, the -COOH acrylic adhesive matrix and the highest fentanyl loading that is physically stable in polyisobutylene matrixes (2%) show a low delivery rate. Part d: Comparative permeation profiles from different adhesive matrixes

Comparative permeation profiles from different adhesive matrixes were performed to establish the differences among the permeation profiles through cadaveric human skin obtained from fentanyl-containing polymeric matrixes formulated with different pressure-sensitive adhesives and the reference product Duragesic®.

For this permeation experiment, the same matrixes used in the dissolution experiment (table 7, part c) were selected. The materials and equipment used, and experimental procedure followed were:

Materials:

Cadaveric human skin.

Surgery Instruments.

Skin Sheer.

Isotonic Phosphate Buffer pH 6.0

Equipment:

Nalia-Chien type cells.

Temperature-controlled Water Bath.

Multishaker. - A suitable HPLC method to assay Fentanyl base from the receptive solution

Procedure:

1. Human Skin Preparation.

The human skin block received is freed from the fat layer. Circular pieces of 2 cm diameter are cut. Each one of these pieces is sliced into 300 μ thin films by means ofthe Skin Sheer. These thin slices keep the skin structure: stratum corneum - epidermis - dermis. 2. Cut the Duragesic® patches into three similar slices using a hand-operated thermosealing machine to assure the hermetic condition of each slice.

3. Clean each skin circle gently with paper and apply the donor face of patches over it ensuring a full skin contact with the adhesive. 4. Build the cells over the multishaker.

5. Add to each cell 5.5 ml of isotonic phosphate buffer ρH=6, a stir bar and close the unit. (Avoid air bubbles to become trapped in the skin surface exposed to the solution).

6. Switch the water bath on and set temperature control to 34°C. 7. Obtain samples from each cell at 8, 24, 32, 48, 56 and 72 h.

8. Perform the assay at least by triplicate for each lot tested.

Results:

Figure 3 shows that formulations with -OH acrylic adhesives or with non- functional acrylic show higher permeation profiles than with -COOH functional groups. Taking into account that the formulations tested do not have any permeation enhancer, the permeation rates obtained were close to that seen with Duragesic®.

EXAMPLE 2: Interaction of fentanyl with enhancers. Twelve mixtures of fentanyl with different permeation enhancers were prepared according to the following procedure. Their chemical stability and the development of colours were analysed.

Procedure of preparation: 1- Weigh 25 g of each enhancer listed in table 8 in individual glass flasks.

2-a) For all enhancers liquid at room temperature, add 25 mg fentanyl base (obtaining a final concentration of lmg/g). Shake manually and then sonicate 10 minutes at room temperature, b) In the case of lauric acid, lauryl alcohol and Myverol 18-99, melt in a water-bath at 45°C, add 25 mg fentanyl base (obtaining a final concentration of lmg/g).

Shake manually and then sonicate 10 minutes at the same temperature. 3- Prepare corresponding controls (no fentanyl base) 4- Aliquot the obtained solutions in 3 ml ampoules.

5- Verify that the ampoules were correctly sealed by immersing them in a solution of methylene blue into a dessicator, apply vacuum for 5 minutes, allowing to stabilise at atmospheric pressure for 1 minute, extract the ampoules and inspect them visually in order to find any blue coloration inside the ampoules.

Results are summarised in table 8.

Table 8: Interactions of fentanyl with enhancers

NA: non- analysed d. = days

The above data shows that fentanyl does not interact chemically with saturated fatty alcohols (1-decanol and lauryl alcohol), and methyl laurate. On the other hand compounds that have acidic groups, oleyl groups, polyols and unsaturated and saturated fatty acid esters (other than methyl laurate) should be avoided in pharmaceutical compositions containing fentanyl. EXAMPLE 3: Alteration of adhesive properties by the use of enhancers

Placebo patches with the addition of 10 % of different enhancers were prepared according to general methodology previously described (see Example 1) and assessed for their adhesive properties. Shear resistance, peel and tack were evaluated.

The equipment and used was a Shear Test Bank (according to PSTC-7) and a Rolling

Ball Tack (according to PSTC-6).

Shear Resistance, Peel and Tack were measured in a controlled room with temperature set to 25°C +1-2 and humidity set to 60% +/- 5, with the following parameters for the tests:

The results obtained are summarised in table 9.

Cohesive Failure O Anchoring Failure.

