Lipid-based dispersions of exemestane for improved dissolution rate and intestinal permeability: in vitro and ex vivo characterization

Abstract

Current study aimed to develop lipid dispersions of poorly water soluble exemestane by employing lipid carriers such as Gelucire 44/14 and TPGS with porous calcium silicate (PCS) as an adsorbent carrier and formulate into a solid dosage form. The lipid dispersions at 1:5 ratio showed the highest solubility and dissolution compared to pure exemestane. Further, the ex vivo intestinal permeation studies showed improved apparent permeability (P_app, cm/s) of exemestane from the lipid dispersions (GLD_1:5 1.3 × 10⁻²cm/s; TLD_1:5 1.8 × 10⁻²cm/s) compared to pure exemestane (0.7 × 10⁻²cm/s). The optimized lipid dispersions (GLD_1:5 and TLD_1:5) were evaluated for scalability to develop into capsules.

Keywords:

• Dissolution
• exemestane
• Gelucire 44/14
• lipid dispersions
• permeability
• vitamin E TPGS

Introduction

Exemestane is a steroidal irreversible aromatase inhibitor used for the treatment of early or advanced breast cancer in postmenopausal women (Goss and von Eichel 2007, Kaufmann et al. 2000). It has a low oral bioavailability (only 42%), which is due to its poor water solubility (80 μg/mL) and low permeability (BCS class IV drug) (Löbenberg and Amidon 2000, Yavuz et al. 2007).

Over few decades, the development of lipid-based formulations and the use of surfactants in self-emulsifying drug delivery systems (SEDDS) and micellar dispersions to improve the solubility and oral bioavailability of poorly water soluble APIs were reported (Eedara et al. 2015, Inugala et al. 2015, Ramasahayam et al. 2015, Singh et al. 2008, Sunkavalli et al. 2016). However, the SEDDS in liquid dosage form suffers with drug leakage, low stability and excipient-capsule incompatibility issues. Thus, in recent past, there has been an amplified interest in the use of surfactants in the development of lipid-based solid dispersions for improved solubility and bioavailability of insoluble APIs (Eedara et al. 2014, Zhang et al. 2014).

Solid dispersion is one of the most assured approaches to improve the solubility and oral bioavailability of poorly water soluble APIs. The various mechanisms proposed for improved solubility are particle size reduction of the API, increase in the surface area (Graig 2002), improved wettability of the API (Taylor and Zografi 1997), and further no energy is required for crystal lattice breakage in solid dispersion for dissolution. In recent times, it has been reported that the solubility of poorly soluble APIs can be increased with an addition of lipid carriers with surfactant characteristics and/or self-emulsifying properties (Karata et al. 2005, Khoo et al. 2000, Serajuddin 1999, Vasanthavada and Serajuddin 2007). Thus, solid lipid dispersions formulated by utilizing amphiphilic surface active carriers have shown to achieve a greater degree of bioavailability for poorly soluble APIs and stabilize the solid dispersion by avoiding drug recrystallization.

The various surfactants reported to improve the solubility and oral bioavailability of poorly water soluble APIs include Gelucire^® 44/14, Gelucire^® 50/13, Inutec^® SP1, poloxamer 407 and vitamin E TPGS (D-α-tocopheryl polyethylene glycol succinate) (Eedara et al. 2014, Majerik et al. 2007, van den Mooter et al. 2006, Yuksel et al. 2003). In the present study, amphiphilic, self-emulsifying, non-ionic waxy lipids such as Gelucire 44/14 (HLB 14; melting point 44 °C) and TPGS (HLB 13; melting point 37–41 °C) were employed in the formulation of lipid dispersions. Gelucire 44/14 is a waxy solid lipid carrier composed of glycerides and long fatty acid esters of polyethylene glycol (PEG) 1500. TPGS is a water-soluble derivative of natural vitamin E, which has an amphiphilic property due to its lipophilic tail of phytyl chain of d-α-tocopherol and a hydrophilic head of PEG 1000. These amphiphilic surface active lipids are reported to improve the oral bioavailability of poorly water soluble drugs by enhancing the aqueous solubility as well as intestinal absorption (Dintaman and Silverman 1999, Kristina et al. 2007).

The aim of this study is to develop free-flowing lipid dispersions of exemestane using Gelucire 44/14 and TPGS as lipids and porous calcium silicate (PCS) as a surface adsorbent carrier. Phase solubility studies were conducted to determine the solubility of exemestane with the increasing amounts of carrier in aqueous solutions. Further, saturation solubility, dissolution, ex vivo intestinal permeation studies were performed for the prepared lipid dispersions to assess the dissolution rate and permeability of poorly soluble exemestane in the presence of the lipid carriers.

Materials and methods

Materials

Exemestane (EXM) was kindly provided by Dr. Reddy’s Laboratories (Hyderabad, India). Gelucire 44/14 and vitamin E TPGS (TPGS) samples were obtained as a generous gift sample from Gattefosse (Saint-Priest Cedex, France) and Isochem (Vert-le-Petit, France), respectively. Porous calcium silicate (Florite^TM RE; PCS) was purchased from Eisai Co., Ltd. (Tokyo, Japan). All other materials were of HPLC or analytical grade.

