Draganoiu Elena Simona

UNIVERSITY OF CINCINNATI May 12 03 _____________ , 20 _____

Elena Simona Draganoiu I,______________________________________________, hereby submit this as part of the requirements for the degree of:

Doctor of Philosophy ________________________________________________

in: Pharmaceutical Sciences (Industrial Pharmacy) ________________________________________________

It is entitled: EVALUATION OF KOLLIDON® SR FOR pH-INDEPENDENT ________________________________________________ EXTENDED RELEASE MATRIX SYSTEMS ________________________________________________

________________________________________________ ________________________________________________

Approved by: Dr. Adel Sakr, Chairperson ________________________ Dr. Hussein Al-Khalidi ________________________ Dr. Bernadette D'Souza ________________________ Dr. Ronald Millard ________________________ Dr. Apryll Stalcup ________________________

EVALUATION OF KOLLIDON® SR FOR pH-INDEPENDENT EXTENDED RELEASE MATRIX SYSTEMS A dissertation submitted to the Division of Research and Advanced Studies of the University of Cincinnati in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Industrial Pharmacy Program Division of Pharmaceutical Sciences College of Pharmacy 2003 by Elena Simona Draganoiu, B.Sc. Pharm. University of Medicine and Pharmacy ‘Gr. T. Popa’ Iasi, Romania

Committee Chair Adel Sakr, Ph.D.

Abstract The characteristics of a new Polyvinylacetate/Povidone based excipient, Kollidon® SR were evaluated for application in extended release matrix tablets. The effects of the following formulation and process variables on tablet properties and drug release were tested: Kollidon® SR concentration in the tablet, addition of external binder for wet granulation, presence of an enteric polymer in the matrix, method of manufacturing and compression force. The similarities in release profiles were evaluated by applying the model independent f2 similarity factor. A pilot bioequivalence study was performed in human volunteers to confirm in vivo the extended release characteristics of the propranolol tablets manufactured with Kollidon® SR.

It was found that Kollidon® SR is suitable for pH-independent extended release matrix tablets. A minimum concentration of 30% polymer was necessary to achieve a coherent matrix, able to extend the release of the incorporated drugs. Increasing the Kollidon® SR concentration in the tablet led to a slower drug release. Drug release followed square root of time dependent kinetics, thus indicating a diffusion-controlled release mechanism. The drug release was influenced by the aqueous solubility of the drug. The drug release rate was faster for wet granulation than direct compression, thus making direct compression the method of choice for manufacturing Kollidon® SR extended release systems. It was found that Kollidon® SR was the main release controlling agent in the presence of an external binder or enteric polymer in the matrix. A significant

reduction in the dissolution rates associated with an increase in tablet hardness was observed during the stability test under accelerated conditions.

The developed propranolol matrix tablets formulation was compared in a pilot bioequivalence study to the reference listed product (Inderal® LA capsules). It was found that the two products were not bioequivalent according to the FDA bioequivalence criteria. The tablets had higher bioavailability than the capsules as shown by higher Cmax and AUC 0-24h. For the developed tablet formulation the higher initial plasma concentrations correlated with the faster initial release observed in vitro.

It was concluded that Kollidon® SR is a potentially useful excipient for the production of pH-independent extended release matrix tablets.

Acknowledgments My gratitude to the committee members for their valuable comments and advice and for guiding my efforts to complete this research. My deepest gratitude to my advisor, Dr. Adel Sakr, for his constant professional, financial and emotional support. Special thanks for giving me the chance to be one of the researchers he has fostered in the Industrial Pharmacy Program (Family), for mentoring my professional steps, for all the meetings I have participated, for the interactions with the professional world I have had through him. Most of all, for his continuous encouragement, confidence in me and friendship. To Dr. Hussein AlKhalidi for guiding me explore the ‘statistics world’ through courses and valuable advice in preparation of the comprehensive exam and dissertation. To Dr. Bernadette D’Souza, my special thanks for her major contribution in the Bioequivalence study and for her kind and warm support. To Dr. Ronald Millard for sharing his knowledge and for being an academic model. To Dr. Apryll Stalcup for providing guidance through the Chemical Separation course and interactions during the analytical work. Special thanks to Dr. Karl Kolter and BASF Germany for the donation of Kollidon® SR and all the support they provided.

To my colleagues Ehab, Hatim, Himanshu, Juan, Julia, Murad, Oliver, Rajesh, Shadi who volunteered for the Bioequivalence study. I highly appreciate their generous support. To Dr. Lubna Izzatullah and Mr. Nosa Ekhator (Veterans Affair Medical Center) for their kind help during the Bioequivalence study. To Dr. Pankaj Desai for allowing me to use some equipment in his lab, and for valuable discussions. Thanks to Murad Melhem for useful suggestions in the bioanalytical work and help with the WinNonlin software. Special thanks to Dr. James Ebel for the great experience of two summer internships in the Procter & Gamble Health Care Research Center, and for his assistance with equipment and advice during my Ph.D. research. To Dr. Ronald Shoup (BAS Analytics) for the loan of analytical equipment without knowing me; his generosity impressed me. To the College of Pharmacy for providing me with the University Graduate Scholarship. To my professors at the University of Cincinnati and at the University of Medicine and Pharmacy, Iasi, Romania, for doing such a good job in educating generations. To my colleagues in Industrial Pharmacy Program, for being such good company and for all I have learnt from them: Julia, Laxmi, Susan, Ehab, Hamid, Hatim, Himanshu, John, Juan, and Mohamed. Special thanks to some of them for their true friendship.

To my Romanian friends from here and home, for being such good friends as one could have and for all the memories we share. To my parents and my family for their love. They are the reason for what I am now. This work is dedicated to them.

Contents 1.

Introduction............................................................................................ 9

1.1.

Extended release matrix systems ............................................................ 9

1.2.

Mechanisms of drug release from matrix systems................................. 11

1.2.1.

Dissolution controlled systems .............................................................. 11

1.2.2.

Diffusion controlled systems .................................................................. 12

1.2.3.

Bioerodible and combination diffusion and dissolution systems ............ 19

1.3.

Impact of the formulation and process variables on the drug release from extended release matrix systems .................................................. 24

1.3.1.

Formulation variables ............................................................................ 24

1.3.2.

Process variables .................................................................................. 37

1.4.

Rationale for studying Kollidon® SR as extended release matrix excipient ................................................................................................ 40

1.5.

Kollidon® SR - background ................................................................... 41

1.6.

Propranolol extended release formulations ........................................... 45

2.

Objective, hypothesis and specific aims........................................... 52

2.1.

Objective................................................................................................ 52

2.2.

Hypothesis ............................................................................................. 52

2.3.

Specific aims ......................................................................................... 52

3.

Experimental ........................................................................................ 54

3.1.

Materials and supplies ........................................................................... 54

3.2.

Equipment ............................................................................................. 57

3.3.

Tablet composition................................................................................. 59

3.4.

Tablet manufacture................................................................................ 61

3.5.

Tablet testing ......................................................................................... 65

3.6.

Experimental design and methodology.................................................. 68

3.6.1.

Propranolol 10 mg tablets...................................................................... 68

-1-

3.6.1.1. Manufacture of propranolol 10mg tablets by direct compression........... 68 3.6.1.2. Manufacture of propranolol 10mg tablets by wet granulation ................ 69 3.6.1.3. Drug release profiles from propranolol 10mg matrix tablets manufactured with Eudragit® RSPO ..................................................... 71 3.6.1.4. Testing of propranolol 10mg tablets....................................................... 71 3.6.2.

Buspirone 10 mg tablets ........................................................................ 71

3.6.3.

Propranolol 80 mg tablets...................................................................... 73

3.6.3.1. Manufacture of propranolol 80mg tablets with 40-60% Kollidon® SR ... 73 3.6.3.2. Testing of propranolol 80mg tablets with 40-60% Kollidon® SR............ 74 3.6.3.3. Manufacture of propranolol 80mg tablets with 70% polymer (Kollidon® SR alone or in combination with Eudragit® L100-55)........... 74 3.6.3.4. Testing

of

propranolol

80mg

tablets

with

70%

polymer

(Kollidon® SR alone or in combination with Eudragit® L100-55)........... 75 3.6.3.5. Testing of Inderal® LA capsules (reference listed drug product) ........... 76 3.6.3.6. Selection of propranolol 80mg formulation for pilot bioequivalence study ...................................................................................................... 76 3.6.3.7. Testing of propranolol 80mg tablets for the pilot bioequivalence study ...................................................................................................... 76 3.6.4.

Pilot bioequivalence study ..................................................................... 77

3.6.4.1. Design and methodology ....................................................................... 77 3.6.4.2. Analysis of propranolol in plasma .......................................................... 81 3.6.4.3. Pharmacokinetic and statistical analysis................................................ 83 4.

Results and Discussions .................................................................... 84

4.1.

Propranolol 10 mg tablets...................................................................... 84

4.1.1.

Effect of Kollidon® SR on drug release from propranolol 10mg tablets manufactured by direct compression ......................................... 84

4.1.2.

Effect of Kollidon® SR on drug release from propranolol 10mg tablets manufactured by wet granulation ............................................... 88

4.1.3.

Effect of external binder on drug release from propranolol 10mg tablets manufactured by wet granulation ............................................... 93

-2-

4.1.4.

Effect of dissolution medium on drug release from propranolol 10mg matrix tablets ............................................................................... 95

4.1.5.

Drug release profiles from matrix tablets with Eudragit® RSPO.......... 103

4.2.

Buspirone 10mg tablets ....................................................................... 105

4.2.1.

Effect of Kollidon® SR and compression force on physical properties and drug release of buspirone 10mg tablets....................... 105

4.2.2.

Effect of dissolution medium on drug release from buspirone 10mg tablets .................................................................................................. 113

4.3.

Propranolol 80mg tablets..................................................................... 118

4.3.1.

Effect of Kollidon® SR and compression force on physical properties and drug release from propranolol 80mg tablets ................ 118

4.3.2.

Effect of dissolution medium on drug release from propranolol 80mg tablets ........................................................................................ 125

4.3.3.

Effect of Kollidon® SR – Eudragit® L100-55 combination on drug release from propranolol 80mg tablets ................................................ 127

4.3.4.

Comparison of the propranolol 80 mg tablet formulations with the reference listed capsule product .......................................................... 133

4.3.5.

Effect of storage conditions on propranolol 80 mg tablets physical properties and drug release................................................................. 141

4.4.

Evaluation of bioequivalence of propranolol 80 mg matrix tablets to Inderal® LA capsules........................................................................... 145

4.4.1.

Analysis of propranolol in plasma ........................................................ 145

4.4.2.

Subjects monitoring during the pilot bioequivalence study .................. 145

4.4.3.

Pharmacokinetic and statistical analysis.............................................. 146

5.

Conclusions ....................................................................................... 160

6.

References ......................................................................................... 162

7.

Appendix 1 ......................................................................................... 171

8.

Appendix 2 ......................................................................................... 197

-3-

List of Figures Figure 1. Schematic representation of a matrix release system ......................... 14 Figure 2. The fronts in a swellable HPMC matrix................................................ 22 Figure 3. Process flow chart for tablets manufactured by direct compression .... 62 Figure 4. Process flow chart for tablets manufactured by wet granulation.......... 63 Figure 5. Effect of Kollidon® SR on drug release in water from propranolol 10mg tablets manufactured by direct compression .............................. 86 Figure 6. Effect of Kollidon® SR on diffusion controlled drug release in water from propranolol 10mg tablets manufactured by direct compression ......................................................................................... 87 Figure 7. Effect of Kollidon® SR on drug release in water from propranolol 10mg tablets manufactured by wet granulation .................................... 90 Figure 8. Effect of Kollidon® SR on diffusion controlled drug release in water from propranolol 10mg tablets manufactured by wet granulation............................................................................................ 91 Figure 9. Effect of external binder on drug release in water from propranolol 10mg tablets with 30% and 50% Kollidon® SR .................................... 94 Figure 10. Effect of dissolution medium on drug release from propranolol 10mg tablets with 30% Kollidon® SR manufactured by direct compression ......................................................................................... 96 Figure 11. Effect of dissolution medium on drug release from propranolol 10mg tablets with 30% Kollidon® SR manufactured by wet granulation............................................................................................ 97 Figure 12. Effect of dissolution medium on drug release from propranolol 10mg tablets with 40% Kollidon® SR manufactured by direct compression ......................................................................................... 98 Figure 13. Effect of dissolution medium on drug release from propranolol 10mg tablets with 40% Kollidon® SR manufactured by wet granulation............................................................................................ 99 Figure 14. Effect of dissolution medium on drug release from propranolol 10mg tablets with 50% Kollidon® SR manufactured by direct compression ....................................................................................... 100 Figure 15. Effect of dissolution medium on drug release from propranolol 10mg tablets with 50% Kollidon® SR manufactured by wet granulation.......................................................................................... 101 Figure 16. Effect of Eudragit® RSPO on drug release in water from propranolol 10mg tablets .................................................................... 104 Figure 17. Effect of Kollidon® SR concentration and compression force on the hardness of buspirone 10mg tablets............................................. 106 -4-

Figure 18. Effect of compression force on drug release from buspirone 10mg tablets with 10 - 30% Kollidon® SR .......................................... 108 Figure 19. Effect of compression force on drug release from buspirone 10mg tablets with 40 - 60% Kollidon® SR .......................................... 109 Figure 20. Effect of Kollidon® SR on drug release from buspirone 10mg tablets ................................................................................................. 111 Figure 21. Effect of Kollidon® SR on diffusion controlled drug release from buspirone 10mg tablets ...................................................................... 112 Figure 22. Effect of dissolution medium on drug release from buspirone 10mg tablets with 30% Kollidon® SR ................................................. 114 Figure 23. Effect of dissolution medium on drug release from buspirone 10mg tablets with 40% Kollidon® SR ................................................. 115 Figure 24. Effect of dissolution medium on drug release from buspirone 10mg tablets with 50% Kollidon® SR ................................................. 116 Figure 25. Effect of dissolution medium on drug release from buspirone 10mg tablets with 60% Kollidon® SR ................................................. 117 Figure 26. Effect of Kollidon® SR and compression force on the hardness of propranolol 80mg tablets ................................................................ 121 Figure 27. Effect of Kollidon® SR and compression force on drug release in water from propranolol 80mg tablets .................................................. 122 Figure 28. Effect of Kollidon® SR on diffusion controlled drug release from propranolol 80mg tablets .................................................................... 124 Figure 29. Effect of dissolution medium on drug release from propranolol 80mg tablets ....................................................................................... 126 Figure 30. Effect of Kollidon® SR and Eudragit® L100-55 combination on drug release in water from propranolol 80mg tablets.......................... 129 Figure 31. Effect of Kollidon® SR and Eudragit® L100-55 combination on drug release in 0.1N HCl from propranolol 80mg tablets .................... 130 Figure 32. Effect of Kollidon® SR and Eudragit® L100-55 combination on drug release in pH 6.8 buffer from propranolol 80mg tablets.............. 131 Figure 33. Propranolol release in water over 48 hours from tablets manufactured with 70% Kollidon® SR................................................ 132 Figure 34. Comparison of drug release from propranolol 80 mg tablets with 60 and 70% Kollidon® SR and Inderal® LA ....................................... 134 Figure 35. Compression and ejection forces recorded during manufacturing of propranolol 80 mg tablets with 65% Kollidon® SR ......................... 136 Figure 36. Comparison of the drug release profiles from propranolol 80mg tablets with 65% Kollidon® SR and Inderal® LA ................................ 138

-5-

Figure 37. Reproducibility of propranolol 80 mg tablets formulation with 65% Kollidon® SR ...................................................................................... 140 Figure 38. Effect of storage on drug release from propranolol 80 mg tablets – ICH long term stability conditions .................................................... 143 Figure 39. Effect of storage on drug release from propranolol 80 mg tablets – ICH accelerated stability conditions ................................................. 144 Figure 40. Plasma levels of propranolol following administration – subject #1 ........................................................................................... 147 Figure 41. Plasma levels of propranolol following administration – subject #2 ........................................................................................... 148 Figure 42. Plasma levels of propranolol following administration – subject #3 ........................................................................................... 149 Figure 43. Plasma levels of propranolol following administration – subject #4 ........................................................................................... 150 Figure 44. Plasma levels of propranolol following administration – subject #5 ........................................................................................... 151 Figure 45. Plasma levels of propranolol following administration – subject #6 ........................................................................................... 152 Figure 46. Plasma levels of propranolol following administration – subject #7 ........................................................................................... 153 Figure 47. Plasma levels of propranolol following administration – subject #8 ........................................................................................... 154 Figure 48. Plasma levels of propranolol following administration (mean ± SEM)................................................................................................... 155

-6-

List of Tables Table 1. Application of matrix for drug delivery systems..................................... 24 Table 2. Pharmacokinetic properties of propranolol ........................................... 47 Table 3. Propranolol 10mg matrix tablets formulation ........................................ 69 Table 4. Propranolol 10mg matrix tablets formulations with external binders ..... 70 Table 5. Propranolol 10mg tablets formulated with Eudragit® RSPO................. 70 Table 6. Formulation of buspirone 10mg tablets................................................. 72 Table 7. Formulation of propranolol 80mg tablets with 40-60% Kollidon® SR ... 74 Table 8. Formulation of propranolol 80mg tablets with 70% polymer ................. 75 Table 9. Stability study design ............................................................................ 77 Table 10. Analytical method for analysis of propranolol in plasma ..................... 82 Table 11. Effect of Kollidon® SR on physical properties of propranolol 10mg tablets manufactured by direct compression ........................................ 84 Table 12. Regression parameters of the diffusion drug release curves for propranolol 10mg tablets manufactured by direct compression............ 85 Table 13. Effect of Kollidon® SR on the physical properties of Propranolol 10mg tablets manufactured by wet granulation .................................... 89 Table 14. Regression parameters of the diffusion drug release curves for propranolol 10mg tablets manufactured by wet granulation ................. 92 Table 15. f2 values - effect of dissolution medium on drug release from propranolol 10mg tablets .................................................................... 102 Table 16. Physical properties of buspirone 10mg tablets ................................. 107 Table 17. f2 values - effect of compression force on drug release from buspirone 10mg tablets ...................................................................... 110 Table 18. Regression parameters of the diffusion drug release curves for buspirone 10mg tablets ...................................................................... 113 Table 19. f2 values – effect of dissolution medium on drug release from buspirone 10mg tablets ...................................................................... 118 Table 20. Effect of compression force and Kollidon® SR concentration on physical properties of propranolol 80mg tablets ................................. 119 Table 21. f2 values – effect of compression force on drug release from propranolol 80mg tablets .................................................................... 123 Table 22. Regression parameters of the diffusion drug release curves in water from propranolol 80mg tablets .................................................. 125 Table 23. Composition of the propranolol 80mg tablets formulation used in the pilot bioequivalence study............................................................. 135 -7-

Table 24. Characteristics of propranolol 80mg tablets used in the pilot bioequivalence study .......................................................................... 137 Table 25. Drug release from the propranolol 80 mg tablets with 65% Kollidon® SR (used for the pilot bioequivalence study) ...................... 139 Table 26. Effect of storage on the hardness of propranolol 80 mg tablets........ 142 Table 27. Pharmacokinetic parameters after administration of propranolol 80mg tablets and Inderal® LA 80mg .................................................. 157 Table 28. Results of the bioequivalence testing using WinNonlin software ...... 158

-8-

1. Introduction

1.1. Extended release matrix systems Extended release dosage forms are formulated in such manner as to make the contained

drug

available

over

an

extended

period

of

time

following

administration. Expressions such as controlled-release, prolonged-action, repeataction and sustained-release have also been used to describe such dosage forms. A typical controlled release system is designed to provide constant or nearly constant drug levels in plasma with reduced fluctuations via slow release over an extended period of time. In practical terms, an oral controlled release should allow a reduction in dosing frequency as compared to when the same drug is presented as a conventional dosage form (Qiu and Zhang, 2000). A matrix device consists of drug dispersed homogenously throughout a polymer matrix.

Two major types of materials are used in the preparation of matrix devices (Venkatraman et al., 2000): ♦

Hydrophobic carriers: •

Digestible base (fatty compounds) – glycerides - glyceryltristearate, fatty alcohols, fatty acids, waxes - carnauba wax (Chiao and Robinson, 1995);

•

Nondigestible

base

(insoluble

plastics)

-

methylacrylate

methylmethacrylate, polyvinyl chloride, polyethylene, ethyl cellulose;

-9-

-

♦

Hydrophilic polymers – methyl cellulose, sodium carboxy methyl cellulose, hydroxypropyl methyl cellulose, sodium alginate, xanthan gum, polyethylene oxide, carbopols.

Matrix systems offer several advantages: •

easy to manufacture

•

versatile, effective, low cost

•

can be made to release high molecular weight compounds

•

since the drug is dispersed in the matrix system, accidental leakage of the total drug component is less likely to occur, although occasionally, cracking of the matrix material can cause unwanted release.

Disadvantages of the matrix systems: •

the remaining matrix must be removed after the drug has been released

•

the drug release rates vary with the square root of time. Release rate continuously diminishes due to an increase in diffusional resistance and/or a decrease in effective area at the diffusion front (Qiu and Zhang, 2000). However, a substantial sustained effect can be produced through the use of very slow release rates, which in many applications are indistinguishable from zero-order (Jantzen and Robinson, 1996).

- 10 -

1.2. Mechanisms of drug release from matrix systems The release of drug from controlled devices is via dissolution or diffusion or a combination of the two mechanisms.

1.2.1. Dissolution controlled systems A drug with slow dissolution rate will demonstrate sustaining properties, since the release of the drug will be limited by the rate of dissolution. In principle, it would seem possible to prepare extended release products by decreasing the dissolution rate of drugs that are highly water-soluble. This can be done by: •

preparing an appropriate salt or derivative

•

coating the drug with a slowly dissolving material – encapsulation dissolution control

•

incorporating the drug into a tablet with a slowly dissolving carrier – matrix dissolution control (a major disadvantage is that the drug release rate continuously decreases with time) (Jantzen and Robinson, 1996).

The dissolution process can be considered diffusion-layer-controlled, where the rate of diffusion from the solid surface to the bulk solution through an unstirred liquid film is the rate-determining step. The dissolution process at steady-state is described by the Noyes-Whitney equation:

D dC = k D ⋅ A ⋅ (Cs − C ) = ⋅ A ⋅ (Cs − C ) h dt

(1)

- 11 -

where: dC/dt dissolution rate kd

the dissolution rate constant (equivalent to the diffusion coefficient divided

by the thickness of the diffusion layer D/h) D

diffusion coefficient

Cs

saturation solubility of the solid

C

concentration of solute in the bulk solution

Equation (1) predicts that the rate of release can be constant only if the following parameters are held constant:

•

surface area

•

diffusion coefficient

•

diffusion layer thickness

•

concentration difference.

These parameters, however, are not easily maintained constant, especially surface area, and this is the case for combination diffusion and dissolution systems (Jantzen and Robinson, 1996).

1.2.2. Diffusion controlled systems Diffusion systems are characterized by the release rate of a drug being dependent on its diffusion through an inert membrane barrier (Higuchi, 1963). Usually, this barrier is an insoluble polymer. In general, two types or subclasses of diffusional systems are recognized: reservoir devices and matrix devices (Jantzen and Robinson, 1996). It is very common for the diffusion-controlled

- 12 -

devices to exhibit a non-zero order release rate due to an increase in diffusional resistance and a decrease in effective diffusion area as the release proceeds. (Venkatraman et al, 2000).

Diffusion in matrix devices In this model, drug in the outside layer exposed to the bathing solution is dissolved first and then diffuses out of the matrix. This process continues with the interface between the bathing solution and the solid drug moving toward the interior. It follows obviously that for this system to be diffusion controlled, the rate of dissolution of drug particles within the matrix must be much faster than the diffusion rate of dissolved drug leaving the matrix (Jantzen and Robinson, 1995). Derivation of the mathematical model to describe this system involves the following assumptions: a) a pseudo-steady state is maintained during drug release; b) the diameter of the drug particles is less than the average distance of drug diffusion through the matrix; c) the diffusion coefficient of drug in the matrix remains constant (no change occurs in the characteristics of the polymer matrix (Jantzen and Robinson, 1995); d) the bathing solution provides sink conditions at all times; e) no interaction occurs between the drug and the matrix;

- 13 -

f) the total amount of drug present per unit volume in the matrix is substantially greater than the saturation solubility of the drug per unit volume in the matrix (excess solute is present) (Chiao and Robinson, 1995); g) only the diffusion process occurs (Qiu and Zhang, 2000).

Depleted Matrix Zone

Drug Solid Drug

Cs

dh

“Ghost” Matrix x=0

x=h

Figure 1. Schematic representation of a matrix release system

For a homogenous monolithic matrix system (Jantzen and Robinson, 1996), corresponding to the schematic in Figure 1 – page 14, the release behavior can be described by the following equation:

dM C = C0 ⋅ dh − s dh 2

(2)

where dM

change in the amount of drug released per unit area

dh

change in the thickness of the zone of matrix that has been depleted of

drug C0

total amount of drug in a unit volume of matrix

Cs

saturated concentration of the drug within the matrix.

- 14 -

From diffusion theory:

dM =

Dm ⋅ Cs ⋅ dt h

(3)

where Dm is the diffusion coefficient in the matrix. By combining equations (2) and (3):

M = [Cs ⋅ Dm ⋅ (2C0 − Cs ) ⋅ t ]1 / 2

(4)

When the amount of drug is in excess of the saturation concentration, (C0 >>Cs)

M = [2Cs ⋅ Dm ⋅ C0 ⋅ t ]1 / 2

(5)

That indicates that the amount of drug released is a function of square root of time. Drug release from a porous monolithic matrix involves the simultaneous penetration of surrounding liquid, dissolution of drug and leaching out of the drug through tortuous interstitial channels and pores. The volume and length of the openings must be accounted for in the drug release from a porous or granular matrix:

M = [ Ds ⋅ Ca ⋅

p ⋅ (2C0 − p ⋅ Ca ) ⋅ t ]1 / 2 T

(6)

where: p

porosity of the matrix

t

tortuosity

Ca

solubility of the drug in the release medium

Ds

diffusion coefficient in the release medium.