The above data shows that the adhesive polymers less affected by the enhancers are Gelva 788 and DT 87-4098. These adhesive matrixes can be formulated with up to 10% of different enhancers without a significant drawback in the adhesive performance. However, in the particular case ofthe matrix obtained by blend of DT 87-4098 and methyl laurate the tack value obtained denotes a matrix with low adhesion and high internal force that will produce an unacceptable final performance in vivo. These attributes are confirmed by shear and peel values. Thus, if possible, methyl laurate should be avoided as permeation enhancer because of this negative effect in the in vivo adhesive performance ofthe adhesive matrix.

EXAMPLE 4: Combination of fatty alcohols enhancers with non-functional acrylic adhesive.

Thirteen patch formulations containing lauryl alcohol or/and 1-decanol (described in table 10) were prepared according to the following procedure. Their chemical and physical stability, as well as in vitro permeation profiles, were analysed.

Procedure of prep aration of formulations :

1. Weigh the Fentanyl base in the glass flasks.

2. Add the adhesive and enhancer. Mix until the solution is homogeneous.

3. Perform the coating in the pilot plant, setting the following parameters:

Perform the coating of each solution on Release Liner 1-5 PESTR Clear 6200

(P2).

5. Laminate with Backing Scotch Pack 1012 with its brilliant side towards the adhesive matrix. 6. The laminates thus obtained are cut in 4 cm length pieces, obtaining patches of approximately 20 cm² area.

7. These patches are stored in aluminium pouches at the following conditions:

CONDITION

55°C / Room Humidity

40°C / 75%R.H.

25°C / 60% R.H.

Table 10: Patch formulations

Part a: Physical stability

Physical stability of the prepared patches was tested in three different conditions. The obtained results are listed in tables 11, 12 and 13 in which OK means "No crystals observed" and FAIL or F means "Crystals observed".

Table 11: Storage Conditions:55 °C / Room Humidity

Table 12: Storage Conditions:40 °C / 75 % R.H.

Patches containing between 2 and 4 % fentanyl and formulated with 1- decanol and/or lauryl alcohol as permeation enhancers, (5-10% or in a mixture 1 : 1 thereof), offer a formulation that was physical stable along the period ofthe study.

Part b: Chemical Stability

Chemical stability ofthe prepared formulations were tested in three different conditions. The obtained results are listed in tables 14, 15 and 16. The results are identify as follows:

- 'TSTon-significant change" (NSC) means that the loss of fentanyl from the initial value is less or equal to 5%.

- "N.A." means that the sample has not been analysed

Table 15: Storage Conditions:40°C / 75% RH.

Table 16: Storage Conditions:25°C / 60% RH.

Chemical stability is good in both types of formulations, i.e. those containing lauryl alcohol and the ones containing 1-decanol.

Part c: Permeation profiles

The permeation experiment was performed using the materials, equipment and procedure of example 1, part d).

Selected patches formulations for the permeation experiment are listed in table 17:

Table 17: Patch formulations for the permeation experiment

Results in figure 4 show a marked difference between 1-decanol and lauryl alcohol as permeation enhancers for fentanyl.

The patch formulated with 10% of lauryl alcohol shows the major ratio of enhancement. The increment ofthe permeation rate reaches 60 % over the value obtained for the formulation used as internal control (Lot 006).

The above data clearly show that fentanyl has good physical and chemical stability in both fatty alcohols tested contained in non-functional acrylic adhesive matrixes. Lauryl alcohol produced an enhancement of fentanyl permeation that is higher than that produced by 1-decanol.

EXAMPLE 5: Comparative permeation profile of formulations with lauryl alcohol and methyl laurate as enhancer.

The permeation experiment was performed using the same materials, equipment and procedure as in example 1, part d).

Selected patches formulations for the permeation experiment are listed in table 18:

Table 18: Patch formulations for the permeation experiment

Results in figure 5 show that lauryl alcohol has a better in vitro performance than methyl laurate as permeation enhancer in the selected adhesive. The formulation containing lauryl alcohol is capable to permeate 20 % more of the drug than the formulation containing methyl laurate within the same period of time.

EXAMPLE 6: Comparative permeation profile between formulations with lauryl alcohol as enhancer and Duragesic®.

The permeation profile obtained from matrixes formulated with 10 % of lauryl alcohol as permeation enhancer and different loads of fentanyl were analysed against the reference product Duragesic®. The permeation experiment was performed using the same materials, equipment and procedure of example 1, part d). Lots to be tested are listed in table 19.

