Multidrug, Anti-HIV Amorphous Solid Dispersions: Nature and

Department of Industrial and Physical Pharmacy, Purdue University, West ... Department of Biological Sciences, Virginia Tech, Blacksburg, Virginia 240...
0 downloads 0 Views 2MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Article

Multidrug, anti-HIV amorphous solid dispersions: nature and mechanisms of impacts of drugs on each other’s solution concentrations Hale Çigdem Arca, Laura I. Mosquera-Giraldo, Durga Dahal, Lynne S. Taylor, and Kevin J. Edgar Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b00203 • Publication Date (Web): 05 Sep 2017 Downloaded from http://pubs.acs.org on September 7, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Molecular Pharmaceutics is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Multidrug, anti-HIV amorphous solid dispersions: nature and mechanisms of impacts of drugs on each other’s solution concentrations Hale Çiğdem Arca1, Laura I. Mosquera-Giraldo2, Durga Dahal3, Lynne S. Taylor2, Kevin J. Edgar1,4* 1

Macromolecules Innovation Institute, Virginia Tech, Blacksburg, VA 24061, USA

2

Department of Industrial and Physical Pharmacy, Purdue University, IN, 47907, USA

3

Department of Biological Sciences, Virginia Tech, Blacksburg, VA, 24061, USA

4

Department of Sustainable Biomaterials, Virginia Tech, Blacksburg, VA 24061, USA

Abstract Drug therapy has been instrumental in prolonging the lives of patients infected by human immunodeficiency virus (HIV). In order to combat development of resistance, therapies involving three or more drugs in combination are recommended by the World Health Organization (WHO) to suppress HIV and prevent development of acquired immune deficiency syndrome (AIDS). It is desirable for multidrug combinations to be coformulated into single dosage forms where possible, to promote patient convenience and adherence to dosage regimens, for which amorphous solid dispersion (ASD) is particularly well-suited. We investigated multi-drug ASDs of three model anti-HIV drugs, ritonavir (Rit), etravirine (Etra), and efavirenz (Efa), in cellulosic polymer matrices. We hypothesized that the presence of multiple drugs would reduce crystallization tendency, thereby providing stable, supersaturating formulations for bioavailability enhancement. We explored new ASD polymers including cellulose acetate suberate (DSSub 0.9, CASub) and cellulose acetate adipate propionate (DSAd 0.9, CAAdP), and control commercial cellulosic polymers including 6-carboxycellulose acetate butyrate (CCAB) and carboxymethyl cellulose acetate butyrate (CMCAB). We succeeded in preparing three drug ASDs containing very high drug loadings (45% drug total; 15% of each drug); each polymer tested was effective at stabilizing the amorphous drugs in the solid phase, as demonstrated by XRD, SEM and DSC studies. In pH 6.8 dissolution studies ASDs released each anti-HIV drug over 8 h, affording supersaturated

ACS Paragon Plus Environment

1

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 33

solutions of each drug, but unexpectedly failing in some cases to reach maximum possible supersaturation. In a second set of dissolution studies (pH 6.8), the cause of the observed solution concentration limitations was investigated by studying release from single- and two-drug ASDs. Concentrations of Rit, Etra and Efa achieved from three-drug ASDs were higher than those achieved from crystalline drugs. Surprisingly however, there was a decrease in the achieved drug concentrations of both Rit and Efa when they dissolved together, while Etra solution concentration was enhanced by the presence of Rit and Efa in the ASD. We demonstrate that these effects have to do primarily with solution phase interactions between the anti-HIV drugs, rather than from the drugs influencing each other’s release rate, and we suggest that such observations may indicate an important, previously inadequately recognized, and general phenomenon for ASDs containing multiple hydrophobic drugs. Key Words: Amorphous solid dispersion, multidrug formulation, nanodroplets, supersaturation, cellulose acetate suberate, carboxymethyl cellulose acetate butyrate, hydroxypropyl methyl cellulose acetate succinate, HIV, AIDS 1. Introduction Human immunodeficiency virus (HIV) is a retrovirus that infected, according to the World Health Organization (WHO), a total of 37 million people globally in 2015, including 2.1 million new HIV infections. In the same year, fully 1.1 million people lost their lives due to acquired immunodeficiency syndrome (AIDS), which is caused by HIV infection.1 No cure is currently available for HIV infection, but fortunately the Herculean efforts of biologists, chemists, and physicians since the first identification of HIV and AIDS in the early 1980s have provided a remarkable level of understanding of the biology of this virus, and a suite of therapeutic drugs that, in most cases, can suppress its harmful effects over the course of a lifetime.2–4 Development of a combination therapy comprising three or more anti-HIV drugs with different drug mechanisms, called Highly Active Anti-Retroviral Therapy (HAART), was a major advance in control of HIV infection and prevention of progression to AIDS, as HAART therapy effectively suppresses HIV, and prevents development of resistant strains.5,6 Development of more

