TRANSDERMAL DRUG DELIVERY MATRIX FOR COADMINISTERING ESTRADIOL AND ANOTHER STEROID

Technical Field This invention is in the field of transdermal drug delivery. More particularly it relates to a matrix type transdermal patch for coadministering estradiol and another steroid wherein the flux of each steroid from the matrix is independent of the concentration of the other in the matrix.

Background Matrix-type transdermal patches are those in which the drug is contained in and released from a polymer matrix. The matrix is typically made of a pressure sensitive adhesive and defines the basal surface of the patch (i.e. the surface affixed to the skin). While more than one drug may be delivered from such a matrix, the respective fluxes of the individual drugs from the matrix typically depend upon the concentration of the other drug(s) in the pressure sensitive adhesive. This is because the concentration of each drug in the matrix affects the solubility of the other drug(s) in the pressure sensitive adhesive.

EPA 89310350.7 (published 1 April 1990) describes a transdermal matrix type patch for administering estradiol and/or esters of estradiol. The pressure sensitive adhesive component of the patch is a copolymer of 2-ethylhexylacrylate (EHA) and N-vinyI-2- pyrrolidone (NVP). This copolymer is said to provide a means for maintaining a relatively high concentration of estradiol in the matrix without estradiol crystallization. This NVP-containing acrylic copolymer adhesive uses two monomers with very different reactivity ratios, so that for all practical purposes, the polymer is likely to have a "block copolymer" structure, with distinct long chain NVP and EHA domains. Beyond the suggestion that estradiol esters may be used as a drug, this application provides no suggestion or data regarding the inclusion of a second, different steroid in the matrix.

Disclosure Of The Invention The invention is a transdermal patch for administering estradiol and another steroid comprising:

a) a backing layer; and b) a matrix layer comprising:

(i) a NVP-containing acrylic copolymer pressure sensitive adhesive; (ii) estradiol; and

(iii) another steroid wherein the flux of said other steroid from the matrix layer is independent of the concentration of estradiol in the matrix layer and the flux of estradiol from the matrix layer is independent of the concentration of the other steroid in the matrix layer. Another aspect of this invention is a method for providing hormone replacement therapy to a woman in need of such therapy comprising applying the above described patch to the skin of said woman.

Brief Description Of The Drawings Figures 1 -12 are graphs of in vitro skin flux data described in the examples, infra. Modes For Carrying Out The Invention

As used herein the term "transdermal" intends percutaneous and transmucosal (e.g. transbuccal) administration, i.e., passage of the drug by diffusion through unbroken skin or mucosa into circulation.

The term "another steroid" intends a steroid other than estradiol or an ester of estradiol. Examples of such other steroids, without limitation, are progesterone, norethindrone acetate, norethindrone, desogestrel, gestodene, norgestrel, levo-norgestrel, testosterone, methyltestosterone and androsteinedione.

The term "flux" intends the in vitro rate of release of steroid per unit area as measured using the procedure described in the examples, infra. The term "independent" intends that the flux of each steroid from the matrix does not vary significantly as the concentration of the other steroid in the matrix varies. Typically the variation in flux, if any, will be in the range of ± 35%.

The pressure sensitive adhesive copolymer component of the matrix is a NVP-containing acrylic copolymer. The NVP constitutes 5 to 50 mol % with other acrylic monomers comprising 40 to 95 mol %. Other monomers typically used in acrylic copolymer adhesives are described in the Background section, supra. EPA 89310350.7 for example discloses a copolymer of NVP and EHA. The EHA constitutes 45 to 80 mol %,

preferably 55 to 70 mol % of the copolymer whereas NVP constitutes 20 to 55 mol %, preferably 30 to 45 mol % of the copolymer.

Estradiol is present in the matrix at about 1 to 20% by weight, preferably about 2 to 12% by weight of the matrix. The other steroid will normally constitute 1 to 20% by weight of the matrix, depending upon the particular steroid involved. For instance, when the other steroid is norethindrone acetate it will typically constitute 10 to 8% by weight of the matrix and when the other steroid is testosterone, it will typically constitute 1 to 10% by weight of the matrix. Although the precise mechanism by which the flux of estradiol is independent of the concentration of the other steroid in the matrix (and vice versa) is not known, it is possible that the "block copolymer" structure of NVP-containing acrylic copolymers may result in each steroid selectively partitioning into a specific block domain and being released from that domain independent of the other steroid.

