Ritonavir–PEG 8000 Amorphous Solid Dispersions: In vitro and In vivo Evaluations

Ritonavir–PEG 8000 Amorphous Solid Dispersions: In Vitro and In Vivo Evaluations DEVALINA LAW,1 ERIC A. SCHMITT,1 KENNAN C. MARSH,1 ELIZABETH A. EVERITT,1 WEILI WANG,1 JAMES J. FORT,2 STEVEN L. KRILL,3 YIHONG QIU1 1

Global Pharmaceutical Research and Development, Abbott Laboratories, R4P7, R1B, 1400 Sheridan Road, North Chicago, Illinois 60064-6254 2

Pharmaceutical Research and Development, Wyeth Healthcare, Richmond, Virginia 23261

3

Phamaceutical Research and Development, Boehringer Ingelheim Pharm, Ridgefield, Connecticut 06877

Received 14 February 2003; revised 11 August 2003; accepted 25 August 2003

ABSTRACT: Ritonavir is a large, lipophilic molecule that is practically insoluble in aqueous media and exhibits an exceedingly slow intrinsic dissolution rate. Although it has favorable lipophilicity, in vitro permeability studies have shown that ritonavir is a substrate of P-glycoprotein. Thus, the oral absorption of ritonavir could be limited by both dissolution and permeability, thereby making it a Class IV compound in the Biopharmaceutics Classification System. Because formulations rarely exert direct influence on local intestinal permeability, the effect of enhanced dissolution rate on oral absorption was explored. More specifically, poly(ethylene glycol) (PEG)–amorphous ritonavir solid dispersions were prepared with different drug loadings, and the in vitro and in vivo performances of the dispersions were evaluated. In vitro dissolution was conducted in 0.1N HCl with a USP Apparatus I. A crossover design was used to evaluate the oral bioavailability of amorphous dispersions relative to crystalline drug in beagle dogs. Intrinsic dissolution measurements of the two solid phases indicated a 10-fold improvement in intrinsic dissolution rate for amorphous ritonavir compared with the crystalline counterpart. In vitro dissolution of ritonavir depended on the solid phase as well as drug loading of the dispersion. In vivo study results indicate that amorphous solid dispersions containing 10–30% drug exhibited significant increases in area under the curve of concentration versus time (AUC) and maximum concentration (Cmax) over crystalline drug. For example, 10% amorphous dispersion exhibited increases of 22- and 13.7-fold in AUC and Cmax, respectively. However, both in vitro dissolution and bioavailability decreased with increasing drug load, which led to the construction of a multiple Level C in vitro– in vivo relationship for this Class IV compound. The established relationship between in vitro dissolution and in vivo absorption can help guide formulation development. ß 2004 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 93:563–570, 2004

Keywords: ritonavir; inhibition; cell lines; amorphous; solid dispersion; bioavailability; in vivo–in vitro correlation (IVIVC)

INTRODUCTION Over the last decade, advances in mechanistic understanding and scale-up technologies have Correspondence to: Yihong Qiu (Telephone: 847-938-5220; Fax: 847-935-1997; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 93, 563–570 (2004) ß 2004 Wiley-Liss, Inc. and the American Pharmacists Association

provided formulation scientists more options of developing solid dispersion formulations for poorly soluble drugs.1–3 These solid dispersion formulations can be broadly divided into two categories: (1) crystalline dispersions in which crystalline drug is dispersed in primarily crystalline polymers e.g., eutectics,4 and (2) amorphous dispersions in which amorphous drug is dispersed in crystalline5 or amorphous polymers.6 In the

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

563

564

LAW ET AL.