Similarly for pseudo-steady state (C0 >>Cs):

- 15 -

M = [2 Ds ⋅ C a ⋅ C 0 ⋅

p 1/ 2 t] T

(7)

The porosity is the fraction of matrix that exists as pores or channels into which the surrounding liquid can penetrate. It is the total porosity of the matrix after the drug has been extracted; it consists of initial porosity due to the presence of air or void space in the matrix before the leaching process begins as well as the porosity created by extracting the drug and the water-soluble excipients. p = pa +

C0

ρ

+

C ex

(8)

ρ ex

where ρ is the drug density and ρex and Cex are the density and the concentration of water-soluble excipient respectively. In a case where no water-soluble excipient is used in the formulation and initial porosity is much smaller than porosity created by drug extraction, total porosity becomes: p=

C0

(9)

ρ

Hence the release equations can be written as:

M = [ Ds ⋅ Ca ⋅

p ⋅ (2C0 − p ⋅ Ca ) ⋅ t ]1 / 2 T

M = [2 Ds ⋅ C a ⋅ C 0 ⋅

(10)

p 1/ 2 t] T

(11)

For purpose of data treatment, equation (6) can be reduced to:

M = k ⋅ t1 / 2

(12)

where k is a constant, so that the amount of drug released versus the square root of time will be linear, if the release of drug from matrix is diffusion-controlled. If

- 16 -

this is the case, one may control the release of drug from a homogeneous matrix system by varying the following parameters:

•

initial concentration of drug in the matrix

•

porosity

•

tortuosity

•

polymer system forming the matrix

•

solubility of the drug (Jantzen and Robinson, 1996, Chiao and Robinson, 1995).

In a hydrophilic matrix, there are two competing mechanisms involved in the drug release: Fickian diffusional release and relaxation release. Diffusion is not the only pathway by which a drug is released from the matrix; the erosion of the matrix following polymer relaxation contributes to the overall release. The relative contribution of each component to the total release is primarily dependent on the properties of a given drug. For example, the release of a sparingly soluble drug from hydrophilic matrices involves the simultaneous absorption of water and desorption of drug via a swelling-controlled diffusion mechanism. As water penetrates into a glassy polymeric matrix, the polymer swells and its glass transition temperature is lowered. At the same time, the dissolved drug diffuses through this swollen rubbery region into the external releasing medium. This type of diffusion and swelling does not generally follow a Fickian diffusion mechanism (Qiu and Zhang, 2000). Peppas (1985) introduced a semi-empirical equation to describe drug release behavior from hydrophilic matrix systems:

- 17 -

Q = k ⋅tn

(13)

where Q is the fraction of drug released in time t, k is the rate constant incorporating characteristics of the macromolecular network system and the drug and n is the diffusional exponent. It has been shown that the value of n is indicative of the drug release mechanism. For n=0.5, drug release follows a Fickian diffusion mechanism that is driven by a chemical potential gradient. For n=1 drug release occurs via the relaxational transport that is associated with stresses and phase transition in hydrated polymers. For 0.5
In order to describe relaxational transport, Peppas and Sahlin (1989) introduced a second term in equation (13): Q = k1 ⋅ t n + k 2 ⋅ t 2 n

(14)

where k1 and k2 are constants reflecting the relative contributions of Fickian and relaxation mechanisms. In the case the surface area is fixed, the value of n should be 0.5 and equation (14) becomes: Q = k1 ⋅ t 0.5 + k 2 ⋅ t

(15)

where the first and second term represent drug release due to diffusion and polymer erosion, respectively (Qiu and Zhang, 2000).

- 18 -

1.2.3. Bioerodible

and

combination

diffusion

and

dissolution

systems Strictly speaking, therapeutic systems will never be dependent on dissolution or diffusion only. In practice, the dominant mechanism for release will overshadow other processes enough to allow classification as either dissolution rate-limited or diffusion-controlled release (Jantzen and Robinson, 1996).

As a further complication these systems can combine diffusion and dissolution of both the drug and the matrix material. Drugs not only can diffuse out of the dosage form, as with some previously described matrix systems, but also the matrix itself undergoes a dissolution process. The complexity of the system arises from the fact that as the polymer dissolves the diffusional path length for the drug may change. This usually results in a moving boundary diffusion system. Zero-order release is possible only if surface erosion occurs and surface area does not change with time.

Swelling-controlled matrices exhibit a combination of both diffusion and dissolution mechanisms. Here the drug is dispersed in the polymer, but instead of an insoluble or non-erodible polymer, swelling of the polymer occurs. This allows for the entrance of water, which causes dissolution of the drug and diffusion out of the swollen matrix. In these systems the release rate is highly dependent on the polymer-swelling rate and drug solubility. This system usually

- 19 -

minimizes burst effects, as rapid polymer swelling occurs before drug release (Jantzen and Robinson, 1996).

With regards to swellable matrix systems, different models have been proposed to describe the diffusion, swelling and dissolution processes involved in the drug release mechanism (Siepman and Kranz, 2000, Siepman et al., 1999a, Siepman et al., 1999b, Siepman et al., 1999c, Peppas and Colombo, 1997, Colombo et al., 1999, Colombo et al., 1996, Colombo et al., 1995, Colombo et al., 1992, Wan et al., 1995). However the key element of the drug release mechanism is the forming of a gel layer around the matrix, capable of preventing matrix disintegration and further rapid water penetration.

When a matrix that contains a swellable glassy polymer comes in contact with a solvent or swelling agent, there is an abrupt change from the glassy to the rubbery state, which is associated with the swelling process. The individual polymer chains, originally in the unperturbed state absorb water so that their endto-end distance and radius of gyration expand to a new solvated state. This is due to the lowering of the transition temperature of the polymer (Tg), which is controlled by the characteristic concentration of the swelling agent and depends on both temperature and thermodynamic interactions of the polymer– water system. A sharp distinction between the glassy and rubbery regions is observed and the matrix increases in volume because of swelling. On a molecular basis, this phenomenon can activate a convective drug transport, thus increasing the

- 20 -

reproducibility of the drug release. The result is an anomalous non-Fickian transport of the drug, owing to the polymer-chain relaxation behind the swelling position. This, in turn, creates osmotic stresses and convective transport effects.

The gel strength is important in the matrix performance and is controlled by the concentration, viscosity and chemical structure of the rubbery polymer. This restricts the suitability of the hydrophilic polymers for preparation of swellable matrices. Polymers such as carboxymethyl cellulose, hydroxypropyl cellulose or tragacanth gum, do not form the gel layer quickly. Consequently, they are not recommended as excipients to be used alone in swellable matrices (Colombo et al., 2000, Colombo et al., 1996).

The swelling behavior of heterogeneous swellable matrices is described by front positions, where ‘front’ indicates the position in the matrix where the physical conditions sharply change. Three fronts are present (Colombo et al., 2000), as shown in Figure 2 – page 22:

•

the ‘swelling front’ clearly separates the rubbery region (with enough water to lower the Tg below the experimental temperature) from the glassy region (where the polymer exhibits a Tg that is above the experimental temperature).

•

the ‘erosion front’, separates the matrix from the solvent. The gel-layer thickness as a function of time is determined by the relative position of the swelling and erosion moving fronts.

- 21 -

•

the ‘diffusion front’ located between the swelling and erosion fronts, and constituting the boundary that separates solid from dissolved drug, has been identified.

During drug release, the diffusion front position in the gel phase is dependent on drug solubility and loading. The diffusion front movement is also related to drug dissolution rate in the gel.

Swelling front Diffusion front

Erosion front

Figure 2. The fronts in a swellable HPMC matrix

Drug release is controlled by the interaction between water, polymer and drug. The delivery kinetics depends on the drug gradient in the gel layer. Therefore, drug concentration and thickness of the gel layer governs the drug flux. Drug concentration in the gel depends on drug loading and solubility. Gel-layer thickness depends on the relative contributions of solvent penetration, chain disentanglement and mass (polymer and drug) transfer in the solvent. Initially solvent penetration is more rapid than chain disentanglement, and a rapid build-

- 22 -

up of gel-layer thickness occurs. However, when the solvent penetrates slowly, owing to an increase in the diffusional distance, little change in gel thickness is observed since penetration and disentanglement rates are similar. Thus gel-layer thickness dynamics in swellable matrix tablets exhibit three distinct patterns. The thickness increases when solvent penetration is the fastest mechanism, and it remains constant when the disentanglement and water penetration occur at a similar rate. Finally, the gel-layer thickness decreases when the entire polymer has undergone the glassy–rubbery transition. In conclusion, the central element of the release mechanism is a gel-layer forming around the matrix in response to water penetration. Phenomena that govern gel-layer formation, and consequently drug-release rate, are water penetration, polymer swelling, drug dissolution and diffusion, and matrix erosion. Drug release is controlled by drug diffusion through the gel layer, which can dissolve and/or erode.

- 23 -

1.3. Impact of the formulation and process variables on the drug release from extended release matrix systems

1.3.1. Formulation variables The physicochemical characteristics of the drug, in particular its aqueous solubility, should be considered in the formulation of a matrix system. According to Qiu and Zhang (2000), the following recommendations apply to matrix systems (Table 1 – page 24):

Table 1. Application of matrix for drug delivery systems Matrix system

Drug delivery mechanism

Drugs not recommended

Diffusion and erosion

Very soluble

Erosion

Freely soluble

Monolithic

Diffusion

Practically insoluble

Multiparticulate

Diffusion

Freely soluble

Erosion/enzymatic degradation

-

Hydrophilic Swellable / erodible Erodible Hydrophobic

Erodible/Degradable Qiu and Zhang, (2000)

- 24 -

Other drug properties affecting system design include drug stability in the system and at the site of absorption, pH-dependent solubility, particle size and specific surface area. Drug particle size Effect of drug particle size on release is important in the case of moderately soluble drugs. Velasco et al. (1999) showed that for a given effective surface area, diclofenac particle size influenced the release rate from hydroxypropyl methyl cellulose (HPMC) tablets. The smallest particle size of drug dissolved more easily when dissolution medium penetrated through the matrix resulting in a greater role for diffusion. The larger particle size dissolved less readily and therefore was more prone to erosion at the matrix surface. A similar dependence was shown for a less soluble drug, indomethacin (Ford et al., 1995).

Hogan (1989) showed that in the case of water-soluble aminophylline or propranolol HPMC-based tablets an increase in drug particle size did not significantly alter the release rate of the drug. A noticeable effect was seen only at a low drug: HPMC ratio and at a large drug particle size (above 250µm) any was seen; in this case, rapid dissolution of the water soluble drug would leave a matrix with low tortuosity and high porosity. Drug: polymer ratio For diclofenac tablets formulated with HPMC, Velasco et al. (1999) showed that an increase in drug: polymer ratio reduced the release rate. This was because an increase in polymer concentration caused an increase in the viscosity of the gel

- 25 -

(by making it more resistant to drug diffusion and erosion) as well as the formation of a gel layer with a longer diffusional path.

Similar findings were reported by Rekhi et al. (1999). Diffusional release of watersoluble drug metoprolol (primarily controlled by the gel thickness) decreased with increasing HPMC incorporation.

By varying the polymer level (Methocel® K4M 10-40%), Nellore et al. (1998) achieved different metoprolol in vitro release profiles.

Sung et al. (1996) demonstrated that changes in HPMC: lactose ratio can be used to produce a wide range of drug (adinazolam mesylate) release rates.

For Ethocel® 100 and Eudragit® RSPO matrices, Boza et al. (1999) showed that an increase in the polymer content resulted in a decrease in the drug release rates due to a decrease in the total porosity of the matrices (initial porosity plus porosity due to the dissolution of the drug). Polymer type Various grades of commercially available HPMC differ in the relative proportion of the hydroxypropyl and methoxyl substitutions; increasing the amount of hydrophilic hydroxypropyl groups lead to a faster hydration: Methocel®K > Methocel®E > Methocel®F. Generally rapid hydrating Methocel®K grade is preferred, especially for highly soluble drugs where a rapid rate of hydration is

- 26 -

necessary. It is important to note that an inadequate polymer hydration rate may lead to dose dumping, due to quick penetration of gastric fluids into the tablet core (Dow Pharmaceutical Excipients, 1996).

In each grade, for a fixed polymer level, the viscosity of the selected polymer affects the diffusional and mechanical characteristics of the matrix. By comparing different Methocel®K viscosity grades, Nellore et al. (1998) found that the higher viscosity gel layers provided a more tortuous and resistant barrier to diffusion, which resulted in slower release of the drug (metoprolol HCl).

Sung et al. (1996) compared different viscosity grades of HPMC (Methocel® K100LV, K15, K100). The fastest release of adinazolam mesilate was achieved for the K100LV formulation. The K4M formulation exhibited a slightly greater drug release than K15M and K100M. Due to the lack of a significant difference in the release profiles between K15M and K100M, the authors suggested a limiting HPMC viscosity of 15000cP, above which if viscosity increased, the release rate would no longer decrease. Similarly, formulations containing higher HPMC viscosity grades had slower HPMC release, but no limiting HPMC viscosity was observed for polymer release.

In a study by Campos-Aldrete and Villafuerte-Robles (1997), for low HPMC concentration (10%) formulations, the lag time was found to be dependent on the viscosity grade. The increasing burst effect produced by higher viscosity grades

- 27 -

was attributed to slower swelling with increasing polymer viscosity, allowing greater time for the dissolution of the drug (metronidazole) before the gel barrier was established. For HPMC concentration of 20% or more, the porosity was a less important factor in the drug release and the effect of viscosity grade was minimized.

In the case of ethyl cellulose, the findings are completely different. The lower viscosity grades of ethylcellulose are more compressible than the higher viscosity grades, resulting in harder tablets and slower release (Katikaneni et al., 1995a, Shileout and Zessin, 1996, Upadrashta et al., 1993).

By comparing Eudragit® RSPO to Ethocel® 100, the release rate of lobenzarit sodium was slower for the Eudragit® based matrix (Boza et al., 1999). The explanation was based on the chemical structure of the polymers. Ethocel® 100 has hydrophilic hydroxyl and ethoxyl groups, which make the matrix water sensitive. Consequently, it was more difficult to control the release of the hydrophilic drug. Eudragit® RSPO is only slightly permeable to water due to its low content of quaternary ammonium groups; therefore it was more suitable for controlling the release of the hydrophilic drug. Polymer particle size Velasco et al. (1999) found that the diclofenac sodium release rate from HPMC tablets decreased as the polymer particle increased. Also, as the HPMC particle size increased, the lag period decreased – the drug release occurred during the

- 28 -

initial dissolution stage, prior to the formation of the gel layer (coarse fraction of HPMC hydrated slower).

Campos-Aldrete and Villafuerte-Robles (1997) found that increasing particle size of HPMC allowed the free dissolution of metronidazole at higher proportion before the gel was established. Decreasing particle size caused a smaller burst effect and induced lag times. The explanation was based on a faster swelling of the smaller particles that allowed a rapid establishment of the gel barrier.

Heng et al. (2001) observed significant effect of HPMC particle size on aspirin release for polymer concentrations up to 20%. A mean HPMC (Methocel® K15M Premium) particle size of 113µm was identified as a critical threshold for the release of aspirin. The drug release rate increased markedly when polymer particle size was increased above 113µm. The release rate was much less sensitive to changes in particle size below 113µm. The aspirin release mechanism followed first order kinetics, when mean HPMC particle size was below 113µm. The release mechanism deviated from first order kinetics, when the mean particle size was above 113µm. Polymer fractions with similar mean particle size but differing size distribution were also found to influence drug release rates but not the release mechanism.

In the case of ethyl cellulose, using a constant compression force and increasing the particle size, caused a decrease in tablet hardness and an increase in

- 29 -

dissolution rate, due to a reduction in the interparticular forces. Erosion occurred for tablets manufactured with ethylcellulose particle size above 120µm (Kakiketeni et al., 1995a). Characterization of tablets prepared using different particle sizes revealed that the porosity increased with increase in particle size (Katiketeni et al., 1995b) and the increase in porosity resulted in a faster drug release. Fillers Nellore et al. (1998) studied the effect of filler (57% of the tablet weight) on a metoprolol formulation at 20% Methocel® K4M level. They concluded that filler solubility had a limited effect on release rate. The release profiles showed a decrease of about 5-7% after 6h, as the filler was changed from lactose to lactose – microcrystalline cellulose then to dicalcium phosphate dihydrate microcrystalline cellulose. Addition of soluble fillers enhanced the dissolution of soluble drugs by decreasing the tortuosity of the diffusion path of the drug, while insoluble fillers like dicalcium phosphate dihydrate got entrapped in the matrix. Also, they assumed that presence of a swelling insoluble filler like microcrystalline cellulose changed the release profile to a small extent due to a change in swelling at the tablet surface.

Changing the filler from 100% dicalcium phosphate dihydrate to 100% lactose resulted in an increase in metoprolol release from Methocel® K100LV tablets at 4, 6 and 12h (Rekhi et al., 1999). This was explained by dissolution of lactose and the consequent reduction in the tortuosity and or gel strength of the polymer.

- 30 -

Similar dissolution profiles were obtained for filler concentration up to 48%. No dose dumping due to stress cracks (Dow Pharmaceutical Excipients, 1996) during gelling were observed in the case of insoluble fillers. Ion-exchange resins Ion exchange resins can be used as release modifiers in matrix formulation containing oppositely charged drugs, based on in situ drug-resin complex formation. Sriwongjanya and Bodmeier (1998) studied the release of cationic drug propranolol from HPMC matrix tablets containing drug without resin (Amberlite® IRP69), drug-resin complex and drug - resin physical mixture. The fastest release was observed for resin free tablets (in all the dissolution media). In the case of drug-resin complex tablets, the drug was not released in water, since there were no counterions in the medium to replace drug ions from the ion exchange resin within the gelled matrix. The drug was released in 0.1N HCl and pH 7.4 phosphate buffer, indicating that the drug release was initiated by an ionexchange process (the counterions present in the dissolution medium diffused through the gel layer to replace the drug, which was then released by diffusion). A similar extended release pattern was obtained by using the physical mixture of drug and resin, which denoted the in situ complex formation within the gelled region. The in situ method is more advantageous with regard to simplifying the manufacturing process compared to the use of the preformed complexes.

- 31 -

The rate of drug binding to the resin increased with decreasing the resin particle size, thus explaining the slower release and the absence of the burst phase with smaller sized resin particles. As the amount of resin increased, the drug release initially decreased, leveling up at a resin level enough to bind the drug in situ. Using a weak cation exchange resin (Amberlite® IRP88), in situ complex formation and release retardation was observed only in pH 7.4 buffer, but not in 0.1N HCl, because of the non-ionization of the carboxyl groups. Comparing different matrix materials, a rapid formation of a strong gel layer was important for the in situ complex formation; drug release decreased in the following order glyceryl palmitostearate > polyethylene oxide 400K > HPMC K15M. For different HPMC sorts, the rate of hydration influenced the release; tablets based on methyl cellulose or HPMC E4M (higher degree of methoxyl group substitution) disintegrated shortly after exposure to the medium because of the slow rate of hydration and the disintegrating effect of the resin (resins have large swelling ability). Similar results were observed for sodium diclofenac and the anion exchange resin cholestiramine. The phenomenon was not observed in case of the non-ionic drug guaifenesin. Surfactants Feely and Davis (1988) characterized the ability of charged ionic surfactants to retard

the

release

of

oppositely

charged

- 32 -

drugs

from

HPMC

tablets

(chlorphemiramine maleate and sodium alkylsulphates, sodium salicylate and cetylpiridinium bromide). The mechanism involved was an in situ drug-surfactant ionic interaction, resulting in a complex with low aqueous solubility, that the release would be more dependent on the matrix erosion than diffusion. The retarding effect was dependent upon the surfactant concentration in the matrix and independent on the surfactant hydrocarbon chain length. The pH of the environment played an important role, by altering the ionization of both the drug and the surfactant. The ionic strength of the dissolution medium affected the action of the resin. Polymeric excipients Feely and Davis (1988) studied the effect of polymeric additives (non-ionic polyethylene glycol 6000 or ethyl cellulose, cationic diethylaminoethyl dextran, anionic

sodium

(chlorpheniramine

carboxymethyl maleate,

cellulose sodium

Na-CMC) salicylate

on and

drug

release

potassium

fenoxymethylpenicillin) from HPMC matrix (85%). Non-ionic polymers (15% of tablet weight) did not significantly alter the release rates. Na-CMC (50% replacement of HPMC) reduced the chlorpheniramine maleate release in pH 7 buffer (near zero order release), but not in an acidic medium. This was explained by a complexation of the drug with the cationic polymer; which was not possible below pH 3, when Na-CMC was in its un-ionized insoluble form. As a result of the complexation, the gel erosion became the prominent release mechanism instead of diffusion.

- 33 -

No interaction occurred between sodium salicylate and Na-CMC (both anionic). In the presence of diethylaminoethyl dextran, sodium salicylate release was slower at pH 7, but not altered at pH 1 (when the drug was present in its unionized form). Overall, the effect of ionic polymers incorporated into HPMC matrices on the release of oppositely charged drugs was small compared to the ion-exchange resins.

Goldberg and Sakr (2003) used the drug-polymer ionic complexation approach in designing oral dosage formulation for controlled release of buspirone. As anionic exchange polymers sodium carboxymethyl cellulose and methacrylic acid / ethylacrylate copolymer were recommended based on the complexation affinity and dispersability in the aqueous environment of the gastrointestinal tract (average molecular weight of less than 500,000). The weight ratio of buspirone to anionic exchange polymer varied between 4:1 and 1:6, preferably between 2:1 and 1:4. In addition to facilitating the controlled release of buspirone, the formulations increased the bioavailability and reduced the inter-individual variability. Therefore, the buspirone-ion exchange polymer HPMC tablets permitted enhanced targeting of therapeutic amounts and effects of the drug.

Takka et al. (2001) studied the effect of the addition of anionic polymers (Eudragit® S, Eudragit® L 100-55, and Na-CMC) on the release of weakly basic

- 34 -

propranolol hydrochloride from HPMC matrices. The interaction between propranolol hydrochloride and anionic polymers influenced the drug release. The HPMC: anionic polymer ratio also affected the drug release. The matrix containing HPMC: Eudragit® L 100-55 (1:1) produced pH-independent extendedrelease tablets.

Bonferoni et al. (1998) used an optimization procedure to determine the HPMC:

λ-carrageenan ratio (34:30) required for a pH-independent release of chlorpheniramine maleate. λ-Carrageenan was added to overcome the increase in diffusion path length and decrease in the release rate associated with HPMC systems. λ-carrageenan was subjected to erosion, which was higher at acidic pH.

Streubel et al. (2000) failed to achieve a pH independent release of weakly basic drugs (verapamil HCl) from matrix tablets (ethylcellulose or HPMC) by adding an enteric polymer HPMCAS (HPMC acetate succinate). The creation of the waterfilled pores at high pH by dissolution of the enteric polymer was expected to accelerate the drug release and thus compensating the effect of the reduced solubility of the drug. However the addition of the HPMCAS to the ethylcellulose matrix reduced the verapamil release both in 0.1N HCl and pH 6.8 buffer compared to the ethyl cellulose solely based matrix. The authors explained this by a reduction of the matrix pore size in case of addition of HPMCAS due to the effect of particle size

- 35 -

difference: 33µm for ethyl cellulose, 6µm for HPMCAS on compaction behavior (larger pores in the ethyl cellulose matrix). As the release mechanism was predominantly diffusion, a reduction of the pore size significantly reduced the release rate. In contrast to ethyl cellulose, no effect was found when adding HPMCAS to the HPMC systems either in 0.1N HCl, or in pH 6.08 buffer. It was considered that this happened because the drug was predominantly released by diffusion through the swollen polymer network and not through the water filled pores. Thus, reduction of the initial porosity of the system was of minor significance in drug release rate. On the other hand, due to its high molecular weight, HPMCAS dissolution in phosphate buffer was hindered by the presence of the HPMC network; pre-existing cavities within the HPMC network could not accommodate diffusing HPMCAS molecules. Addition of organic acids In order to overcome the pH dependent release of a weakly basic drug (verapamil HCl) from matrix tablets, Streubel et al. (2000) added organic acids, which were expected to create a constant acidic microenvironment inside the tablets. Substances selected (fumaric, sorbic and adipic acid) had high acidic strength (low pKa value) and relatively low solubility in 0.1N HCl. These acids dissolved rather slowly and remained in the tablets during the entire period of drug release. Independent of the pH of the dissolution medium, the pH inside the tablet was acidic and thus the solubility of the weakly basic drug was high. In addition, at high pH, the organic acids acted as pore formers. The release rates

- 36 -

obtained for both ethyl cellulose and HPMC matrices were pH-independent. Among the three acids, fumaric acid showed the best results, due to the lowest pKa value.

1.3.2. Process variables

Compression force It has been reported (Velasco et al., 1999) for HPMC tablets, that although the compression force had a significant effect on tablet hardness, its effect on drug release from HPMC tablets was minimal. It could be assumed that the variation in compression force should be closely related to a change in the porosity of the tablets. However, as the porosity of the hydrated matrix is independent of the initial porosity, the compression force seems to have little influence on drug release. The influence of compression force could only be observed in the lag time (Velasco et al., 1999). Tablets made at the lowest crushing strength (compression force 3kN) with Methocel®K4M showed an initial burst effect due to an initial partial disintegration. Once the polymer was swollen, the dissolution profiles became similar to those tablets compressed to a higher crushing strength.