ACS Paragon Plus Environment

2

Page 3 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

potent drugs with fewer side effects has made patient compliance easier, but the daily requirement for multiple pills is still challenging for patients who do not feel ill.7,8 Non-nucleoside reverse-transcriptase inhibitors (NNRTI), such as etravirine (Etra) and efavirenz (Efa), bind to and inhibit the reverse-transcriptase enzyme, which is essential for retroviral replication. Etra, a second generation NNRTI used for the treatment of HIV-1, has a high genetic barrier to viral resistance, so that clinically significant resistance develops only as a result of a large number of critical mutations; Etra therefore continues to be effective even in the presence of common NNRTI mutations.9 Ritonavir (Rit) is used to inhibit Cytochrome P450-3A4 (CYP3A4) enzymes that metabolize many protease inhibitors, thereby increasing their efficacy significantly. Thus, HAART therapy attacks HIV through multiple mechanisms, enhancing efficacy and retarding the development of resistant strains. Although HAART formulations were first approved more than a decade ago,10 HIV treatment is still problematic. In developing countries, the cost of anti-HIV medications is not affordable for many patients, and support for treatment from foundations and governments has not proved adequate to get necessary drug therapy to all patients. Additional therapeutic challenges include patient non-compliance to lifelong, multi-pill therapeutic regimes in environments where medical care is difficult, and potential for significant side effects.11 Oral administration is essential for practical, daily, lifelong treatment of HIV patients. Most of the more than 25 anti-HIV drugs approved by FDA12,10 for oral administration have low solubility in water (0.1-10 µg/mL), limiting the rate and the extent of drug absorption. While other methods have been used to increase solubility of anti-HIV drugs,13–16 amorphous solid dispersion (ASD) has particular value for enhancing drug solution concentration, leading to enhanced bioavailability. ASD works by trapping the drug in a polymer matrix, creating a metastable, molecular dispersion of drug in polymer, thereby eliminating the drug crystal lattice energy as a barrier to drug dissolution. Because this methodology creates supersaturated solutions, rather than enhancing thermodynamic solubility (as cyclodextrin and solvent approaches do, for example)17, ASDs also enhance the transport rate across the enterocytes and thus reduce both

ACS Paragon Plus Environment

3

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 33

solubility and permeation barriers to bioavailability. Since ASDs are often prepared starting with an organic solvent solution of drug and polymer, then converting to particulate ASD by methods including spray-drying or film casting, it would seem to be particularly well-suited to formation of multi-drug molecular dispersions in one or more polymers (melt extrusion is another popular ASD formation technique, which could be equally well-suited to formation of multi-drug dispersions). Our starting hypothesis was that formation of ASDs containing multiple drugs should be advantageous, since the presence of the other drug(s) would tend to inhibit crystallization of any one-drug component, thereby enhancing ASD stability against crystallization18 that could in turn permit higher drug loading, and/or faster drug release. Confirmation of this hypothesis would be particularly valuable for HIV drug treatment, since it could permit reduction in drug usage and expense, potentially expanding the number of patients treated for a given amount of costly drug. In addition, successful multidrug formulations could enhance patient compliance, and the reduced doses might even diminish certain side effects. We chose Rit, Efa, and Etra (Fig. 1) as model drugs for this study due to their suitability for oral administration, low solubility, low bioavailability, and ready availability, realizing that this particular three-drug combination is not necessarily the ideal anti-HIV drug treatment regime. CH3 O

Br O

O

H N

O N

S

N H

N OH

O

H

N CH3

NC S

NH2 N

CH3

F3C

N Cl

HN

N

N CN

Ritonavir

Etravirine

H

O

Efavirenz

Fig. 1. Chemical structures of antiviral drugs investigated. We chose to explore ASDs with cellulose acetate suberate (CASub) and cellulose acetate adipate propionate (CAAdP) (Fig. 2), which were designed and developed in the Edgar and Taylor laboratories as high-performance ASD polymers.19,20 Each of these cellulose ω-carboxyalkanoates has high solubility in organic solvents, high glass transition temperature (Tg), and demonstrated ability to form amorphous dispersions with at least