case of crystalline dispersions, the eutectic point determines the upper limit for drug loading.7 The limit of drug loading, however, is not as clear-cut with amorphous dispersions. In our previous work,8 ritonavir [5S-(5R*,8R*,10R*,11R*))10hydroxy-2-methyl-5-(1-methylethyl)-1-(2-(1-methylethyl)-4-thiazolyl)-3,6-dioxo-8,11-bis(phenylmethyl)-2,4,7,12-tetraazatridecan-13-oic acid, 5thiazolylmethyl ester], a human immunodeficiency virus (HIV) protease inhibitor indicated for the treatment of AIDS (Fig. 1), was used as a model compound for studying solid dispersions. Crystalline ritonavir is practically insoluble in aqueous media (400 mg/mL in 0.1N HCl, 1 mg/mL at pH 6.8, 378C), highly lipophilic (log D ¼ 4.3, pH 6.8), and exhibits an extremely slow intrinsic dissolution rate (0.03 mg/cm2-min in 0.1N HCl). A systematic study of the physicochemical properties of ritonavir and the influence of the carrier poly(ethylene glycol) (PEG) on ritonavir properties showed that PEG–ritonavir dispersions were two component systems containing amorphous ritonavir dispersed in crystalline PEG.8 These dispersions were physically stable for >2 years, and showed improved in vitro dissolution rates compared to that of crystalline ritonavir. The present studies were undertaken to (1) determine drug permeation characteristics, (2) evaluate the effect of drug load on in vitro dissolution rate and oral bioavailability in an animal model, and (3) explore the relationship between in vitro dissolution and in vivo performance.

EXPERIMENTAL Materials Ritonavir was obtained from Specialty Products Division of Abbott Laboratories, and radiolabeled ritonavir was obtained from the Department of Drug Metabolism, Abbott Laboratories. Poly(ethylene glycol) 8000 (PEG 8000; Carbowax,

Figure 1. Chemical structure of ritonavir. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

Union Carbide, Danbury, CT), absolute ethanol (McCormick Distilling Company, Weston, MO), and purified and deionized water (Millipore, Bedford, MA) were used in this study. Vinblastine and cyclosporin A were purchased from USP (Rockville, MD). Radiolabeled vinblastine was purchased from Amersham (Piscataway, NJ). Cell culture studies used Caco-2 cells obtained from ATCC (Rockville, MD), HTS TranswellTM 24well polycarbonate filter plates (Corning-Costar, Cambridge, MA.), fetal bovine serum (FBS; Hyclone Labs, Logan UT), Hank’s buffered saline solution (HBSS), Dulbecco’s Modified Eagle Medium (DMEM), penicillin/streptomycin, L-glutamine (Gibco, Grand Island, NY), and lucifer yellow (Sigma-Aldrich, St. Louis, MO) were used as received. All other solvents were HPLC grade, and the chemicals were of analytical reagent grade. High-Performance Liquid Chromatography (HPLC) The HPLC methods for dissolution rate determination8 and analysis of plasma samples9 have been described previously. Intrinsic Dissolution Rate (IDR) Amorphous or crystalline ritonavir (
65 mg) was compressed at 1000 lb in a hydraulic press (model C, Carver Laboratory Press, Menomonee Falls, WI) using a 30-s dwell time. The method and the apparatus have been described previously.10 The dissolution studies were performed in 0.1N HCl maintained at 378C. Samples were withdrawn at predetermined intervals, replaced with dissolution medium, and analyzed by HPLC. Permeability Studies Caco-2 cells were seeded onto 24-well HTS TranswellTM polycarbonate filter plates and maintained in DMEM þ10% FBS þ2 mM L-glutamine for 14–28 days. Plates were incubated at 378C in 5% CO2/95% air. For permeability studies, cell culture media were removed and replaced with HBSS at pH 7.4. Donor solutions were prepared by diluting stock solutions of drug (radiolabeled or unlabeled ritonavir) or model compounds (caffeine, inulin, cyclosporin A, and vinblastine) with pH 6.8 buffered HBSS and lucifer yellow. The receiving chamber contained pH 7.4 HBSS. Permeation was monitored for both the apical-tobasolateral (A-to-B) and the basolateral-to-apical (B-to-A) directions. In vitro P-glycoprotein (Pgp) transport/inhibition properties of ritonavir were

EVALUATION OF RITONAVIR–PEG 8000 SOLID DISPERSIONS

also investigated in a bidirectional manner using radiolabeled ritonavir along with vinblastine (a known substrate of Pgp) and cyclosporin A (an inhibitor of Pgp). Radioactive caffeine and inulin were used as control compounds. Samples from the receiving and donor chambers were analyzed by liquid scintillation counting (LSC). Lucifer yellow concentrations were determined with a fluorescence plate reader. Caco-2 membrane integrity was monitored after each experiment by determining the flux of lucifer yellow. A flux value of <0.25%/h indicated intact monolayer.11 Apparent permeability (Papp) was obtained by dividing the amount accumulated per time by the membrane surface area and the drug concentration in donor solution. Papp values were ranked according to Yee; 12 that is, low Papp (<1  106 cm/s), moderate Papp (¼1  10  106 cm/s), or high Papp (>10  106 cm/s) correspond to predictions of poor, moderate, and well-absorbed compounds, respectively.