High performance liquid chromatography analysis

Exemestane content in the samples was analyzed using the high performance liquid chromatography (HPLC) system (LC-10 AT, isocratic, Shimadzu, Kyoto, Japan) equipped with UV spectrophotometric detector (SPD-10 AVP) and Lichrospher C₁₈ column (5 μm, 4.6 × 250 mm) maintained at 25 °C (Eedara et al. 2013). The mobile phase composed of double distilled water and methanol (20:80% v/v) was pumped at a flow rate of 1 mL/min. A 20 μL sample was manually injected and analyzed at a detection wavelength of 250 nm.

Phase solubility studies

Phase solubility studies were conducted by adding an excess quantity of exemestane to the glass vials containing 10 mL of the carrier (Gelucire 44/14 and TPGS) aqueous solutions at 0, 5, 10, 15, 20 and 25% w/v concentrations (Higuchi and Connors 1965). These vials were enclosed with a parafilm and shaken in a thermostatically controlled water bath for a period of 48 h at 37 °C. After equilibration for 6 h at room temperature, the samples were centrifuged at a speed of 10,000 rpm for 15 min. Supernatant samples were diluted with methanol, and drug solubility was measured by HPLC analysis. The phase solubility diagrams were plotted using the carrier concentration in aqueous solution versus the amount of exemestane solubilized in carrier solutions.

The drug-carrier complex apparent stability constant (K_s) was calculated using the following equation, where slope and intercept were obtained from the straight line equation of the phase solubility diagram (Higuchi and Connors 1965). (1)

The Gibbs free energy of transfer (ΔG°_tr) of exemestane from plain water without any carrier to the aqueous solutions of carriers showed enhanced drug solubilization with increasing amounts of carrier and was calculated using the following equation: (2) where R is the general gas constant (8.31 J K⁻¹mol⁻¹), T is the temperature (°K), S₀/S_S is the ratio of molar solubility of exemestane in aqueous solution of the carriers to solutions without carrier, respectively (Bandari et al. 2013).

Preparation of lipid dispersions

Lipid dispersions were prepared by the solvent evaporation method with exemestane and lipid carrier (Gelucire 44/14 and TPGS) in the weight ratios of 1:1, 1:3 and 1:5. Accurately weighed quantities of drug and lipid carrier were dissolved by adding 15 mL of ethanol/dichloromethane (1:1 ratio) mixture in a round bottom flask with manual gentle shaking for 5 min to get a clear solution. To this drug-carrier solution, PCS with the same quantity as the lipid carrier was added and dispersed manually by gentle shaking. The round bottom flask with drug carrier dispersion was attached to the rotary flash evaporator (Heidolph, Germany) and the organic solvent mixture was removed under vacuum at 35 °C with continuous rotation at a speed of 50 rpm. The solid powder was removed from round bottom flask and dried in a desiccator for a period of 24 h for complete removal of solvent traces. The obtained dry powder was passed through a 250 μm sieve with mesh size #60 and stored in air tight glass container at room temperature for further evaluation.

Preparation of physical mixtures

Blank lipid dispersions without drug were prepared as explained earlier by taking equal quantities of lipid carriers (Gelucire 44/14 and TPGS), and PCS using the same solvent composition. The physical mixtures in the ratio of optimized lipid dispersions were prepared by mixing the drug gently with accurately weighed quantities of blank lipid dispersions using a spatula for 10 min in a glass mortar. All prepared physical mixtures were passed through a sieve with mesh size #60 (250 μm) and stored in air tight glass container at room temperature until further evaluation.

Saturation solubility studies

All the lipid dispersions and physical mixtures prepared using Gelucire 44/14 and TPGS were added to the glass vials containing distilled water (10 mL) with vortex mixing in excess quantities. The sealed glass vials containing dispersed aqueous samples were shaken in a thermostatically controlled water bath for a period of 48 h at 37 °C. After equilibration at room temperature for 6 h, the samples were centrifuged at a speed of 10,000 rpm for 15 min. Supernatant samples were diluted using methanol and drug solubility was measured by HPLC analysis.

In vitro dissolution studies

Exemestane dissolution from pure exemestane, all the batches of lipid dispersions and physical mixtures (equivalent to 25 mg) was determined in distilled water as dissolution medium (900 mL) at 37 ± 0.5 °C using USP (XXIV) type II apparatus (Electrolab, Mumbai, India) at a rotational speed of 50 rpm. Cumulative amount of drug released into the dissolution medium was measured by collecting the samples (5 mL) at designated time intervals over a dissolution period of 120 min. Respective volumes of fresh distilled water maintained at 37 ± 0.5 °C was added immediately after every sampling point to maintain the dissolution medium volume constant. All the collected samples were filtered by passing through a membrane filter (0.45 μm) and filtrates were assayed for exemestane using HPLC after appropriate dilution using methanol.