ACS Paragon Plus Environment

4

Page 5 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

some of these model drugs. We chose to compare them to currently or formerly commercial cellulosic polymers that have promise in ASD, namely carboxymethyl cellulose acetate butyrate (CMCAB)21 and 6-carboxycellulose acetate butyrate (CCAB).22 CMCAB has been shown repeatedly to be an effective ASD polymer, while the chemical features of CCAB (high Tg, good solubility, carboxyl groups for release trigger and to enhance specific interactions with drug molecules) are of interest for that purpose in our laboratory (Fig. 2). O O

O

O

O

HO

n

O

O

H

O

(CH2)4 O

O

HO2C

HO2C HO2C

O

HO

O

CAAdP

O

O

CMCAB

H

n

O

O

O O

O

O

O HO

n

O

O (CH2)6

HO

H

O

O

O O

n

H

O O

HO2C

CASub

CCAB

Fig. 2. Chemical structures of CMCAB, CASub, CAAdP, and CCAB. These structures are not meant to convey regioselective substitution; specific substituent locations are merely for convenience and clarity of depiction (except for CCAB carboxyl, which is exclusively at C-6). ASDs of hydrophobic drugs, like those used in our study, can be plagued by poor drug release, especially at high drug loadings. Surfactants are often useful for improving drug release from ASDs.23 D-α-Tocopheryl polyethylene glycol 1000 succinate (TPGS, Fig. 3) is a nonionic surfactant that is also a weak inhibitor of P-glycoprotein (P-gp). TPGS forms micelles above its critical micelle concentration (0.02 wt%), and can thereby help to solubilize some lipophilic actives in aqueous media. This solubilization, as well as in some cases the P-gp inhibitory activity of TPGS, can increase oral bioavailability.24–26

ACS Paragon Plus Environment

5

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 33

Fig. 3. Chemical structure of D-α-tocopherol poly(ethylene glycol) succinate (TPGS). Table 1. Polymer physicochemical properties Polymer Type

DS (CO2H)

DS (Other)

Solubility Parameter (MPa1/2)

Polymer Tg (° C)

3-Drug ASD Tg (° C)

CAAdP

0.90

Pr 2.09 Ac 0.04

22.79

116

58

CASub

0.90

Ac 1.82

23.72

101

58

CMCAB

0.30

CM 0.33 Bu 1.64 Ac 0.44

23.03

141

77

CCAB

0.28

Bu 1.62 Ac 0.06

24.44

134

74

We report herein investigations to test our hypothesis by attempts to form ASDs of three model anti-HIV drugs (Etra, Efa and Rit) with the selected cellulosic polymers, CMCAB, CCAB, CASub, and CAAdP, and to evaluate the extent and rate of release of each drug from the ASD, as well as the extent of supersaturation achieved. We report also the unexpected drug interactions that we encountered, the influence of added TPGS, the consequences of those interactions with regard to supersaturation of each individual drug, and our initial investigations of the mechanism leading to the observed effects. This work is part of our overall efforts to design novel, high-performance ASD polymers to solve performance issues with commercial polymers and enable broader use of ASD to enhance

ACS Paragon Plus Environment

6

Page 7 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

effectiveness of existing drugs, and reduce the number of pipeline drug failures due to bioavailability issues.

2. Experimental 2.1 Materials Etravirine, ritonavir, and efavirenz were purchased from Attix Pharmaceuticals (Toronto, Canada). Cellulose acetate propionate (CAP-504-0.2; DS (acetate) = 0.04; DS (propionate) = 2.09; Mn = 15,000)21, cellulose acetate (CA-320S; DS (acetate) = 1.82; Mn = 50,000)27, CMCAB (CMCAB 641-0.2; DS (butyrate) = 1.64; DS (acetate) = 0.44; DS (carboxymethyl) = 0.33; approximate Mw 22,000)21, CCAB (cellouronic acid acetate butyrate or 6-carboxy cellulose acetate butyrate; DS (butyrate) = 1.62; DS (acetate) = 0.06; DS (CO2H) = 0.28; Mw 252,000), and D-α-tocopheryl polyethylene glycol 1000 succinate (TPGS) were from Eastman Chemical Company (Kingsport, Tennessee). Cellulose ester DS information is by 1H NMR spectroscopy and molecular weight information is by size exclusion chromatography, as described in prior publications28. Polymers were dried in a vacuum oven at 50° C overnight before use. Acetonitrile (HPLC-grade), tetrahydrofuran (THF), benzyl alcohol, toluene, dichloromethane, N, Ndimethylformamide (DMF), ethanol, potassium phosphate monobasic, and sodium hydroxide (NaOH) were purchased from Fisher Scientific and used as received. Suberic acid, sebacic acid, adipic acid, methyl ethyl ketone (MEK), 1,3-dimethyl-2imidazolidinone (DMI) (dried over 4Å molecular sieves), p-toluenesulfonic acid (PTSA), triethylamine (Et3N) and oxalyl chloride were purchased from ACROS Organics. Palladium hydroxide was purchased from Sigma Aldrich. Water was purified by reverse osmosis and ion exchange using the Barnstead RO pure ST (Barnstead/Thermolyne, Dubuque, IA, USA) purification system. 2.2 Methods 2.2.1 Synthesis of CAAdP (DS 0.9) and CASub (DS 0.9): CAAdP was synthesized as previously reported,27,27 as summarized in the Supplementary Material.