565

In Vivo Study Eight beagle dogs weighing 10–15 kg were fasted overnight prior to dosing but were permitted free access to water. Food was returned at the end of the sampling interval (12 h after dosing). Approximately 30 min prior to drug administration, each dog received a 100-mg/kg subcutaneous (sc) dose of histamine. The reference formulation was crystalline ritonavir. The study utilized a crossover design to evaluate the three amorphous dispersions and the reference crystalline mixture. All formulations were administered as hard gelatin capsules at a dose of 5 mg/kg. A washout period of
1 week separated each of the four dosing periods. Sequential blood samples were obtained from each dog for 12 h after dosing. Plasma was separated by centrifugation and frozen at 208C until assayed by HPLC with low wavelength ultraviolet (UV) detection.9 The data from one dog were not included in the results because the dog experienced emesis after dosing.

Solid Dispersion Preparation The preparation and solid-state characterization of amorphous solid dispersions containing 10, 20, and 30% ritonavir in PEG have been described previously.8 Briefly, a solvent evaporation–fusion method was used such that on removal of alcohol at 758C, ritonavir precipitated as an amorphous phase. On subsequent cooling, the carrier PEG crystallized. The dispersions were then ground with a mortar and pestle, and sifted. Particles between 149 and 420 mm were filled into hard gelatin capsules and used for further study. Powder X-ray diffraction (PXRD), hot-stage microscopy, and differential scanning calorimetry (DSC) were used to characterize the dispersions.8 In Vitro Dissolution Tests The in vitro dissolution of amorphous ritonavir dispersions was compared with that of the crystalline drug. The crystalline drug, PEG, and the dispersions were sifted, and 149–420-mm particles were used to fill the capsules for dissolution studies. The tests were performed in 900 mL of 0.1N HCl at 37  0.58C using a USP apparatus I (basket method) at 50 rpm. Dissolution testing was conducted on an appropriate weight of dispersion to give 100 mg of ritonavir. Samples were withdrawn at predetermined intervals, filtered through a 0.45-mm syringe filter, and assayed by HPLC.

RESULTS AND DISCUSSION Amorphous Ritonavir and PEG–Ritonavir Dispersions Previous work has shown the apparent solubility of amorphous ritonavir to be 10-fold higher than that of the crystalline phase.8 Typically, such an increase in solubility would translate into an increased dissolution rate. However, when a metastable amorphous material is exposed to aqueous media, both dissolution and hydration may occur simultaneously. The hydrated phase would not only become plasticized and tend to crystallize but, in the case of ritonavir, it would also form a gel. Because the goal for preparing these amorphous dispersions is to improve the dissolution rate, a rapid crystallization rate may lead to limited utility. Under dry conditions, amorphous ritonavir has low mobility and large configurational entropy,13 both of which contribute to very slow crystallization rate. However, when exposed to moisture, the plasticized phase is expected to have faster crystallization kinetics. Furthermore, the extent of gelation would depend on the particle size of the hydrated species. Therefore, the impact of increased solubility on dissolution rate was investigated by maintaining a constant surface area. Results, shown in Figure 2, indicate a markedly improved intrinsic dissolution rate that parallels the observed enhancement JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

566

LAW ET AL.

ritonavir is a lipophilic molecule;8 thus, formulations with enhanced dissolution rate should exhibit improved oral absorption. Biopharmaceutics Characteristics of Ritonavir

Figure 2. Intrinsic dissolution rates of ritonavir in 0.1N HCl at 378C: (a) amorphous (0.3 mg/cm2-min, n ¼ 3) and (b) crystalline (0.03 mg/cm2-min, n ¼ 4).

in apparent solubility for amorphous ritonavir. The dissolution experiments were carried out for
30 min, and a linear relationship was observed for the amount dissolved as a function of time. This result suggests that the dissolution rate of the amorphous ritonavir is faster than that of crystalline ritonavir. Therefore, if surface area is similar, formulations containing amorphous ritonavir are expected to show at least 10-fold faster dissolution than formulations of crystalline drug, provided that the excipient does not have a detrimental effect. PEG–ritonavir solid dispersion formulations primarily consist of two phases.8 The PEG in the dispersions is primarily crystalline and ritonavir is amorphous. Because the hydrophilic carrier PEG dissolves very rapidly in aqueous media, the polymer should have minimal impact on the dissolution of amorphous ritonavir. Furthermore,