Rekhi et al. (1999) reported similar findings, i.e. changes in compression force or crushing strength appeared to have minimal effect on drug release from HPMC matrix tablets once a critical hardness was achieved. Increased dissolution was

- 37 -

only observed when the tablets were too soft and it was attributed to the lack of powder compaction or consolidation (3kP). Tablet shape Rekhi et al. (1999) showed that the size and shape of the tablet for the matrix system undergoing diffusion and erosion might impact the drug dissolution rate. Modification of the surface area for metoprolol tartrate tablets formulated with Methocel® K100LV from the standard concave shape (0.568sq. in.) to caplet shape (0.747 sq. in.) showed an approximately 20-30% increase in dissolution at each time point. Furthermore they recommended that for maximum maintenance of controlled release characteristics, tablet matrices should be as near spherical as possible to produce minimum release rate.

The release rate of the drug (theophylline) from erodible hydrogel matrix tablets (HPMC E50) having different geometrical shapes (compressed under the same compression force) was found (Karasulu et al., 2000) to be the highest on triangular tablets and successively in order of decreasing amounts on halfspherical and cylindrical tablets. This was attributed to heterogenous erosion of the matrices.

Siepman et al. (1999b) showed that varying the aspect ratio (radius/height) of the HPMC tablets is the very easy and effective tool to modify the release rate of the matrix system. Release rate for tablets with the same volume was higher for flat shape (ratio = 20) than regular cylinders (ratio 2) and almost rod-shaped

- 38 -

cylinders (ratio 0.2). The reason for this phenomenon was the difference in the surface area of the tablets. They proposed a new mathematical model that can be applied to calculate the optimal aspect ratio and size of a cylindrical tablet to achieve a desired profile. The model takes into account Fickian diffusion of water in and drug out of the tablets and swelling; it does not take into account dissolution and it cannot be applied for water insoluble drugs, which are released by dissolution process. Model applicability in predicting the dissolution rates was confirmed for water-soluble drugs (propranolol HCl and chlorpheniramine maleate) (Siepmann et al., 2000). Tablet size For tablets having the same aspect ratio and drug concentration, Siepman et al. (1999b) found that the tablet size had a very strong influence on the release rate; within 24 hours, 99.8% was released from the small tablets, 83.1% from the medium size and 50.9% from the large tablets. The explanation was based on the higher surface area referred to the volume for the small tablets than for the large ones. In addition, the diffusion pathways were much longer in large tablets than in small ones. Thus the relative amount of drug released versus time was much higher for small tablets. The variation of the size of the tablet was an effective tool to achieve a desired release.

- 39 -

1.4. Rationale for studying Kollidon® SR as extended release matrix excipient Due to technological accessibility, manufacturing capability and cost of the monolithic drug delivery systems, the pharmaceutical industry has placed a lot of emphasis on the design and development of these formulations. However, there are some distinct disadvantages of some of these matrix formers, which complicate the development and production of matrix tablets. Some of these include lack of flowability of polymers hampering the direct compression process, poor compressibility of the polymer forms resulting in tablets of low hardness, the influence of pH values or ionic strengths on the release profiles, burst effect and diminishing release rate with time. These disadvantages limit the application of currently used polymers and require development and evaluation of new polymers for extended release matrices.

Therefore evaluation of newly available matrix materials for their ability to promote pH-independent extended release of drugs is much warranted. Kollidon® SR was introduced to the pharmaceutical market recently, and thus its evaluation constitutes a novel research topic for the pharmaceutical industry.

- 40 -

1.5. Kollidon® SR - background Polyvinylacetate/Povidone based polymer (Kollidon® SR) is a relatively new extended release matrix excipient. It consists of 80% Polyvinylacetate and 19% Povidone in a physical mixture, stabilized with 0.8% sodium lauryl sulfate and 0.2% colloidal silica.

Polyvinylacetate – homopolymer of vinyl acetate. It is obtained by emulsion polymerization. Description: water white, clear solid resin, soluble in benzene and acetone, insoluble in water or emulsion readily diluted with water (Ash and Ash, 1995). Polyvinylacetate is a very plastic material that produces a coherent matrix even under low compression forces. Regulatory status: diluent in color additive mixtures for food use exempt from certification, food additive (21CFR73).

Povidone (polyvinylpyrrolidone) – white amorphous hygroscopic powder, soluble in water (Ash and Ash, 1995). It has good binding properties both under dry or wet conditions. Due to its hygroscopicity, Povidone promotes water uptake and facilitates diffusion and drug release (Shivanand and Sprockel, 1998). Manufacture US

Patent

6,066,334

describes

the

manufacture

procedure

for

the

polyvinylacetate / povidone redispersible polymer powders and their application

- 41 -

as binder at 0.5-20% (of the tablet weight), when the active ingredients are released within a time of 0.1 to 1.0 hour.

The

redispersible

polymer

powders

are

manufactured

by

emulsion

polymerization of vinyl acetate followed by addition of polyvinylpyrrolidone (as 10-50w/w solution) and spray- or freeze-drying. The polymerization takes place at temperature of 60-80°C and results in shear-stable fine-particle dispersion. The k value of the polymers should be in the range from 10-350, preferably 5090. To prevent particles caking together, silica (spraying aid) is added to the dispersion before spraying. Spray drying is done in spray towers (with disks or nozzles) or in fluid beds. Physicochemical properties Description: white or slightly yellowish, free flowing powder; Particle size distribution: average particle size of about 100µm; Molecular weight of polyvinyl acetate 450 000; Bulk density: within the range of 0.30-0.45g/ml; 0.37g/ml (Ruchatz et al., 1999); Tap density: 0.44g/ml (Ruchatz et al., 1999); Flowability: good flow properties with a response angle below 30° (BASF, 1999), 21° (Ruchatz et al., 1999). Solubility: Polyvinylacetate is insoluble in water. Povidone gradually dissolves in water; in tablets it acts as a pore-former. pH: 3.5-5.5.

- 42 -

The manufacturer generally claims for Kollidon® SR good compressibility and drug release independent of the dissolution medium (pH and salt/ion content) and rotation speed. Compressibility results were published for propranolol 160mg tablets (drug: polymer 1:1). The compression force did not affect the drug release profile. The pH-independent release was also tested for caffeine (BASF, 1999).

Pathan and Jalil (2000) evaluated Kollidon® SR as matrix excipient for Theophylline tablets. Tablets containing 20-70% theophylline showed Higuchian release kinetics; the release rates increased exponentially with the drug loading. The increase in compressional force from 20kN to 60kN caused a slight linear decrease in the release rate. Annealing of the tablets for 24 hours at temperatures of 45 and 55°C showed a slight decrease in the release rate compared to the room temperature.

Shao et al. (2001) reported the effect of accelerated stability conditions on diphenhydramine HCl tablets prepared with Kollidon® SR. A decrease in dissolution rate along with an increase in tablet hardness was noticed for tablets with high level of Kollidon® SR (>37%) prepared without diluents or with 15% diluent (lactose, Emcompress®). At 25% Emcompress®, no changes occurred. Such changes were not observed for tablets stored at 25°C/ 60%RH or cured at 60°C for at least one hour.

- 43 -

Rock et al. (2000) evaluated different additives: diacetyl-tartaric acid diglyceride ester, pectin, stearic acid and methyl hydroxyethyl cellulose for optimization of caffeine release from Kollidon® SR -based matrix tablets. Stearic acid retarded the initial drug release in acidic medium due to its hydrophobic character, but failed to accelerate it in neutral medium. Diacetyl-tartaric acid diglyceride ester, methyl hydroxyethyl cellulose and pectin reduced the initial drug release and intensified the dissolution after the pH change.

Flick et al. (2000) showed the applicability of Kollidon® SR in hot melt technology using acetaminophen.

Regulatory status: Kollidon® SR is not a pharmacopoeial or NF listed additive. In 2001, BASF filled a DMF (drug master file) for this product with the FDA (FDA Drug Master Files).

- 44 -

1.6. Propranolol extended release formulations Propranolol HCl is a racemic mixture of dextrorotary and levorotary forms of 1(Isopropylamino)-3-(1-naphthyloxy)-2-propanol hydrochloride

It is a white, odorless crystalline powder, readily soluble in water and ethanol (1:20), soluble in 0.1 N HCl: 220 mg/ml and in pH 7.4 phosphate buffer: 254 mg/ml (Siepmann and Kranz, 2000); pKa=9.5 (Avdeef et al., 2000). Propranolol is a highly lipophilic (log Kp=3.65), non-selective beta-adrenergic antagonist, which interacts with beta1 and beta2 receptors with equal affinity, lacks intrinsic sympathomimetic activity, possesses membrane stabilizing activity and does not block alpha-adrenergic receptors (Goodman and Gilman, 2001). Propranolol is almost completely absorbed from the gastrointestinal tract, by passive non-stereoselective diffusion (Goodman and Gilman, 2001, Mehvar and Brocks, 2001). The absorption takes place from both the proximal and distal intestine, making it a good candidate for extended release dosage forms (Buch and Barr, 1998). Propranolol is subjected to an extensive and highly variable hepatic first pass metabolism, with a reported systemic bioavailability between 15 and 23% (Cid et al., 1986, Walle et al., 1986). Propranolol binds (90% of the dose) to both albumin and α1-acid glycoprotein in plasma stereoselectively, resulting in higher free fraction of S(-) propranolol in plasma. The age and gender

- 45 -

of the patients do not appear to have a substantial effect on the protein binding of propranolol enantiomers (Mehvar and Brocks, 2001). Peak effect occurs after 1-2 hours and can vary up to seven fold after oral administration due to individual variations in hepatic metabolic activity (Shand et al., 1970). Propranolol is metabolized by three main pathways of ring hydroxylation (40% of the dose), side chain oxidation (35-40%) and glucuronidation (remaining 20-25% of the dose), the metabolism being overall stereoselective for the less active R(+) enantiomer, resulting in a higher plasma concentrations of the S(-) enantiomer (Buch and Barr, 1998, Mehvar and Brocks, 2001). The metabolism is affected by genetic polymorphism for both CYP1A and CYP2D6 isozymes in the liver (Mehvar and Brocks, 2001). The biologic half-life is approximately four hours (Shand et al., 1970, Mehvar and Brocks, 2001). No conclusive results were reported for the effect of the input rate on the ratio of the enantiomers in plasma (Mehvar and Brocks, 2001). The cardiac beta-blocking activity of propranolol resides in S(-) enantiomer, which is x100 times more potent than the R(+) enantiomer (Mehvar and Brocks, 2001). Due to relatively short plasma half-life, propranolol conventional tablets are administrated at 6 to 8 hours intervals. Such frequent drug administration may reduce patient compliance and thus therapeutic efficacy (Serlin et al., 1983). Several extended release systems have been developed in order to enable daily administration of the drug and a 24-hour maintained beta-adrenoceptor blockade.

- 46 -

There are some reported problems associated with propranolol extended release (ER) formulations. Besides the variable propranolol bioavailability (due to first pass degradation, influence of food, ethnic factor, other medication), ER formulations generally exhibit a lower systemic bioavailability than the conventional tablets (Table 2 – page 47). This is due to a slower/poor absorption and higher first pass effect or sometimes due to an underestimation of the area under the plasma concentration-time curve due to limited blood sampling or low analytical sensitivity (Nace and Wood, 1987). However similar bioavailability for ER and conventional products has been reported (Bottini et al., 1983, Dunn et al., 1985). Unlike conventional formulations of propranolol, absorption of propranolol from the ER formulations has been shown to be unaffected by food or stimulation of gastrointestinal motility by coadministration of metoclopramide (Nace and Wood, 1987).

Table 2. Pharmacokinetic properties of propranolol Formulation

Extent of absorption (%of dose)

Bioavailability (% of dose)

Interpatient variation in plasma level

β-Blocking plasma concentration

Protein binding (%)

Immediate release

>90%

30

20 fold

50-100ng/ml

93

Extended release

>90%

20

10-20 fold

20-100ng/ml

93

(Frishman and Jorde, 2000)

- 47 -

The apparent elimination half-life of ER propranolol formulations ranges from 8 to 11 hours, or approximately 2 to 3 times that of conventional propranolol. This marked increase in the apparent half-life is due to continued absorption of the drug from the gastrointestinal tract (Nace and Wood, 1987). Currently it is well accepted that once daily ER propranolol is as effective as conventional immediate release propranolol given in divided doses. Once daily dosing of ER products produced relatively constant plasma concentrations and prolonged beta-adrenoblockade and offers the potential for improved patient compliance in the treatment of hypertension and prevention of angina. Different studies showed that single daily doses of ER propranolol produce significant blockade of cardiac beta-adrenoceptors throughout a 24-hour dose interval, as assessed by inhibition of exercise-induced tachycardia (Perucca et al., 1984, McAinsh et al., 1978, Serlin et al., 1983, Garg et al., 1987, Lalonde et al., 1987). The time course and degree of beta-adrenoblockade were similar to those obtained with conventional propranolol given in divided doses and correlated well with plasma concentrations. Shanks (1984) suggested that 15-20% inhibition of exercise-induced tachycardia is necessary for therapeutic cardiac betaadrenoceptor blockade. Propranolol ER produced a significant fall in blood pressure throughout the 24-hour dosing interval, although no correlation could be established between propranolol concentrations and hypotensive effects. This lack of correlation could be attributed to the multiple mechanisms involved in the antihypertensive action, including effects on the renin-angiotensin system and central nervous system (Nace and Wood, 1987).

- 48 -

Takahashi et al. (1990) showed no significant differences in the hepatic metabolism of propranolol administered as extended release capsules 60mg once-a-day or immediate release tablets 20mg x 3/day. The observed differences in the area under the curve for propranolol, 4-hydroxy-propranolol glucuronide and naphthoxylactic acid after the administration of the two products were explained by the lower absorption and consequently lower bioavailability of the ER capsules compared to the immediate release tablets. Bioavailability of a 160mg slow release formulation following single dose administration was about one third that of the conventional preparation (Drummer et al., 1981). Garg et al. (1987) showed that area under the curve and the peak concentration was lower for two propranolol long-acting formulations (80mg and 160mg) than for the conventional tablets; in addition the elimination half-life was longer (9 hours) for the extended release products than for conventional propranolol (4 hours). In a crossover study performed in healthy subjects, bioavailability of propranolol 160mg as extended release capsules was 52% for single dose and 54% for steady state compared to the regular tablet formulation (Straka et al., 1987). Mean bioavailabilities of extended release Duranol® capsules (single dose in the morning) and Inderal® conventional formulation (two doses: morning and evening) were similar despite prolonged absorption time for the sustained action capsules (Bottini et al., 1983). In a study with extended release propranolol (Elanol® 120mg, Inderal® LA 160mg) and conventional Inderal® (40mgx3/day) single doses of controlled

- 49 -

release preparations produced a smoother drug serum level profile with lower and delayed peak times. At steady state, all regimens ensured relatively sustained serum levels and a stable degree of pharmacological effect. Dose corrected AUC decreased in the following order: Elanol® > Inderal® > Inderal® LA. These results demonstrated that long acting formulations of propranolol can be developed which are not necessarily associated with reduced bioavailability secondary to enhanced first pass metabolism (Perucca et al., 1984). The bioavailability of Inderal® LA (80, 160 and 240mg once daily for 4 days) was proportional to the dose administrated as sustained action capsules. Steady state was attained after two doses (Dvornik et al., 1983). For different extended release formulations, the peak blood level and AUC decreased as the dissolution time increased and the half-lives were inversely proportional to the dissolution rate. The lowering of the systemic bioavailability as the dissolution time increased, was assumed to be caused by an increased metabolism of propranolol (McAinsh et al., 1981) An attempt to develop plastic matrix tablets was done in 1974 by Grundy et al. The matrix consisted on propranolol 125 mg embedded in an insoluble matrix of Pevikon D-42-P (polyvinyl chloride, 273 mg). The formulation had a satisfactory in vitro release profile (50% of the dose in 3 hours, at 100rpm). However, when administered in dogs, the in vivo release profile was unsatisfactory (the drug was not completely released from the matrix) (Grundy et al., 1974). Single entity

- 50 -

extended release formulations of propranolol were therefore abandoned in favor of multiparticulate systems. All the propranolol extended release formulations currently in use in the United States are capsules (Electronic Orange Book, 2002). Therefore, this research represents a novel approach in development of extended release propranolol dosage forms by formulating them as tablets. The reference listed product, Inderal® LA, consists of hard gelatin capsules containing film coated spheroids each comprising propranolol hydrochloride in admixture with microcrystalline cellulose. The drug containing spheroids are in turn coated with ethylcellulose alone or in combination with hydroxypropyl methylcellulose and/or plasticizer; the semipermeable membrane allows drug to diffuse at a controlled rate (US Patent 4, 138, 475).

- 51 -

2. Objective, hypothesis and specific aims

2.1. Objective The objective of this study is to evaluate Kollidon® SR as extended release matrix forming excipient.

2.2. Hypothesis Kollidon® SR promotes in vitro pH-independent extended release of drugs.

2.3. Specific aims Studying the effect of the following variables on the tablet properties and in vitro release of drugs from matrix tablets based on Kollidon® SR:

♦

♦

♦

Formulation variables:

•

polymer concentration

•

external binder addition in the wet granulation process

•

enteric polymer addition.

Process variables:

•

method of manufacturing (direct compression, wet granulation)

•

compression force.

Dissolution medium.

- 52 -

Evaluation for one developed tablet formulation of its bioequivalence to an extended release reference listed product.

- 53 -

3. Experimental

3.1. Materials and supplies Acetonitrile HPLC grade (Fisher Scientific, Fair Lawn NJ, USA) Ammonio methacrylate copolymer type B NF Eudragit® RSPO (Rohm, Darmstadt, Germany) Buspirone HCl (Brantford Chemicals Inc., Brantford Ontario, Canada) Citric acid (Sigma Chemicals, St. Louis MO, USA) Colloidal silicon dioxide (Aerosil® 200, Degussa, Parsippany NJ, USA) Dibasic calcium phosphate dihydrate (Emcompress®, Penwest, Patterson NY, USA) DL propranolol hydrochloride 99% (Acros Organics, Fair Lawn NJ, USA) Ethyl acetate HPLC grade (Fisher Scientific, Fair Lawn NJ, USA) Hydrochloric acid (Fisher Chemicals, Fair Lawn NJ, USA) Inderal® LA (Ayerst Laboratories Inc., Philadelphia PA, lot # 9010268, expiration date 07/2003) Magnesium stearate (Mallinckrodt Chemical Inc., St. Louis MO, USA) Methacrylic acid copolymer type C NF Eudragit® L100-55 (Rohm, Darmstadt, Germany) Microcrystalline cellulose (Emcocel® 90M, Penwest, Patterson NY, USA) o-Phosphoric acid 85% HPLC grade (Fisher Scientific, Fair Lawn NJ, USA)

- 54 -

Polyvinyl acetate and Povidone based excipient, Kollidon® SR (BASF, Ludwigshafen, Germany) Polyvinyl acetate dispersion, Kollicoat® SR 30D (BASF, Ludwigshafen, Germany) Polyvinyl pyrrolidone, Kollidon® 30 (BASF, Ludwigshafen, Germany) Potassium phosphate monobasic (Fisher Scientific, Fair Lawn NJ, USA) Pronethalol hydrochloride (Tocris, Ellisville MO, USA) Propranolol hydrochloride (Wychoff Chemicals, South Haven MI, USA) Sodium chloride (Mallinckrodt Chemical Inc., St. Louis MO, USA) Sodium chloride Injection USP 0.9% 10ml (American Pharmaceutical Partners Inc., Los Angeles CA, USA) Sodium hydroxide (Fisher Chemicals, Fair Lawn NJ, USA) Sodium phosphate dibasic anhydrous (Fisher Chemicals, Fair Lawn NJ, USA) Triethylamine (Fisher Scientific, Fair Lawn NJ, USA) Water HPLC grade (Fisher Scientific, Fair Lawn NJ, USA)

BD Vacutainer Lithium Heparin 5ml (Becton Dickinson, Franklin Lakes NJ, USA) Clear glass threaded vials 1.5dr. (Fisher Scientific, Pittsburgh PA, USA) Full flow filters 35µm (VanKel Technology Group, Carry NC, USA) Glass inserts 250µl for HPLC vials (Agilent Technologies, Palo Alto CA, USA) High-density polyethylene bottles 60cc, 90cc (Selco Inc., Anaheim CA, USA) IEC Centra-8R Centrifuge (International Equipment Company, Needham Heights, MA, USA)

- 55 -

Luer Adapter Venoject (Terumo Corp., Tokyo, Japan) Metachem Inertsil ODS-3 5µm 250x4.6mm HPLC Column with MetaGuard 4.6mm Inertsil ODS-3 5µm (Metachem Technologies Inc., Torrance CA, USA) Millipore MF TM 0.45µm membrane filters (Millipore Corp., Bedford MA, USA) Millipore swinnex disks filter holders 25mm (Millipore Corp., Bedford MA, USA) Polypropylene conical tubes 15ml (Becton Dickinson, Franklin Lakes NJ, USA) Polypropylene flat top microcentrifuge tubes 2ml (Fisher Scientific, Pittsburgh PA, USA) Redi-Tip general purposes (200-1000µ) (Fisher Scientific, Pittsburgh PA, USA) Septa Target (National Scientific Company, Duluth GA, USA) Serological Disposable pipettes 5ml (Fisher Scientific, Pittsburgh PA, USA) Target Vials 2ml (National Scientific Company, Duluth GA, USA) Terumo Needles 20gx11/2’’ (Terumo Medical Corp., Elkton MD, USA) Terumo Syringes 2ml, 5ml, 10ml (Terumo Medical Corp., Elkton MD, USA) USA standard testing sieves (Gilson Company, Worthington OH, USA)

- 56 -

3.2. Equipment Accumet 1002 pH meter (Fisher Scientific, Fair Lawn NJ, USA) Balances PB1502, AB104 (Mettler Toledo International, Greifensee Switzerland) HPLC Beckman System Gold 126 Solvent Module with 507e Autosampler (Beckman Coulter, Fullerton CA, USA) and Waters 474 Scanning Fluorescence Detector (Waters Corporation, Milford MA, USA) Computrac Moisture Analyzer MAX 50 (Arizona Instrum., Phoenix AZ, USA) Dissolution Tester VK7000 (VanKel Technology Group, Carry, NC, USA) coupled to a Spectrophotometer DU 640 (Beckman Coulter, Fullerton CA, USA) Espec Humidity Cabinet LHL112 (Tabai Espec Corp, Osaka, Japan) Hardness Tester (Key International Inc., Englishtown NJ, USA) Integrapette Digital Pipette 1000µl, 20µl (Liquid Handling Systems, Indianapolis IN, USA) Isotemp Incubator 655D (Fisher Scientific, Pittsburg PA, USA) Planetary Mixer (Kitchen Aid, St. Joseph MI, USA) ReactiTherm III TM with Heating Stirring Module Reacti Vap TM III (Pierce, Rockford IL, USA) Rotary tablet press Manesty D3B (Manesty Machines Ltd., Liverpool, UK) Ultra Low Temperature Freezer Sanyo (Sanyo Electric Biomedical Co. Osaka, Japan) Micrometer Starrett (Starrett, Athol MA, USA) Stirrer/Hot plate PC 620 (Corning Inc., New York NY, USA)

- 57 -

Tumbling Mixer Turbula T2G (Glen Mills Inc., Maywood NJ, USA) Ultrasonic Cleaner FS30 (Fisher Scientific, Pittsburg PA, USA) Integrator Varian 4270 (Varian Inc., Palo Alto CA, USA) Vortex Genie (Scientific Industries Inc., Springfield MA, USA) Software Beam Spider (Hottinger Baldwin Messtechnik, Darmstadt, Germany) DU Data Capture 600-7000 (Beckman Coulter, Fullerton CA, USA) WinNonlin 4.0.1 (Pharsight Corporation, Mountain View CA, USA) SAS software systems for Windows Release 8.02 (SAS Institute Inc, Cary NC, USA)

- 58 -

3.3. Tablet composition Two water-soluble model drugs were used in the experiments. Propranolol HCl (section 1.6 – page 45) Buspirone hydrochloride : (MW 421.97) 8-[4-[4-(2-Pyrimidinyl)-1-piperazinyl]butyl]-8-azaspiro[4,5]decane-7,9-dione hydrochloride - racemic mixture.

Buspirone HCl is a white crystalline powder, soluble in water, pKa1=4.12, pKa2=7.32 (Takacs-Novak and Avdeef, 1996).

The hydrophilic Polyvinylpyrrolidone (Kollidon® 30) and the hydrophobic polyvinylacetate dispersion (Kollicoat® SR 30D) were added in the wet granulation experiments, as external binders.

Ammonio methacrylate copolymer (Eudragit® RSPO), a direct compressible matrix-forming polymer with extended release properties was used for comparative studies (section 3.6.1.3 – page 71)

A mixture of dibasic calcium phosphate dihydrate (Emcompress®) and microcrystalline cellulose (Emcocel® 90M) in ratio 1:1 was used as tablet filler.

- 59 -

This ratio was selected based on literature data (Sakr et al., 1988, Sakr et al., 1987) and preliminary experiments.

Other tablets components were colloidal silicon dioxide (Aerosil® 200) as a glidant and magnesium stearate as a lubricant.

Eudragit® L100-55 was added to some propranolol 80mg formulations to see the effect of the addition of an enteric polymer on the drug release (section 3.6.3.4 – page 75).