ACS Paragon Plus Environment

7

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

Page 8 of 33

H NMR of monobenzyl adipate (CDCl3); δ1.68 (m, 4H), 2.36 (m, 4H), 5.09 (s, 2H),

and 7.32 (m, 5H). 1

H NMR of monobenzyl adipoyl chloride (CDCl3): δ1.73 (m, 4 H), 2.39 (t, 2 H), 2.90

(t, 2 H), 5.12 (s, 2 H), 7.32 (m, 5 H). Similar procedures were followed to synthesize monobenzyl suberoyl chloride (1H NMR (CDCl3): δ1.34 (m, 4H), 1.66 (m, 4H), 2.36 (t, 2H), 2.86 (t, 2H), 5.12 (s, 2H), 7.35 (m, 5H)). 1

H NMR of CAAdP monobenzyl ester (CDCl3): δ1.02-1.20 (m, COCH2CH3 of

propionate), 1.66 (broad s, COCH2CH2CH2CH2CO of adipate), 2.16-2.35 (m, COCH2CH3 of propionate, COCH3 of acetate and COCH2CH2CH2CH2CO of adipate), 3.25-5.24 (cellulose backbone), 5.10 (CH2C6H5), 7.33 (CH2C6H5). DS by 1H NMR (CDCl3): adipate 0.9, propionate 2.09, acetate 0.04. Characterization of final products:

1

H NMR CAAdP (DMSO): δ1.02-1.20 (m,

COCH2CH3 of propionate), 1.66 (broad s, COCH2- CH2CH2CH2CO of adipate), 2.162.35 (m, COCH2CH3 of propionate, COCH3 of acetate and COCH2CH2CH2CH2CO of adipate), 3.25-5.24 (cellulose backbone). DS by 1H NMR: adipate 0.9, propionate 2.09, acetate 0.04. 1

H NMR CASub (DMSO): δ1.2 (COCH2CH2CH2CH2CH2CH2CO of suberate), 1.4-1.6

(COCH2CH2CH2CH2CH2CH2CO

of

suberate),

2.10–2.46

(COCH2CH2CH2CH2CH2CH2CO of suberate, and COCH3 of acetate), and 3.00–5.20 (cellulose backbone). DS by 1H NMR: suberate 0.9, acetate 1.82.

2.2.2 Preparation of ASDs by Solvent Casting and Grinding: Each formulation was prepared to total 1 g material, including polymer and drug(s). Three-drug formulations (0.15 g of each drug and 0.55 g of polymer), two-drug formulations, (0.15 g of each drug and 0.70 g of polymer) and single drug formulations (0.15 g drug and 0.85 g polymer)

ACS Paragon Plus Environment

8

Page 9 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

were each dissolved in 50 mL THF at room temperature. Three-drug samples that included surfactant were prepared with 0.15 g of each drug, 0.15 g of TPGS, and 0.4 g polymer, and were likewise dissolved in 50 mL THF at room temperature. From each solution a film was cast upon a Teflon surface; each film was air-dried at 50 ° C. Each film was then peeled from the Teflon surface and cryogenically ground by placing into a Micro-Mill® (Scienceware®, Wayne, NJ 07470 USA) apparatus with dry ice, then milled for 5 min in order to prepare the dispersion in powder form.