Solid state characterization

Fourier transform infrared spectroscopy

Fourier transform infrared spectroscopy (FTIR) spectra of pure exemestane, Gelucire 44/14, TPGS, PCS, optimized lipid dispersions and their respective physical mixtures were assessed in the scanning range of 4000–500 cm⁻¹ by FTIR spectrophotometer (Spectrum GX-FT-IR, Perkin Elmer, Waltham, MA). Each powder sample (2 mg) was gently mixed with IR grade dry potassium bromide (200 mg). The prepared powder mixture was compressed into a pellet using a hydraulic press and scanned at a speed of 4 scans/s.

Differential scanning calorimetry

Thermal behavior of pure exemestane, Gelucire 44/14, TPGS, PCS, optimized lipid dispersions and their respective physical mixtures was investigated using a differential scanning calorimetry (DSC; TA-60WSI, Shimadzu, Kyoto, Japan) calibrated with indium. About 5 mg of the powder sample was placed into a flat bottomed standard aluminum pan (Shimadzu DSC-60, Kyoto, Japan) and press sealed using standard aluminum lid. All the samples were heated from 30 °C to 240 °C (10 °C/min) under a nitrogen gas flow of 50 mL/min.

X-ray powder diffraction

The crystallinity of the drug in the optimized lipid dispersions and their respective physical mixtures was determined by Bruker D8 advance X-ray diffractometer (D8 Advance, Bruker AXS Inc., Karlsruhe, Germany) with Cu-Kα radiation operated at a voltage of 40 kV and current of 35 mA. Each powder sample was loaded in the sample holder, gently pressed using a clean glass slide and scanned over an angular range of 3–50° (2θ).

Scanning electron microscopy

Surface morphology of the pure exemestane, Gelucire 44/14, TPGS, PCS, blank lipid dispersions, optimized lipid dispersions and their respective physical mixtures was assessed using scanning electron microscope (SEM; JEOL Ltd., Tokyo, Japan). Powder samples were sprinkled onto a double-sided carbon tape that was fixed to an aluminum holder and were sputter coated with platinum (3–5 nm) under vacuum.

Ex vivo intestinal permeation studies

Ex vivo intestinal permeation studies were performed with prior approval from the Institutional Animal Ethical Committee, St. Peter’s Institute of Pharmaceutical Sciences, Hanamkonda. All the animals (male Wistar rats) were habituated in isolated cages under controlled room conditions with free access to water and food. Before the experiment, the rats were fasted overnight with free access to water.

Ex vivo intestinal permeation studies were conducted in triplicate to determine the permeation characteristics of exemestane in the presence of carriers used in the preparation of lipid dispersions through the rat intestine from mucosal to serosal direction. Overnight fasted male Wistar rats (250–300 g) were sacrificed by cervical dislocation method after anesthesia (thiopental sodium; 50 mg/kg; intraperitonial bolus). Abdominal region of the rats was cut open by making about 3 cm midline longitudinal incision and immediately after that jejunum portion was gently excised from the small intestine region. From the excised jejunum portion, approximately 10 cm of the medial jejunum was selected and rinsed gently by passing 5 mL of saline several times using an oral gavage to remove intestinal contents. After completing the removal of intestinal contents, 2 mL samples of pure drug, lipid dispersions (equivalent to 5 mg) in phosphate-buffered saline (PBS; pH 7.4) were added into one end of the ligated jejunum segments and other end was also secured tightly with a silk thread. Each intestinal sac after verification for no leakage was placed in a glass beaker containing 50 mL of PBS (pH 7.4; 37 °C) and permeation study was conducted for a period of 120 min with continuous aeration (5–10 bubbles/min). Aliquot of serosal medium (2 mL) was collected at predetermined time intervals (0, 15, 30, 45, 60, 90 and 120 min) with equal volumes of fresh medium replacements to maintain the volume constant and analyzed for the amount of drug permeated using HPLC.

After completion of the study, length of the intestinal sac under study was measured to estimate the apparent permeability (P_app, cm/s) using ex vivo permeation data, i.e. cumulative amount permeated and time (Neupane et al. 2014).

Preparation and characterization of capsules

The optimized lipid dispersions were formulated into capsules (Table 1) by mixing accurately weighed quantities of dispersions (equivalent to 25 mg of exemestane) with microcrystalline cellulose (Avicel^® PH 202) and crospovidone for 10 min to obtain a homogenous mixture, followed by sieving through a 40-mesh screen. To this mixture, magnesium stearate and talc were added, and gently mixed for few minutes. Hard gelatin capsules (size ‘0’) were filled with the above powder blend equivalent to a single dose of exemestane.

Table 1. Composition of exemestane and lipid dispersion capsules.

	Quantity (mg) for each capsule
Ingredients	CEXM	CGLD(1:5)	CTLD(1:5)
Exemestane	25		–
GLD1:5	–	275	–
TLD1:5	–		275
Crospovidone	30	30	30
Magnesium stearate	8	8	8
Talc	6	6	6
Microcrystalline cellulose	331	81	81
Total	400	400	400

Flow properties of the optimized lipid dispersions and its respective capsule mixtures were assessed to investigate the feasibility and scalability aspects of these powders into solid dosage form, i.e. capsules. Angle of repose, which is a direct indication of material flowability, was determined by the conventional funnel method (Carr 1965).