In addition to the copolymer and the steroids, the matrix may also contain one or more skin permeation enhancers. Examples of enhancers that may be used, without limitation, include saturated and unsaturated fatty acids and their esters,, alcohols, monoglycerides, acetate, diethanolamides and N, N-dimethylamides, such as oleic acid, propyl oleate, isopropyl myristate, glycerol monooleate, glycerol monolaurate, methyl laurate, lauryl alcohol, lauramide diethanolamide and combinations thereof. Saturated and unsaturated sorbitan esters, such as sorbitan monooleate and sorbitan monolaurate may also be used. Other conventional additives used in matrix type patches may also be included in the matrix. Such additives include, without limitation, tackfiers, fillers or other additives that affect the adhesive properties of the matrix and additives such as glycerin, that reduce skin irritation, and additives that affect the solubility of the steroids in the copolymer. The matrix may be formulated by mixing the adhesive (which is typically obtained in solution), estradiol, other steroid, permeation enhancer (if necessary) and other additives (if desired) in appropriate proportions, casting the mixture onto a substrate (e.g. a release liner), drying the cast layer to remove the solvent, and laminating a backing layer on to the dried polymer matrix. The backing layer will typically be occlusive. Release liner and backing layer materials are well known in the transdermal patch art.

The invention is further illustrated by the following examples. These examples are not intended to limit the invention in any manner.

EXAMPLES EXAMPLE 1

(A) Norethindrone Acetate {NEA) only matrices:

Matrix laminates containing norethindrone acetate (NEA, Schering AG, Berlin, Germany) were fabricated as follows. The percent solid adhesive of an EHA/NVP acrylic copolymer adhesive (TSR Adhesive, Sekisui Chemical Co., Japan), was determined by weighing a small amount of adhesive solution in a preweighed aluminum dish. The solvent was evaporated by overnight drying in a convection oven at 70°C, and the dish was reweighed. The percent solids was calculated by dividing the dry weight by the wet weight and multiplying by 100. Known amounts of TSR adhesive solution were weighed into glass bottles. From the weight of the adhesive solution and the percent solid adhesive, the amount of adhesive in the solution was calculated. Appropriate quantities of NEA and sorbitan monooleate permeation enhancer (ARLACEL 80, ICI Americas, Wilmington, Delaware) were added to yield various compositions as shown in Table I below (Formulations 1-3), all percentages being calculated on a dry weight basis. Each glass bottle was then tightly capped, sealed with laboratory film (PARAFILM "M", American National Can Company, Greenwich, CT)), and rotated overnight.

About 8 ml of the drug/sorbitan monooleate/TSR solution was then dispensed on a release liner (siliconized polyester release liner, Release Technologies, Inc., W. Chicago, Illinois), and cast with a 10 mil gap casting knife. This cast mixture was dried in a convection oven at 70 °C for 15 minutes to yield a dry film approximately 2.0 mil thick. A backing film (polyethylene backing film, 3M Corp., St. Paul, Minnesota) was then laminated onto the dry adhesive film using a rubber roller. This matrix laminate was used for in vitro skin flux measurements which were performed as described below.

In vitro skin flux studies were conducted using modified Franz diffusion cells. Heat separated human epidermal membrane was cut into rectangular strips. The matrix laminates (described above) were cut into circular punches of 0.71 cm ² surface area. After the release liner was peeled and discarded, the circular punches were laminated onto the stratum corneum surface of the epidermal membrane. Each piece of the skin-punched

matrix sandwich was loaded between the donor and receiver compartments of a diffusion cell, with the epidermal side facing the receiver compartment, and clamped in place. The receiver compartment was then filled with 0.02% sodium azide solution, and the cell was then placed in a circulating water bath calibrated to maintain the skin surface temperature at 32 ± 1 °C. At predetermined intervals, the entire contents of the receiver compartment was collected for drug quantitation, and the receiver compartment was refilled with fresh receptor medium, taking care to eliminate any air bubbles at the skin/solution interface. The cumulative amount of drug permeated per unit area at any time t (Q _t μg/cm ² ) was determined as follows: t

Q _t = ∑{C„ * V) / A

where C _n is the concentration (mg/ml) of drug in the receiver sample for the corresponding sample time, V is the volume of fluid in the receiver chamber (-6.3 cm ³ ), and A is the diffusional area of the cell (0.64 cm ² ). The slope of the best fit line to the Qt vs. t plot gives the steady state flux (J _ss , μg/cm ² /h); the intercept of this line on the time axis gives the lag time (t _L , h).

Three formulations (Table I, Formulations 1-3), with progressively increasing NEA loading (1.5-6 % w/w) along with a control formulation were evaluated for in vitro skin flux as described above on the same donor skins. The purpose of the control formulation was to minimize inherent skin to skin variability and allow for better elucidation of trends in the results. The in vitro drug fluxes from the test formulations were normalized on an individual skin basis relative to the fluxes from the control formulation which was run simultaneously on the same donor skins in this and subsequent experiments. This normalization procedure significantly minimized inter skin variability and allowed for easy comparison of relative flux performance between formulations in this and subsequent experiments. The NEA fluxes obtained for the Formulations 1-3 and the control formulation are summarized in Table II. The normalized flux ratios are plotted in Figure 1.