2.2.3 Ritonavir, Efavirenz and Etravirine Analysis by High-Performance Liquid Chromatography (HPLC): Drug analyses employed an Agilent 1200 series HPLC consisting of a quaternary pump, online degasser, autosampler, and Agilent Chemstation LC 3D software. Analyses were conducted in reversed phase mode using an Eclipse XDB-C18 column (4.6 × 150 mm i.d., particle size 5 µm). A gradient analytical method was used employing acetonitrile and 0.05 M phosphate buffer (pH 5.55). The acetonitrile proportion was 40 % for 1 min, then raised to 60 % over 1 min and held for 14 min, then reduced to 40 % in 1 min, remaining at 40 % for 4 min. Flow rate was 1.5 mL/min, column temperature was 25 ° C, sample injection volume was 5 µL, UV detection was at 240 nm, and retention times observed were 10.49 min, 11.57 and 13.80 for Rit, Efa and Etra, respectively. 2.2.4. 1H NMR: Sample (1-5 mg) was dissolved in 1 mL NMR solvent (CDCl3 or (CD3)2SO) and a 0.7 mL portion transferred to the NMR tube. Proton NMR spectra were acquired on an INOVA 400 spectrometer operating at 400 MHz at room temperature using 32 scans, with a 1 second relaxation delay.

2.2.5 Powder X-ray Diffraction (XRD): X-ray powder diffraction patterns were measured using a Shimadzu XRD 6000 diffractometer (Shimadzu Scientific Instruments, Columbia, Maryland) on film samples. The instrument was calibrated relative to a silicon standard, which has a characteristic peak at 28.44° 2θ. Divergence and scattering slits were set at 1.0 mm, and the receiving slit was set at 10 iris. The experiments were conducted with a scan range from 10° to 50° 2θ.

ACS Paragon Plus Environment

9

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

2.2.6 Differential Scanning Calorimetry (DSC): DSC analyses were performed on a TA Instruments Q200. Dry samples (5 mg) were loaded in Tzero™ aluminum pans. Each sample was equilibrated at -50 ◦C and then heated to 190 ◦C at 20 ◦C/min. Samples were then quench cooled to -50 ◦C, then reheated to 190 ◦C at 20 ◦C/min. Tg values were recorded as the step-change inflection point from second heating scans. 2.2.7 Scanning Electron Microscopy (SEM): Particle size and morphology were analyzed on an LEO 1550 field emission scanning electron microscope (FESEM). The samples were spread on double faced adhesive tape and coated with a very thin gold palladium layer (sputter coater Cressington 208HR) for 1 min. 5kV was used for excitation. 2.2.8. Determination of experimental amorphous solubility: Supersaturated solutions of each individual drug were prepared by adding a specific amount of the organic stock solution to 15 mL of 100 mM potassium phosphate buffer pH 6.8 with 5 µg/mL HPMCAS-MF at 37 ˚C and 300 rpm. Total concentrations of the supersaturated solutions were 80 µg/mL EFA, 80 µg/mL RTV and 30 µg/mL ETR. Each solution was centrifuged at 35,000 rpm for 20 min to remove undissolved drug, using an Optima L-100 XP ultracentrifuge equipped with Swinging-Bucket Rotor SW 41 Ti (Beckman Coulter, Inc., Brea, CA). The supernatant concentration, which represents the free drug in solution, was measured by HPLC.

2.2.9. In Vitro Drug Release from ASDs: Dissolution studies were conducted in 250 mL jacketed flasks, using circulating ethylene glycol/water (1:1) to maintain 37 ° C. Phosphate buffer (0.05 M, pH 6.8, 100 mL) was charged to each flask; volume was kept constant by replacing each aliquot (1 mL) withdrawn by an equal volume of fresh buffer solution. All dissolution experiments were conducted under non-sink conditions that would permit supersaturation. Efa has the highest aqueous solubility among the drugs studied, therefore Efa concentration was set to afford supersaturation of 25X theoretical solubility and the other two drugs were tested at the same concentration as Efa.

ACS Paragon Plus Environment

10

Page 11 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

2.2.9.1. Experiment A: Dissolution Study to Evaluate Drug Release Profiles: Drugcontaining ASD (166 mg) or 25 mg of each pure drug (25 mg Efa, 25 mg Etra and 25 mg Rit together) was added to the pH 6.8 phosphate buffer (0.05 M, 100 mL) in the dissolution flask and magnetically stirred at 400 rpm. Samples (1 mL each) were withdrawn every 0.5 h during the first 2 h, then every 1 h for the following 6 hours. Each sample was centrifuged at 13,000 rpm for 10 min, then 0.5 mL supernatant from the top of the centrifuge vial was transferred to an HPLC vial and analyzed immediately. Two-drug experiments were conducted with a similar protocol where 166 mg ASD (containing 25 mg each of 2 drugs) was placed into a dissolution flask and sampling was by the same protocol as above. Control experiments were carried out involving dissolution of each corresponding one-drug ASD (each containing 25 mg drug). An additional control involved adding two separate one-drug ASDs to the same dissolution flask (25 mg of drug in each ASD), and measuring dissolution of both simultaneously, again using the same sampling protocol as above.