In vitro dissolution of capsules

In vitro dissolution studies for the formulated capsules were performed in triplicate using distilled water as dissolution medium by maintaining all experimental conditions as described for dissolution studies of lipid dispersions over a period of 120 min. Dissolution samples (5 mL) collected at designated time intervals were filtered by passing through a membrane filter (0.45 μm) and filtrates obtained were assayed for exemestane using HPLC after appropriate dilution by methanol. All the in vitro dissolution parameters were calculated as per our earlier report (Gangishetty et al. 2015, Kallakunta et al. 2013).

Statistical analysis

Statistical assessment was performed using GraphPad prism 5 software (GraphPad Software, San Diego, CA) by the one-way analysis of variance (ANOVA). The significance of difference between formulations was calculated by Student–Newman–Keuls (compare all pairs) with P < 0.05.

Results and discussion

Phase solubility studies

The phase solubility of exemestane in distilled water and in the presence of lipid carriers (Gelucire 44/14 and TPGS) at different concentrations in distilled water is graphically presented in Figure 1 and all the computed solubility parameters are listed in Table 2. The aqueous solubility of exemestane in distilled water was found to be 0.08 ± 0.01 mg/mL signifying that it is a practically water insoluble drug and this value is consistent as stated in the literature (Singh et al. 2008, Yavuz et al. 2010).

Figure 1. Phase solubility diagram for exemestane in the presence of Gelucire 44/14 and TPGS in distilled water at 37 ± 0.5 °C (mean ± SD; n = 3).

Table 2. Phase solubility of the exemestane (mg/mL) at various concentrations of Gelucire 44/14 and TPGS.

	Gelucire 44/14		TPGS
Conc. of carrier (%w/v)	Conc. of exemestane	ΔG°tr (kJ/mol)	Conc. of exemestane	ΔG°tr (kJ/mol)
0	0.08 ± 0.01	–	0.08 ± 0.01
5	0.60 ± 0.03	−5.28 ± 0.12	0.87 ± 0.03	−6.24 ± 0.10
10	1.02 ± 0.03	−6.65 ± 0.08	1.43 ± 0.03	−7.53 ± 0.05
15	1.28 ± 0.04	−7.25 ± 0.08	1.92 ± 0.041	−8.29 ± 0.06
20	1.77 ± 0.05	−8.08 ± 0.07	2.40 ± 0.04	−8.86 ± 0.04
25	2.10 ± 0.05	−8.52 ± 0.06	3.18 ± 0.05	−9.59 ± 0.04
Ks	492.9		598.9
R^2	0.993		0.992
Slope	0.08		0.12
Type of curve	AL		AL

K_s, stability constant; ΔG°_tr, Gibbs free energy of transfer; A_L, linear type curve.

Among the various solubility parameters calculated, the linearity of phase solubility curve (i.e. solubility of exemestane versus concentration of the carrier) was considered to evaluate the solubility of exemestane with an increasing carrier concentration and to determine the maximum drug solubilization. Figure 1 shows A_L-type phase solubility diagrams with both the lipid carriers evaluated indicating the linear solubility of exemestane with increasing concentration of Gelucire 44/14 and TPGS in water, which may be due to the micellar solubilization nature of surfactant carriers (Kawakami et al. 2004, Sheu et al. 2003). This was further assessed using the coefficient of determination (R²) values obtained from the phase solubility curves. The R² values obtained from the phase solubility curves with both the carriers were greater than 0.99 indicating a positive influence of carriers on exemestane aqueous solubility. Further, these results were supported by the slopes of the linear solubility curves as the greater the slope value the higher the increase in aqueous solubility of the drug (Bandari et al. 2014).

The aqueous solubility of exemestane increased up to 3.18 mg/mL in the presence of TPGS and was found to be the highest, followed by 2.10 mg/mL in the presence of Gelucire 44/14 with a slope less than one suggesting A_L type solubility curves (Higuchi and Connors 1965). The calculated ΔG°_tr values for both of the carriers at varying amount were found to be negative, indicating exemestane solubility enhancement by the presence of the carriers and also the reaction is favorable with increasing amount of the carriers. This might be due to the improved wettability of the hydrophobic drug and a reduced interfacial tension between the drug and water by the presence of carriers. The ΔG°_tr values were more negative for TPGS compared to Gelucire 44/14 indicating greater solubilization of exemestane in the presence of TPGS. Further, the apparent stability constant (K_s) of exemestane with TPGS (598.9 mL/g) and Gelucire 44/14 (492.9 mL/g) were found to be in the ideal range of 100–1000 mL/g signifying the formation of stable complex with the lipid carriers evaluated. Phase solubility study results indicate that both of the carriers have strong solubility enhancement capacity for poorly water soluble exemestane.