- 60 -

3.4. Tablet manufacture Tablets were manufactured by direct compression or wet granulation, according to the process flow presented in Figure 3 – page 62 and Figure 4 – page 63, respectively and then stored in airtight high-density polyethylene (HDPE) bottles till further testing.

Direct Compression - Process flow:

•

The corresponding amounts of drug and Kollidon® SR were accurately weighed.

•

The powders were screened using screen #35.

•

The screened powder was transfered into the turbula mixer jar and mixed for 5 minutes.

•

The corresponding amounts of Emcocel® 90M, Emcompress®, Aerosil® 200 (and Eudragit® L100-55 for some formulations) were accurately weighed, screened through screen #35, added to the turbula jar and mixed for 10 minutes.

•

The corresponding amount of magnesium stearate was accurately weighed and mixed with the powder in the turbula jar for additional 3 minutes.

•

The powder was compressed into tablets using an instrumented tablet press and tablets were collected during compression for in-process testing (weight and hardness).

- 61 -

Drug Kollidon® SR

Mixing 5 Minutes

(sieve #35)

Fillers Glidant

Mixing 10 Minutes

(Turbula Mixer)

(sieve #35)

Lubricant (sieve #35)

Final Mixing 3 Minutes

Compression(*)

(*)

(Manesty D3B Press)

in process control of tablets’ weight and hardness recording of the compression and ejection forces (Beam Spider Software)

Figure 3. Process flow chart for tablets manufactured by direct compression

- 62 -

Drug Kollidon® SR

Mixing 5 Minutes

(sieve #35)

(Turbula Mixer) Fillers (sieve #35)

Distilled water or Binder dispersion

Glidant Lubricant

Mixing 10 Minutes

Granulation

(Planetary Mixer)

Wet screening

(# 12 mesh)

Drying (1.5%)

(Oven 40°C)

Dry screening

(# 18 mesh)

Final Mixing 3 Minutes

Compression(*)

(Turbula Mixer)

(Manesty D3B Press)

(*) in process control of tablets’ weight and hardness recording of the compression and ejection forces (Beam Spider Software) Figure 4. Process flow chart for tablets manufactured by wet granulation

- 63 -

Wet Granulation - Process flow:

•

The corresponding amounts of drug and Kollidon® SR were accurately weighed.

•

The powders were screened using screen #35.

•

The screened powder was transferred into the turbula mixer jar and mixed for 5 minutes.

•

The corresponding amounts of Emcocel® 90M, Emcompress® were accurately weighed, screened through screen #35, added to the turbula jar and mixed for 10 minutes.

•

The powder mixture was transferred to the planetary mixer and granulated with water or binder dispersion.

•

The wet mass was passed through a #12 sieve and the resulting granules were placed on trays for drying into the oven at 40°C to a moisture content of 1.5%.

•

The dried granules were passed through a #18 sieve.

•

The dried granules and the corresponding amount of magnesium stearate and Aerosil® 200 were accurately weighed and then mixed in the turbula jar for additional 3 minutes.

•

The mixture was compressed into tablets using an instrumented tablet press and tablets were collected during compression for in-process testing (weight and hardness).

- 64 -

3.5. Tablet testing Tablet weight variation - twenty tablets from each batch were individually weighed and the average weight and relative standard variation were reported. Thickness - was determined for 10 pre-weighed tablets of each batch using a micrometer and the average thickness and relative standard variation were reported. Hardness - was determined for 10 tablets (of known weight and thickness) of each batch; the average hardness and relative standard variation were reported. Uniformity of dosage units was assessed according to the USP requirements <905> for content uniformity. The batch meets the USP requirements if the amount of the active ingredient in each of the 10 tested tablets lies within the range of 85% to 115% of the label claim and RSD is less than or equal to 6%. According to the USP criteria, if one of these conditions is not met, additional 20 tablets need to be tested. Not more than 1 unit of the 30 tested should be outside the range of 85% and 115% of the label claim and no unit outside the range of 75% to 125% of label claim; also RSD should not exceed 7.8%. In vitro drug release In vitro drug release was performed for the manufactured tablets according to the USP 25 “Dissolution procedure” <711>, over a 24-hour period, using an automated dissolution system. A minimum of 6 tablets per batch were tested. Method A - Apparatus 2 (paddle) was used at 50rpm, with 1000ml dissolution medium at 37°C; the UV absorbance of the dissolution medium was measured at

- 65 -

0.5, 1, 2, 3, 4, 6, 8, 10, 12, 16, 20 and 24 hours. The release was calculated using a standard solution. The drug release was tested in different dissolution media: distilled water, 0.1N HCl and USP 25 pH=6.8 phosphate buffer. The pH range (pH = 1.2 – 6.8) was chosen to reflect the physiologic conditions of the gastrointestinal tract.

Method B - in addition to the general method (method A), some of the propranolol 80mg tablet batches were tested according to the USP dissolution method required in the propranolol 80mg extended release capsules USP monograph (apparatus 1, 100rpm, 900ml, first 1.5 hours pH 1.2 buffer, then pH 6.8 buffer). Additional sampling times (1.5 and 14 hours) were included in establishing the dissolution profile.

Different dissolution profiles were compared to establish the effect of formulation or process variables or dissolution medium on the drug release. The dissolution similarity was assessed using the FDA recommended approach (f2 similarity factor). This model independent mathematical approach was described by Moore and Flanner (1996): n

f 2 = 50 ⋅ log{[1 + (1 / n)∑ (Rt − Tt ) 2 ] − 0.5 ⋅ 100}

(16)

t =1

where Rt and Tt are the cumulative percentage dissolved at each of the selected n time points of the reference and test product respectively Factor f2 is inversely proportional to the average squared difference between the two profiles, with emphasis on the larger difference among all the time-points.

- 66 -

The transformation is such that the f2 equation takes values less or equal to 100. When the two profiles are identical, f2=100. An average difference of 10% at all measured time points results in an f2 value of 50 (Shah et al., 1998). FDA has set a public standard of f2 value between 50 -100 to indicate similarity between two dissolution profiles. To use mean data for extended release products, the coefficient of variation for mean dissolution profile of a single batch should be less than 10% (FDA, 1997b). The average difference at any dissolution sampling point should not be greater than 15% between the tested and reference products (FDA, 1997a). Because f2 values are sensitive to the number of dissolution time points, for extended release products only one point past the plateau of the profiles should be used in the calculation (FDA, 1997a). The dissolution profiles were fitted using the Higuchi model (for drug release up to 60%) and the R2 was reported. For the dissolution profiles, which confirmed this diffusion model, the slopes of the curves were used to compare the release rates.

- 67 -

3.6. Experimental design and methodology

3.6.1. Propranolol 10 mg tablets Literature data were available just on Kollidon® SR application as extended release excipient for high dose drug matrix tablets manufactured by direct compression method (BASF, 1999, Pathan and Jalil, 2000, Rock et al., 2000). In this dissertation the suitability of Kollidon® SR was evaluated for low-dose drug extended release systems, using propranolol HCl (10mg) as a model drug. A full factorial design was applied to study the effect of polymer concentration (10-50% w/w of the tablet weight) and the method of manufacture (direct compression and wet granulation) on tablet properties and drug release. Tablets were manufactured by direct compression or wet granulation (section 3.6.1.1 – page 68, section 3.6.1.2 – page 69). For the wet granulation technology, the effect of the addition of an external hydrophilic or hydrophobic binder was investigated.

3.6.1.1. Manufacture of propranolol 10mg tablets by direct compression All ingredients in their specified ratios as mentioned in Table 3 – page 69 were blended in a turbula mixer and tablets manufactured by direct compression method (process flow chart - Figure 3, page 62) to a target weight of 133.33mg/tablet and hardness of about 10 KP, using 7 mm round punches.

- 68 -

Table 3. Propranolol 10mg matrix tablets formulation Ingredients

Formula 1

Formula 2

Formula 3

Formula 4

Formula 5

7.50

7.50

7.50

7.50

7.50

Kollidon® SR

10.00

20.00

30.00

40.00

50.00

Emcocel® 90M

40.75

35.75

30.75

25.75

20.75

Emcompress®

40.75

35.75

30.75

25.75

20.75

Aerosil® 200

0.50

0.50

0.50

0.50

0.50

Magnesium stearate

0.50

0.50

0.50

0.50

0.50

100.00

100.00

100.00

100.00

100.00

Propranolol HCl

Total

(% of the tablet weight)

3.6.1.2. Manufacture of propranolol 10mg tablets by wet granulation For wet granulation, the blends were granulated in a planetary mixer by adding distilled water (Formulations 1-5, Table 3 – page 69) or binder dispersion (Formulations 6-9, Table 4 – page 70) and the tablets were compressed to a target weight of 133.33mg/tablet and hardness of about 10 KP, using 7 mm round punches (process flow chart – Figure 4 – page 63). Reproducibility batches were manufactured under the same conditions.

- 69 -

Table 4. Propranolol 10mg matrix tablets formulations with external binders Ingredients

Formula 6

Formula 7

Formula 8

Formula 9

7.50

7.50

7.50

7.50

Kollidon® SR

30.00

30.00

50.00

50.00

Kollidon® 30

5.00

-

5.00

-

-

5.00

-

5.00

Emcocel® 90M

28.25

28.25

18.25

18.25

Emcompress®

28.25

28.25

18.25

18.25

Aerosil® 200

0.50

0.50

0.50

0.50

Magnesium stearate

0.50

0.50

0.50

0.50

Total

100

100

100

100

Propranolol HCl

Kollicoat® SR30D

(% of the tablet weight)

Table 5. Propranolol 10mg tablets formulated with Eudragit® RSPO Ingredients

Eudragit® RSPO 30%

Eudragit® RSPO 40%

Eudragit® RSPO 50%

Propranolol HCl

7.50

7.50

7.50

Eudragit® RSPO

30.00

40.00

50.00

Emcocel® 90M

30.75

25.75

20.75

Emcompress®

30.75

25.75

20.75

Aerosil® 200

0.50

0.50

0.50

Magnesium stearate

0.50

0.50

0.50

100.00

100.00

100.00

Total (% of the tablet weight)

- 70 -

3.6.1.3. Drug release profiles from propranolol 10mg matrix tablets manufactured with Eudragit® RSPO Kollidon® SR was replaced in direct compression with a polymethacrylate polymer, Eudragit® RSPO at 30, 40 and 50% concentration levels and the drug release profiles in distilled water were compared (Table 5– page 70).

3.6.1.4. Testing of propranolol 10mg tablets Tablets were tested for physical properties and in vitro drug release according to the USP 25 (apparatus 2) paddle method at 50 rpm in 1000 ml of distilled water maintained at 37±0.5°C (Method A – section 3.5, page 65). The effect of dissolution medium on drug release was tested for the formulations with 30, 40 and 50% Kollidon® SR. The release profiles in three different dissolution media, distilled water, USP pH 6.8 phosphate buffer and 0.1N Hydrochloric acid (method A – section 3.5, page 65) were compared using the FDA recommended approach (f2 similarity factor). The applicability of the diffusional release mechanism (Higuchi time square model) was assessed.

3.6.2. Buspirone 10 mg tablets Buspirone HCl was selected as model drug in this set of experiments, based on its solubility in water, basic character and low-dose loading (10mg). Buspirone HCl has a short and variable biological half-life (2-3 hours), and high

- 71 -

first pass metabolism (Sakr and Andheria, 2001, Mahmood and Sahajwalla, 1999). Two independent variables, polymer level as formulation parameter and compression force, as process parameter were tested. A full factorial design at three levels of compression force (1000, 2000, 3000lbs) and six levels of polymer concentration (10-60%) was used. Also a control batch without the polymer was manufactured. Buspirone and fillers were optimally mixed with Kollidon® SR at various concentrations

and

directly

compressed

into

capsule-shaped

tablets

(0.185 x 0.426 in) of label claim 10 mg Buspirone (tablet weight 160.0mg), under standardized conditions, according to the process flow chart presented in Figure 3 – page 62. Table 6. Formulation of buspirone 10mg tablets Ingredient (%)

KSR 0%

KSR 10%

KSR 20%

KSR 30%

KSR 40%

KSR 50%

KSR 60%

Buspirone HCl

6.25

6.25

6.25

6.25

6.25

6.25

6.25

Kollidon® SR

0.00

10.00

20.00

30.00

40.00

50.00

60.00

Emcocel® 90M

46.375 41.375 36.375 31.375 26.375 21.375 16.375

Emcompress®

46.375 41.375 36.375 31.375 26.375 21.375 16.375

Aerosil® 200

0.50

0.50

0.50

0.50

0.50

0.50

0.50

Magnesium stearate

0.50

0.50

0.50

0.50

0.50

0.50

0.50

Total

100.00 100.00 100.00 100.00 100.00 100.00 100.00

(% of the tablet weight)

- 72 -

The physical properties of the tablets and the drug release in water, 0.1N HCl and pH 6.8 phosphate buffer were tested and evaluated as mentioned in section 3.5 – page 65.

3.6.3. Propranolol 80 mg tablets This set of experiments was designed to evaluate the potential of Kollidon® SR as matrix former for propranolol 80mg tablets (high dose), knowing that the development of monolithic extended release matrices for high dose highly soluble drugs presents a challenge.

3.6.3.1. Manufacture

of

propranolol

80mg

tablets

with

40-60%

Kollidon® SR The experimental design was a full factorial for two factors at three levels each: polymer concentration (40, 50 and 60%) (Table 7 – page 74) and compression force (1000, 2000, 3000lbs). Tablets were manufactured by direct compression according to the process flow chart presented in Figure 3 – page 62, using bisect capsule shaped punches (0.185 x 0.0.426 in) to a target weight of 225mg/tablet.

- 73 -

Table 7. Formulation of propranolol 80mg tablets with 40-60% Kollidon® SR Ingredient (%)

40% KSR

50% KSR

60% KSR

Propranolol HCl

35.55

35.55

35.55

Kollidon® SR

40.00

50.00

60.00

Emcocel® 90M

11.725

6.725

1.725

Emcompress®

11.725

6.725

1.725

Aerosil® 200

0.50

0.50

0.50

Magnesium stearate

0.50

0.50

0.50

Total

100.0

100.0

100.0

(% of the tablet weight)

3.6.3.2. Testing of propranolol 80mg tablets with 40-60% Kollidon® SR Tablets were tested for physical properties and drug release in distilled water, 0.1N HCl and pH 6.8 buffer (method A – section 3.5, page 65). The released amounts were plotted as function of square root of time, to determine the mechanism of drug release. A model independent approach using similarity factor f2 was used to compare the dissolution profiles.

3.6.3.3. Manufacture of propranolol 80mg tablets with 70% polymer (Kollidon® SR alone or in combination with Eudragit® L100-55) In this set of experiments, propranolol 80mg tablets were formulated by increasing the polymer level up to 70% of the tablet weight and/or partial replacement of Kollidon® SR with of an enteric polymer (Eudragit® L100-55) (Table 8 – page 75). As a result of increasing the polymer level, tablet weight had to be increased to allow 70% polymer addition. - 74 -

The objective was to study the effect of addition of an enteric polymer on drug release and to modify the release to be close / similar to the USP requirements for extended release propranolol capsules. Tablets were manufactured by direct compression, using bisect capsule shaped punches (0.220 x 0.500 in) to a target weight of 275.86mg (~276mg) and hardness 10-15 kP (flow chart Figure 3 – page 62).

Table 8. Formulation of propranolol 80mg tablets with 70% polymer Ingredient (%)

70% KSR

65% KSR 5% Eudragit® L100-55

60% KSR 10% Eudragit® L100-55

Propranolol HCl

29.00

29.00

29.00

Kollidon® SR

70.00

65.00

60.00

-

5.00

10.00

Aerosil® 200

0.50

0.50

0.50

Magnesium stearate

0.50

0.50

0.50

Total

100.0

100.0

100.0

Eudragit® L100-55

(% of the tablet weight)

3.6.3.4. Testing of propranolol 80mg tablets with 70% polymer (Kollidon® SR alone or in combination with Eudragit® L100-55) Tablets were tested for physical properties and drug release in various media (method A – section 3.5, page 65) and according to the USP method for propranolol extended release capsules (method B – section 3.5, page 65). The release data obtained were plotted as a function of square root of time, to

- 75 -

determine the mechanism of drug release. A model independent approach using similarity factor f2 was used to compare the dissolution profiles.

3.6.3.5. Testing of Inderal® LA capsules (reference listed drug product) Inderal® LA (lot #9010268), the reference listed product, was tested for the drug release in different media according to method A and method B (section 3.5, page 65). This step was necessary because the reference listed product served as comparison for some of the developed matrix tablet formulations.

3.6.3.6. Selection

of

propranolol

80mg

formulation

for

pilot

bioequivalence study The release profiles (obtained according to method B – section 3.5, page 65) of the developed matrix tablet formulations with 60 and 70% Kollidon® SR were compared to Inderal® LA and it was decided to formulate and manufacture tablets using an intermediate polymer level (65%).

3.6.3.7. Testing of propranolol 80mg tablets for the pilot bioequivalence study For

the

selected

formulation

(65%

Kollidon® SR),

the

following

characteristics/tests were performed: physical properties of the tablets, content uniformity, propranolol release – method A and B (section 3.5, page 65), reproducibility, effect of storage on tablet hardness and in vitro drug release.

- 76 -

Effect of storage on tablet physical properties and drug release

Propranolol 80mg tablets with 65% Kollidon® SR were stored in HDPE bottles in the presence of desiccant under different storage conditions (FDA, 2001 ICH Q1A, FDA, 1997 ICH Q1C). At predetermined time points, the tablets were sampled and tested for physical properties and drug release (Table 9 – page 77).

Table 9. Stability study design Study

Long-term

Storage condition

25 ± 2°C / 60± 5%RH

Accelerated 40 ± 2°C / 75± 5%RH

Frequency of testing

Tests performed

0, 1, 3, 6, 9 months

Appearance, weight, thickness, hardness, drug release – method B

0, 3 and 6 months (9 months included, although not required by ICH

Appearance, weight, thickness, hardness, drug release – method B

3.6.4. Pilot bioequivalence study

3.6.4.1. Design and methodology The relative bioavailabilities of the selected propranolol 80mg extended release matrix tablets and reference listed drug product Inderal® LA 80mg were evaluated in a pilot bioequivalence study, according to a protocol (# 01-6-19-1) approved by the University of Cincinnati Institutional Review Board (Appendix 1page171). The study design was a randomized cross-over single-dose twoperiod open-label two-treatment, with a wash out period of one week. - 77 -

According to the 21 CFR 320.31 b., the study did not require an IND submission because it was designed to assess the bioavailability / bioequivalence in humans of single dose of an approved non-new chemical entity (propranolol hydrochloride) and the dose did not exceed the maximum single dose specified in the labeling of the drug product that is the subject of an approved new drug application or abbreviated new drug application. Additionally, correspondence with the Food and Drug Administration - Office of Generic Drugs was submitted to the Institutional Review Board to support the protocol approval. The pilot study was conducted in compliance with the requirements for IRB review and informed consent (21 CFR parts 56 and 50, respectively) and with the requirements concerning the promotion and sale of drug (21 CFR 312.7). The study did not invoke 21 CFR 50.24. It was performed under medical supervision at the University of Cincinnati and Veterans Affairs Hospital facilities in Cincinnati. Ten volunteers underwent a screening procedure 2-6 days prior to the first testing period and 8 subjects who met inclusion criteria, provided written consent were enrolled in the study and randomized to one of the two dosing sequences (SAS software). Inclusion Criteria

•

Healthy, male and female subjects between the ages of 18 – 65 years inclusive

•

Subjects must be outpatients at the time of screening

- 78 -

•

Subjects must be on no chronic medications (prescription or OTC) and must be medication-free for a period of at least one week prior to the first test day and throughout the duration of the study

•

Subjects must be off any investigational drug for a period of at least 3 months prior to the entry in the study

•

Subjects must be in good health as determined by medical history, routine physical examination, ECG and clinical laboratory tests

•

Subjects must be free of significant psychiatric illness

•

Subjects must be willing and able to provide written informed consent.

Exclusion Criteria

•

Subjects with a history or evidence of clinically significant and currently relevant

hematological,

pulmonary,

renal,

dermatological,

hepatic,

oncological

gastrointestinal, or

neurological

endocrine, illness,

and

alcoholism. •

Subjects with a history of cardiovascular disease, including hypotension, hypertension, heat block, congestive heart failure, angina pectoris, bypass surgery, or myocardial infarction

•

Subjects with clinically significant abnormalities on the electrocardiogram at screening

•

Pregnant and breast-feeding women were not eligible

•

Subjects using concomitant drugs

•

Subjects with known allergy to propranolol

•

Subjects with clinically significant emotional problems - 79 -

•

Subjects unable and/or unlikely to comprehend and follow the study protocol.

Screening examinations

•

Routine physical examination and medical history

•

Safety examination – ECG before the treatment, blood pressure, pulse and temperature

•

Laboratory examination – complete blood count with differential, hepatic and renal profiles.

Treatments

♦

Test product - propranolol 80mg developed extended release matrix tablets.

♦

Reference product - Inderal® LA 80mg capsules (reference listed product, innovator product).

Methodology

Subjects were admitted as outpatients in the morning (7.30 am) of the first day of each period and after the insertion of the catheter, they received a single dose of the drug (test or reference product) at 8.00am (0 hour of the test). Subjects were in the facility until 8.00pm (after the 12-hour blood sample was withdrawn). Subjects returned the second day of each period at 8.00 am and 2.00pm for the 24-hour and 30-hour blood sample withdrawal. Meals and Food Restrictions. Subjects fasted for at least 12 hours prior to the

dose administration. Prior to and during each study phase subjects were allowed water as desired except for one hour before and after drug administration. Subjects received lunch at 1.00pm. Subjects abstained from alcohol for 24 hours

- 80 -

prior to each study period and until after the last sample from each period was collected. Use of tobacco and caffeine was not allowed for 24 hours prior to each study period and until after the last sample from each period was collected. Subject monitoring. The blood pressure and pulse rate were monitored prior to

dosing and at the sampling times. The treatment effects on blood pressure and pulse rate at every time point were tested by one-way ANOVA. Subjects had their weight measurements taken and recorded at each period. Subjects were advised to avoid the use of prescription and OTC medications. Blood samples

During each period, 12 venous blood samples were taken from the antecubital veins in heparinized vacutainers as follows: •

Day 1 - at 0 (pre-dose), and at 1, 2, 3, 4, 5, 6, 8, 10, 12 hours post-dose (using catheter hep-lock)

•

Day 2 - at 24, 30 hours post-dose (by direct venipuncture).

The plasma was separated by centrifugation (3000g x 15minutes) and then, transferred to the labeled tubes and promptly frozen. The samples were stored at –70°C, until analyzed.

3.6.4.2. Analysis of propranolol in plasma Propranolol was analyzed in plasma by a reverse phase HPLC - fluorescence detection method, developed based on published data (Drummer et al., 1981, Braza et al., 2000, Rekhi et al., 1995). The method parameters are presented in Table 10 – page 82.

- 81 -

To 1.0ml spiked plasma or sample, 0.1ml 1M NaOH, 0.1 ml Pronethalol 600ng/ml (internal standard) and 5ml ethyl acetate were added. The mixture was vortexed for 15sec and then centrifuged at 3000 rpm for 3 minutes. 4ml of the supernatant were transferred to disposable vials and evaporated to dryness at 40°C using nitrogen steam. The residue was reconstituted in 0.5ml mobile phase, vortexed for 10sec and 100µl were injected on the column.

Table 10. Analytical method for analysis of propranolol in plasma Column

Metachem Inertsil ODS-3 5µm 250x4.6mm HPLC Column with MetaGuard 4.6mm Inertsil ODS-3 5µm

Mobile phase

Acetonitrile : Water with 1.2% (w/v) triethylamine and pH adjusted to 3 with 85% orthophosphoric acid =30 : 70

Flow rate

1.0ml/min

Injection volume

100µl

Detection Method

Fluorescence detection excitation wavelength 280nm, emission wavelength 333nm

Linearity. Daily standard curves were prepared by spiking plasma with

propranolol HCl solution in water to obtain the following final concentrations: 2, 4, 10, 20, 40, 100ng/ml. Calibration curves were generated by plotting the ratio of areas of propranolol / internal standard versus ratio of the concentrations of the two components. The calibration curve was considered linear for values of the correlation coefficient above 0.99.

- 82 -

Accuracy. Three concentrations within the linearity range (2, 20, 100ng/ml) were

prepared by spiking the plasma with the corresponding amount of propranolol solution and internal standard and analyzed. Accuracy was calculated as percentage of measured (recovered) concentration to theoretical values. Intra- and inter-day variability. The intra-day variability was determined by

analyzing three replicates of spiked plasma at three different concentrations (2, 20, 100ng/ml). For inter-day variability, the samples were prepared and injected into the column on two consecutive days.

3.6.4.3. Pharmacokinetic and statistical analysis The values of the concentrations were natural log-transformed and a noncompartmental pharmacokinetic model was applied to calculate Cmax, area under the concentration time curves from 0-24h (AUC 0-24h) and 0-∞ (AUC 0-∞) for each subject and formulation (WinNonlin). The resulting data were statistically analyzed by a non-parametric test (Wilcoxon) for carry-over (residual) effects, sequence and treatment effects (SAS software). The bioequivalence of the two formulations was tested using the following model (WinNonlin software - bioequivalence wizard): Y = intercept + sequence + treatment + period Random effect = subject (sequence).

- 83 -

4. Results and Discussions

4.1. Propranolol 10 mg tablets

4.1.1. Effect of Kollidon® SR on drug release from propranolol 10mg tablets manufactured by direct compression Propranolol 10mg tablets were manufactured with different concentrations of Kollidon® SR, (10, 20, 30, 40 and 50% of tablet weight) – section 3.6.1.1, page 68. Tablets were uniform in weight and thickness and their hardness increased as the concentration of polymer in the formulation increased (Table 11 – page 84).