2.2.9.2 Experiment B: Evaluation of Drug-Drug Solution Interactions: A single drug ASD (166 mg) was dispersed into PBS buffer solution (0.05 M, 100 mL, pH 6.8) and stirred 8 h at 400 rpm. Afterwards, another single drug ASD (166 mg) was added to the buffer and aliquots were withdrawn at 1, 2 and 8 hours thereafter. Each sample was centrifuged at 13,000 rpm for 10 min, then 0.5 mL supernatant from the top of the centrifuge vial was transferred to an HPLC vial and analyzed immediately.

2.2.10 Dynamic Light Scattering The size of species present in supersaturated solutions evolved by dissolution of select ASDs was measured using a Nano-Zetasizer and dispersion technology software (DTS) (Malvern Instruments, Westborough, MA). 12 mm square polystyrene disposable cuvettes were used for particle size analysis. 12 mg of ASD (15 % drug: 85% CMCAB) was added to 10 mL of phosphate buffer pH 6.8, 100 Mm. The samples were stirred for 6 hours, and filtered through a 0.45 µm glass filter to remove any undissolved ASD particles prior to analysis.

ACS Paragon Plus Environment

11

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

3. Results and Discussion 3.1 ASD Preparation and Characterization Solid dispersions of the anti-HIV drugs Etra, Efa and Rit were prepared by solvent casting and subsequent micro-milling of the resulting films, using cellulose derivative matrices (CMCAB, CCAB, CAAdP (DS 0.9) and CASub (DS 0.9)) (Fig. 2). The onedrug dispersions were prepared with 15% drug loads, in order to determine whether each drug could form a miscible dispersion with each polymer, and whether such ASDs would afford supersaturated solutions of the individual drugs. Key physical properties and categories of each of these poorly soluble drugs are summarized in Table 2. We initially explored

whether

the

one-drug

dispersions

were

amorphous

by

solid-state

characterization, using XRD, DSC and SEM, comparing as appropriate versus pure drug. Etra, Efa and Rit are crystalline drugs, showing sharp, characteristic XRD peaks (Fig. 4). XRD diffraction patterns (see Fig. S1) of the three one-drug formulations showed no sharp peaks; instead each exhibited an amorphous halo, as shown in Fig. S1. Furthermore all two- (Fig. S3) and three-drug (CCAB and CMCAB Fig. 5; CASub and CAAdP, Fig. S2) dispersions also showed amorphous haloes but no crystalline diffraction peaks, confirming that all of these dispersions are amorphous to the limits of the XRD technique.

ACS Paragon Plus Environment

12

Page 13 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Fig. 4 XRD diffractograms of crystalline Etra, Efa and Rit.

Fig. 5 XRD diffractograms of 3-drug ASDs (15% Etra, 15% Efa and 15% Rit) using CMCAB, and CCAB matrix polymers, as well as with 15% TPGS in CMCAB.

ACS Paragon Plus Environment

13

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

Table 2. Physicochemical properties of anti-HIV drugs used. Property

Etra

Efa

Rit

Molecular weight (g mol-1)

435

316

721

pKa

3.529

10.2

1.8 and 2.620

Log P

5.2

4.7

5.6

Melting point (° C)

260

13929

12329

Tg of amorphous drug (° C)

9930

3331

5032

ND*

19.833

20.633

0.83 ± 0.05

18.7 ± 0.4

27.8 ± 0.9

≤1

9.8

1.5

Drug category

NNTRI

NNRTI

PI

BCS

Class IV

Class II

Class IV

Recommended dose

200 mg

600 mg

600 mg

Theoretical amorphous -1

solubility (µg mL ) Experimental amorphous solubility (µg mL-1) Experimental crystalline solubility (µg mL-1)

*Etra decomposes upon melting, and hence the value corresponding to the enthalpy of fusion is unreliable. As a consequence, it was not possible to determine Etra theoretical amorphous solubility.