Saturation solubility studies

The saturation solubility studies data reveal that solubility of exemestane increased from both the physical mixtures and the lipid dispersions prepared using Gelucire 44/14 and TPGS as lipid carriers (Figure 2). The extent of improved solubility of exemestane from 1:5 ratio of TPGS was 39.13-folds and its corresponding physical mixture was 11.4-fold. Similarly, the extent of improved solubility of exemestane in 1:5 ratio of Gelucire 44/14 was 13.9-fold and its respective physical mixture was 3.8-folds. This suggests the improved solubility of poorly soluble exemestane in Gelucire 44/14 and TPGS would be attributable to improved wettability, reduced surface free energy at the solid interface by the surfactant nature of the lipid carriers and self-emulsifying ability of the carriers (Eedara et al. 2014).

Figure 2. Saturation solubility of exemestane from pure exemestane, physical mixtures and lipid dispersions composed of Gelucire 44/14 and TPGS at various drug: carrier weight ratios (mean ± SD; n = 3).

In vitro dissolution studies of lipid dispersions

Figure 3 shows the exemestane dissolution from pure drug and lipid dispersions and physical mixtures in distilled water as a medium. The dissolution rate of the exemestane from pure exemestane was low due to its poor wetting property and low aqueous solubility. The high surface free energy leads to the predominating cohesive forces over adhesive forces between drug and water, which prevents the development of an interface. Whereas the dissolution rate of exemestane from the lipid dispersions was significantly higher and rapid as compared with pure drug.

Figure 3. In vitro dissolution profiles of pure exemestane, optimized lipid dispersions at 1:5 ratio and their respective physical mixture at 1:5 ratio composed of Gelucire 44/14 and TPGS as lipid carriers (mean ± SD; n = 3).

Tables 3 and 4 show the increased dissolution of exemestane in highest ratio of the lipid dispersions developed using Gelucire 44/14 and TPGS compared to its corresponding physical mixtures and pure exemestane. Percent drug release data at 15 min (Q₁₅) revealed that the drug dissolution was about 16.2-fold and 16.4-fold in lipid dispersions of Gelucire 44/14 and TPGS, respectively, where as its physical mixtures were 4.2 and 10.7 times more than pure exemestane. Furthermore, the initial dissolution rate (IDR) data also show the highest initial dissolution rate from GLD_1:5 and TLD_1:5 lipid dispersions compared with its respective physical mixtures and pure exemestane. The various other parameters like dissolution efficiency at 15 min and 60 min, i.e. DE₁₅, DE₆₀, mean dissolution time (MDT), mean dissolution rate (MDR), relative dissolution rate (RDR) also showed a similar trend of improved solubility and dissolution rate of exemestane with the highest ratio of both GLD_1:5 and TLD_1:5 compared with physical mixtures and pure exemestane.

Table 3. Summary of dissolution parameters for lipid dispersions and physical mixtures of exemestane prepared using Gelucire 44/14 (mean ± SD; n = 3).

	Dissolution parameters
Formulationcode	Q15	Q60	DE15	DE60	MDT	MDR	IDR	RDR	t50% (min)
EXM	4.8	26.86	2.4	13.4	39.36	0.24	0.32	–	>120
GPM1:5	20.3	36.4	10.1	24.2	28.82	0.54	1.35	1.27	>120
GLD1:1	33.4	52.43	16.7	36.9	20.66	0.85	2.22	1.67	50
GLD1:3	56.9	66.63	28.5	54.2	13.78	1.33	3.80	2.12	<15
GLD1:5	77.6	86.58	38.8	72.0	13.04	1.78	5.17	2.76	<15

Q₁₅ and Q₆₀: Percent of drug released at 15 and 60 min.

DE₁₅ and DE₆₀: Dissolution efficiency at 15 and 60 min.

MDT, MDR, IDR, RDR and t_50% (min): mean dissolution time, mean dissolution rate, initial dissolution rate, relative dissolution rate and time taken to release 50% of the drug, respectively.

Table 4. Summary of dissolution parameters for lipid dispersions and physical mixtures of exemestane prepared using TPGS (mean ± SD; n = 3).

	Dissolution parameters
Formulationcode	Q15	Q60	DE15	DE60	MDT	MDR	IDR	RDR	t50% (min)
EXM	4.8	26.9	2.4	13.4	39.4	0.2	0.3	–	>120
TPM1:5	51.2	65.5	25.6	51.4	16.4	1.2	3.4	2.1	<15
TLD1:1	58.7	68.8	29.3	55.5	19.6	1.4	3.9	2.4	<15
TLD1:3	70.7	77.1	35.3	65.0	18.4	1.6	4.7	2.6	<15
TLD1:5	78.7	91.5	39.4	74.5	15.8	1.8	5.3	3.0	<15

Q₁₅ and Q₆₀: Percent drug released at 15 and 60 min.

DE₁₅ and DE₆₀: Dissolution efficiency at 15 and 60 min.

MDT, MDR, IDR, RDR and t_50% (min): mean dissolution time, mean dissolution rate, initial dissolution rate, relative dissolution rate and time taken to release 50% of the drug, respectively.