Results:

As is shown in figure 6, the average flux between 24 to 72 hours was 1.55 μg/cm²/h fo DΓr lloott 004488,, 22..0044 μμgg//ccmm²²//hh ffoorr lloott 004419, 3.24 μg/cm²/h for lot 050 and 2.62 μg/cm7h for the reference product DuragesictJ

EXAMPLE 7: Use of myristyl alcohol as co-enhancer.

Twelve formulations were prepared (see table 20) to evaluate the use of myristyl alcohol as permeation enhancer or co-enhancer of fentanyl according to the following procedure. Then, their physical stability, chemical stability, adhesiveness and permeation profile were evaluated.

Preparation of Formulations:

1. Weigh the fentanyl base in the glass flasks.

2. Add the adhesive and enhancer. Mix till total dissolution.

3. Perform the coating in the pilot plant, setting the following parameters:

4. Perform the coating of each solution on the Release Liner 1-5 PESTR Clear (P2) 6200.

5. Laminate with Backing Scotch Pack 1012 with its brilliant face in contact with the . adhesive matrix.

6. The laminates thus obtained are cut in 40 mm length pieces, obtaining patches of approximately 20 cm² area.

7. These patches are stored in aluminum pouches at the following conditions:

CONDITION

55°C / Room Humidity

40°C / 75%R.H.

25°C / 60% RH. Table 20: Formulations with myristyl and/or lauryl alcohol

TS: Total solids (expressed as percentage).

Part a: Physical stability

Physical stability was tested in three different conditions. The obtained results are listed on tables 21, 22 and 23 in which OK means "No crystals observed" and FAIL or F means "Crystals observed".

Table 22: Storage Conditions:40 °C / 75 % R.H.

Part b: Chemical stability

Chemical stability was tested in three different conditions. The obtained results are listed in tables 24, 25 and 26. The results are identify as follows:

- "Non-significant change" (NSC) means that the loss of fentanyl from the initial value is less or equal to 5%. "Significant change" (SC) means loss of potency greater than 5%.

- "N.A." means that the sample has not been analysed Table 24: Storage Conditions:55 °C / Room Humidity

Table 26: Storage Condition:25°C / 60% R.H

All the formulations tested showed good chemical stability

Part c: Adhesiveness

The adhesive properties ofthe formulations detailed in Table 20 were analysed. The equipment and procedure used were the following: Equipment:

Shear Test Bank (according to PSTC-7).

Rolling Ball Tack (according to PSTC-6). Test:

Assess Shear Resistance, Peel and Tack in a controlled room with temperature set to 25°C ± 2 and humidity set to 60% ± 5, settling the following parameters for the tests:

Results in table 27 are reported as the mean ofthe individual data.

Table 27: Adhesive properties of matrixes with lauryl and/or myristyl alcohol

* Cohesive failure

According to these results, the formulations with the best adhesive performance are those containing 10% of lauryl alcohol without myristyl alcohol and regardless ofthe percentage of fentanyl. Part d: Permeation

The permeation experiment was performed using the materials, equipment and procedure of example 1, part d).

Table 28: Lots used in the permeation experiment

Results:

As shown in figure 7, the combination of two saturated fatty alcohols with different number of carbons like myristyl and lauryl alcohol do not produce a phenomena of synergism when used as permeation promoters for fentanyl. Moreover, the results obtained indicate that the enhancing effect of 10% of lauryl alcohol on the permeation of a 4% matrix cannot be improved to a great extent by addition of myristyl alcohol.

EXAMPLE 8: Wearing Test of Placebo TDS on Humans.

A placebo patch containing lauryl alcohol as permeation enhancer was tested on 7 healthy volunteers, with evaluations at 24 h, 48 h and 72 h after patch application. The degree of adhesion, glue residue and skin reaction of placebo patches were evaluated on an open label 3-day study. Subjects were applied one placebo patch in the external side ofthe right arm, avoiding hairy zones.

Following patch removal, additional observations at 30 min and 24 h were conducted to assess local irritation induced by the systems.

As a subjective measure of treatment acceptability, the grade of distress produced by patch removal was recorded in visual analogue scales (NAS). Placebo patch of 20 cm² (rounded shape, transparent backing), Lot 060 was manufactured according to the following formula:

Table 29: Placebo patch used in the wearing test

Assessment of Skin Reactions

The irritation observed was classified according to the following scoring system:

Extent of Erythema

0 - none

1 - < 50%) of occluded area

2 - > 50%) of occluded area

3 - Exceeds the occluded area

The grade of all observed events were classified using the following scale: 0 - none present 1 - mild: annoying, but not requiring medical attention, nor Umiting daily activities.