ACS Paragon Plus Environment

14

Page 15 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Fig. 6. DSC thermograms of three-drug CMCAB and TPGS three-drug CMCAB dispersions. Thermal analysis is also a useful tool for quantifying drug-polymer miscibility; partial miscibility can lead to instability since concentrated drug domains can promote recrystallization. If polymer and the drug(s) are miscible, a single Tg that ranges between the Tg values of pure components should be recorded. Theoretical Tg values were calculated according to the Gordon-Taylor equation:

Tgmix =

(Tg1W1 + Tg2W2 ) (W1 + W2 )

where Tg is the glass transition temperature, and W1 and W2 are the weight fractions of the components. Thermal properties of the crystalline drugs are presented in Table 2. DSC thermograms of the drug combinations show (Fig 6) the absence of drug melting endotherms or crystallization exotherms which, along with the single Tg values, suggest complete drug-polymer miscibility in each 3-drug dispersion prepared, consistent with the XRD results. In the three-drug CMCAB dispersion containing TPGS, there is some

ACS Paragon Plus Environment

15

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 33

hint of additional endothermic behavior in the vicinity of 40 ° C; melting of the PEG portion of TPGS has been observed by DSC at about that temperature34. Experimentally obtained Tg values were close to theoretical predictions for single drug formulations, but showed significant negative deviations from predictions for three-drug formulations (Supplementary Material). Such deviations are often associated with stronger drug-drug associations than polymer-drug associations34, and may indicate somewhat greater affinity of these quite hydrophobic drugs for each other than for the polymer. 3.2 Drug Release Profiles at pH 6.8 Having established that all the dispersions were amorphous, even with all three drugs present and even with as high as 45% drug content, we investigated the dissolution properties of the anti-HIV drugs from three-drug ASDs at small intestine pH in an attempt to confirm our hypothesis that the presence of an additional drug would help to stabilize each drug against crystallization and thereby enhance supersaturation. Considering that one of the aims of this study was to evaluate the ability of the polymer to stabilize supersaturated drug solutions against recrystallization, the amount of dispersion used was chosen to allow supersaturated solutions of each drug, assuming that all the drugs were completely released. Dissolved drugs were quantified by an HPLC method that afforded baseline peak separation of the three anti-HIV drug peaks. 3.2.1 Drug supersaturation from three-drug ASDs Results from dissolution experiments employing the ternary drug ASDs are presented in Figure 7. Each includes the appropriate crystalline drug control for comparison. Dissolution results of ASDs prepared with commercial polymers CMCAB and CCAB are presented in Fig. 7a, b, and c, while Fig 7d, e, and f display dissolution of ASDs prepared with new cellulose ω-carboxyalkanoate polymers CAAdP and CASub. We discuss the results by individual drug to aid in deconvolution of the complex phenomena involved. Although the final Rit solution concentration from CMCAB, CCAB and CAAdP 3-drug ASDs is higher than Rit crystalline solubility, the value is well short of maximum attainable Rit supersaturation (Fig. 7a and d).19 The supersaturation ratio observed was 1.5-fold for CMCAB, 2-fold for CCAB and 1.7-fold for CAAdP. Supersaturation was

ACS Paragon Plus Environment

16

Page 17 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

stable throughout the 8 h experiment in the presence of either CCAB or CMCAB, significantly in excess of the time the formulation would typically spend in the small intestine. We were surprised to observe only slight Rit supersaturation from CASub 3drug ASDs (CASub is slightly more hydrophilic than is CAAdP). Rit solution concentration from CAAdP ASD was 1.45 fold higher than that from crystalline drug, but CASub ASD did not perform better than crystalline Rit in spite of its amorphous nature (Fig 7d). This suboptimal Rit supersaturation from our 3-drug ASDs (note that maximum Rit supersaturation (ca. 20 µg/mL) has been previously reported from single drug ASDs)28 was our first indication that these systems were more complex than first anticipated and could reveal unexpected phenomena. Crystalline Efa has considerably higher aqueous solubility (9.8 µg/mL) than Rit, consistent with its higher calculated solubility parameter. Keeping in mind that the amorphous to crystalline solubility ratio for efavirenz is only approximately 2-fold, ASD is not expected to afford a large Efa solution concentration improvement. Solution concentration was actually higher from crystalline Efa than for Efa ASDs prepared with CMCAB, CCAB or CASub polymers (5.0, 5.8, and 6.7 µg/mL, respectively, Fig. 7b and 7e); this is counterintuitive considering that amorphous drug typically has higher solubility than the crystalline form. On the other hand, CAAdP 3-drug ASD provided approximately the same Efa solution concentration (9.9 µg/mL) as from crystalline drug. Overall, dispersion in the 3-drug polymer ASDs investigated did not enhance Efa solution concentration as we had hypothesized. The three-drug CAAdP ASD was most generally effective among all the three-drug formulations (without surfactant), since it afforded Rit and Etra concentrations higher than those from the crystalline drugs, and Efa concentration similar to that achieved from the drug itself (Table 3). ASD of Efa has been reported previously (ca. two-fold supersaturation has been reported from ASD with the methyl/butyl/dimethylaminoethyl acrylate copolymer Eudragit EPO35). Etra has been formulated in ASD with hydroxypropyl methyl cellulose (HPMC) for clinical use36, and roughly 7 fold supersaturation has been reported in the presence of HPMC or HPMCAS37.