These results suggest that the exemestane in lipid dispersions either transformed from crystalline state to amorphous form or present at molecular dissolved state was confirmed from XRPD and SEM analysis. However, the amphiphilic nature of lipids such as Gelucire 44/14 and TPGS (HLB 14 and 13) further confirmed improved solubility and dissolution of exemestane in dispersions, which was in accordance with our previous reports (Eedara et al. 2013). Furthermore, these carriers are reported to have self-emulsifying characteristic feature and the ability to form micelles, which contribute to the improved solubility by micellar solubilization (Kawakami et al. 2004, Sheu et al. 2003) and particle size reduction that would increase surface exposure of exemestane with dissolution medium.

Solid state characterization

Fourier transform infrared spectroscopy

Fourier transform infrared (FTIR) studies were performed to investigate the physicochemical interactions between drug, lipid carriers and PCS used in the lipid dispersions based on characteristic drug peaks absence or shifting. The FTIR spectra of pure exemestane, PCS, Gelucire 44/14, TPGS, physical mixtures (GPM_1:5 and TPM_1:5) and optimized lipid dispersions (GLD_1:5 and TLD_1:5) are shown in Figure 4. The FTIR spectrum of pure exemestane (Figure 4(A)) showed intense, well defined characteristic absorption peaks at 1730 cm⁻¹ (C = O stretching) and 1657 cm⁻¹ (C = C stretching). Figure 4(B) shows the FTIR spectrum of PCS without any sharp absorption peaks. The lipid excipients Gelucire 44/14 (Figure 4(C)) and TPGS (Figure 4(D)) showed principle absorption peaks at 1736 cm⁻¹ (C = O stretching) and 1102 cm⁻¹ (C-O stretching). The infrared spectrum of optimized lipid dispersions (GLD_1:5 and TLD_1:5; Figure 4(E) and (F)) and their respective physical mixtures (GPM_1:5 and TPM_1:5; Figure 4(G) and (H)) showed the characteristic peaks of exemestane and excipient with drug peaks at their respective positions, i.e. at 1730 cm⁻¹ and 1657 cm⁻¹. However, the C–H stretching peaks of both drug (934 cm⁻¹ and 597 cm⁻¹) and excipients (941 cm⁻¹ and 1435 cm⁻¹) are slightly shifted in the optimized lipid dispersions and their respective physical mixtures indicating physical interactions. Absence of any new extra peaks in the FTIR spectra of formulations indicates lack of chemical interaction.

Figure 4. Fourier transform infrared (FTIR) spectra of pure exemestane (A), porous calcium silicate (B), Gelucire 44/14 (C), TPGS (D), physical mixture at 1:5 ratio GPM_1:5 (E), TPM_1:5 (F), optimized lipid dispersions at 1:5 ratio GLD_1:5 (G) and TLD_1:5 (H).

Differential scanning calorimetry

The DSC thermograms (Figure 5) of the exemestane, PCS, Gelucire 44/14, TPGS, physical mixtures (GPM_1:5 and TPM_1:5) and optimized lipid dispersions (GLD_1:5 and TLD_1:5) allow to resolve physical state and thermotropic phase transition behavior of the drug in the formulations. The DSC thermogram of pure exemestane (Figure 5(A)) showed sharp endothermic phase transition peak at 195.7 °C, corresponding to its melting point with an enthalpy of 33.8 J/g, which indicates the presence of exemestane in crystalline form. Similarly, the endothermic peaks at 41.7 °C and 35.5 °C shown by DSC thermograms of Gelucire 44/14 (Figure 5(C)) and TPGS (Figure 5(D)) indicate the melting points of respective lipids. However, PCS (Figure 5(B)) exhibited no observable endothermic peak over the entire range of temperature studied. The endothermic peaks of lipid at respective melting point without representative peak of exemestane was observed for lipid dispersions (Figure 5(G) and (H)), suggesting the transformation of crystalline drug to amorphous state or dissolution of drug in the molten lipids with low melting points. Ability of the lipid excipients in maintaining the drug in dissolved state and inhibiting recrystallization of the drug was further confirmed by X-ray diffraction studies. However, the respective physical mixtures (Figure 5(E) and (F)) showed melting endothermic peaks of lipids with drug peak at around 190 °C. The degree of crystallinity of drug in the physical mixtures was found to be less than 5%, indicating that only a very small amount of the drug is present in its crystalline form, which may be due to the dissolution of crystalline drug in the molten lipid carriers.

Figure 5. Differential scanning calorimetry (DSC) thermograms of pure exemestane (A), porous calcium silicate (B), Gelucire 44/14 (C), TPGS (D), physical mixture at 1:5 ratio GPM_1:5 (E) , TPM_1:5 (F), optimized lipid dispersions at 1:5 ratio GLD_1:5 (G) and TLD_1:5 (H).