Table 11. Effect of Kollidon® SR on physical properties of propranolol 10mg tablets manufactured by direct compression Kollidon® SR

Weight (mg)

Thickness (mm)

Hardness (kP)

Average

RSD

Average

RSD

Average

RSD

10% KSR

131.49

0.577

3.897

0.172

4.14

14.324

20% KSR

132.75

0.725

3.923

0.173

6.47

7.252

30% KSR

132.78

0.001

4.007

0.407

8.91

11.524

40% KSR

133.87

1.264

4.152

1.077

11.54

10.024

50% KSR

133.68

0.001

4.224

0.488

13.12

10.300

- 84 -

It was found that increasing polymer concentration up to 40%, significantly decreased the drug release rate in water, sustaining the release of the highly water soluble drug incorporated at low dose for a longer period of time (dissolution data for all the experimental batches were reproducible n=6, RSD<3% and hence only the average values were plotted). There was no significant difference between the formulations containing 40% and 50% of the polymer content f2>50 (Figure 5 – page 86).

The regression parameters of the drug release curves for formulations with 3050% polymer content are indicated in Table 12 – page 85 and the plot of percent drug released versus square root of time is illustrated in Figure 6 – page 87. The high correlation coefficient (above 0.99) obtained indicates a square root of time dependent release kinetics. Thus, as the data fitted the Higuchi model, it confirmed a diffusion drug release mechanism.

Table 12. Regression parameters of the diffusion drug release curves for propranolol 10mg tablets manufactured by direct compression Kollidon® SR %a

Slope (n)

Intercept (l)

r2

30

45.582

-7.771

0.999

40

23.404

3.356

0.997

22.947

-2.099

0.999

50 1/2

a

(Q=n*t +l) Kollidon® SR as percentage of total tablet weight

- 85 -

110

100

90

80

%released

70

10% KSR 20% KSR 30% KSR 40% KSR 50% KSR

60

50

40

30

20

10

0 0

4

8

12

16

20

24

time (hr)

Figure 5. Effect of Kollidon® SR on drug release in water from propranolol 10mg tablets manufactured by direct compression

- 86 -

110 100 90 80

%released

70 60 30% KSR 40% KSR

50

50% KSR

40 30 20 10 0 0

1

2

3

4

√t (√hr)

Figure 6. Effect of Kollidon® SR on diffusion controlled drug release in water from propranolol 10mg tablets manufactured by direct compression

- 87 -

It is suggested that the main driving force for the drug release in case of watersoluble drug like propranolol hydrochloride from the matrix tablets is the infiltration of release medium. As the tablet is introduced into the medium, water penetrates into the matrix and povidone leaches out to form pores through which the drug may diffuse out. Also, as observed in Figure 6 – page 87, as the polymer level in the formulation is increased, drug diffusion is slowed due to the lower porosity and higher tortuosity of the matrix. Thus polyvinylacetate, which is a very plastic material, produces a coherent matrix, sustaining the drug release from the tablet matrix. Similarly, Ruchatz et al 1999 reported that caffeine was released from Kollidon® SR matrix tablets by diffusion over more than 16 hours. The matrix remained intact during the dissolution test due to the water-insoluble polyvinylacetate.

4.1.2. Effect of Kollidon® SR on drug release from propranolol 10mg tablets manufactured by wet granulation The application of Kollidon® SR for tablets manufactured by wet granulation using distilled water as granulating medium, was studied (section 3.6.1.2 – page 69). Tablets were uniform in weight, thickness and hardness ( Table 13 – page 89).

- 88 -

Table 13. Effect of Kollidon® SR on the physical properties of Propranolol 10mg tablets manufactured by wet granulation Kollidon® SR

Weight (mg)

Thickness (mm)

Hardness (kP)

Average

RSD

Average

RSD

Average

RSD

10% KSR

135.96

1.065

3.998

0.105

8.37

4.141

20% KSR

134.84

1.419

4.029

0.273

9.40

6.784

30% KSR

134.80

0.001

4.079

0.270

11.11

8.055

40% KSR

135.49

1.570

4.133

0.511

12.81

11.709

50% KSR

133.30

0.002

4.207

0.663

11.40

7.725

- 89 -

110

100

90

80

%released

70

60

10% KSR 20% KSR 30% KSR 40% KSR 50% KSR

50

40

30

20

10

0 0

2

4

6

8

10

12

time (hr)

Figure 7. Effect of Kollidon® SR on drug release in water from propranolol 10mg tablets manufactured by wet granulation

- 90 -

110 100 90 80

%released

70 60 30% KSR 40% KSR

50

50% KSR

40 30 20 10 0 0

1

2

3

4

√t (√hr)

Figure 8. Effect of Kollidon® SR on diffusion controlled drug release in water from propranolol 10mg tablets manufactured by wet granulation

- 91 -

The drug release in water is shown in Figure 7 – page 90 and the Higuchi plots in Figure 8 – page 91. By comparing the slopes of Higuchi plots as an indicator for release rate, it can be seen that wet granulation (Table 14 – page 92) produced a faster release than direct compression (Table 12– page 85).

Table 14. Regression parameters of the diffusion drug release curves for propranolol 10mg tablets manufactured by wet granulation Kollidon® SR %a

Slope (n)

Intercept (l)

r2

30

42.438

-7.037

0.999

40

37.774

-5.167

0.999

50

53.380

-16.238

0.994

(Q=n*t1/2+l) aKollidon® SR as percentage of total tablet weight

In contrast to the direct compression method, in tablets manufactured by wet granulation, increasing the polymer concentration from 30 to 50%, produced a faster rate of drug release from the matrix. The regression parameters for Higuchi model are presented in Table 14 – page 92 and the change in release profiles is indicated by the varying slope values for the square root of time plots. This behavior could be attributed to a faster penetration of waterfront into the matrix, leading to a formation of more porous structure in the matrix. The povidone in the polymer would have deposited on the polyvinylacetate particles during granulation, thus localizing as discrete granules between polyvinylacetate particles, leading to a faster channeling action. The lower tortuosity and higher

- 92 -

water penetration due to an increase in the volume of povidone at 50% polymer content, could also lead to a faster drug release rate. As Kollidon® SR was not studied before for wet granulation applications, no literature data were available for comparison purposes.

4.1.3. Effect of external binder on drug release from propranolol 10mg tablets manufactured by wet granulation The effect of the addition of an external binder in the granulating medium, on the drug release rate from formulations containing 30 and 50% Kollidon® SR content was evaluated and the release profiles are as shown in Figure 9 – page 94. The two binders studied at 5% concentration levels were water-soluble Kollidon® 30 and Kollicoat® SR30D aqueous dispersion with hydrophobic properties. No significant change in drug release profiles (f2 >50) was observed at 30% Kollidon® SR level. At a concentration of 50% Kollidon® SR, additional external binder did not slow the release as expected. None of the two binders used could compensate for the reduced interaction of the hydrophobic polyvinylacetate with the other hydrophilic components from the tablets (reduced interaction caused by exposure of the polymer during the wet granulation process) (Mulye and Turco, 1994). The results indicated that Kollidon® SR was primarily controlling the drug release rate.

- 93 -

110

100

90

80

%released

70

30% KSR / water

60

30% KSR / 5% Kollidon 30 50

30% KSR / 5% Kollicoat SR30D 50% KSR / water

40

50% KSR / 5% Kollidon 30 30

50% KSR / 5% Kollicoat SR30D

20

10

0 0

2

4

6

8

10

12

time (hr)

Figure 9. Effect of external binder on drug release in water from propranolol 10mg tablets with 30% and 50% Kollidon® SR

- 94 -

4.1.4. Effect of dissolution medium on drug release from propranolol 10mg matrix tablets Drug release from tablets with 30, 40 and 50% Kollidon® SR was tested in three different dissolution media: distilled water, USP pH 6.8 phosphate buffer and 0.1N hydrochloric acid (Figure 10 - Figure 15, pages 96 - 101). On applying the similarity factor, f2, to compare the dissolution in 0.1N HCl or pH 6.8 buffer to the release in water, values of above 50 were obtained indicating the similarity of the release profiles (Table 15 – page 102). Drug release from matrix systems is influenced by the aqueous solubility of the drug and matrix behavior at different pH. Propranolol has a pKa=9.5 (Avdeef et al., 2000) and an acceptable solubility over the physiologic pH range: 220 mg/ml in 0.1 N HCl and 254 mg/ml in pH 7.4 phosphate buffer (Siepmann and Kranz, 2000). Kollidon® SR contains no ionic groups, hence it is inert to drug substances and its solubility and hydration are not influenced by pH. As a result, the drug release was pH-independent and it was concluded that Kollidon® SR is suitable for the manufacturing of pH-independent extended release matrix tablets, on the condition that drug solubility does not drastically change with the pH.

- 95 -

110

100

90

80

% released

70

60 water 0.1N HCl

50

pH 6.8 buffer 40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 10. Effect of dissolution medium on drug release from propranolol 10mg tablets with 30% Kollidon® SR manufactured by direct compression

- 96 -

110

100

90

80

% released

70

60 water 0.1N HCl

50

pH 6.8 buffer

40

30

20

10

0 0

2

4

6

8

10

12

time (hr)

Figure 11. Effect of dissolution medium on drug release from propranolol 10mg tablets with 30% Kollidon® SR manufactured by wet granulation

- 97 -

110

100

90

80

% released

70

60 water 0.1N HCl

50

pH 6.8 buffer 40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 12. Effect of dissolution medium on drug release from propranolol 10mg tablets with 40% Kollidon® SR manufactured by direct compression

- 98 -

110

100

90

80

% released

70

60

50

water 0.1N HCl pH 6.8 buffer

40

30

20

10

0 0

2

4

6

8

10

12

time (hr)

Figure 13. Effect of dissolution medium on drug release from propranolol 10mg tablets with 40% Kollidon® SR manufactured by wet granulation

- 99 -

110

100

90

80

% released

70

60 water 0.1N HCl

50

pH 6.8 buffer

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 14. Effect of dissolution medium on drug release from propranolol 10mg tablets with 50% Kollidon® SR manufactured by direct compression

- 100 -

110

100

90

80

% released

70

60 water

50

0.1N HCl pH 6.8 buffer

40

30

20

10

0 0

2

4

6

8

10

12

time (hr)

Figure 15. Effect of dissolution medium on drug release from propranolol 10mg tablets with 50% Kollidon® SR manufactured by wet granulation

- 101 -

Table 15. f2 values - effect of dissolution medium on drug release from propranolol 10mg tablets Formulation

f2 (0.1N HCl – water)

f2 (pH 6.8 buffer – water)

30% Kollidon® SR direct compression

86.62

82.81

30% Kollidon® SR wet granulation

72.99

53.77

40% Kollidon® SR direct compression

94.30

72.33

40% Kollidon® SR wet granulation

82.06

55.45

50% Kollidon® SR direct compression

72.10

88.77

50% Kollidon® SR wet granulation

75.44

84.31

- 102 -

4.1.5. Drug release profiles from matrix tablets with Eudragit® RSPO Kollidon® SR was replaced in direct compression with a polymethacrylate polymer, Eudragit® RSPO. Tablets with 30, 40 and 50% polymer levels were manufactured and the drug release profiles in distilled water were compared. The drug release was faster (Figure 16 – page 104), with about 80-100% of propranolol released in the first 1-2 hours, which was attributed to a rapid and complete erosion of the matrix (disintegration time for all Eudragit® RSPO formulations tested was less that 10 minutes). This was a result of a low cohesiveness of the powder during compression, the maximum hardness which could be achieved (under maximum compression force) was between 4-6kP.

- 103 -

110

100

90

80

%released

70

60

50

Eudragit RSPO 30% Eudragit RSPO 40% Eudragit RSPO 50%

40

30

20

10

0 0

2

4

6

8

10

12

time (hr)

Figure 16. Effect of Eudragit® RSPO on drug release in water from propranolol 10mg tablets

- 104 -

4.2. Buspirone 10mg tablets

4.2.1. Effect of Kollidon® SR and compression force on physical properties and drug release of buspirone 10mg tablets Buspirone tablets were found uniform in weight and thickness and had high mechanical strength, even under the lowest applied compression force (Table 16 – page 107). Increasing the compression force from 1000lbs to 2000lbs significantly increased the hardness of the tablets, but further increase above 2000lbs did not significantly change the hardness of the tablets with 40 - 60% Kollidon® SR (p>0.05). Increasing the polymer concentration increased the hardness, mainly due to the polyvinylacetate component, which is a very plastic material (Figure 17 – page 106).

Increasing the compression force from 1000lbs to 2000 lbs reduced the release rate in water, but compression forces above 2000 lbs did not significantly change the drug release profile f2>50 (Figure 17 - Figure 19, pages 106 - 109; Table 17 – page 110).

- 105 -

25

20

0% KSR 10% KSR

15 Hardness (kP)

20% KSR 30% KSR 40% KSR 50% KSR 60% KSR 10

5

0 0

1000

2000

3000

4000

Compression force (lbs)

Figure 17. Effect of Kollidon® SR concentration and compression force on the hardness of buspirone 10mg tablets

- 106 -

Table 16. Physical properties of buspirone 10mg tablets Compression Force 0% KSR

10% KSR

20% KSR

30% KSR

40% KSR

50% KSR

60% KSR

Weight (mg)

Thickness (mm) Hardness (kP)

Average RSD Average

RSD Average RSD

1000lbs

160.74

0.995

3.102

0.204

4.41

5.286

2000lbs

160.20

1.196

2.811

0.615

9.15

4.349

3000lbs

160.73

0.855

2.748

0.537

11.27

5.751

1000lbs

158.40

0.710

3.037

0.222

7.12

8.551

2000lbs

162.51

0.551

2.919

0.441

11.71

2.744

3000lbs

162.24

0.844

2.863

0.595

13.86

2.102

1000lbs

161.08

0.901

3.127

0.905

11.00

4.791

2000lbs

161.22

0.584

2.996

0.322

12.56

11.624

3000lbs

158.39

0.563

2.914

0.289

14.94

2.913

1000lbs

159.31

1.358

3.218

0.353

9.80

5.931

2000lbs

161.10

0.482

3.101

0.238

15.16

3.381

3000lbs

162.22

0.735

3.083

0.485

18.44

2.387

1000lbs

161.82

0.720

3.588

0.729

8.08

5.887

2000lbs

160.03

0.374

3.194

0.219

17.63

2.593

3000lbs

157.94

1.477

3.128

0.795

18.07

5.250

1000lbs

158.77

0.463

3.621

0.087

9.07

2.801

2000lbs

160.86

0.544

3.314

0.432

20.87

3.873

3000lbs

159.14

1.082

3.262

0.853

21.70

4.334

1000lbs

161.47

0.714

3.739

0.320

12.02

2.280

2000lbs

158.90

0.843

3.376

0.348

22.14

2.998

3000lbs

156.03

0.69

3.325

0.622

21.73

5.079

- 107 -

110

100

90

80

%released

70

60

10% KSR - 1000lbs 10% KSR - 2000lbs 10% KSR - 3000lbs 20% KSR - 1000lbs 20% KSR - 2000lbs 20% KSR - 3000lbs 30% KSR - 1000lbs 30% KSR - 2000lbs 30% KSR - 3000lbs

50

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 18. Effect of compression force on drug release from buspirone 10mg tablets with 10 - 30% Kollidon® SR

- 108 -

110

100

90

80

%released

70

60

50

40

40% KSR - 1000lbs 40% KSR - 2000lbs 40% KSR - 3000lbs 50% KSR - 1000lbs 50% KSR - 2000lbs 50% KSR - 3000lbs 60% KSR - 1000lbs 60% KSR - 2000lbs 60% KSR - 3000lbs

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 19. Effect of compression force on drug release from buspirone 10mg tablets with 40 - 60% Kollidon® SR

- 109 -

Table 17. f2 values - effect of compression force on drug release from buspirone 10mg tablets Formulation

f2 value (2000 lbs – 3000 lbs)

30% Kollidon® SR

94.45

40% Kollidon® SR

81.45

50% Kollidon® SR

86.54

60% Kollidon® SR

98.50

Consequently, further testing was carried out for tablets compressed at 2000 lbs. By increasing Kollidon® SR concentration in the tablets, drug diffusion was slowed down due to the lower porosity and higher tortuosity of the matrix. Consequently, the drug release rate significantly decreased, prolonging the release of the buspirone up to 24 hours (Figure 20 – page 111). A minimum Kollidon® SR concentration of 30% was necessary in order to achieve a coherent matrix and an extended drug release.

The release of drug dispersed in the matrix systems fitted the Higuchi model (Table 18 – page 113), which denoted a diffusion-controlled mechanism (Figure 21 – page 112).

- 110 -

110

100

90

80

%released

70

60

50

40

KSR 0% KSR10% KSR20% KSR30% KSR40% KSR50% KSR60%

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 20. Effect of Kollidon® SR on drug release from buspirone 10mg tablets

- 111 -

110

100

90

80

%released

70

60

50

40 30% KSR

30

40% KSR 50% KSR 60% KSR

20

10

0 0

1

2

3

4

5

√t (√hr)

Figure 21. Effect of Kollidon® SR on diffusion controlled drug release from buspirone 10mg tablets

- 112 -

Table 18. Regression parameters of the diffusion drug release curves for buspirone 10mg tablets Kollidon® SR %a

Slope (n)

Intercept (l)

r2

30

32.702

-1.262

0.969

40

18.086

7.598

0.972

50

15.387

8.571

0.985

60

10.774

7.859

0.986

(Q=n*t1/2+l) aKollidon® SR as percentage of total tablet weight

4.2.2. Effect of dissolution medium on drug release from buspirone 10mg tablets Although Kollidon® SR is a non-ionic polymer, buspirone release rate at each polymer level in the three dissolution media varied (Table 19 – page 118); the fastest release was obtained in 0.1N HCl and the slowest in pH 6.8 phosphate buffer (Figure 22 - Figure 25, pages 114 - 117). This was attributed to the pHdependent solubility of buspirone, which is a basic drug (pKa1=4.12, pKa2=7.32). It was concluded that although Kollidon® SR can promote a pH-independent release, the drug release is also a function of drug solubility.

- 113 -

110

100

90

80

%released

70

60

0.1N HCl water pH 6.8

50

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 22. Effect of dissolution medium on drug release from buspirone 10mg tablets with 30% Kollidon® SR

- 114 -

110

100

90

80

%released

70

60

50

0.1N HCl water pH 6.8

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 23. Effect of dissolution medium on drug release from buspirone 10mg tablets with 40% Kollidon® SR

- 115 -

110

100

90

80

%released

70

60

50

0.1N HCl water pH 6.8

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 24. Effect of dissolution medium on drug release from buspirone 10mg tablets with 50% Kollidon® SR

- 116 -

110

100

90

80

%released

70

60

50

40

0.1N HCl water pH 6.8

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 25. Effect of dissolution medium on drug release from buspirone 10mg tablets with 60% Kollidon® SR

- 117 -

Table 19. f2 values – effect of dissolution medium on drug release from buspirone 10mg tablets Formulation

f2 (0.1N HCl – water)

f2 (pH 6.8 buffer – water)

30% Kollidon® SR

51.03

41.16

40% Kollidon® SR

48.50

44.31

50% Kollidon® SR

48.80

46.41

60% Kollidon® SR

47.24

57.04

(Q=n*t1/2+l) aKollidon® SR as percentage of total tablet weight

4.3. Propranolol 80mg tablets

4.3.1. Effect of Kollidon® SR and compression force on physical properties and drug release from propranolol 80mg tablets Previous results suggested a minimum Kollidon® SR concentration of 30% is necessary for a coherent matrix, able to extend the drug release. Considering this previous finding and the higher drug dose to be used (80mg) the minimum concentration of Kollidon® SR for this set of experiments was 40%. Tablet formulations with 80mg propranolol and 40-60% Kollidon® SR were evaluated with regard to the robustness of the release to variations in compression forces, which may occur during manufacturing. The resultant tablets were uniform in weight, thickness and hardness, as shown in Table 20 – page 119.

- 118 -

Table 20. Effect of compression force and Kollidon® SR concentration on physical properties of propranolol 80mg tablets Compression force

Weight (mg)

Thickness (mm)

Hardness (kP)

Average

RSD

Average

RSD

Average

RSD

1000lbs

222.62

0.800

5.075

0.104

5.79

8.824

2000lbs

223.40

0.534

4.761

0.459

10.62

4.587

3000lbs

223.92

0.797

4.576

0.537

16.80

5.383

1000lbs

222.77

0.325

5.339

0.399

6.12

8.035

2000lbs

223.08

0.905

4.820

0.366

16.28

5.005

3000lbs

223.02

0.726

4.722

0.676

20.12

4.036

1000lbs

224.07

1.196

5.616

0.172

5.53

11.690

2000lbs

224.62

1.198

4.941

0.544

18.90

5.918

3000lbs

222.97

0.842

4.873

0.774

21.27

5.323

40% KSR

50% KSR

60% KSR

A change in the drug release due to variation in the compression force during the manufacturing process is a significant disadvantage. It is the formulator’s task to assure that minor changes in the formulation and process variables that may occur during the manufacturing process will not result in alteration of the product performance.

- 119 -

For tablet formulations containing 40 - 60% Kollidon® SR, it was observed that while changes in the compression forces from 1000 to 2000 lbs produced an increase in tablet hardness (Figure 26 – page 121) and a slight decrease in dissolution rate (not significant according to the f2 similarity factor, Table 21 – page 123), further increase to 3000 lbs did not affect the drug release profiles (Figure 27 – page 122). Therefore a robust delivery system was attained at compression force above 2000 lbs and this represented a definite advantage of these formulations. Increasing the Kollidon® SR concentration in the tablet led to an increase in tablet hardness, as shown in Figure 26 – page 121. Release profiles of the tablets that were formulated with 40-60% Kollidon® SR and compressed under 2000 lbs are shown in Figure 28 – page 124. It was found that the drug release was faster at 40% polymer levels, and further increase from 50 to 60% did not significantly change the release rate.

- 120 -

25

20

15 Hardness (kP)

40% KSR 50% KSR 60% KSR

10

5

0 0

1000

2000

3000

4000

Compression force (lbs)

Figure 26. Effect of Kollidon® SR and compression force on the hardness of propranolol 80mg tablets

- 121 -

110

100

90

80

%released

70

40% KSR 1000lbs 40% KSR 2000lbs 40% KSR3000lbs 50% KSR 1000lbs 50% KSR 2000lbs 50% KSR 3000lbs 60% KSR 1000lbs 60% KSR 2000lbs 60% KSR 3000lbs

60

50

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 27. Effect of Kollidon® SR and compression force on drug release in water from propranolol 80mg tablets

- 122 -

Table 21. f2 values – effect of compression force on drug release from propranolol 80mg tablets Formulation

f2 (1000lbs – 2000lbs)

f2 (2000lbs – 3000 lbs)

40% Kollidon® SR

53.24

57.91

50% Kollidon® SR

44.96

86.19

60% Kollidon® SR

40.92

84.72

The release was diffusion controlled (Higuchi mechanism) as confirmed by the data presented in Table 22 – page 125. When porous hydrophobic polymers drug delivery systems are placed in contact with a dissolution medium, the release of the drug must be preceded by the drug dissolution in water filled pores and by diffusion through the water filled channels. The geometry and the structure of the pore network are important to the drug release process (Gurny et al., 1982). The insoluble polyvinylacetate component of the Kollidon® SR is considered to give a coherent matrix in which the drug is dispersed and the release occurred by diffusion through the pore formed by gradually dissolving povidone. Consequently, the release rate is dependent on the porosity and tortuosity of the tablets. At lower polymer levels, the diffusion occurred faster due to lower porosity of the matrix, while increasing the polymer concentration led to a slower release until the matrix achieved its maximum tortuosity and minimum porosity. All the tablets remained intact during the 24-hour dissolution test.

- 123 -

110 100 90 80

%released

70 60

40% KSR 50% KSR 60% KSR

50 40 30 20 10 0 0

1

2

3

4

√t (√hr)

Figure 28. Effect of Kollidon® SR on diffusion controlled drug release from propranolol 80mg tablets

- 124 -

Table 22. Regression parameters of the diffusion drug release curves in water from propranolol 80mg tablets Kollidon® SR %*

Slope (n)

Intercept (l)

r2

40

49.336

1.7028

0.998

50

33.329

3.3383

0.995

60

32.245

2.4016

0.997

(Q=n*t1/2+l) aKollidon® SR as percentage of total tablet weight

4.3.2. Effect of dissolution medium on drug release from propranolol 80mg tablets Drug release from matrix systems is influenced by the aqueous solubility of the drug and matrix behavior at different pH. Propranolol has a pKa=9.5 (Avdeef et al., 2000) and an acceptable solubility over the physiologic pH range: 220 mg/ml in 0.1 N HCl and 254 mg/ml in pH 7.4 phosphate buffer (Siepmann and Kranz, 2000). Kollidon® SR contains no ionic groups and is therefore inert to drug substances and pH of the dissolution medium. The release rates at every polymer level were virtually pH independent, as confirmed by the almost superimposable release curves in pH 6.8 buffer and 0.1N HCl (Figure 29 – page 126) and f2 values greater that 50 (66.51, 73.38 and 64.95 for Kollidon® SR 40%, 50% and respectively 60%). This confirmed the findings in case of propranolol 10mg tablets, regarding the ability of Kollidon® SR to provide a pH-independent release, depending on the drug solubility (Draganoiu et al., 2001).