As can be seen from the data presented in Figure 1 , a 4 fold increase in the drug loading results in a proportional 4 fold increase in the flux. The normalized in vitro NEA fluxes therefore display linear and Fickian dependence on the drug concentration between

1.5-6 % w/w loading in matrices made with the NVP containing TSR acrylic copolymer adhesive.

TABLE I

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID TSR NEA E2 ARLACEL Adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

1. Lot # 040695-1 88.5 1.5 0.0 10.0

2. Lot # 040695-2 87.0 3.0 0.0 10.0

3. Lot # 040695-3 84.0 6.0 0.0 10.0

Control Formulation

4. Control # 95Z007 77.8 6.0 6.2 10.0 NEA Control Formulation

TABLE II

CUMULATIVE NEA PERMEATION IN 96 hr (Q96-μg/cm ² /96 hr)

Formulation ID # of Q96* Ratio ⁺ skins/cells

Test Formulations

1. Lot # 040695-1 6/24 8.4 ± 2.5 0.27 ± 0.06

2. Lot # 040695-2 6/24 17.3 ± 3.8 0.55 ± 0.10

3. Lot # 040695-3 6/24 32.9 ± 8.9 1.00 ± 0.08

* Q96-Cumulatιve amount permeated from test formulation in 96 hr

+ Skin flux normalized relative to Control formulation on an individual skin basis Control = 32 0 + 7 9

(B) Estradiol (E2) only matrices:

Matrix laminates containing estradiol (E2, Berlichem, Wayne, New Jersey) were prepared as described above in Example 1 (A) except that E2 was used as the drug instead of NEA. Necessary amounts of E2 were pre-dissolved in iso-propyl alcohol (IPA) and added to the casting solution to yield various compositions as shown in Table III below (Formulations 5-7).

TABLE III

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID TSR NEA E2 ARLACEL Adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

5. Lot # 040695-4 87.0 0.0 3.0 10.0

6. Lot # 040695-8 83.8 0.0 6.2 10.0

7. Lot # 040695-12 81.0 0.0 9.0 10.0

Control Formulation

8. Control # 94Z003 93.5* 0.0 1.5 5.0 E2 Control Formulation

* Formulation made with DUROTA 87-2070 Adhesive, National Starch and Chemical Company. Bridgewater, NJ

These formulations were evaluated for in vitro skin flux along with an E2 control formulation (Formulation 8). The in vitro skin fluxes for the three test formulations and the control formulation on the same skins are presented in Table IV below. The normalized flux ratio are plotted in Figure 2.

As can be seen from the data presented in Figure 2, a 3 fold increase in the drug loading results in a proportional 3 fold increase in the flux. The normalized in vitro E2 fluxes therefore display linear and Fickian dependence on the drug concentration between 3-9 % w/w loading in matrices made with the NVP containing TSR acrylic copolymer adhesive.

- θ -

TABLE IV

CUMULATIVE E2 PERMEATION IN 96 hr (Q96-μg/cm ² /96 hr)

Formulation ID # of Q96* Ratio ^4" skins/cells

Test Formulations

5. Lot # 040695-4 6/24 6.1 ± 1.7 0.29 ± 0.06

6. Lot # 040695-8 6/24 13.4 ± 4.0 0.64 ± 0.08

7. Lot # 040695-12 6/24 20.9 ± 5.6 l .O ± O. lό

* Q96-(Λιmulatιve amount permeated from test formulation in 96 hr

+ Skin flux normalized relative to Control formulation on an individual skin basis Contiol 20 6 + 4 4

(C) NEA/E2 co flux matrices:

Matrix laminates containing both E2 and NEA in combination were prepared as described above in Example 1 (A). Necessary amounts of E2 were pre-dissolved in iso- propyl alcohol (IPA) and added to the casting solution along with NEA and sorbitan monooleate to yield various compositions as shown in Table V below (Formulations 9-17).

These formulations were evaluated for in vitro skin flux along with NEA and E2 control formulations (Formulations 4 and 8 respectively) described above in Examples 1(A) and 1(B) respectively. The in vitro skin fluxes for the test formulations and the control formulations on the same skins are presented in Table VI. The normalized NEA and E2 flux ratio are plotted in Figures 3 and 4 respectively.