X-ray powder diffraction studies

The physical state (i.e. crystallinity) of the exemestane in optimized lipid dispersions (GLD_1:5 and TLD_1:5) and their respective physical mixtures (GPM_1:5 and TPM_1:5) was investigated by powder X-ray powder diffraction studies. Figure 6 shows typical XRPD spectra of pure exemestane, PCS, blank lipid dispersions, optimized lipid dispersions and respective physical mixtures ranging from 3° to 50° (2ϴ). As anticipated, the pure exemestane (Figure 6(A)) showed sharp peaks due to its characteristic crystalline structure at diffraction angles of 8.2, 5.5 and 4.8 A°. Whereas the spectra of optimized lipid dispersions (Figure 6(G) and (H)) are identical to the blank lipid dispersions (Figure 6(C) and (D)) without characteristic exemestane peaks demonstrating the absence of crystalline drug in the prepared lipid dispersions. Thus, the crystalline state of exemestane was changed to amorphous state or molecularly dissolved state in the lipid dispersions prepared with Gelucire 44/14 and TPGS. However, the XRPD spectra of physical mixtures (Figure 6(E) and (F)) showed characteristic drug peaks at respective positions with decreased intensity compared to pure drug. The decreased intensity of drug peaks in the physical mixtures may be due to dilution of drug by the presence of other carriers.

Figure 6. X-ray powder diffractograms (XRPD) of pure exemestane (A), porous calcium silicate (B), blank lipid dispersions of Gelucire 44/14 (C), TPGS (D), physical mixture at 1:5 ratio GPM_1:5 (E), TPM_1:5 (F), optimized lipid dispersions at 1:5 ratio GLD_1:5 (G) and TLD_1:5 (H).

Scanning electron microscopy

Figure 7 shows surface morphology of pure exemestane, Gelucire 44/14, TPGS, PCS, blank lipid dispersions, optimized lipid dispersions (GLD_1:5 and TLD_1:5) and respective physical mixtures (GPM_1:5 and TPM_1:5) as examined by scanning electron microscope. Pure exemestane (Figure 7(A)) was characterized by crystals of irregular size and shape. Figures 7(B) and (C) show waxy lipid nature of the carriers such as Gelucire 44/14 and TPGS, respectively. Figure 7(D) shows porous rough surfaced powder particles of PCS made of calcium silicate flakes arranged irregularly like petals (Ozeki et al. 2011, Yuasa et al. 1994). From SEM images of blank lipid dispersions (Figure 7(E) and (F)), it is evident that porous surface of the PCS was covered by lipid matrix of Gelucire 44/14 and TPGS, respectively. Whereas physical mixture shows (Figure 7(G) and (H)) the exemestane crystals adhered to the surface of the PCS surface coated with lipid matrix. The optimized lipid dispersions (Figure 7(I) and (J)) show PCS particles surface concealed with lipid matrix without any visible exemestane drug crystals, which confirms the presence of drug in dissolved form in the lipid matrix or transformation of crystalline drug to amorphous form. Furthermore, from SEM images it is apparent that Gelucire 44/14 and TPGS were having good solubilizing capacity for exemestane without any recrystallization. The adsorption of lipids containing dissolved drug over the porous carrier with greater surface area has improved the dissolution rate of poorly water soluble drug, i.e. exemestane.

Figure 7. Scanning electron microscopic (SEM) images of pure exemestane (A), Gelucire 44/14 (B), TPGS (C), porous calcium silicate (D), blank lipid dispersions of Gelucire 44/14 (E), TPGS (F), physical mixture at 1:5 ratio GPM_1:5 (G), TPM_1:5 (H) and optimized dispersions at 1:5 ratio, GLD_1:5 (I) and TLD_1:5 (J).

Ex vivo permeation study

In the recent drug development process, the estimation of pharmacokinetic parameters of a drug utilizing its in vitro dissolution and absorption data is recognized as a fundamental approach. The bioavailability of an orally administered drug depends on its dissolution behavior as well as its permeability through the gastro intestinal (GI) membranes. The oral bioavailability of a poorly water soluble drug is improved by enhancing its dissolution behavior. However, the additives used in the formulation to improve the dissolution rate also show some influence on the absorption through the GI membranes (Fenyvesi et al. 2011, Frank et al. 2014). Therefore, we conducted ex vivo intestinal permeation studies to determine the permeation characteristics of exemestane in the presence of carriers used in the preparation of lipid dispersions through the rat intestine from mucosal to serosal direction.

The P_app values for the pure exemestane and lipid dispersions were found to be 0.7 × 10⁻²cm/s, 1.3 × 10⁻²cm/s (GLD_1:5) and 1.8 × 10⁻²cm/s (TLD_1:5), respectively. The improved permeation may be due to the enhanced dissolution by the presence of amorphous form of poorly water soluble drug in the lipid dispersions and permeation enhancing property of the lipids used in the production of dispersions (Dintaman and Silverman 1999, Fernandez et al. 2008, Yu et al. 1999). Moreover, the amphiphilic lipid carriers in the lipid dispersions could enormously increase the contact area with the GI tract, which results in the improved absorption characteristic of the drug (Bikiaris et al. 2005, Odon et al. 2011). In our previous study, we evaluated the influence of Gelucire 44/14 and TPGS on the enhancement of intestinal absorption and bioavailability of fexofenadine hydrochloride (Eedara et al. 2013). Improved dissolution and permeation through the intestinal membranes may lead to the improved oral bioavailability of exemestane. However, further evaluation of in vivo pharmacokinetic behavior of the developed lipid dispersions is necessary to determine their oral bioavailability.