- 125 -

110

100

90

80

%released

70

60

50

40% KSR 0.1N HCl 40% KSR pH 6.8 50% KSR 0.1N HCl 50% KSR pH 6.8 60% KSR 0.1N HCl 60% KSR pH 6.8

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 29. Effect of dissolution medium on drug release from propranolol 80mg tablets

- 126 -

4.3.3. Effect of Kollidon® SR – Eudragit® L100-55 combination on drug release from propranolol 80mg tablets Addition of an anionic polymer to matrix tablets is a widely used approach to modulate the drug release and /or to promote a pH independent release (Streubel et al., 2000). In acidic medium, the enteric polymer is insoluble and acts as a part of the matrix thus contributes to the retardation of the drug release. In buffer media, the enteric polymer dissolves and loosens the matrix structure, thus increasing the porosity and permeability of the dosage form and compensating for the reduction in the diffusion rate. The effect of partial replacement (5 or 10% of the tablet weight) of Kollidon® SR with Eudragit® L100-55, while keeping constant the total matrix forming agent concentration (70% of the tablet weight) was investigated. Eudragit® L100-55 is a methacrylic acid copolymer insoluble at pH below 5.5. As expected, the release rates in water and 0.1N HCl were slightly reduced (Figure 30 – page 129, Figure 31 – page 130). This was because Eudragit® L100-55 is insoluble in water or 0.1N HCl, so it acted as a diffusion barrier. Surprisingly, the same phenomenon was observed in pH 6.8 buffer (Figure 32 – page 131). Possible explanations reside in a hindered dissolution of the enteric polymer due to the polyvinylacetate network (Streubel et al., 2000) and also in cationic drug - anionic polymer interaction (Takka et al., 2001, Streubel et al., 2000, Chang and Bodmeier, 1997, Feely and Davis, 1988).

- 127 -

A 48-hour dissolution test was performed in distilled water for the formulation containing 70% Kollidon® SR to verify the hypothesis that the non-released propranol at the end of the first 24 hours was still present in the matrix. 100% of the label claim was released at the end of the 48-hour interval, compared to 7% released after 24 hours and thus confirming the hypothesis.

- 128 -

110

100

90

80

%released

70

60

50

40

30

70%KSR

20

65%KSR+ 5%Eudragit 60%KSR+ 10%Eudragit

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 30. Effect of Kollidon® SR and Eudragit® L100-55 combination on drug release in water from propranolol 80mg tablets

- 129 -

110

100

90

80

%released

70

60

50

40

30

70%KSR

20

65%KSR+ 5%Eudragit 60%KSR+ 10%Eudragit

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 31. Effect of Kollidon® SR and Eudragit® L100-55 combination on drug release in 0.1N HCl from propranolol 80mg tablets

- 130 -

110

100

90

80

%released

70

60

50

40

30

70%KSR

20

65%KSR+ 5%Eudragit 60%KSR+ 10%Eudragit

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 32. Effect of Kollidon® SR and Eudragit® L100-55 combination on drug release in pH 6.8 buffer from propranolol 80mg tablets

- 131 -

110 100 90 80

%released

70 60 50 40 30 20 10 0 0

4

8

12

16

20

24

28

32

36

40

44

48

time (hr)

Figure 33. Propranolol release in water over 48 hours from tablets manufactured with 70% Kollidon® SR

- 132 -

4.3.4. Comparison of the propranolol 80 mg tablet formulations with the reference listed capsule product Currently all extended release propranolol products available in the United States are capsules and Inderal® LA is the reference listed product (RLD, innovator). The product is formulated as capsules containing coated pellets. Although the condition for pharmaceutical equivalence is not met (due to difference in dosage forms capsules versus tablets), Inderal® LA was used as reference product in developing matrix tablet formulations. By evaluating the release profiles obtained according to the USP dissolution method (method B – section 3.5, page 65) for propranolol 80mg tablets with 60 and 70% Kollidon® SR and the reference listed capsule product (Figure 34 – page 134), it was found that the initial release was faster for the tablets than for the capsules, while at the later dissolution stages the release profile for the innovator product was intermediate to the tablet profiles. Thus, it was decided to formulate and manufacture tablets using an intermediate polymer level (65%) and this formulation was used in the pilot bioequivalence study.

- 133 -

110 100 90 80

%released

70 60 50 40

70%KSR 60%KSR Inderal LA

30 20 10 0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 34. Comparison of drug release from propranolol 80 mg tablets with 60 and 70% Kollidon® SR and Inderal® LA

- 134 -

The composition of the selected formulation (65% Kollidon® SR) is presented in Table 23 – page 135.

Table 23. Composition of the propranolol 80mg tablets formulation used in the pilot bioequivalence study Ingredient

Manufacturer / Lot #

Amount (mg) / tablet

Percent / tablet

Propranolol HCl (BP)

BASF C20011001

80.000

29.0

Kollidon® SR

BASF16-9006

179.309

65.0

Emcompress®

Penwest A20E

6.8965

2.5

Emcocel® 90M

Penwest 9D5H1

6.8965

2.5

Aerosil® 200

DegussaD10221D

1.3793

0.5

Magnesium stearate

Malinckrodt C19408

1.3793

0.5

275.86

100.0

Total

*all the materials were certified by the manufacturers for human use.

An example of the compression and ejection forces recorded during the manufacturing is shown in Figure 35 - page136.

- 135 -

Compression Force [lb] Average: 1967.17 lb St. Dev. 113.87 lb Rel. SD 5.79 %

1967.17 1934.57 1901.96 2108.46 2206.28 1978.04 1858.49 1880.23 1847.62 1988.91

Ejection Force [lb] Average 111.95 lb St. Dev. 3.03 lb Rel. SD. 2.71 %

115.02 115.02 109.97 108.11 116.22 114.75 111.16 111.30 108.64 109.30 Figure 35. Compression and ejection forces recorded during manufacturing of propranolol 80 mg tablets with 65% Kollidon® SR

- 136 -

The resulting tablets were uniform in weight, thickness and hardness and passed the USP criteria for the Content Uniformity (Table 24 – page 137).

Table 24. Characteristics of propranolol 80mg tablets used in the pilot bioequivalence study Characteristics

Average

RSD

Tablet weight (mg)

277.91

0.861

Tablet thickness (mm)

4.884

0.308

Tablet hardness (kP)

14.12

4.847

Content uniformity

95.194

2.534

Compared to the reference-listed product, the drug release from the matrix tablets was faster in the initial stage (Figure 36 – page 138). This can be attributed

to

differences

in

the

formulation

and

release

mechanism

(multiparticulate versus monolithic system). The burst effect observed with the tablets could be explained by the propranolol trapped on the surface of the matrix and released immediately upon activation in the dissolution medium. This is a common reported phenomenon for matrix systems (Krajacic and Tucker, 2003, Huang and Brazel, 2001, Bodea and Leucuta, 1997). During the buffer stage, the developed product met the USP requirements for propranolol release (Table 25 – page 139) and was similar to the innovator product, as determined with the similarity factor (f2=60.60).

- 137 -

110

100

90

80

%released

70

60

50

40

65%KSR Inderal LA

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 36. Comparison of the drug release profiles from propranolol 80mg tablets with 65% Kollidon® SR and Inderal® LA

- 138 -

Table 25. Drug release from the propranolol 80 mg tablets with 65% Kollidon® SR (used for the pilot bioequivalence study) Time (hr)

Propranolol 80mg tablets

USP requirements

(65% Kollidon® SR)

1.5

32.69%

NMT 30%

4

49.64%

35-60%

8

63.23%

55-80%

14

77.74%

70-95%

24

91.22%

81-110%

The formulation was robust and reproducible, as shown by the drug release profiles from batches manufactured on different days (Figure 37 – page 140). This formulation was used in the pilot bioequivalence study and was tested for stability of the dissolution profiles under different storage conditions.

- 139 -

110

100

90

80

% released

70

60

50

40

batch 1 batch 2 batch 3

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 37. Reproducibility of propranolol 80 mg tablets formulation with 65% Kollidon® SR

- 140 -

4.3.5. Effect of storage conditions on propranolol 80 mg tablets physical properties and drug release Propranolol 80 mg tablets with 65% Kollidon® SR were tested to see the effect of storage conditions (long term or accelerated ICH testing conditions) on tablet physical properties and drug release (method B – section 3.5, page 65). No change in the dissolution profile was observed for tablets stored under longterm stability conditions for a period of up to nine months. A change in the dissolution profile was observed for tablets stored at 40°C/75% RH for more than 3 months. The reduction in the dissolution rate continued after six months, the time period recommended for conducting accelerated stability studies; it was also observed at nine months testing point. The change in the dissolution profile observed just in case of the tablets stored under accelerated conditions could be attributed to the amorphous nature of polyvinylacetate coupled with its unusually low glass transition temperature of 28–31°C, which imparts certain unique characteristics to the matrix. Such a change in dissolution profile is usually indicative of polymer structural relaxation. These results are in agreement with published data. Shao et al. (2001) observed a reduction in the dissolution rate for the diphenhydramine - Kollidon® SR formulation stored at 40°C/75%RH, as a result of polyvinylacetate relaxation. A post-compression curing step (1-18 hours at 60°C) was found to be critical in stabilizing the release rates of tablets containing high levels (≥47 %w/w) of Kollidon® SR.

- 141 -

The change in the dissolution rate of propranolol tablets was accompanied by an increase in tablet hardness. The increase in hardness was significantly higher for accelerated conditions compared to the long term conditions (p<0.05), as seen in Table 26 – page 142. Tablets stored under accelerated conditions became yellow after 6 months storage.

Table 26. Effect of storage on the hardness of propranolol 80 mg tablets Time

25°C/60%RH

40°C/75%RH

Initial

14.12±0.68

14.12±0.68

1 month

15.54±0.43

20.06±0.46 * **

3 months

16.39±0.99 *

21.06±0.70 * **

6 months

16.57±0.63 *

29.51±0.32 * **

9 months

16.97±0.59 *

ND

(*) significantly different from the initial at 0.05 level (Tukey procedure) (**) significantly different from the long term conditions at 0.05 level (Tukey procedure) ND – could not be determined (above the hardness tester maximum capacity); tablets were plastically deformed.

- 142 -

110

100

90

80

%released

70

60

50

initial 1 month 3 months 6 months 9 months

40

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 38. Effect of storage on drug release from propranolol 80 mg tablets – ICH long term stability conditions

- 143 -

110

100

90

80

%released

70

60

50

40

initial 1 month 3 months 6 months 9 months

30

20

10

0 0

2

4

6

8

10

12

14

16

18

20

22

24

time (hr)

Figure 39. Effect of storage on drug release from propranolol 80 mg tablets – ICH accelerated stability conditions

- 144 -

4.4. Evaluation of bioequivalence of propranolol 80 mg matrix tablets to Inderal® LA capsules

4.4.1. Analysis of propranolol in plasma The calibration curves generated by plotting the ratio of areas of propranolol to internal standard (pronethalol) versus concentration ratio of the two components were linear over the concentration range of 2 - 100ng/ml, (correlation coefficient > 0.99). Accuracy calculated as percentage of measured (recovered) concentration to theoretical values for three concentrations within the linearity range (2, 20, 100ng/ml) was in the range of 89-115%. The intra- and inter-day variability determined by using three replicate analyses of spiked plasma at three different concentrations (2, 20, 100ng/ml) were 9.60, 6.29, 3.94%, respectively 10.46, 6.68, 2.82%.

4.4.2. Subjects monitoring during the pilot bioequivalence study Subjects enrolled in the study had their body weight recorded each period and their vital signs were monitored at each blood drawing (Appendix 2 – page 197). It was found that at each time point the treatment effects on blood pressure and pulse rate were not significant (p>0.05), as tested by one-way ANOVA procedure. No significant adverse effects were reported during and post - study.

- 145 -

4.4.3. Pharmacokinetic and statistical analysis Based on FDA recommendation for assessing the bioequivalence for a previously approved molecular entity, a cross-over single dose non-replicate fasting study was performed for propranolol 80mg developed tablets and the reference listed product Inderal® LA (FDA, 2002). The single dose study is considered to be more sensitive in addressing the primary question of bioequivalence, i.e. release of the drug substance from the product into the systemic circulation. The multiple dose study is not recommended by the FDA even in the instances where nonlinear kinetics is present. The parent drug propranolol was measured in plasma (rather than the metabolites) because the concentration-time profile of the parent drug is more sensitive to changes in formulation performance than a metabolite, which is more reflective of metabolite formation, distribution and elimination (FDA, 2002). Propranolol plasma concentrations obtained after administration of the developed matrix tablets and Inderal® LA are graphically displayed for each subject in Figure 40 - Figure 47, pages 147 - 154. The mean results are shown in Figure 48 page 155.

- 146 -

70 Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 40. Plasma levels of propranolol following administration – subject #1

- 147 -

70 Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 41. Plasma levels of propranolol following administration – subject #2

- 148 -

70 Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 42. Plasma levels of propranolol following administration – subject #3

- 149 -

70 Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 43. Plasma levels of propranolol following administration – subject #4

- 150 -

70 Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 44. Plasma levels of propranolol following administration – subject #5

- 151 -

70 Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 45. Plasma levels of propranolol following administration – subject #6

- 152 -

70

Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 46. Plasma levels of propranolol following administration – subject #7

- 153 -

70 Inderal LA 80mg Propranolol 80mg tablets

60

Propranolol (ng/ml)

50

40

30

20

10

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 47. Plasma levels of propranolol following administration – subject #8

- 154 -

Propranolol 80 mg tablets

40

Inderal LA 80 mg

35

Propranolol (ng/ml)

30

25

20

15

10

5

0 0

2

4

6

8

10 12 14 16 18 20 22 24 26 28 30 time (hr)

Figure 48. Plasma levels of propranolol following administration (mean ± SEM)

- 155 -

A good agreement was found between the in vitro drug release and propranolol plasma concentration for the first twelve hours post-dosing, as may be seen in Figure 5 – page 86 and Figure 48 – page 155. The developed matrix tablets which had faster initial release produced higher plasma concentrations compared to the reference listed product. The difference in plasma concentrations was significant at 2, 3, 4 and 5 hours post dosing (P<0.05). These results confirm those of McAinsh et al. (1981) who reported for different extended release propranolol formulations that the peak blood level and AUC decreased as the dissolution was slower. McAinsh et al. (1981) explained the lowering of the systemic bioavailability as the dissolution time increases by an increased metabolism of propranolol.

The calculated AUC0-24h, AUC0-∞ and Cmax for each subject are presented in Table 27 – page 157. Testing the AUC0-24h, AUC 0-∞ and Cmax by non-parametric Wilcoxon twosample test – NPAR1WAY procedure for variable SUM, showed no significant carry over (residual) effect (p= p=0.8852, p=0.8852 and p=1.000 respectively). Testing for period effect by Wilcoxon two-sample test – NPAR1WAY procedure for variable XOVERDIF proved that the period did not significantly affect the responses, i.e. AUC 0-24h, AUC 0-∞ and Cmax (p=0.6650, p=1.000, respectively p=0.3123).

- 156 -

Table 27. Pharmacokinetic parameters after administration of propranolol 80mg tablets and Inderal® LA 80mg Subject

Propranolol 80mg tablets

Inderal® LA

AUC 0-24h

AUC 0-∞

Cmax

AUC 0-24h

AUC 0-∞

Cmax

1

210.91

239.21

18.47

187.00

467.26

17.99

2

284.84

346.27

20.59

257.78

495.71

19.28

3

393.99

404.06

66.92

358.41

694.92

28.09

4

398.29

433.93

28.75

166.35

259.14

10.91

5

241.12

251.82

23.89

237.32

1659.82

15.98

6

812.39

1064.99

54.94

360.21

900.28

33.24

7

604.80

830.07

74.30

677.34

880.97

50.45

8

502.18

830.90

62.76

315.18

659.68

20.36

Analysis of variance was carried out to test for the treatment effect. The treatment effect was not significant with regard to AUC 0-24h and 0-∞ (p=0.1070, p=0.3094) at a 5% level of significance. Tablets and capsules produced similar 24 hour- and total drug exposure. Analysis of Cmax showed significant difference between the two treatments at a 5% level of significance (p=0.0107). According to the FDA two products are considered bioequivalent if the 90% confidence interval for the ratio of the averages (population geometric means) of the measures for the test and reference (Cmax, AUC) falls within a BE limit, usually 80-125% for the ratio of the product averages (FDA 2001). By applying

- 157 -

this criterion, the two products tested (propranolol 80mg matrix tablets and Inderal® LA) were not bioequivalent with regards to Cmax, AUC0-24h, AUC 0-∞ (Table 28 – page 158). The tablets produced higher Cmax and 24-hour drug exposure than the capsules.

Table 28. Results of the bioequivalence testing using WinNonlin software Cmax

AUC 0-24h

AUC 0-∞

Probab. < 80

0.1317

0.0296

0.8465

Probab. > 125

0.8674

0.9677

0.1530

Maxim probability

0.8674

0.9677

0.8465

Total probability

0.9991

0.9974

0.9995

AH p value

0.7357

0.9381

0.6934

Power

0.0999

0.1

0.0999

Thus, Cmax was higher for the tablets than the capsules as tested by both ANOVA and FDA criterion for bioequivalence. Testing by ANOVA for the area under the curve did not show a significant treatment effect, while testing according to the FDA criterion revealed that the two products were not bioequivalent. This difference in results could be explained by the high variability of propranolol plasma concentration and the reduced number of subjects included in the pilot study. The variability of propranolol plasma concentrations is known to be due to the high intersubject variability in the hepatic metabolism of the drug (Bottini et al., 1983, Flouvat et al., 1989, Lalonde et al., 1987, Perucca - 158 -

et al., 1984). As the study could not be performed on a larger number of subjects because of the limited resources, it was designated as a pilot study. Similar sample size (6-9 subjects) was used in other studies on the bioavailability / bioequivalence of propranolol extended release formulations (Bottini et al., 1983, Lalonde et al., 1987, Perucca et al., 1984, Rekhi et al., 1996). To account for the variability, a non-parametric test was used for the analysis of variance. It is concluded that according to the FDA criteria the two products were not bioequivalent and the tablets had higher bioavailability as shown by Cmax and AUC 0-24h than the capsules. This conclusion applies for the mean results and for each subject. The initial faster release observed in vitro in case of the developed matrix tablets was reflected in vivo by the higher plasma concentrations for up to 12 hours (statistically significant for up to 5 hours).

- 159 -

5. Conclusions A minimum concentration of 30% polymer was necessary to achieve a coherent matrix, able to extend the release of the incorporated drugs. Increasing the Kollidon® SR concentration in the tablet led to an increase in the tablet hardness and a slower drug release. Drug release followed square root of time dependent kinetics, thus indicating a diffusion-controlled release mechanism. Although Kollidon® SR promoted pH-independent drug release, the drug release was dependent on the solubility at various pHs.

The drug release rate was faster for wet granulation than for direct compression, thus making direct compression the method of choice for manufacturing Kollidon® SR extended release systems.

Kollidon® SR was the main release controlling agent in the presence of an external binder or enteric polymer in the matrix.

A significant reduction in the dissolution rates associated with an increase in tablet hardness was observed during stability testing under accelerated conditions, but not under long term conditions. Based on this finding, the recommended storage conditions are at 25°C / 60%RH or lower.

- 160 -

The developed propranolol 80mg extended release formulation was found to have higher bioavailability than the reference listed product capsules, as shown by higher Cmax and AUC 0-24h. For the developed tablet formulation, the higher initial plasma concentration was correlated with the faster initial release observed in vitro. Thus, according to the FDA bioequivalence criteria, the two products were not bioequivalent.

Based on the above, it is concluded that Kollidon® SR is a potentially useful excipient for the production of pH-independent extended release matrix tablets.

- 161 -

6.

References

Ash, M, Ash, I., 1995. Handbook of Pharmaceutical Additives, Gower Publishing Limited. Avdeef, A., Berger, C. M., Brownell, C., 2000. pH-metric solubility. Part 2. Correlation between the acid-base titration and the saturation shake flask solubility-pH methods. Pharm. Res. 17, 85-89. BASF, 1999. Technical information for Kollidon® SR, BASF AG, Ludwigshafen/ Rh., Germany. Bodea, A., Leucuta, S.E., 1997. Optimization of hydrophilic matrix using a Doptimal design. Int. J. Pharm. 153, 247-255. Bonferoni, M.C., Rossi, S., Ferrari, F., Bertoni, M., Caramella, C. et al., 1998. On the employment of λ carrageenan in a matrix system. Part 3. Optimization of a carrageenan-HPMC hydrophilic matrix. J. Contr. Rel. 51, 231-239. Bottini, P.B. Caulfield, E. M.., Devane, J.G., Geoghegan, E. J., Panoz, D. E. 1983. Comparative oral bioavailability of conventional propranolol tablets and a new controlled absorption propranolol capsule. Drug Dev. Ind. Pharm. 9, 1475-1493. Boza, A., Caraballo, I., Alvarez-Fuentes, J., Rabasco, A. M., 1999. Evaluation of Eudragit RS-PO and Ethocel 100 matrices for the controlled release of lobenzarit disodium. Drug Dev. Ind. Pharm. 25, 229-233. Braza, A.J, Modamio, P, Marino, E.L., 2000. Two reproducible and sensitive liquid chromatographic methods to quantify atenolol and propranolol in human plasma and determination of their associated analytical error functions. J Chromat. B. 738, 225-31. Buch, A., Barr, W.H., 1998. Absorption of propranolol in humans following oral, jejunal, and ileal administration. Pharm. Res. 5, 953-7. Campos-Aldrete, M. E., Villafuerte Robles, L., 1997. Influence of the viscosity grade and the particle size of HPMC on metronidazole release from matrix tablets. Eur. J. Pharm. Biopharm. 43, 173-178. Chang, C.M., Bodmeier, R., 1997. Swelling of and drug release from monoglyceride-based drug delivery systems. J. Pharm. Sci. 86, 747-52.

- 162 -

Chiao, C.S.L., Robinson, J.R., 1995. Sustained Release Drug Delivery Systems in Remington - The science and Practice of Pharmacy 19th Ed. Mack Publishing Company. Cid, E., Mella, F., Lucchini, L., Carcamo, M., Monasterio, J., 1986. Plasma concentrations and bioavailability of propranolol by oral, rectal and intravenous administration in man. Biopharm. Drug Disp. 7, 559-566. Code of Federal Regulations, Title http://www.gpo.gov/nara/cfr/index.html

21

Food

and

Drugs,

Colombo, P., Bettini, R., Santi, P.A., Peppas, N.A., 2000. Swellable matrices for controlled drug delivery: gel-layer behaviour, mechanisms and optimal performance. Pharm. Sci. Tech. Today 6, 198-204. Colombo, P., Bettini, R., Peppas, N.A., 1999. Observation of swelling process and diffusion front position during swelling in hydroxypropyl methylcellulose (HPMC) matrices containing a soluble drug. J. Contr. Rel. 61, 83-91. Colombo, P., Bettini, R., Santi, P., De Ascentiis, A., Peppas, N.A., 1996. Analysis of the swelling and release mechanisms from drug delivery systems with emphasis on drug solubility and water transport. J. Contr. Rel. 39, 231-237. Colombo, P., Bettini, R., Massimo, G., Catellani, P.L., Peppas, N.A., et al. 1995. Drug diffusion front movement is important in drug release control from swellable matrix tablets. J. Pharm. Sci. 84, 991-997. Colombo, P., Catellani, P.L., Peppas, N. A., Maggi, L., Conte, U., 1992. Swelling characteristics of hydrophilic matrices for controlled release: new dimensionless number to describe the swelling and release behavior. Int. J. Pharm. 88, 99-109. Dow Pharmaceutical Excipients, 1996. Formulating for controlled release with Methocel Premium cellulose ethers. The Dow Chemical Company, Midland, Michigan. Draganoiu, E.; Andheria, M.; Sakr, A., 2001. Evaluation of the new polyvinylacetate/povidone excipient for matrix sustained release dosage forms. Pharm. Ind. 63, 624-629. Drummer, O.H., McNeil, J., Pritchard, E., Louis, W.J., 1981. Combined high performance liquid chromatographic procedure for measuring 4hydroxypropranolol and propranolol in plasma: pharmacokinetic measurements following conventional and slow-release propranolol administration. J. Pharm. Sci. 70, 1030-1032. Dunn, J.M., Groth, P.E., DeSimone, A., 1985. Once Daily Propranolol, Lancet 8465, 1183. - 163 -

Dvornik, D., Kraml, M., Dubuc, J., Coelho, J., Novello, L.A., et al. 1983. Relationship between plasma propranolol concentrations and dose of longacting propranolol (Inderal® LA). Curr. Ther. Res. 34, 595-605. Electronic Orange Book, http://www.fda.gov/cder/ob/default.htm, updated June 2002 FDA, 2002. Bioavailability and Bioequivalence Studies for Orally Administered Drug Products - General Considerations. Draft Guidance. US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. FDA, 2001. Statistical Approaches to Establishing Bioequivalence. Guidance for Industry. US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. FDA, 2001. ICH Q1A Stability Testing of New Drug Substances and Products Guidance for Industry. US Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research, ICH. FDA, 1997. ICH Q1C Stability Testing for New Dosage Forms. Guidance for Industry. US Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research, ICH. FDA, 1997a. Modified Release Solid Oral Dosage Forms. Scale-Up and Postaproval Changes: Chemistry, Manufacturing, and Controls, In Vitro Dissolution Testing and In Vivo Bioequivalence Documentation. Guidance for Industry. US Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research. FDA, 1997b. Extended Release Solid Oral Dosage Forms Development, Evaluation And Application Of In Vitro-In Vivo Correlation. Guidance for Industry. US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. FDA Drug Master Files, http://www.fda.gov/cder/dmf/index.htm Feely, L.C., Davis, S.S., 1988. Influence of polymeric excipients on drug release from hydroxypropylmethylcellulose matrices. Int. J. Pharm. 41, 83-90. Feely, L.C., Davis, S.S., 1988. Influence of surfactants on drug release from hydroxypropylmethylcellulose matrices. Int. J. Pharm. 44, 131-139.