TABLE V

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID TSR NEA E2 ARLACEL

Adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

9. Lot # 040695-5 85.5 1.5 3.0 10.0

10. Lot # 040695-6 84.0 3.0 3.0 10.0

1 1. Lot # 040695-7 81.0 6.0 3.0 10.0

12. Lot # 040695-9 82.3 1.5 6.2 10.0

13. Lot # 040695-10 80.8 3.0 6.2 10.0

14. Lot # 040695-1 1 77.8 6.0 6.2 10.0

15. Lot # 040695-13 79.5 1.5 9.0 10.0

16. Lot # 040695-14 78.0 3.0 9.0 10.0

17. Lot # 040695-15 75.0 6.0 9.0 10.0

Control Formulations

4. Control # 95Z007 77.8 6.0 6.2 10.0 NEA Control Formulation

8. Control # 94Z003 93.5* 0.0 1.5 5.0 E2 Control Formulation

[Ormulation made with DUROTAK 87-2070 Adhesive, National Starch and Chemical Company, Bπdgewatcr. NJ

As can be seen from the data in Figure 3, a 4 fold increase in the NEA loading results in a proportional 4 fold increase in the flux of NEA. Similarly, a 3 fold increase in E2 loading results in a proportional 3 fold increase in E2 flux (Figure 4). The normalized in vitro NEA and E2 fluxes therefore display linear and Fickian dependence on the drug concentration in the presence of each other in matrices made with the NVP containing TSR acrylic copolymer adhesive over the range of drug loadings investigated (0-6 % NEA loading and 0-9 % E2 loading).

TABLE VI

CUMULATIVE NEA & E2 PERMEATION IN 96 hr (Q96 -μg/cm*V96 hr)

#of

Formulation ID skins/cells NEA data E2 data

Q96* 096 Ratio 096* Q96 Ratιo ⁺ (Control)* (( ontrol)

X

Test Formulations

9 Lot # 040695-5 7/28 77±36 257 X 110 029 54±24 187± 028-i

± 68 003 002

10 Lot #040695-6 7/28 154±63 257 ± 110 06± 57i 23 187 ± 030± 008 68 002

11 Lot # 040695-7 7/28 276 J 257 ± 110 11 ± 54±25 187± 028±

110 009 68 004

12 Lot #040695-9 5/20 70± 19 278 ± 7.7 025 1014 158 ± 065±

± 27 55 010

003

13 Lot #040695-10 5/20 142±42 278 t 77 051 104± 158-i 066±

± 35 55 003

005

14 1 ot# 040695- 11 5/20 278 ± 77 278±77 1 O± 102 i 158 i 066±

00 30 55 006

15 Lot #040695-13 6/24 89±26 374 ±223 028 203 ± 247 ± 091-i

± 75 163 030

010

16 Lot #040695-14 6/24 183 ±87 374 ±223 053 230- 247 ± 099±

A 135 163 017

012

17 Lot #040695-15 6/24 324 ± 374 ±223 090 213- 247 ± 089-t

180 ± 132 163 017

012

* 096-Cumulatιve amount permeated from test formulation in 96 hi x Q96 (Control)-Cumulatιve amount permeated from Control formulation in 96 hr on same skins as the test formulations

+ Skin flux normalized relative to Control formulation on an individual skin basis

The slope of each of the three linear regression lines (normalized data for NEA fluxes in the presence of E2, Figure 3) was compared statistically to the slope of the

regression line for the NEA formulations without E2 using a Student's t-test. The results indicated that there was no statistically significant difference (p> 0.10) in slopes between each of the three linear regression lines (for NEA formulations containing E2) relative to the slope of the regression line for the NEA formulations without E2. This confirms that the presence of E2 in the matrix does not affect the flux of NEA in systems made with the

NVP containing TSR acrylic copolymer adhesive.

The slope of each of the three linear regression lines (normalized data for E2 fluxes in the presence of NEA, Figure 4) was compared statistically to the slope of the regression line for the E2 formulations without NEA using a Student's t-test. The results indicated that there was no statistically significant difference (p> 0.10) in slopes between each of the three linear regression lines (for E2 formulations containing NEA) relative to the slope of the regression line for the E2 formulations without NEA. This confirms that the presence of NEA in the matrix does not affect the flux of E2 in systems made with the NVP containing TSR acrylic copolymer adhesive. The above data clearly shows that in systems prepared with the NVP containing acrylic copolymer adhesive, TSR, over the range of drug loadings investigated (0-6 % NEA loading and 0-9 % E2 loading), the flux of each steroid depends only on its concentration and is not affected by the presence of the other steroid.