The physicochemical properties of ritonavir render it a poorly soluble drug with a favorable lipophilicity (log D) for passive transport. To further evaluate the permeation potential, in vitro permeability of ritonavir across Caco-2 cells was determined. The A-to-B transport data summarized in Table 1 indicate that ritonavir has moderate intestinal permeability. The high B-toA transport suggests a high efflux of ritonavir resulting from a net secretive transport across the Caco-2 cells. This observation is consistent with the findings from other laboratories14 and is in part due to the active transporter Pgp. This result is further confirmed using cyclosporin A, which is known to inhibit the efflux of Pgp substrates. The absorptive flux-to-secretory flux ratio for ritonavir was modulated toward unity with cyclosporin A.15 Vinblastine has been well-characterized as an avid substrate of the MDR1 glycoprotein. Over the past 5 years, in vitro evidence of the clinically relevant drug–drug interaction between vinblastine and ritonavir has been reported in the HIV and oncology literature.16–18 Data in Table 2 confirm that ritonavir is also a Pgp inhibitor that shows concentration-dependent inhibition of efflux of vinblastine. At a concentration of 10 mM, ritonavir was approximately as potent as cyclosporin A at inhibiting the efflux. Given that ritonavir is poorly soluble and moderately permeable, it merits designation in the Biopharmaceutics Classification System19 as a Class IV drug. Drugs in this class typically exhibit poor and variable absorption, not only because

Table 1. Papp Values in the Absorptive (A-to-B) and Secretory (B-to-A) Directions in Caco-2 Permeation Studies of Ritonavir Permeant [14C]Caffeine (50 mM) [3H]Inulin (50 mM) [14C]Ritonavir (5 mM) [14C]Ritonavir þ Cyclosporin A [14C]Ritonavir (10 mM) [14C]Ritonavir þ Cyclosporin A

ABa Papp (106 cm/s  SD)

BAb Papp (106 cm/s  SD)

48.84  0.38 0.21  0.07 8.95  0.87 18.95  0.85 9.50  1.02 16.61  0.88

NDc ND 19.15  0.70 15.82  2.06 17.44  1.70 15.08  1.28

a

AB, Apical-to-basal direction. BA, Basolateral-to-apical direction. ND, Not determined.

b c

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

EVALUATION OF RITONAVIR–PEG 8000 SOLID DISPERSIONS

567

Table 2. Papp Values in the Absorptive (A-to-B) and Secretory (B-to-A) Directions in Pgp Inhibition Studies of Ritonavir

Permeant

A-to-Ba Papp (106cm/s  SD)

B-to-Ab Papp (106 cm/s  SD)

BA/AB Papp Ratio

0.88  0.13 2.16  0.08 1.44  0.06 1.97  0.08

10.46  0.22 7. 28  0.15 7.84  0.74 6.28  0.72

11.9 3.4 5.4 3.2

[3H]Vinblastine (5 mM) [3H]Vinblastine þ Cyclosporin A (10 mM) [3H]Vinblastine þ Ritonavir (5 mM) [3H]Vinblastine þ Ritonavir (10 mM) a

AB, apical-to-basal direction. BA, basolateral-to-apical direction.

b

of dissolution rate/solubility and permeabilitylimited absorption, but also because of the various physicochemical, formulation, and in vivo variables that affect dissolution and permeation. A review of the literature reveals that these drugs often match the currently evolving structure– activity relationship profile for high-affinity ligands of Pgp20–22 These large molecules with high log P values have a propensity for engaging and being effluxed by the membrane-bound apical transporter Pgp on the human enterocyte and the human Caco-2 cell. These properties complicate the in vivo absorptive transport characteristics of ritonavir. However, because Pgp efflux is saturable, the maintenance of high luminal drug concentrations by increasing the solubility and/or dissolution rate is one approach for improving oral absorption of Class IV compounds like ritonavir.

is evident that amorphous ritonavir dispersions in PEG show a significantly improved dissolution rate. The total amount released depends on the drug load. To maximize the benefit obtained by incorporating the high-energy phase, knowledge that there is an apparent upper limit for drug loading is important. In Vivo Absorption The dog was used as an animal model for studying absorption of these immediate-release oral dosage forms because of similarities between dog and human gastrointestinal (GI) physiology.23 In general, Class IV compounds are often subject to significant food effect because food induces changes in the GI tract (e.g., mixing, bile flow, and splanchnic blood flow). As a result, minor differences in absorption between formulations may