- 164 -

Flick, D., Fraunhofer, W., Rock, T.C., Kolter, K., 2000. Melt granulation with Kollidon SR for the manufacture of sustained release tablets. Proceed. Int’l. Symp. Control. Rel. Bioact. Mater. 27, Controlled Release Society Inc. Ford, J.L., Rubinstein, M.H., Hogan, J.E., 1985. Dissolution of a poorly water soluble drug, indomethacin, from Hydroxypropylmethylcellulose controlled release tablets. J. Pharm. Pharmacol. 37, 33P. Frishman, W.H., Jorde, U. 2000. β-Adrenergic Blockers in Oparil, S., Weber, M.A. Hypertension: A Companion to Brenner and Rector’s The Kidney, W.B. Saunders Company. Garg, D.G., Jallad, N.S., Mishriki, A., Chalavarya, G., Kraml, M. et al., 1987. Comparative Pharmacodynamics and Pharmacokinetics of Conventional and Long-Acting propranolol J. Clin. Pharmacol. 27, 390-396. Goldberg, A., Sakr, A., 2003. Pharm. Ind. in press. Goodman and Gilman’s, 2001, The Pharmacological Basis of Therapeutics, 10th Ed., Gilman, A., Hardman, J., Limbird, L. (eds), McGraw-Hill Press. Grundy, R.U., McAinsh, J., Taylor, D.C. 1974 The effect of food on the in vivo release of propranolol from a PVC matrix tablet in dog, J.Pharm. Pharmacol. 26 Suppl., 65P. Gurny, R., Doelker, E., Peppas, N.A., 1982. Modeling of sustained release of water-soluble drugs from porous, hydrophobic polymers. Biomat. 3, 27-32. Heng, P.W.S., Chan, L.W., Easterbrook, M.G., Li, X., 2001. Investigation of the influence of mean HPMC particle size and number of polymer particles on the release of aspirin from swellable hydrophilic matrix tablets. J. Contr. Rel. 76, 39-49. Higuchi T., 1963. Mechanism of Sustained Action Medication: Theoretical Analysis of Rate of Release of Solid Drugs Dispersed in Solid Matrices. J. Pharm. Sci. 52, 1145-1149. Hogan, J.E., 1989. Hydroxypropylmethylcellulose sustained release technology. Drug Dev. Ind. Pharm. 15, 975-999. Huang, X., Brazel, C.S., 2001. On the importance and mechanisms of burst release in matrix-controlled drug delivery systems J Contr. Rel. 73, 121-36. Jantzen, G.M., Robinson, J.R., 1996. Sustained- and Controlled-Release Drug Delivery Systems in Modern Pharmaceutics, 3rd Ed (Banker G., Rhodes, C. Edts). Marcel Dekker Inc.

- 165 -

Karasulu, H.Y., Ertan, G., Kose, T., 2000. Modeling of theophylline release from different geometrical erodible tablets. Eur. J. Pharm. Biopharm. 49, 177-182. Katikaneni, P.R., Upadrashta, S.M., Neau, S.H., Mitra, A.K., 1995a. Ethylcellulose matrix controlled release tablets of a water soluble drug. Int. J. Pharm. 123, 119-125. Katikaneni, P.R., Upadrashta, S.M., Rowlings, C.E., Neau, S.H., Hileman, G.A., 1995b. Consolidation of ethylcellulose: effect of particle size, press speed, and lubricants. Int. J. Pharm. 117, 13-21. Krajacic, A., Tucker, I.G., 2003. Matrix formation in sustained release tablets: possible mechanism of dose dumping. Int. J. Pharm. 251, 67-78. Lalonde, R.L., Pieper, J.A., Straka, R.J., Bottorff, M.B., Mirvis, D.M., 1987. Propranolol pharmacokinetics and pharmacodynamics after single doses and at steady-state. Eur. J. Clin. Pharmacol. 33, 315-8. Mahmood, I., Sahajwalla, C., 1999. Clinical Pharmacokinetics and Pharmacodynamics of Buspirone, an Anxiolytic Drug. Clin. Pharmacokin. 36, 277-287. McAinsh, J., Baber, N.S., Holmes, B.F., Young, J., Ellis, S.H., 1981. Bioavailability of sustained release propranolol formulations. Biopharm Drug Disp. 2, 39-48. McAinsh, J., Baber, N.S., Smith, R., Young, J. 1978. Pharmacokinetic and pharmacodynamic studies with long-acting Propranolol. Br. J. Clin. Pharmacol. 6, 115-21. Mehvar, R., Brocks, D.R., 2001. Stereospecific pharmacokinetics and pharmacodynamics of beta-adrenergic blockers in humans. J. Pharm. Pharm. Sci. 4, 185-200. Moller, H., 1983. Release in vitro and in vivo and bioavailability of propranolol from sustained release formulations. Acta Pharm. Technol. 29, 287-294. Moore, J.W., Flanner, H.H., 1996. Mathematical Comparison of curves with an emphasis on in vitro dissolution profiles. Pharm. Tech. 20(6), 64-74. Mulye, N. V., Turco, S. J., 1994. Matrix type tablet formulation for controlled release of highly water soluble drugs. Drug Dev. Ind. Pharm. 20, 2633-43. Nace, G.S., Wood, A.J., 1987. Pharmacokinetics of long acting propranolol. Implications for therapeutic use. Clin. Pharmacokinet. 13, 51-64.

- 166 -

Nellore, R.V., Rekhi, G.S., Hussain, A.S., Tillman, L.G., Augsburger, L.L., 1998. Development of metoprolol tartrate extended release matrix tablet formulations for regulatory policy consideration. J. Contr. Rel. 50, 247-256. Pathan, S.I., Jalil, R., 2000. Evaluation of Sustained Release Matrix Tablets of Theophylline Prepared from a New Rate Retarding Polyvinyl Acetate/Povidone (Kollidon SR) Polymer. Poster AAPS Annual Meeting, Indianapolis. Peppas, N.A., Colombo, P., 1997. Analysis of drug release behavior from swellable polymer carriers using the dimensionality index. J. Contr. Rel. 45, 35-40. Peppas, N.A., 1985. Analysis of Fickian and Non-Fickian Drug Release from Polymers. Pharm. Acta Helv. 60 (4), 110-111. Peppas, N.A., Sahlin, J.J., 1989. A simple equation for the description of solute release. III. Coupling of diffusion and relaxation. Int. J. Pharm. 57, 169-172. Perucca, E., Grimaldi, R., Gatti, G., Caravaggi, M., Frigo, G.M. et al., 1984. Pharmacokinetic and pharmacodynamic studies with a new controlled release formulation of propranolol in normal volunteers: comparison with other commercially available formulations. Br. J. Clin. Pharm. 18, 37-43 Qiu, Y., Zhang, G., 2000. Research and Development Aspects of Oral Controlled-Release Dosage Forms in Handbook of Pharmaceutical Controlled Release Technology (Wise, D. L. Edt), Marcel Dekker Inc Rekhi, G.S., Jambhekar, S.S., Souney, P.F., Williams, D.A., 1995. A fluorimetric liquid chromatographic method for the determination of propranolol in human serum/plasma. J. Pharm Biomed Anal. 13, 1499-505. Rekhi, G.S., Nellore, R.V., Hussain, A.S., Tillman, L.G., Augsburger, L.L., et al., 1999. Identification of critical formulation and processing variables for metoprolol tartrate extended-release (ER) matrix tablets. J. Contr. Rel. 59, 327-342. Rock, T.C., Steenpaß, T., Ruchatz, F., Kolter, K., 2000. Methods to reduce the initial release rate of sustained release tablets containing Kollidon SR. Proceed. Int’l. Symp. Control. Rel. Bioact. Mater. 27. Controlled Release Society Inc. Ruchatz, F., Kolter, K., Wittermer, S., 1999. Kollidon  SR – a new excipient for sustained release matrices. Proceed. Int’l. Symp. Control. Rel. Bioact. Mater. 26. Controlled Release Society Inc.

- 167 -

Sakr, A., Andheria, M., 2001. A comparative multidose pharmacokinetic study of buspirone extended-release tablets with a reference immediate-release product. J. Clin. Pharmacol. 41, 886-894. Sakr, A., Sidhom, M., 1988. Direct compression of diuretic tablets. Part 2. Effectiveness of cross linked carboxymethylcellulose in directly compressed hydrochlorothiazide tablets. Pharm. Ind. 50, 105-107. Sakr, A., Vales, G., Jimenez, W., 1987. Direct compression of diuretic tablets. Part 1. Preliminary report. Pharm. Ind. 49, 401-404. Serlin. M.J., Orme, M. L’E., MacIver, M., Sibeon, R.G., Breckenridge, A.M., 1983. The Pharmacodynamics and pharmacokinetics of conventional and longacting propranolol in patients with moderate hypertension. Br. J. Clin. Pharm. 15, 519-527. Shah, V.P., Tsong, Y., Sathe, P., Williams, R.L., 1999. Dissolution profile comparison using similarity factor, f2. Dissol. Tech. 6, 15. Shah, V.P., Tsong, Y., Sathe, P., 1998. In vitro dissolution profile comparison statistics and analysis of the similarity factor, f2. Pharm. Res. 15, 889-896. Shand, D.G., Nuckolls, E.M., Oates, J.A., 1970. Plasma propranolol levels in adults with observations in four children. Clin. Pharm. Ther. 11, 112-120. Shanks, R.G., 1984. Studies with long acting propranolol. Postgrad. Med. J. 60 Suppl 2, 61-8. Shao, Z.J., Farooqi, M. I., Diaz, S., Krishna, A.K., Muhammad, N.A., 2001. Effects of formulation variables and postcompression curing on drug release from a new sustained release matrix material: polyvinylacetate povidone. Pharm. Dev. Techol. 6, 247-254. Shileout, G., Zessin, G., 1996. Investigation of ethylcellulose as a matrix former and a new method to regard and evaluate the compaction data. Drug Dev. Ind. Pharm. 22, 313-319. Shivanand, P., Sprockel, O.L., 1998. Controlled porosity drug delivery system. Int. J. Pharm. 167, 83-96. Siepmann, J., Kranz, H., 2000. Calculation of the required size and shape of hydroxypropyl methylcellulose matrices to achieve desired drug release profiles. Int. J. Pharm. 200, 151-164. Siepmann, J., Kranz, H., Bodmeier, R., Peppas, N.A., 1999a. HPMC matrices for controlled drug delivery: new model combining diffusion, swelling, and dissolution mechanisms and predicting the release kinetics. Pharm. Res. 16, 1748-1756. - 168 -

Siepmann, J., Lecomte, F., Bodmeier, R., 1999b. Diffusion-controlled drug delivery systems: calculation of the required composition to achieve desired release profiles. J. Contr. Rel. 60, 379-389. Siepmann, J., Podual, K., Sriwongjanya, M., Peppas, N.A., Bodmeier, R., 1999c. New model describing the swelling and drug release kinetics from hydroxypropyl methylcellulose tablets. J. Pharm. Sci. 88, 65-72. Sriwongjanya, M., Bodmeier, R., 1998. Effect of ion exchange resins on drug release from matrix tablets. Eur. J. Pharm. Biopharm. 46, 321-327. Straka, R.J., Lalonde, R.L., Pieper, J.A., Bottorff, M. B., Mirvis, D.M., 1987. Nonlinear pharmacokinetics of unbound propranolol after oral administration. J. Pharm. Sci. 76, 521-524 Streubel, A., Siepmann, J., Dashevsky, A., Bodmeier, R., 2000. pH independent release of a weakly basic drug from water insoluble and soluble matrix tablets. J. Contr. Rel. 67, 101-110. Sung, K.C., Nixon, P.R., Skoug, J.W., Ju, T.R., Patel, M.V., et al., 1996. Effect of formulation variables on drug and polymer release from HPMC-based matrix tablets. Int. J. Pharm. 142, 53-60. Takacs-Novak, K., Avdeef, A., 1996. Interlaboratory study of log P determination by shake-flask and potentiometric methods J. Pharm Biomed Anal. 14, 140513. Takahashi, H., Ogata, H., Warabioka, R., Kashiwada, K., Someya, K., et al. 1990. Decreased absorption as a possible cause for the lower bioavailability of a sustained-release propranolol. J. Pharm. Sci. 79, 212-215. Takka, S., Rajbhandari, S., Sakr, A., 2001. Effect of anionic polymerws on the release of propranolol hydrochloride from matrix tablets. Eur. J. Pharm. Biopharm. 52, 75-82. United States Pharmacopeia&National Formulary 25th Ed. The United States Pharmacopeial Convention Inc., 2002. Upadrashta, S.M., Katikaneni, P.R., Hileman, G.A., Keshary, P.R., 1993. Direct compression controlled release tablets using ethylcellulose matrices. Drug Dev. Ind. Pharm. 19, 449-460. US Patent 4, 138, 475. US Patent 6, 066, 334. Velasco, M.V., Ford, J.L., Rowe, P., Rajabi-Siahboomi, A.R., 1999. Influence of drug: hydroxypropylmethylcellulose ratio, drug and polymer particle size and

- 169 -

compression force on the release of diclofenac sodium from HPMC tablets. J. Contr. Rel. 57, 75-85. Venkatraman, S., Davar, N., Chester, A., Kleiner, L., 2000. An Overview of Controlled-Release Systems in Handbook of Pharmaceutical Controlled Release Technology (Wise, D. L. Edt), Marcel Dekker Inc. Walle, T., Walle, U.K., Olanoff, L.S., Conradi, E.C., 1986. Partial metabolic clearances as determinants of the oral bioavailability of propranolol. Br. J. Clin. Pharm. 22, 317-323. Wan, L.S., Heng, P.W., Wong, L.F., 1995. Matrix swelling: simple model describing extent of swelling of HPMC matrices. Int. J. Pharm. 116, 159-168.

- 170 -

7. Appendix 1 Study of the Bioavailability of Two Extended Release Propranolol HCl Dosage Forms – Research Protocol # 06-19-01

- 171 -

Research Protocol - page 1

To: UCMC Institutional Review Board Chairperson

From:

Principal Investigator

Coinvestigators:

Bernadette D’Souza, M.D. Associate Director of Clinical Affairs, Associate Professor [email protected] Phone (513) 475-6326 Fax (513) 475-6379 Mail Location 116A Department of Veterans Affairs, Medical Center Mental Health Care Line 3200 Vine Street, Cincinnati OH 45220 Adel Sakr, Ph.D., Professor & Director, Industrial Pharmacy Graduate Program, College of Pharmacy, University of Cincinnati Thomas Geracioti, Jr., M.D., Associate Professor and Vice Chair, Department of Psychiatry, College of Medicine, University of Cincinnati Elena Draganoiu, Graduate Student, Industrial Pharmacy Graduate Program, College of Pharmacy, University of Cincinnati

Study of the Bioavailability of Two Extended Release Propranolol HCl Dosage Forms

Department Chair Approval: Daniel Acosta Jr., Ph.D. Dean College of Pharmacy

06/11/2001 Revised 11/07/01

- 172 -

Research Protocol - page 2

RESEARCH PROTOCOL OUTLINE A. B. C. D.

E.

F. G. H. J.

Specific aims Significance Background Information Preliminary Studies Experimental Design and Methods 1. Subjects a. Criteria for subject selection Inclusion Criteria Exclusion Criteria b. Screening examinations 2. Source of subject population 3. Research Protocol a. Methodology • Study design • Products studied • Drug assignment • Study visits • Screening • Study test days • Drug administration • Blood samples • Meals and food restrictions • Subject monitoring b. Analysis • Method of analysis • Pharmacokinetic and statistical analysis c. Setting and laboratory facilities Human subjects 1. Recruitment 2. Risks and benefits 3. Payment 4. Subject costs 5. Consent form Estimated period of time to complete the study Study schedule Funding References Consent form

- 173 -

3 3 4 7 9 9 9 9 9 10 10 11 11 11 11 11 12 12 12 12 13 13 13 14 14 14 14 15 15 15 15 15 15 16 16 16 17 19

Research Protocol - page 3

RESEARCH PROTOCOL A.

Specific aims

As part of a Ph.D. Dissertation Research (E. Draganoiu – College of Pharmacy, University of Cincinnati), a systematic pharmaceutical technology research resulted into an in vitro extended release tablet formulation for Propranolol HCl. The specific aim of this protocol is to compare the bioavailability of the developed Propranolol HCl extended release (ER) tablets with the leading commercial brand in the US market. This will validate and complete the research objectives of the Ph.D. Dissertation of E. Draganoiu.

B.

Significance

All the Propranolol extended release formulations currently in use are hard gelatin capsules, containing small spheroids each containing Propranolol hydrochloride dispersed in an insoluble matrix. The drug containing spheroids are in turn coated with a semipermeable membrane, which allow drug to diffuse at a controlled rate. This study is an attempt to formulate and deliver Propranolol as directly compressed extended release tablets. The drug is homogenously dispersed through the Polyvinylacetate-Povidone matrix and the drug release follows the diffusional mechanism (Jantzen, Robinson 1996, Draganoiu et al, 2001). Compared to capsule manufacturing (spheronization followed by coating) the tablet manufacturing technology by direct compression is easier and more efficient. The in vitro dissolution test conducted for tablet in different media should provide an extended release of the drug over 24 hours.

- 174 -

Research Protocol - page 4

Background Information Propranolol is almost completely absorbed from the gastrointestinal tract, but it is subjected to an extensive and highly variable hepatic first pass metabolism, with a reported systemic bioavailability between 15 and 23% (Cid et al, 1986, Walle et al, 1986). Peak effect occurs after 1-2 hours and can vary up to seven fold after oral administration due to individual variations in hepatic metabolic activity (Shand et al, 1970). The biologic half-life is approximately four hours. Due to relatively short plasma half-life, Propranolol conventional tablets are administrated at 6 to 8 hours intervals. Such frequent drug administration may reduce patient compliance and thus therapeutic efficacy (Serlin et al, 1983). Several sustained release systems have been developed in order to enable daily administration of the drug and a 24 hours maintained beta-adrenoceptor blockade.

Propranolol extended release systems should fulfill two objectives. Firstly to achieve an effective plasma concentration through the dosing interval, while avoiding potentially toxic peak concentration or ineffective plasma concentration that might occur with conventional formulations and secondly to produce a pharmacological effect as effective, at least, as the conventional drug given at more frequent dosing interval. Different formulations have been tested in vitro and in vivo in comparison to conventional tablets for these claims. (Serlin et al, 1983) However there are some problems associated with Propranolol ER formulations. Besides the variable Propranolol bioavailability (first pass degradation, influence of food, ethnic factor, other medication), ER formulations exhibit a significantly lower systemic bioavailability than the conventional tablets. This is due to a slower absorption and higher first pass effect. Pharmacokinetic Properties of Propranolol (Frishman and Jorde, 2000) Formulation Extent of Bioavailability Interpatient β-Blocking variation in plasma absorption (%of dose) plasma (%of dose) concentration level Immediate >90% 30 20 fold 50-100ng/ml release Extended >90% 20 10-20 fold 20-100ng/ml release

Protein binding (%) 93 93

In a crossover single oral dose study (Takahashi et al, 1990) on healthy subjects who received 60 mg Propranolol as sustained release capsules (Inderal LA) or as conventional tablets, significant differences (parallel decreases for Inderal LA release compared to conventional tablets) were observed in area under the curve of Propranolol hydrochloride, Propranolol glucuronide and naphtoxylactic acid and in the amounts of all metabolites excreted in urine. Therefore it was

- 175 -

Research Protocol - page 5

concluded that the hepatic metabolism of Propranolol would not be affected by the slower absorption at a single dose of 60mg. Bioavailability of a 160mg slow release formulation following single dose administration was about one third that of the conventional preparation (Drummer et al, 1981) Garg et al (1987) showed that for two Propranolol long-acting formulation (80mg and 160mg) the area under the curve and the peak concentration were significantly less compared to the conventional tablets; in addition the elimination half-life was longer (9 hours) than for conventional Propranolol (4hours). In a crossover study on healthy subjects with Propranolol 160mg daily for 7 days mean bioavailability of sustained release capsules relative to regular tablet formulation was 52% for single doses and 54% for steady state (Straka et al, 1987). For one-day therapy with sustained release Duranol capsules (single dose in the morning) and Inderal conventional formulation (two doses morning and evening) it was found that the relative bioavailabilities were similar despite prolonged absorption time for the sustained action capsules (Bottini et al, 1983) In a study with sustained release Propranolol (Elanol 120mg, Inderal LA 160mg) and conventional Inderal (40mgx3/day) single doses of controlled release preparations produced a smoother serum level profile with lower and delayed peak times (dose corrected AUC lower for Inderal LA than for Elanol). At steady state all regimen ensured relatively sustained serum levels and a stable degree of pharmacological effect. Dose corrected AUC decreased in order Elanol>Inderal>Inderal LA. These results demonstrated that long acting formulations of Propranolol can be developed which are not necessarily associated with reduced bioavailability secondary to enhanced first pass metabolism (Perucca et al, 1984). The bioavailability of Inderal LA (80, 160 and 240mg once daily for 4 days) was proportional to the dose administrated as sustained action capsules. Steady state was attained after 2 doses. (Dvornik et al, 1983) For two different sustained release formulations (Dociton retard and Propranolol 160 Stada) there were found analogous associations between in vitro and in vivo dissolution after 4 hours (Moller, 1983). For different sustained release formulation, the peak blood level and AUC decrease as the dissolution time increase; the half-life are inversely proportional to the dissolution rate. The lowering of the systemic bioavailability as the dissolution time increases is thought to be due to an increased metabolism of Propranolol (McAinsh et al, 1981) - 176 -

Research Protocol - page 6

An attempt to develop plastic matrix tablets was done in 1974 by Grundy et al. The matrix consisted on Propranolol 125 mg embedded in an insoluble matrix of Pevikon D-42-P (polyvinyl chloride, 273 mg). The formulation had a satisfactory in vitro release profile (50% of the dose in 3 hours, at 100rpm). However when administered in dogs, the in vivo release profile was unsatisfactory (the drug was not completely released from the matrix) (Grundy et al, 1974). Single entity extended release formulations of Propranolol were therefore abandoned in favor of multiparticulate systems. *Sustained release, Extended release, Controlled release, Long-acting forms are alternative terms used by various researchers to describe the modified release dosage forms (excluding delayed release).

- 177 -

Research Protocol - page 7

C.

Preliminary Studies

The dissolution test is the best in vitro predictor for in vivo product performance. Among the dissolution test specifications, the dissolution profile comparison seems to be more precise than single point estimate approach to characterize the drug product (O’Hara et al, 1998). For extended release formulation FDA recommends that the dissolution profile should be evaluated by using a multipoint profile, with adequate sampling at different time points (for example at 1, 2 and 4 hours and every two hours after, until either 80% of the drug is release or an asymptote is reached). The regulatory accepted method for comparison of the dissolution profiles is a model independent mathematical approach described by Moore and Flanner (1996), which is know as f2 (similarity factor) equation. n

f 2 = 50 ⋅ log{[1 + (1 / n)∑ (Rt − Tt ) 2 ] − 0.5 ⋅ 100} t =1

Where Rt and Tt are the cumulative percentage dissolved at each of the selected n time points of the reference and test product respectively. Factor f2 is inversely proportional to the average squared difference between the two profiles, with emphasis on the larger difference among all the time-points. The transformation is such that the f2 equation takes values less or equal to 100. The value of f2 is 100 when the test and reference mean profiles are identical. The factor f2 measures the closeness between the two profiles. When the two profiles are identical, f2=100. An average difference of 10% at all measured time points results in an f2 value of 50 (Shah et al, 1998). FDA has set a public standard of f2 value between 50-100 to indicate similarity between two dissolution profiles. In vitro release data of the Propranolol tablets and reference Inderal LA capsules in different media (0.1N HCl and pH 6.8 buffer) are shown.

- 178 -

Research Protocol - page 8

Propranolol release from Inderal LA cps and Propranolol Tb 110 100 90 80 70 60 50

Inderal LA cps 0.1N HCl

40

Inderal LA cps pH 6.8

30

Propranolol Tb 0.1N HCl

20

Propranolol Tb pH 6.8

10 0 0

2

4

6

8

10

12

14

16

18

20

22

24

The dissolution profiles in 0.1N HCl for the two formulations are almost super imposable. The similarity is confirmed by f2 values at all tested points greater than 50. In case of using pH 6.8 USP phosphate buffer as dissolution medium, the dissolution profiles meet the FDA criteria for similarity (f2 values greater than 50 at all tested points). For the developed tablet formulation the release is slighter slower than for the marketed product. This difference of in vitro release should be tested for in vivo significance, knowing that in some cases formulation with significantly different in vitro release rates exhibit equal bioavailability (Sakr, Andheria 2001)

- 179 -

Research Protocol - page 9

D.

1.

Experimental Design and Methods

Subjects

A total number of 8 subjects capable of giving informed consent will be studied. Subjects will be healthy male and female volunteers and will be studied as outpatients. All samples from all subjects will be analyzed. Informed consent written will be obtained from each subject prior to entry into the study. Criteria for subject selection Inclusion Criteria

•

Healthy, male and female subjects between the ages of 18 – 65 years inclusive

•

Subjects must be outpatients at the time of screening

•

Subjects must be on no chronic medications (prescription or OTC) and must be medication-free for a period of at least one week prior to the first test day and throughout the duration of the study

•

Subjects must be off any investigational drug for a period of at least 3 months prior to the entry in the study

•

Subjects must be in good health as determined by medical history, routine physical examination, ECG and clinical laboratory tests

•

Subjects must be free of significant psychiatric illness

•

Subjects must be willing and able to provide written informed consent.