EXAMPLE 2

(A) NEA only matrices:

NEA only matrices were prepared as described in Example 1(A) except that the adhesive used was DUROTAK 87-2516 (an acrylic copolymer adhesive containing EHA, vinyl acetate and hydroxyethyl acrylate, National Starch and Chemical Co, Bridgewater, NJ). This adhesive does not contain N-vinyl-2- pyrrolidone. Necessary amounts of NEA and sorbitan monooleate were dissolved in the adhesive solution to yield various final compositions as shown in Table VII below (Formulations 1-4).

TABLE VII

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID DURO ^' fAK NEA E2 ARLACEL 87-2516 adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

1. Lot # 013096-1 88.0 2.0 0.0 10.0

2. Lot # 013096-2 86.0 4.0 0.0 10.0

3. Lot # 013096-3 84.0 6.0 0.0 10.0

4. Lot # 013096-4 82.0 8.0 0.0 10.0

Control Formulation

5. Control # 95Z098 77.6 6.0 6.4 10.0 NEA/E2 Control Formulation

* TSR adhesive, Sekisui Chemical Co., Osaka Japan

These formulations were evaluated for in vitro skin flux along with a control formulation (Formulation 5). The in vitro skin fluxes for the three test formulations and the control formulation on the same skins are presented in Table VIII below. The normalized flux ratios are plotted in Figure 5.

As can be seen from the data presented in Figure 5, a 4 fold increase in the drug loading results in a proportional 4 fold increase in the flux. The normalized in vitro NEA

fluxes therefore display linear and Fickian dependence on the drug concentration between 2-8 % w/w loading in matrices made with DUROTAK 87-2516 adhesive.

TABLE VIII

CUMULATIVE NEA PERMEATION IN 24 hr (Q24-μg/cm ² /24 hr)

Formulation ID # of Q24* Ratio ^4" skins/cells

Test Formulations

1. Lot # 013096-1 3/12 1.8 ± 0.7 0.5 ± 0.2

2. Lot # 013096-2 3/12 3.5 ± 1.1 0.9 ± 0.3

3. Lot # 013096-3 3/12 5.4 ± 0.7 1.4 ± 0.4

4. Lot # 013096-4 3/12 8.6 ± 4.7 2.3 ± 1.8

Q24-Cumulatιvc amount permeated from test formulation in 24 hr

Skin flux normalized relative to Control formulation on an individual skin basis Control = 4 1 + 1.3

(B) E2 only matrices:

Matrix laminates containing E2 were prepared as described above in Example 1 (A) except that E2 was used as the drug instead of NEA and DUROTAK 87-2516 was used as the adhesive instead of TSR. Necessary amounts of E2 were pre-dissolved in IPA and added to the casting solution to yield various compositions as shown in Table IX below (Formulations 6-8).

TABLE IX

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID DUROTAK NEA E2 ARLACEL

87-2516 adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

6. Lot # 013096-5 89.0 0.0 1.0 10.0

7. Lot # 013096-10 88.0 0.0 2.0 10.0

8. Lot # 013096-15 86.0 0.0 4.0 10.0

Control Formulation

5. Control # 95Z098 77.6* 6.0 6.4 10.0 NEA/E2 Control Formulation

^♦ TSR adhesive, Sekisui Chemical Co., Osaka, Japan

These formulations were evaluated for in vitro skin flux along with a E2 control formulation (Formulation 5). The in vitro skin fluxes for the three test formulations and the control formulations on the same skins are presented in Table X. The normalized flux ratios are plotted in Figure 6.

As can be seen from the data presented in Figure 6, the normalized in vitro E2 fluxes increase linearly with drug concentration between 1-4 % w/w loading in matrices made with DUROTAK 87-2516 adhesive.

TABLE X

CUMULATIVE E2 PERMEATION IN 24 hr (Q24-μg/cm ² /24 hr)

Formulation ID # of Q24* Ratio ⁺ skins/cells

Test Formulations

6. Lot # 013096-5 3/12 3.5 ± 1 - 1 1.4 ± 0.5

7. Lot # 013096-10 3/12 6.2 ± 1.1 2.4 ± 0.3

8. Lot # 013096-15 3/12 9.0 ± 3.0 3.5 ± 1.0

* Q24-Cumulatιve amount permeated from test formulation in 24 hr

+ Skin flux normalized relative to Control formulation on an individual skin basis Control = 2 5 + 6

(C) NEA/E2 co flux matrices

Matrix laminates containing both E2 and NEA in combination were prepared as described above in Example 1 (A) except that the adhesive used was DUROTAK 87-2516 instead of TSR. Necessary amounts of E2 was pre-dissolved in IPA and added to the casting solution along with NEA and sorbitan monooleate to yield various compositions as shown in Table XI below (Formulations 9-20).