ACS Paragon Plus Environment

17

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 33

Rit and Efa were released immediately from CAAdP and CASub ASDs, and the supersaturation achieved was stable for the 8h duration of the experiment. On the other hand, Etra was released slowly and continuously from the CAAdP ASD. Etra is the least water-soluble of the three anti-HIV drugs examined, solubility indeed being so low as to be hard to measure reliably. Etra thermodynamic solubility at pH 6.8 was below our HPLC detection limit, and has been described as less than 1 µg/µL in the literature.29 Etra solution concentration from the CAAdP ASD (2 µg/mL, Table 3) was higher than that of the crystalline drug (p value 0.012) and from the CASub ASD (which did not differ from pure Etra). On the other hand, Etra solution concentration from CMCAB 3-drug ASD was significantly higher than that achieved from crystalline drug (p value 0.035), showing steady release throughout the experiment and reaching 0.74 and 0.12 µg/mL after 8h from CMCAB and CCAB ASDs, respectively. These concentrations are modest to be sure, but are quantifiable by our method and show improvement in the indicated cases versus that from crystalline Etra. No Etra recrystallization was observed from these ASD formulations over the 8h duration of these experiments.

ACS Paragon Plus Environment

18

Page 19 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Fig. 7. Dissolution profile for crystalline drugs and 3-drug ASD (15 wt % of each drug in each ASD), all dissolution experiments run at pH 6.8, 37° C. In each dissolution study, error bars indicate one standard deviation (n = 3). (a, b, c) Rit/Efa/Etra/CMCAB ASD, Rit/Efa/Etra/CCAB

ASD,

and

Rit/Efa/Etra/TPGS/CMCAB

ASD.

(d,

e,

f)

Rit/Efa/Etra/CASub, and Rit/Efa/Etra/CAAdP ASDs.

ACS Paragon Plus Environment

19

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 33

Table 3. Maximum solution concentrations from 3-drug ASDs (all in µg/mL) Polymer matrix

Rit

Efa

Etra

CCAB

3.21

4.98

0.12

CMCAB

2.39

5.79

0.74

CAAdP

2.57

9.85

1.93

CASub

1.70

6.69

0.0

All of the cellulosic polymers tested are relatively hydrophobic materials, which favors strong interaction with hydrophobic drugs like these anti-HIV drugs, but can also restrain release rates. Therefore we hypothesized that addition of a surfactant to the formulation might enhance release rates and improve solution concentrations from these three-drug ASDs. We tested this hypothesis with CMCAB, preparing a dispersion containing 15 wt% of each the three drugs, 15 wt% TPGS, and 40% CMCAB. Since each formulation is a closed system, when one adds something, one must take away something else. We choose in this case to add TPGS, and subtract CMCAB; therefore each formulation has added hydrophilic TPGS but also a lower CMCAB/drugs ratio. Note that each buffer solution, after TPGS dissolves from the ASD, contains roughly 0.025 wt % TPGS, or slightly in excess of its critical micelle concentration (0.02 wt %)25. The TPGScontaining dispersion was shown to be amorphous by XRD and DSC (Figs. 4 & 6). We carried out dissolution studies with this formulation, comparing solution concentration of each drug vs. that from crystalline drug and from the three drug CMCAB formulation without TPGS (Figs. 7a-c). For each drug, the CMCAB ASD containing TPGS surfactant afforded significantly higher solution concentration than either crystalline drug or ASD without TPGS. Rit solution concentration was approximately doubled (to 4.9 µg/mL) vs. the Rit/CMCAB ASD without TPGS, and tripled vs. crystalline Rit. Efa solution concentration was not enhanced by CMCAB ASD alone, but was increased from 8.9 to 12.9 µg/mL by the presence of TPGS surfactant; note that concentration was still

ACS Paragon Plus Environment

20

Page 21 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

increasing at the 8h termination of the experiment. Most strikingly, solution concentration of Etra was increased from below quantification limit (