In Vitro Dissolution of Amorphous Dispersions Dissolution profiles of amorphous solid dispersion formulations with varying drug loadings are compared with that of crystalline drug in Figure 3. The release rates from amorphous dispersions are significantly higher than that of the reference crystalline dispersion. These differences between the crystalline and the amorphous release rates are attributed to the difference in dissolution rate between the two phases (Fig. 2). The results in Figure 3 also indicate decreasing release rates with increasing drug loads. Amorphous ritonavir is phase separated from crystalline PEG in these dispersions;8 therefore, it is likely that the decreasing dissolution rate results from increased particle size of amorphous ritonavir with increased drug loading. Furthermore, it is noted that incomplete drug release becomes more pronounced with higher drug loading, which may be attributed to a gel-forming propensity of amorphous ritonavir discussed previously.8 Nevertheless, it

Figure 3. In vitro dissolution in 0.1N HCl of (a) physical mixture containing crystalline ritonavir–PEG at 10:90, and amorphous ritonavir in PEG solid dispersions at concentrations of (b) 10%, (c) 20%, and (d) 30% (w/w). Dissolution was determined by the USP I method (50 rpm, 378C). The data for 20% dispersion is an average of two runs; others are three runs. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

568

LAW ET AL.

Table 3. Bioavailability Parameters of Ritonavir Dispersions in Beagle Dogsa Formulation (drug loading) Cmax (mg/mL) Tmax (h) A (crystalline) B (10%) C (20%) D (30%) a

0.40  0.24 5.48  0.64 4.42  0.80 3.12  0.30

1.5  0.3 1.6  0.3 1.9  0.4 1.8  0.3

AUC0–1 (mg  hr/mL) 0.64  0.34 14.04  1.49 11.46  1.75 7.00  1.03

Results are expressed as mean  SD (n ¼ 7).

disappear on food intake. Therefore, the dosing regimen for the present study was undertaken with fasting dogs for increased differentiation. Dogs were pretreated with histamine to reduce the gastric pH and create an environment that more closely mimics human physiology.24,25 The mean plasma concentration–time curves following a single oral administration of each formulation are shown in Figure 4. Estimates of maximum concentration (Cmax), time to reach Cmax (Tmax), and area under the concentration– time curve (AUC) are provided in Table 3. The increases in average AUC values of 10, 20, and 30% (w/w) solid dispersions over the AUC value of the formulation containing crystalline ritonavir

were 22.04-, 17.98-, and 10.98-fold, respectively. The large difference between crystalline and amorphous ritonavir may be partly attributed to solubility-limited absorption with crystalline ritonavir. Ritonavir has a high dose/solubility ratio (50 mg/1 mg/L ¼ 50 L in dog), and the amount of water available for dissolution may be insufficient at any given dissolution site in vivo. Therefore, the dissolution time for the entire dose is probably longer than the residence time of the drug particles in the absorption region of upper GI tract, resulting in low bioavailability for crystalline drug. The situation for amorphous dispersion is different because of significantly increased apparent solubility. Because ritonavir is a substrate and inhibitor of Pgp, it is also possible that more rapid dissolution and hence the presence of higher local drug concentration in the intestinal lumen resulted in higher intestinal absorption. Consistent with the findings from in vitro tests, amorphous dispersions resulted in much higher bioavailability compared with crystalline drug as reflected by both AUC and Cmax values. The relationship between drug loading and oral absorption was in the same rank order as that observed in the in vitro tests. The results in Figures 3 and 4 indicate that use of <20% drug loading is more desirable for improving oral absorption. Finally, it should be pointed out that the coefficients of variation in both AUC and Cmax are significantly lower for amorphous dispersions than for crystalline drug. Relationship Between In Vitro and In Vivo Parameters

Figure 4. Mean plasma concentration–time profiles of ritonavir after a single oral dose to beagle dogs. Ritonavir was administered in (a) the crystalline form or the amorphous in PEG solid dispersions at concentrations of (b) 10%, (c) 20%, and (d) 30% (w/w). JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