TABLE XI

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID DUROTAK NEA E2 ARLACEL 87-2516 Adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

9. Lot # 013096-6 87.0 2.0 1.0 10.0

10. Lot # 013096-7 85.0 4.0 1.0 10.0

1 1. Lot # 013096-8 83.0 6.0 1.0 10.0

12. Lot # 013096-9 81.0 8.0 1.0 10.0

13. Lot # 013096-1 1 86.0 2.0 2.0 10.0

14. Lot # 013096-12 84.0 4.0 2.0 10.0

15. Lot # 013096-13 82.0 6.0 2.0 10.0

16. Lot # 013096-14 80.0 8.0 2.0 10.0

17. Lot # 013096-16 84.0 2.0 4.0 10.0

18. Lot # 013096-17 82.0 4.0 4.0 10.0

19. Lot # 013096-18 80.0 6.0 4.0 10.0

20. Lot # 013096-19 78.0 8.0 4.0 10.0

Control Formulation

5. Control # 95Z098 77.6* 6.0 6.4 10.0 NEA/E2 Control Formulation

*TSR Adhesive, Sekisui Chemical Co., Osaka, Japan.

These formulations were evaluated for in vitro skin flux along with a control formulation (Formulation 5) described above in Examples 3(A) and 3(B). The in vitro skin fluxes for the test formulations and the control formulations on the same skins are presented in Table XII. The normalized NEA and E2 flux ratios are plotted in Figures 7 and 8 respectively.

As can be seen from the data in Figure 7, a 4 fold increase in the NEA loading did not result in a proportional 4 fold increase in the flux of NEA. Similarly, a 4 fold increase in E2 loading did not result in a proportional 4 fold increase in E2 flux (Figure 8). The normalized in vitro NEA and E2 fluxes therefore do not display linear and Fickian

dependence on the steroid concentration in the presence of each other in matrices made with DUROTAK 87-2516 adhesive over the range of drug loadings investigated (0-8 % NEA loading and 0-4 % E2 loading).

TABLE XII

CUMULATIVE NEA & E2 PERMEATION IN 24 hr (Q24 -μg/cm ² /24 hr)

#of

Formulation ID skins/cells NfcA data F2 data

Q24* 024 Ratio 024* Q24

(Control)" (Control) ^λ

Test Formulations

9 Lot #013096-6 3/12 46± 13 76±22 06±01 53± 15 33± 10 16±03

10 Lot #013096-7 3/12 94±21 76±22 12±02 53± 12 33 ± 10 16 ± 02

11 lot #013096-8 3/12 138±48 76±22 18±05 51 ± 17 33± 10 15±03

12 Lot #013096-9 3/12 175±56 76±22 23±06 51 + 17 33± 10 I 5±03

13 Lot #013096-11 3/12 25 ± 05 61 ±26 04±01 59± I 3 28± 12 22 + 06

14 Lot #013096-12 3/12 48± 12 61 ±26 08±03 53 14 28± 12 20±07

15 lot #013096-13 3/12 77± 18 61 ±26 14±05 55± 13 28± 12 21 ±08

16 lot #013096-14 3/12 78±27 61 ±26 13 ± 06 42± 16 28± 12 16±08

17 Lot #013096-16 3/12 28±07 81 ±24 04 ± 004 128± 27 32±08 40 ± 04

18 Lot #013096-17 3/12 56± 15 81±24 07±02 118± 30 32±08 38± 12

19 Lot #013096-18 3/12 74± 15 81 ± 24 09±02 96± 19 32±08 31 ±06

20 Lot #013096-19 3/12 82±22 81 ±24 11 ±02 69± 14 32±08 22±04

* 024-Cumulatιve amount permeated from test formulation in 24 hr x Q24 (C'ontrol)-Cumulatιve amount permeated from Control formulation in 24 hr on same skins as the test formulations

+ Skin flux normalized relative to Control formulation on an individual skin basis

The above data clearly shows that in vitro E2 and NEA fluxes are influenced by the presence of each other, do not follow Fickian laws of diffusion, and are not proportional to the steroid concentration in the matrix laminates made with DUROTAK 87-2516 adhesive, over the range of steroid concentrations investigated (0-8 % NEA loading and 0-4% E2 loading). The independent flux of the two steroids in the presence of each other and proportionality in skin flux as a function of steroid in the matrix is apparently unique to NVP containing acrylic copolymer adhesive.

EXAMPLE 3

(A) Testosterone (TS) only matrices:

Testosterone (TS, Upjohn Company, Kalamazoo, MI) only matrices were prepared as described in Example 1(A) except that the steroid used was TS instead of NEA. Necessary amounts of TS were pre-dissolved in IPA and added to the casting solution to yield various compositions as shown in Table XIII below (Formulations 1-3).