Exclusion Criteria

•

Subjects with a history or evidence of clinically significant and currently relevant hematological, renal, hepatic, gastrointestinal, endocrine, pulmonary, dermatological, oncological or neurological illness, and alcoholism. - 180 -

Research Protocol - page 10

•

Subjects with a history of cardiovascular disease, including hypotension, hypertension, heat block, congestive heart failure, angina pectoris, bypass surgery, or myocardial infarction

•

Subjects with clinically significant abnormalities on the electrocardiogram at screening

•

Pregnant and breast-feeding women are not eligible

•

Subjects using concomitant drugs

•

Subjects with known allergy to Propranolol

•

Subjects with clinically significant emotional problems

•

Subjects unable and/or unlikely to comprehend and follow the study protocol

Screening examinations: Routine physical examination and medical history Safety examination – ECG before the treatment, blood pressure, pulse and temperature Laboratory examination – complete blood count with differential, hepatic and renal profiles 2. Source of subject population Normal healthy volunteers who meet all inclusion criteria.

- 181 -

Research Protocol - page 11

3.

Research Protocol

a.

Methodology

Study design

This study will be a cross-over single-dose two-period open-label study which will compare the absorption of Propranolol from two dosage forms: Propranolol 80mg extended release capsules (InderalLA) and Propranolol 80mg extended release tablets, administrated oral, under fasting conditions. Subjects will undergo a screening procedure 2-6 days prior to the first test day. If all inclusion and exclusion criteria are met, subjects will be randomized to one of the two dosing sequence. Subjects will report to the outpatient facility in the morning of the first day of each period and will receive a single dose of the drug (capsule or tablet). Products studied A)

Propranolol ER tablet formulation - test product (Industrial Pharmacy

Laboratory, UC) B)

Propranolol ER capsule formulation – reference product (Inderal LA,

Manufacturer Ayerst Laboratories Inc., lot # 9010268, expiration date 07/2003) Drug Assignment: The subjects will be assigned to two dosing sequence as follows: Period 1

Period 2

4 subjects

A

B

4 subjects

B

A

The administration sequence will be assigned randomly. Subjects will be monitored for 30 hours after each dose. There will be 2 test periods separated by one-week washout.

- 182 -

Research Protocol - page 12

Study visits Screening

Routine physical examination and medical history; body weight will be recorded Vital signs – blood pressure (100-140 mm Hg systolic /70-90 mm Hg diastolic), pulse 60-100 beats/min Electrocardiogram before the treatment Laboratory examination – complete blood count with differential, hepatic and renal profiles Study test days

Period 1

Day 1 Day2

Time between periods: One week Period 2

Day 1 Day 2

Subjects will be admitted in the morning (7.30 am) of the first day of each period and after the insertion of the catheter, they will receive a single dose of the drug (treatment A or B) at 8am (0 hour of the test). Subjects will be in the facility until 8pm (after the 12 hours blood sample is withdrawn). Subjects will return the second day of each period at 7.30 am for the 24h and 30h blood sample withdrawal. There will be 2 test periods separated by one-week washout. Drug Administration Treatment A: Propranolol 80mg ER tablet (1 tablet) oral at 0 hour with 240ml of room temperature water Treatment B Propranolol 80mgER capsule (1 capsule) oral at 0 hour with 240ml of room temperature water

- 183 -

Research Protocol - page 13

Blood samples: During each period, 12 venous blood samples will be taken in heparinized vacutainers as follows: Day 1 - at 0 (predose), and at 1, 2, 3, 4, 5, 6, 8, 10, 12h (using catheter hep-lock) after drug administration Day 2 - at 24, 30h (by direct venipuncture) after drug administration (Singh, Jambhekar, 1996). The plasma will be separated, transferred to the labeled tubes and promptly frozen. The samples will be stored frozen at –20C, until analyzed. Meals and Food Restrictions: Subjects shall fast for at least 12 hours prior to the dose administration. Prior to and during each study phase subjects are allowed to water as desired except for one hour before and after drug administration After drug administration, subjects will receive lunch at 1pm. Subjects should abstain from alcohol, for 24 hours prior to each study period and until after the last sample from each period is collected (alcohol increases plasma clearance rate). Abuse of tobacco, caffeine is not allowed for 24 hours prior to each study period and until after the last sample from each period is collected. Subject monitoring The blood pressure and pulse rate will be monitored prior to dosing and at the sampling times. Subjects will have their weight measurements taken and recorded at check-in, each period. Subjects will be advised to avoid the use of prescription and OTC medications and alcohol

- 184 -

Research Protocol - page 14

Method of analysis: All Propranolol samples obtained from the test and reference product will be analyzed by the same HPLC method coupled with fluorescence or UV detection. The measurement of only the Propranolol concentration is performed, assuming that the concentration-time profile of the parent drug is more sensitive to changes in formulation than a metabolite (FDA Guidance on Bioequivalence). The validation, linearity and sensitivity of the method will be conducted before the study is started. Pharmacokinetic and statistical analysis: Cmax, tmax – obtained direct from the data, t1/2 (terminal half-life) AUC 0-t and AUC 0-∞ Bioavailability of the test and reference product will be tested by two one-sided ttest, by computing a 90% CI for the ratio of the mean response (AUC and Cmax).

c.

Setting and laboratory facilities

The study will be sponsored by the University of Cincinnati and conducted at the Veterans Affairs Medical center (VAMC), Outpatient Clinic facilities. The screening procedure will be conducted at the investigator’s office and laboratory at the VAMC. The Propranolol plasma analysis will be performed in the laboratories of the College of Pharmacy

- 185 -

Research Protocol - page 15

E.

Human subjects

1. Recruitment Advertisement for study enrollment will be posted in the News Record and on boards within the East Campus (see attached advertisement). Healthy volunteers responding to the advertisement will be recruited, after explaining the purpose and protocol of the study and given the informed consent. A screening procedure will be done before enrolment in the study. 2. Risks and benefits After administration of Propranolol adverse effects have been rare, mild and transient: bradycardia, insomnia, weakness, fatigue, nausea, vomiting, epigastric distress, abdominal cramping, diarrhea, constipation. 80mg is a relatively low dose (the usual maintenance dose is 120-240mg/day and it may be increased in some cases up to 640mg/day). 3.

Payment – subjects will not be directly remunerated for the participation in

the study, but compensation consisting of educational materials (up to $200/participant) will be available for each qualified participant 4.

Subject costs:

Funds are not available to cover the costs of any ongoing medical care and the subjects remain responsible for the cost of non-research related care. Tests, procedures and other costs incurred solely for purposes of research will be the financial responsibility of the sponsor. 5. Consent form – is attached as a separate document

- 186 -

Research Protocol - page 16

F.

Estimated period of time to complete the study (approximately 6

weeks): Subject recruitment – 14days Pre-study screening – 7 days First treatment – 2 days Wash-out period – 7 days (during this time the samples from the first treatment will be analyzed) Second treatment - 2 days Analysis of the samples – 2 days Data analysis – 7 days Study schedule

Screen -6 0 x

Days Consent Screening History Physical examination Weight check Electrocardiogram Lab exams Vital signs Administration Study Drug Propranolol plasma analysis

Period 1 1-2

x x x x x x

G.

Wash-out 7

Period 2 1-2

x

x x x

x x

Funding

Industrial Pharmacy Graduate Program, University of Cincinnati through its funds (Industrial Pharmacy Account), will support the costs of this study.

- 187 -

Research Protocol - page 17

REFERENCES Bottini, P. B.; Caulfield, E. M.; Devane, J. G.; Geoghegan, E. J.; Panoz, D. E. 1983. Comparative oral bioavailability of conventional Propranolol tablets and a new controlled absorption Propranolol capsule. Drug Dev. Ind. Pharm. (9) 1475-1493 Cid, E.; Mella, F.; Lucchini, L.; Carcamo, M.; Monasterio, J., 1986. Plasma concentrations and bioavailability of Propranolol by oral, rectal and intravenous administration in man. Biopharm. Drug Disp (7) 559-566 Draganoiu, E., Andheria, M., Sakr, A. Evaluation of a New Polyvinyl acetate/ Povidone Excipient for Matrix Sustained Release Dosage Forms. Accepted for publication in Pharm. Ind. Drummer, O. H.; McNeil, J.; Pritchard, E.; Louis, W. J. 1981. Combined high performance liquid chromatographic procedure for measuring 4-hydroxyPropranolol and Propranolol in plasma: pharmacokinetic measurements following conventional and slowrelease Propranolol administration. J. Pharm. Sci. (70) 1030-1032 Dvornik, D.; Kraml, M.; Dubuc, J.; Coelho, J.; Novello, L. A.; et al. 1983. Relationship between plasma Propranolol concentrations and dose of long-acting Propranolol (Inderal LA). Curr. Ther. Res. (34) 595-605 Frishman, W.H., Jorde, U. 2000 β-Adrenergic Blockers in Oparil, S., Weber, M.A. Hypertension: A Companion to Brenner and rector’s The Kindey, pp.590-594, W.B. Saunders Company Garg, D.G., Jallad, N.S., Mishriki, A., Chalavarya, G., Kraml, M. et al 1987. Comparative Pharmacodynamics and Pharmacokinetics of Conventional and Long-Acting Propranolol J. Clin. Pharmacol. (27) 390-396 Grundy, R. U., McAinsh, J., Taylor, D.C. 1974 The effect of food on the in vivo release of Propranolol from a PVC matrix tablet in dog, J.Pharm. Pharmacol. (26 Suppl.), 65P Jantzen, G.M., Robinson, J.R. Sustained and Controlled Release Drug Delivery Systems in Banker, G.S., Rhodes, C. (eds.) Modern Pharmaceutics, 3rd ed., pp 575-610, Marcel Dekker (1996). McAinsh, J., Baber NS Holmes BF Young J Ellis SH 1981. Bioavailability of sustained release Propranolol formulations. Biopharm Drug Disp. (2) 39-48. Moller, H. 1983. Release in vitro and in vivo and bioavailability of Propranolol from sustained release formulations. Acta Pharm. Technol. (29) 287-294 Moore, J.W., Flanner, H.H., 1996. Mathematical Comparison of curves with an emphasis on in vitro dissolution profiles. Pharm. Tech. 20(6), 64-74. O’Hara, T, Dunne, A, Butler, J., Devane, J., 1998. A review of methods used to compare dissolution profile data. Pharm. Sci. Tech. Today. (5) 214-223.

- 188 -

Research Protocol - page 18

Perucca, E.; Grimaldi, R.; Gatti, G.; Caravaggi, M.; Frigo, G. M. et al. 1984. Pharmacokinetic and pharmacodynamic studies with a new controlled release formulation of Propranolol in normal volunteers: comparison with other commercially available formulations. Br. J. Clin. Pharm. (18) 37-43 Rekhi, G.S., Jambhekar, S.S. 1996. Bioavailability and In-vitro-/in-vivo Correlation for Propranolol Hydrochloride Extended-release Bead Products Prepared Using Aqueous Polymeric Dispersions. J. Pharm. Pharmacol. (48) 1276-1284 Sakr, A., Andheria, M. 2001. Pharmacokinetics of Buspirone Extended Release Tablets: A Single Dose Study. Accepted for publication in J. Clin Pharmacol. Serlin. M.J., Orme, M. L’E., MacIver, M., Sibeon, R.G., Breckenridge, A.M. 1983. The Pharmacodynamics and pharmacokinetics of conventional and long-acting Propranolol in patients with moderate hypertension. Br. J. Clin. Pharm. (15) 519-527 Shah, V.P., Tsong, Y., Sathe, P., 1998. In vitro dissolution profile comparison - statistics and analysis of the similarity factor, f2. Pharm. Res. (15) 889-896. Shand, D.G., Nuckolls, E.M., Oates, J.A. 1970. Plasma Propranolol levels in adults with observations in four children. Clin. Pharm. Ther. (11) 112-120 Straka, R. J.; Lalonde, R. L.; Pieper, J. A.; Bottorff, M. B.; Mirvis, D. M. 1987. Nonlinear pharmacokinetics of unbound Propranolol after oral administration. J. Pharm. Sci. (76) 521-524 Takahashi, H.; Ogata, H.; Warabioka, R.; Kashiwada, K.; Someya, K.; et al 1990. Decreased absorption as a possible cause for the lower bioavailability of a sustainedrelease Propranolol. J. Pharm. Sci. (79) 212-215 Walle, T.; Walle, U. K.; Olanoff, L. S.; Conradi, E. C. 1986. Partial metabolic clearances as determinants of the oral bioavailability of Propranolol. Br. J. Clin. Pharm. (22) 317-323 ***FDA 1997 Guidance for Industry Modified Release Solid Oral Dosage Forms ScaleUp and Postaproval Changes: Chemistry, Manufacturing, and Controls, In Vitro Dissolution Testing and In Vivo Bioequivalence Documentation. Guidance for Industry. US Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research. ***FDA 1997 Guidance for Industry Extended Release Solid Oral Dosage Forms Development, Evaluation And Application Of In Vitro-In Vivo Correlation. US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. *** Martindale - The Extra Pharmacopoeia 32nd Ed. The Royal Pharmaceutical Society London, 1999 *** Physicians' Desk Reference 49th Ed. Medical Economics, 2000. *** United States Pharmacopeia&National Formulary 24th Ed. The United States Pharmacopeial Convention, Inc., 1999.

- 189 -

Research Protocol - page 19

Consent to participate in a Research Study Study of the Bioavailability of Two Extended Release Propranolol HCl Dosage Forms

College of Pharmacy, University of Cincinnati Sponsor

Bernadette D’Souza, M.D.

(513) 475-6326

Principal Investigator

Phone number

Adel Sakr, Ph.D Thomas Geracioti, Jr., M.D. Elena Draganoiu Coinvestigators

INTRODUCTION

Before agreeing to participate in this study, it is important that the following explanation of the proposed procedures be read. It describes the purpose, procedures, benefits, risks, discomforts and precautions of the study. It also describes alternative procedure available and the right to withdraw from the study at any time. I have been told that no guarantee or assurance can be made as to the results. I have also been told that refusal to participate in this study will not influence standard treatment available to me. I,

have been asked to participate in the research

study under the direction and medical supervision of Dr. Bernadette D’Souza. Other professional persons associated with the study may assist or act for him/her. This research is sponsored by the College of Pharmacy, University of Cincinnati. I will be one of 8 subjects to participate in this trial. - 190 -

Research Protocol - page 20

PURPOSE

The purpose of this research study is to evaluate how a new preparation of an extended release Propranolol tablet behaves in the body when taken by healthy volunteers who are under fasting conditions and to compare the blood concentrations of Propranolol when taken as a once-a-day tablet versus a marketed once-a-day capsule. Propranolol is a beta-adrenergic blocker that is currently used for the treatment of high blood pressure, anginal chest pain, some types of heart beat irregularities, and post heart attacks. It is approved for use in the United States DURATION

My participation in this study will last for approximately 14days. PROCEDURE

I have been told that during the course of the study, the following will occur: Initially a physician will take my medical history, perform a physical examination, check my body weight and record my electrocardiogram. For testing purposes, approximately two teaspoons of blood will be drawn from a vein in my arm. The results of my tests and physical examination will be kept confidential and disclosed only as required by law. All procedures will be completed within a seven day timeframe to determine my eligibility for the study. The study consists of two periods separated by one week. On Day One of each period I will come to the outpatient facility at 7.30 am and I will remain there for at least 12 hours after dosing. Because this study is performed under fasting conditions, I will be required not to eat anything after 8 pm of the evening before the test day. I will be able to eat lunch at 1 pm on Day One. I may drink water as desired except within one hour before and after receiving the study medication. A catheter with a lock (hep-lock) will be inserted into a vein in my arm before the drug administration and will be kept at site for 12 hours after administration.

- 191 -

Research Protocol - page 21

During each period I will be given 80 mg of Propranolol either in the form of a capsule or in the form of a tablet with 240 ml of water. During each period I will have twelve (12) blood samples drawn, each sample being about 5 ml or 1 teaspoon. Ten samples will be drawn on Day One through the hep-lock catheter from 8 am to 8 pm. The other two samples will be drawn by direct venipuncture on Day Two at 8 am and 2 pm. EXCLUSION

I should not participate in this study if any of the following apply to me: I am under 18 or over 65 years of age. I have a medical condition (hematological, renal, hepatic, gastrointestinal, endocrine, pulmonary, dermatological, oncological or neurological, alcoholism), requiring medical treatment, medications or care I have a history of cardiovascular disease I take concomitant drugs I am allergic to Propranolol I am a pregnant or lactating woman I have participated in another drug study or any other study using the same drug within the last three months. I have also been informed and understand that: I should be free of all over the counter preparations for one week before starting the study and during the entire study I should not drink alcoholic beverages for 24 hours prior to dosing at each period I am not allowed to smoke from 1 hour prior to until 12 hours after dosing I should refrain to from eating or drinking any caffeine-containing products (chocolate, tea, coffee, cola) for at least 12 hours prior and 12 hours after dosing. RISKS / DISCOMFORTS

I have been told that the study described above may involve certain risks and discomforts, as well as the possibility for unforeseen risks. The 80 mg daily dose - 192 -

Research Protocol - page 22

of Propranolol which I will be taking is relatively low as compared to the normal maintenance dose of 120 to 240mg/day. The most frequently reported adverse events with Propranolol have been gastrointestinal discomfort, loss of appetite, nausea, vomiting, diarrhea, and abdominal pain. Other less frequently reported adverse events are: decreased circulation to the extremities, congestive heart failure, sleep disturbances, dizziness, fatigue and breathing problems. On rare occasions, rash and allergic reactions to Propranolol have been reported. There is a risk of bruising on my arm at the intravenous sites used for drawing blood samples. In case I experience any adverse effects, I will contact Dr. D’Souza at (513) 4756326 to obtain necessary medical treatment. PREGNANCY

If I am a woman of childbearing potential, I will not participate in this research study unless I have a negative pregnancy test and am using an approved form of birth control. I agree to inform the investigator immediately if: 1) I have any reason to suspect pregnancy; 2) I find that circumstances have changed and that there is now a risk of becoming pregnant; or 3) I have stopped using the approved form of birth control. BENEFITS

I have been told that I will receive no payment from my participation in this study, but my participation may help health care practitioners better understand the release and absorption of the study drugs after administration. I will also receive educational materials up to a value of $200. ALTERNATIVES

The study will evaluate the rate of absorption of the studied drugs in healthy volunteers and it is not intended for treatment of a medical condition. As such, - 193 -

Research Protocol - page 23

there are no alternative treatments or procedures that are advantageous to me, the study participant. NEW FINDINGS

I have been told that I will receive any new information during the course of the study concerning significant findings that may affect my willingness to continue participating in the study. CONFIDENTIALITY

Every effort will be made to maintain the confidentiality of my study records. Agents of the United States Food and Drug Administration, representatives of the UCMB – IRB, the investigator and coinvestigators or sponsor will be allowed to inspect sections of my medical and research records related to this study. The data from the study may be published; however I will not be identified by name. My identity will remain confidential unless disclosure is required by law. FINANCIAL COSTS TO THE SUBJECT

Funds are not available to cover the costs of any ongoing medical care and I remain responsible for the cost of non-research related care. Tests, procedures and other costs incurred solely for purposes of research will not be my financial responsibility. COMPENSATION IN CASE OF INJURY

If I am injured as result of research, I will contact Dr. D’Souza at (513) 475-6326 or the Chairman of the Institutional Review Board at (513) 558-5259. The University

of

Cincinnati

Medical

Center

makes

decisions

concerning

reimbursement for medical treatment for injuries occurring during, or caused by participation in biomedical or behavioral research. In the event I become ill or injured as a direct result of my participation in the research study, necessary medical care will be available to me and the University, at its discretion, will pay medical expenses necessary to treat such injury (1) to the extent I am not - 194 -

Research Protocol - page 24

otherwise reimbursed by my medical or hospital insurance or by third party or governmental programs providing such coverage and (2) provided I have used the study drug as directed by the study doctor in accordance with the study protocol. Financial compensation for such things as lost wages, disability or discomfort due to injury during research is not routinely available. PAYMENT TO PARTICIPANTS

I have been told that I will be compensated for my participation in this study with educational materials up to a value of $200 RIGHT TO REFUSE OR WITHDRAW

It has been explained to me that my participation is voluntary and I may refuse to participate, or may discontinue my participation AT ANY TIME, without penalty or loss of benefits to which I am otherwise entitled. I have also been told that the investigator has the right to withdraw me from the study AT ANY TIME. I have been told that my withdrawal from the study may be for reasons solely related to me (e.g. not following study-related directions from the investigator; a serious adverse reaction) or because the entire study has been terminated. I have been told that the sponsor has the right to terminate the study or the investigator’s participation in the study at any time.

OFFER TO ANSWER QUESTIONS

This study has been explained to my satisfaction by

and

my questions were answered. If I have any others questions about this study, I may call Dr. D’Souza at (513) 475-6326. If I have any questions about my rights as a research subject, I may call UCMC IRB Chairperson at (513) 558-5259. If research related injury occurs, I will call Dr. D’Souza at (513) 475-6326.

- 195 -

Research Protocol - page 25

LEGAL RIGHTS

Nothing in this consent form waives any legal rights I may have nor does it release the investigator, the sponsor, the institution or its agents from liability for negligence. I HAVE READ THE INFORMATION PROVIDED ABOVE, I VOLUNTARILY AGREE TO PARTICIPATE IN THIS STUDY. AFTER IT IS SIGNED, I WILL RECEIVE A COPY OF THIS CONSENT FORM.

Subject Signature

Date

Signature and Title of Person Obtaining Consent and

Date

Identification of Role in the Study

Principal Investigator Signature

Date

Revised 11/07/01

- 196 -

8. Appendix 2 Subjects monitoring during the pilot bioequivalence study - period 1 (blood pressure and heart rate) Time postdosing (hr)

0 1 2 3 4 5 6 8 10 12 24 30 Weight

Subject (treatment) 1 (CPS)

2 (CPS)

3 (TB)

4 (TB)

5 (TB)

6 (CPS)

7 (CPS)

8 (TB)

149/80 74 146/87 66 134/78 63 155/80 56 146/83 61 156/82 56 141/77 72 143/73 68 138/75 65 131/72 63 149/78 62 146/69 79 197

148/83 86 126/71 72 128/78 66 127/69 51 118/65 50 131/64 60 131/69 85 134/66 66 136/68 61 145/72 62 136/68 65 149/73 83 164

114/68 69 113/68 62 106/52 64 106/53 56 103/63 58 104/64 62 109/57 64 107/57 73 112/56 66 109/55 71 103/64 68 115/62 57 170

140/70 68 129/58 61 116/71 61 128/75 55 115/60 58 137/56 54 131/63 75 142/51 62 142/64 59 145/58 65 137/66 61 149/70 67 217

113/70 72 113/66 78 97/69 54 106/65 60 110/68 62 119/76 76 121/71 84 91/48 76 98/78 86 104/64 73 112/68 76 127/70 90 117

121/68 75 116/67 66 113/71 60 112/62 62 98/60 61 98/74 66 94/56 73 103/61 70 123/70 73 112/65 74 103/64 72 129/77 80 163

114/66 59 126/66 55 101/61 55 101/61 49 104/58 49 102/61 50 95/54 58 142/118 53 114/70 56 116/67 54 119/72 61 122/65 58 155

111/72 86 99/71 80 87/64 74 99/64 74 86/57 67 96/61 70 96/57 74 94/57 74 86/58 72 96/61 76 94/62 83 93/61 90 116

TB – Propranolol 80mg matrix tablets with 65% Kollidon® SR CPS - Inderal® LA 80mg capsules

- 197 -

Subjects monitoring during the pilot bioequivalence study - period 2 (blood pressure and heart rate) Time postdosing (hr) 0 1 2 3 4 5 6 8 10 12 24 30 weight

Subject (treatment) 1 (TB) 136/83 62 144/76 60 120/66 52 124/71 52 127/74 55 119/55 55 140/70 67 140/78 71 136/76 71 145/73 71 155/70 70 152/76 71 201

2 (TB) 129/66 73 100/75 74 133/69 59 130/75 70 124/72 60 124/62 57 138/68 76 137/63 79 140/75 73 132/64 71 122/76 68 140/75 96 165

3 (CPS) 107/62 65 113/74 64 112/64 64 116/65 61 110/64 58 107/76 62 108/64 72 117/61 65 122/69 68 110/62 68 134/67 62 110/60 75 169

4 (CPS) 130/58 65 130/67 60 134/67 57 136/77 62 147/69 57 127/62 60 112/59 76 136/52 65 150/65 61 146/56 67 128/64 62 160/61 72 218

5 (CPS) 112/70 67 118/72 86 111/77 68 108/64 62 107/69 69 108/69 67 105/61 70 117/62 67 114/65 70 110/63 63 100/60 70 115/59 66 125

6 (TB) 114/69 77 106/70 72 105/68 64 98/69 67 95/58 66 106/67 66 102/58 73 96/53 72 105/75 71 92/74 68 115/70 70 122/59 69 166

TB – Propranolol 80mg matrix tablets with 65% Kollidon® SR CPS - Inderal® LA 80mg capsules

- 198 -

7 (TB) 126/65 62 113/73 60 124/71 50 115/70 49 111/68 51 102/67 48 112/65 66 105/52 60 119/61 57 105/59 58 125/76 57 123/67 62 156

8 (CPS) 105/62 91 94/58 76 94/67 80 94/60 74 92/59 74 82/56 76 96/60 86 83/49 81 92/61 80 105/68 85 92/57 80 96/55 87 117