Feasibility to formulate lipid dispersions into capsules/in vitro dissolution of capsules

To assess the practicability and scalability of developed lipid dispersions into solid dosage forms, the optimized lipid dispersions (GLD_1:5 and TLD_1:5) were formulated into capsule dosage forms. The micromeritic characteristics, such as angle of repose for lipid dispersions (GLD_(1:5) 44.0 ± 1.6, TLD_(1:5) 43.0 ± 1.3) and capsule formulations (C_GLD(1:5) 32.0 ± 2.1, C_TLD(1:5) 28.0 ± 1.9) revealed that micromeritics were improved for capsule formulations because of the presence of glidants and lubricants within the capsule formulations. Finally, the dissolution of exemestane from capsule dosage form evaluated by in vitro dissolution studies (Figure 8) revealed that drug release of exemestane from C_GLD(1:5) and C_TLD(1:5) were higher than pure exemestane capsules (C_EXM). At the end of dissolution study, the dissolution efficiency values (DE₆₀) were found to be 27.4 for C_EXM, and 68.9 and 76.0 for C_GLD(1:5) and C_TLD(1:5) capsules, respectively. This dissolution efficiency along with other dissolution parameters, such as Q_60, MDT, IDR, RDR, MDR and t_50% (Table 5) revealed improved dissolution from the capsules dosage forms developed using lipid dispersions. Nevertheless the dissolution behavior of exemestane from lipid dispersions was similar when compared with its respective capsule dosage forms. The dissolution parameters like MDT, MDR, IDR and t_50% further confirm the similar dissolution profiles of lipid dispersions and capsules. Finally, the dissimilarity factor (f₁) value and similarity factor (f₂) value calculated between GLD_(1:5), TLD_(1:5) and their corresponding capsule formulations were observed to be 3.6, 1.9 and 74.1, 77.9, respectively. Thus, the f₁ and f₂ values are within the limits, confirming the similar dissolution profile between lipid dispersions and capsules. The dissolution data along with the micromeritic characteristic features confirm the feasibility of development of lipid dispersions into solid dosage forms.

Figure 8. In vitro dissolution profiles of capsules composed of pure exemestane, optimized lipid dispersions at 1:5 ratio (mean ± SD; n = 3).

Table 5. Summary of dissolution parameters for exemestane capsules prepared using lipid dispersions (mean ± SD; n = 3).

	Dissolution parameters
Capsules	Q60	DE60	MDT	MDR	IDR	RDR	t50% (min)	f1	f2
CEXM	38.9	27.4	31.6	0.6	1.4	–	>120	–	–
CGLD(1:5)	83.1	68.9	14.3	1.7	5.0	1.8	<15	3.6	74.1
CTLD(1:5)	90.4	76.0	15.7	1.9	5.6	2.0	<15	1.9	77.9

Q₆₀: percent of drug released at 60 min.

DE₆₀: Dissolution efficiency at 60 min.

MDT, MDR, IDR, RDR, t_50% (min), f₁ and f₂: mean dissolution time, mean dissolution rate, initial dissolution rate, relative dissolution rate, time taken to release 50% of the drug, dissimilarity factor and similarity factor, respectively.

Conclusions

The lipid dispersions of exemestane with Gelucire 44/14 and vitamin E TPGS were formulated and further developed into capsule dosage form. Improved solubility, dissolution rate and intestinal permeability of exemestane were observed in the presence of the carriers evaluated compared to pure drug alone. However, the lipid dispersions composed of TPGS exhibited significantly higher dissolution rates and intestinal permeation than the lipid dispersions of Gelucire 44/14. The FTIR studies of both carriers revealed no chemical interaction within the carriers and exemestane. Further, DSC, XRPD and SEM studies of lipid dispersions revealed the loss of crystalline nature of exemestane in the presence of Gelucire 44/14 and TPGS. The P_app values for the pure exemestane and lipid dispersions were found to be 0.7 × 10⁻²cm/s, 1.3 × 10⁻²cm/s (GLD_1:5) and 1.8 × 10⁻²cm/s (TLD_1:5), respectively, indicating improved permeation of drug from the lipid dispersions composed of Gelucire 44/14 and TPGS compared to exemestane alone. The micromeritic characteristics and in vitro dissolution of capsule formulation studies showed the feasibility of lipid dispersions into capsule dosage form with similar dissolution profile between lipid formulations and developed capsules.

Funding information

The corresponding author is thankful for financial support from Science & Engineering Research Board (File No. SB/YS/LS-119/2013), Department of Science and Technology, New Delhi, India.

Acknowledgements

The authors are grateful to Dr. Reddy’s Laboratories, Hyderabad, India, for the gift sample of exemestane, Gattefosse and Isochem France for providing lipid carriers Gelucire 44/14 and vitamin E TPGS. The authors also thank Mr. T. Jayapal Reddy, Chairman, St. Peter’s Institute of Pharmaceutical Sciences, Hanamkonda for providing the necessary facilities.

Disclosure statement

The authors report no conflicts of interest.