TABLE XIII

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID TSR TS E2 ARLACEL

Adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

1. Lot # 012496-1 87.5 2.5 0.0 10.0

2. Lot # 012496-2 86.25 3.75 0.0 10.0

3. Lot # 012496-3 85.0 5.0 0.0 10.0

Control Formulation

4. Control # 95Z082 75.75 3.75 10.5 10.0 TS Control Formulation

These formulations were evaluated for in vitro skin flux using a formulation containing TS as a control (Formulation 4). The in vitro skin fluxes for the three test formulations and the control formulations on the same skins are presented in Table XIV. The normalized flux ratios are plotted in Figure 9.

As can be seen from the data presented in Figure 9, a 2 fold increase in the drug loading results in a proportional 2 fold increase in the flux. The normalized in vitro TS fluxes therefore display linear and Fickian dependence on the steroid concentration between 2.5-5 % w/w loading in matrices made with TSR adhesive.

TABLE XIV

CUMULATIVE TS PERMEATION IN 24 hr (Q24-μg/cm ² /24 hr)

Formulation ID # of Q24* Ratio ⁺ skins/cells

Test Formulations

1. Lot # 012496-1 3/12 15.6 ± 6.6 0.63 ± 0.09

2. Lot # 012496-2 3/12 25.7 ± 13.4 1.01 ± 0.20

3. Lot # 012496-3 3/12 31.8 ± 15.3 1.28 ± 0.29

* 024-Cumulatιve amount permeated from test formulation in 24 hr

+ Skin flux normalized relative to Control formulation on an individual skin basis Control = 25 8 + 10 5

(B) E2 only matrices

Matrix laminates containing E2 were prepared as described above in Example 1 (A) except that E2 was used as the drug instead of NEA. Necessary amounts of E2 were pre- dissolved in IPA and added to the casting solution to yield various compositions as shown in Table XV below (Formulations 5-7).

These formulations were evaluated for in vitro skin flux along with a E2 control formulation (Formulation 8). The in vitro skin fluxes for the three test formulations and the control formulation on the same skins are presented in Table XVI. The normalized flux ratio's are plotted in Figure 10.

As can be seen from the data presented in Figure 10, a 3.5 fold increase in the steroid concentration results in a proportional 3.5 fold increase in the flux. The normalized in vitro E2 fluxes therefore display linear and Fickian dependence on the steroid concentration between 3-10.5 % w/w loading in matrices made with TSR adhesive.

TABLE XV

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID TSR TS E2 ARLACEL Adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

5. Lot #012496-4 87.0 0.0 3.0 10.0

6. Lot #012496-8 84.0 0.0 6.0 10.0

7. Lot #012496-12 79.5 0.0 10.5 10.0

Control Formulation

8. Control #95Z135 93.5* 0.0 1.5 5.0 E2 Control Formulation

* Formulation made with DUROTAK 87-2070 Adhesive, National Starch and Chemical Company, Bπdgewater, Nl

TABLE XVI

CUMULATIVE E2 PERMEATION IN 24 hr (Q24-μg/cm ² /24 hr)

Formulation ID #of Q24* Ratio ^4" skins/cells

Test Formulations

5. Lot #012496-4 3/12 2.6 ±0.7 0.28 ± 0.06

6. Lot #012496-8 3/12 5.4 ±0.9 0.60 ±0.14

7. Lot #012496-12 3/12 9.3 ± 1.6 1.02± 0.18

* 024-Cumulatιve amount permeated from test formulation in 24 hr

+ Skin flux normalized relative to Control formulation on an individual skin basis Control ^: 113+20

(C) TS/E2 co flux matrices:

Matrix laminates containing both E2 and TS in combination were prepared as described above in Example 1 (A). Necessary amounts of E2 and TS were pre-dissolved in iso-propyl alcohol (IPA) and added to the casting solution along with sorbitan monooleate to yield various compositions as shown in Table XVII below (Formulations 9-17).