Establishing a relationship between in vivo absorption and in vitro dissolution can be useful in justifying in vitro test conditions for product development.26 Because faster dissolution in vitro corresponded to more rapid absorption, the in vitro/in vivo relationship (IVIVR) of ritonavir was explored in more detail, even though ritonavir is a Class IV compound. Comparisons between percent released at 5 min (Q5min,) or at 60 min (Q60min) and Cmax (or AUC) values obtained from the in vivo study are shown in Figure 5 and Table 4. Linear relationships between the in vivo and in vitro parameters indicate a multiple Level C correlation,26 suggesting dissolution rate-limited absorption. The data for the crystalline drug were not included in the analysis because crystalline drug and amorphous dispersions are different systems. The results in Figure 5 show that

EVALUATION OF RITONAVIR–PEG 8000 SOLID DISPERSIONS

569

Table 4. Results of Linear Fitting of AUC or Cmax versus Q5min or Q60min for Ritonavir Amorphous Dispersions Parameter IVIVR for AUC IVIVR for Cmax

Q60min

Q5min 2

AUC ¼ 0.272 Q5min þ 4.047 R ¼ 0.9993 Cmax ¼ 0.090 Q5min þ 2.092 R2 ¼ 0.9955

crystalline drug does not follow the IVIVR of the amorphous system. In summary, the established relationships between in vitro and in vivo parameters indicate that fast and complete dissolution resulting from improved apparent solubility of the amorphous ritonavir is critical for enhanced oral absorption.

CONCLUSIONS Permeation studies indicate that ritonavir has moderate permeability and is a substrate for Pgp efflux. In vivo, it is possible that this Pgp efflux could limit the oral absorption or ritonavir, thereby making ritonavir a Class IV compound according to the Biopharmaceutics Classification System. The present study demonstrates that the oral bioavailability of ritonavir is improved significantly by preparation of amorphous solid dispersions. Whether the absorption of ritonavir is dissolution-limited or limited by poor permeability due to Pgp efflux, the improved bioavailability can be explained by the higher luminal concentrations of the drug when dosed as an

Figure 5. Relationships between AUC or Cmax versus Q5min or Q60min of ritonavir amorphous dispersions.

AUC ¼ 0.325 Q60min þ 18.566 R2 ¼ 0.9799 Cmax ¼ 0.109 Q60min þ 5.505 R2 ¼ 0.9977

amorphous dispersion. The extent of improvement depends on drug loading, with low drug loading giving greater improvement in dissolution and bioavailability. Lastly, despite its Class IV status, a preliminary multiple Level C in vitro– in vivo correlation was obtained, which can be used to guide formulation development.

ACKNOWLEDGMENTS The authors thank the Pharmaceutical Analysis and Stability Center of Abbott Laboratories for providing the HPLC method, and Leah Antonucci and Pamela Watson for skilled technical support with the Caco-2 cell assays.

REFERENCES 1. Breitenbach J. 2002. Melt extrusion: From process to drug delivery technology. Eur J Pharm Biopharm 54:107–117. 2. Law D, Schmitt EA, Qiu Y, Wang W, Fort JJ, Krill S. 1999. Amorphous and eutectic solid dispersion systems of model lipophilic compound, fenofibrate. AAPS Pharm Sci 1:S–525. 3. Serajuddin ATM. 1999. Solid dispersion of poorly water-soluble drugs: Early promises, subsequent problems, and recent breakthroughs. J Pharm Sci 88:1058–1066. 4. Lacoulonche F, Chauvet A, Masse J, Egea MA, Gracia ML. 1998. An investigation of FB interactions with poly(ethylene) glycol 6000, poly(ethylene) glycol 4000 and poly-e-caprolactone by thermoanalytical and spectroscopic methods and modeling. J Pharm Sci 87:543–551. 5. Zerrouk N, Chemtob C, Arnaud P, Toscani S, Dugue J. 2002. In vitro and in vivo evaluation of carbamazepine-PEG 6000 solid dispersions. Int J Pharm 225:49–62. 6. Ford JL, Rubinstein MH. 1978. Phase equlibria and dissolution rates of indomethacin-polyethylene glycol 6000 solid dispersions. Pharm Acta Helv 53: 93–98. 7. Law D, Wang W, Schmitt EA, Long MA. 2002. Prediction of poly(ethylene) glycol-drug eutectic compositions using an index based on the van’t Hoff equation. Pharm Res 19:315–321. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

570

LAW ET AL.