2 - moderate: tolerable, but requiring medical management, partially limiting daily activities. 3 - severe;. intolerable, requiring device removal and medical management, restraining completely the subject's daily activities.

Results: a) Patch Adhesion

The numbers in table 30 are percentage of the patch adhered to the skin at each evaluation.

Table 30: Percentage of patch adhesion to the skin

Except with FS, all patches become dislodged around the patch perimeter.

Six of 7 patches showed an acceptable adhesion to skin during 72 h of use (total adhesion / < 20% detachment in borders). Only one showed significant detachment (80%), according to the data in Table 30. The patches resisted repeated contact with warm water and those that become slightly dislodged were able to stick again to skin after it was dry.

Only two patches presented slight ridging during use that did not affect patch performance. In one case, the ridging was able to be re-stuck to the skin.

b) Cold Flow

The data in table 31 was organized as follows: extent of cold flow, colour and % of area involved. Table 31 : Cold Flow observed in the wearing test

Four patches out of 7 presented an adhesive residue (1 mm of extent in average) around patch perimeter following patch removal. This residue was easy to be removed using soap and lukewarm water in all cases.

c) Skin Reactions and Symptoms

Table 32: Skin reactions and symptoms

*: in the form of little dots evenly distributed in the treated zone.

Er (x): erythema (grade)

The 0.5 score was assigned to equivocal reactions observed through the patch.

Estimation ofthe Irritation Index

If each "no" response is considered as 0 in table 33, the averages ofthe individual data at 72 h and afterwards are:

Table 33: Irritation index

Taking into account the score for erythema obtained by averaging the individual data from 72 h (pre-removal), and 30 minutes and 24 h following patch detachment, it is clearly shown that this placebo formulation has a low potential for irritation (0.69 in a 0 to 4 scale - Draize method) to human skin.

d) Discomfort induced during patch removal

The grade of discomfort induced by patch removal was low, obtaining 26.3 ± 5.8 mm in a scale ranging from 0 (painless) to 100 mm (extremely painful).

EXAMPLE 9: Fentanyl plasma levels and irritation potential in rabbits.

Fentanyl plasma levels and irritation potential in rabbits were tested using a patch according to the present invention (Lot 114) against the reference product Duragesic® 25. Also, the estimation ofthe fentanyl delivered was calculated by the analysis of fentanyl remnants in the used patches.

Lot 114 has the same formulation composition and was manufactured using the same methodology employed for lot 050. The patches according to the present invention used in this experiment had 10 cm² of active area while Duragesic® 25 had 10 cm² of active area in a 20 cm² of total area of device. The number of rabbits

(White New Zealand) was 16, divided in two groups: i) 8 rabbits that received one lot 114 patch and ii) 8 rabbits that received one Duragesic® 25 (Lot 0017102). Exposure duration to patches was 72 h. and all animals survived the study.

Blood samples were obtained from the animal's ear veins at Oh, 3h, 6h, 24h, 48h and 72 h following patch application, for fentanyl analytics. The plasma fentanyl levels were determined in the individual samples collected, except for those obtained with significant patch lifting (more than 30%) and/or significant hair growth below the TDS (17%> of samples for lot 114 and 35% of samples for Duragesic® 25). Plasma fentanyl levels were assessed in individual samples using g GC-MS. Evaluation of skin was performed at 24, 72 and 96 (post-removal) hours.

The indexes of dermal irritation (DII) Pre-Removal of TDS and Post-Removal of TDS were calculated for each patch.

Interpretation ofthe results: the indexes obtained were interpreted following ISO/WD 10993-10 (1998):

The recovered patches were returned to their individual primary containers and kept at -20°C until analysis with intact units of both lot 114 and Duragesic® 25. The fentanyl remnants in used patches were assessed by an HPLC method, using the intact patch units as reference controls.

The calculation performed to obtain the fentanyl release rate for each TDS was:

Fentanyl release rate = mg fentanyl in intact TDS - mg fentanyl in used TDS

72h

Results a) fentanyl plasma levels:

As seen in Figure 8, the individual fentanyl plasma levels obtained with the lot 114 10 cm² patch are similar to those obtained with Duragesic® 25 in all sampling times. The dispersion ofthe data of both patches was similar at 3h, 6 h and 72h, and somewhat lower for lot 114 at 24h and for Duragesic® 25 at 48 h.

Also, the fentanyl data obtained in this trial enabled the estimation ofthe area under the concentration-time curve (mean AUC), maximum concentration of fentanyl (mean Cmax) and time to peak (mean Tmax) using the mean fentanyl profile calculated for each TDS.