TABLE XVII

COMPOSITION OF FORMULATIONS EVALUATED

Formulation ID TSR TS E2 ARLACEL

Adhesive 80

% w/w % w/w % w/w % w/w

Test Formulations

9. Lot # 012496-5 84.5 2.5 3.0 10.0

10. Lot # 012496-6 83.25 3.75 3.0 10.0

1 1. Lot # 012496-7 82.0 5.0 3.0 10.0

12. Lot # 012496-9 81.5 2.5 6.0 10.0

13. Lot # 012496-10 80.25 3.75 6.0 10.0

14. Lot # 012496-1 1 79.0 5.0 6.0 10.0

15. Lot # 012496-13 77.0 2.5 10.5 10.0

16. Lot # 012496-14 75.75 3.75 10.5 10.0

17. Lot # 012496-15 74.5 5.0 10.5 10.0

Control Formulations

4. Control # 95Z082 75.75 3.75 10.5 10.0 TS Control Formulation

8. Control # 95Z 135 93.5* 0.0 1.5 5.0 E2 Control Formulation

* Formulation made with DUROTAK 87-2070 Adhesive. National Starch and Chemical Company, Bridgewater, NJ

These formulations were evaluated for in vitro skin flux along with TS and E2 control formulations (Formulations 4 and 8 respectively) described above in Examples 2(A) and 2(B) respectively. The in vitro skin fluxes for the test formulations and the

control formulations on the same skins are presented in Table XVIII below. The normalized TS and E2 flux ratios are plotted in Figures 11 and 12 respectively.

As can be seen from the data in Figure 1 1, a 2 fold increase in the TS loading results in a proportional 2 fold increase in the flux of TS. Similarly, a 3.5 fold increase in E2 concentration results in a proportional 3.5 fold increase in E2 flux (Figure 12). The normalized in vitro TS and E2 fluxes therefore display linear and Fickian dependence on the steroid concentration in the presence of each other in matrices made with the NVP containing TSR acrylic copolymer adhesive over the range of steroid concentrations investigated (0-5 % TS and 0-10.5 % E2). The slope of each of the three linear regression lines (normalized data for TS fluxes in the presence of E2, Figure 1 1) was compared statistically to the slope of the regression line for the TS formulations without E2 using a Student's t-test. The results indicated that there was no statistically significant difference (p> 0.10) in slopes between each of the three linear regression lines (for TS formulations containing E2) relative to the slope of the regression line for the TS formulations without E2. This confirms that the presence of E2 in the matrix does not affect the flux of TS in systems made with the NVP containing TSR acrylic copolymer adhesive.

The slope of each of the three linear regression lines (normalized data for E2 fluxes in the presence of TS, Figure 12) was compared statistically to the slope of the regression line for the E2 formulations without TS using a Student's t-test. The results indicated that there was no statistically significant difference (p> 0.10) in slopes between each of the three linear regression lines (for E2 formulations containing TS) relative to the slope of the regression line for the E2 formulations without TS. This confirms that the presence of TS in the matrix does not affect the flux of E2 in systems made with the NVP containing TSR acrylic copolymer adhesive.

The above data clearly shows that over the range of steroid concentrations investigated (0-5 % TS loading and 0-10.5 % E2), in vitro E2 and TS fluxes are independent of each other, follow Fickian laws of diffusion, and are proportional to the steroid concentration in the matrix laminates made with the NVP containing acrylic copolymer adhesive, TSR. The independence in fluxes for the two steroids in the presence of each other and proportionality in in vitro skin flux as a function of steroid concentration in the matrix again appears to be unique to the NVP acrylic copolymer adhesive.

TABLE XVIII CUMULATIVE TS & E2 PERMEATION IN 24 hr (Q24 -μg/cm ² /24 hr)

#of

Formulation ID skins/ cells TS Data E2 Data

Q24* Q24 Ratio Q24* Q24 Ratιo ⁺ (ControI) ^λ (Control)"

Test Formulations Lot #012496-5 3/12 142±61 230 ± 93 060 ± 01 14 + 05 52+ 18 027 ± 004

10 Lot #012496-6 3/12 242 + 90 230 + 93 106 ± 02 15±05 52+ 18 029 ± 006

11 Lot #012496-7 3/12 281 ± 95 230 + 93 124 + 02 14 + 04 52+ 18 026 ± 004

12 lot #012496-9 3/10 132 + 29 232 ± 84 058 + 02 26 + 05 47+16 057 + 014

13 Lot #012496-10 3/10 182 + 45 232 ± 84 081+03 24 + 06 47+16 054 ± 020

14 Lot #012496-11 3/11 263 ± 79 232 ± 84 113 + 02 27i 07 47+16 058 + 012

15 Lot #012496- 13 3/12 144 + 78 225+100 061 ±01 62 + 32 59 + 24 099 + 019

16 lot #012496-14 3/12 214+ 109 225+100 093 ± 01 62 + 31 59±24 099 ± 015

171 ot# 012496-15 3/12 266+106 225+100 120+03 60 + 22 59 + 24 10 ± 028

* 024-Cumulatιve amount permeated from test formulation in 24 hr

\ Q24 (Control)-Cumulatιvc amount permeated from Control formulation in 24 hr on same skins as the test formulations

+ Skin fluλ normalized relative to Control formulation on an individual skin basis