8. Law D, Krill SL, Schmitt EA, Fort JJ, Qiu Y, Wang W, Porter WR. 2001. Physicochemical considerations in the preparation of amorphous ritonavir/ PEG 8000 solid dispersions. J Pharm Sci 90(8): 1015–1025. 9. Marsh KC, Eiden E, McDonald E. 1997. Determination of ritonavir, a new HIV protease inhibitor, in biological samples using reversed-phase highperformance liquid chromatography. J Chromatography B 704:307–313. 10. Semonelli AP, Mehta SC, Higuchi WI. 1969. Dissolution rate of high energy polyvinylpyrrolidone (PVP)-sulfathiazole coprecipitates. J Pharm Sci 58:538–549. 11. Hilgers AR, Conradi RA, Burton PS. 1990. Caco-2 cell monolayers as a model for drug transport across the intestinal mucosa. Pharm Res 7:902–910. 12. Yee S. 1997. In vitro permeability across Caco-2 cells (colonic) can predict in vivo (small intestinal) absorption in man-fact or myth. Pharm Res 14: 763–766. 13. Zou D, Zhang GGZ, Law D, Grant DJW, Schmitt EA. 2002. Physical stability of amorphous pharmaceuticals: importance of configurational thermodynamic quantities and molecular mobility. J Pharm Sci 91:1863–1872. 14. Aungst BJ, Nguyen NH, Bulgarelli JP, Oates-Lenz K. 2000 The influence of donor and reservoir additives on Caco-2 permeability and secretory transport of HIV protease inhibitors and other lipophilic compounds. Pharm Res 17:1175–1180. 15. Mah SM, Rubas W. 1996. Vinblastine, a substrate for P-glycoprotein, flux across Caco-2 monolayers and rabbit jejunum. In: Proceedings of the 23rd International Symposium on Controlled Release of Bioactive Materials, pp 255–256. 16. Washington CB, Duran GE, Man MC, Sikic BI, Blaschke TR. 1998. Interaction of anti-HIV protease inhibitors with the multidrug transporter P-glycoprotein (P-gp) in human cultured cells. J Acquired Immune Deficiency Syndrome Human Retrovirol 19(3):203–209. 17. van der Spandt ICJ, van Fessem JMK, Voorwinden LH, Gaillard PJ, de Boer AG, Breimer DD. 2000.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 3, MARCH 2004

18.

19.

20.

21.

22.

23.

24.

25.

26.

Anti-microtubule drugs have disruptives effects on the in vitro blood-brain barrier after inhibition of P-glycoprotein and can be used to characterize P-glycoprotein modulators and substrates. Fundam Clin Pharmacol 14(1):57. Dupuis ML, Tombesi M, Cianfriglia M. 2002. Modulation of the multidrug resistance (MDR) phenotype in CEM MDR cells simultaneously exposed to anti HIV-1 protease inhibitor (PI’s) and cytotoxic drugs. Annali Dell’Instituto Superiore di Sanita 38(4):387–392. FDA Guidance for Industry ‘‘Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate-Release Solid Dosage Forms Based on a Biopharmaceutics Classification System’’; 8/2000. Gutmann H, Fricker GT, Michael MS, Beglinger C, Drewe J. 1999. Evidence for different ABC-transporters in Caco-2 cells modulating drug uptake. Pharm Res 16:402–407. Crowe A, Lemaire M. 1998. In vitro and in situ absorption of SDZ-RAD using a human intestinal cell line (Caco-2) and a single pass perfusion model in rats: Comparison with rapamycin. Pharm Res 15:1666–1672. Nerurkar MM, Burton PS, Borchardt RT. 1996. The use of surfactants to enhance the permeability of peptides through Caco-2 cells by inhibition of an apically polarized efflux system. Pharm Res 13: 528–534. Dressman JB, Yamada K. 1991. Animal models for oral drug absorption, in pharmaceutical bioequivalence. New York: Marcel Dekker, pp 235–266. Kahlson G, Rosengren E, Svahn D, Thunberg R. 1964. Mobilization and formation of histamine in the gastric mucosa as related to acid secretion. J Physiol 174:400–416. Akimoto M, Nagahata N, Furuya A, Fukushima K, Higuchi S, Suwa T. 2000. Comparison of canine and human gastrointestinal physiology. Eur J Pharm Biopharm 49:99–102. FDA Guidance for Industry ‘‘Extended Release Oral Solid Dosage Forms: Development, Evaluation and Application of In vitro/In vivo Correlations’’; 9/1997.