For calculation purposes, the concentrations reported by the analytical team as N.D. (non-detectable) were considered as 0 ng/ml. Results are shown in table 34. Table 34: Mean pharmacokinetics parameters for Fentanyl in rabbits

b) Irritation

Two approaches were used to estimate the DII ofthe tested patches: weighed and maximum.

Table 35: Dermal irritation index DII

Both products tested showed a similar performance.

c) Fentanyl release rate The calculations were performed with remnant data belonging to those

TDS that were less than 30% detached during use and without hair growth development between the patch and the skin.

The approximate quantity of fentanyl released is shown in table 36 as mean and (ISD), as well as the calculated fentanyl release rate in μg/h (obtained by dividing the fentanyl released by 72 h) : Table 36: Fentanyl released in rabbits

The fentanyl release rate of Lot 114 10 cm² was close similar to that of Duragesic® 25.

EXAMPLE 10: Pharmacokinetics in healthy volunteers

Following current Good Clinical Practice (cGCP), fentanyl plasma levels in humans produced by a patch according to the present invention (Lot 20801) were compared against the reference product Duragesic® 25 Also, the estimation of fentanyl delivered by the patches was calculated by the analysis of fentanyl remnants in the used patches. Irritation potential, adhesiveness and vital signs were recorded. Lot 20801 has the same formulation composition and was manufactured using the same methodology employed for Lot 050. The patches according to the present invention used in this experiment had 10.6 cm² of active area while Duragesic® 25 (Lot 0017102) had 10 cm² of active area in a 20 cm² of total area device. The design ofthe experiment was the following: Open label, single dose, randomised, active controlled, two period crossover pilot pharmacokinetics study.

Six healthy subjects were enrolled and exposed to the patches according to the present invention and Duragesic® 25 patches during 72 h, using one patch per period of treatment, and allowing a washout interval of 14 days between treatments.

A blood sample was obtained immediately before patch application (0 h for fentanyl assays). Each patch remained applied for 72 h. During patch use, blood samples were obtained at 2-4 h intervals during the first 24 h, and at 6 h intervals from 30 h to 72 h. Then, the patch was removed from the skin and additional blood samples were drawn at 4, 8, 12, 24, 48 and 72 h, to characterise drug decay after each system use. Results: Only five volunteers completed the study, the sixth participant refused to participate in the second period ofthe study alleging reasons not related to the medication received but to the hospital stay.

Figure 9 shows the mean values (± SEM) obtained. The patch according to the invention gave steady values from the 16^th hour onwards. The reference patch gave a similar profile although somewhat higher. This difference in the amount delivered was not statistically significant.

While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope ofthe invention.

References

1. S. D. Roy, M. Gutierrez, G. L. Flynn ad G. W. Clary, Journal of Pharmaceutical Sciences, 85, 5, 1996, pp. 491-495.

2. R. Miguel, J.M. Kretzer, D. Reinhart, P.S. Sebel, J. Bowie, G. Freedman and J. B. Eisenkraft, Anesthesiology, 83, 1995, pp. 470-477.

3. P. Fiset, C. Cohane, S. Browne, S. C. Brand and S. L. Shafer, Anesthesiology, 83, 1995, pp. 459-469.

Claims

Claims

1. A matrix-type transdermal delivery device having a) a pressure sensitive acrylic adhesive matrix formed by a non-functional copolymer in which fentanyl is incorporated, b) a backing layer, and c) a release liner wherein said matrix includes lauryl alcohol as enhancer of percutaneous permeation.

2. A device according to claim 1 wherein the fentanyl is present in the matrix in the range of 2-5% by weight of total content excluding adhesive solvents.

3. A device according to claim 2 wherein the fentanyl is present in the matrix in the range of 3.5-4.5% by weight of total content excluding adhesive solvents

4. A device according to any one of claims 1 to 3 wherein the lauryl alcohol is present in the matrix in the range 2-15% by weight of total content excluding adhesive solvents.

5. A device according to claim 4 wherein the lauryl alcohol is present in the matrix in the range 5-10% by weight of total content excluding adhesive solvents.

6. A device according to claims 1 to 5 wherein the overall thickness ofthe device is not greater than 100 μm

7. A device according to claim 6 wherein the overall thickness ofthe device is not greater than 80 μm.

8. A device according to any one of claims 1 to 7 wherein the non-functional copolymer is a copolymer of 2-ethylhexylacrylate and vinylacetate.