Assessment of in Vivo Performance of Lipid-Based Formulations: Correlation between in Vitro Drug Release Profiles and in Vivo Absorption Rate Profiles

Abstract

The purpose of this study was to assess the in vivo absorption enhancement effects of lipid-based formulations (LBFs) through in vitro release studies. The type IIIA-MC (medium-chain) and type IIIA-LC (long-chain) formulations containing a Biopharmaceutics Classification System (BCS) Class II drug (dipyridamole or ketoconazole) were used as model LBFs. The type IIIA-MC formulation, but not the type IIIA-LC formulation, showed a higher initial absorption rate than the control suspension for both model drugs in rats. An in vitro side-by-side chamber system coupled with a lipid digestion model was used to measure free drugs, available for intestinal absorption, that are released from a model LBF. The profiles of free drug concentration on the donor side were determined by calculating the ratio of permeation rate (LBF/suspension) at every sampling interval. The in vitro free drug concentration was immediately supersaturated when the digestion of type IIIA-MC formulation was initiated for both drugs, which would cause the initially high absorption rate in rats. In contrast, the free concentration of the type IIIA-LC formulation became lower than the equilibrium solubility over time for both drugs. Overall, the profiles of in vitro free concentrations were consistent with those of in vivo absorption rates for both drugs and all LBFs. These findings would help predict the in vivo performance and establish an in vitro–in vivo correlation (IVIVC) of LBFs.

INTRODUCTION

Lipid-based formulations (LBFs) can improve the fraction of dose absorbed (F_a) of poorly water-soluble drugs by maintaining the solubilized state in the gastrointestinal (GI) tract.^1,2) LBFs are categorized into four classes following the composition and proportion of materials (lipid, surfactant, co-solvent, and co-surfactant).³⁾ The composition of the type I formulation is 100% lipids, on the other hand, that of the type IV is 100% surfactant/co-solvent/co-surfactant (0% lipids). The type II and III formulations are comprised of lipids and surfactants/co-solvents/co-surfactants, which are called self-(micro)emulsifying drug delivery systems (SEDDS or SMEDDS). The type II formulations contain low hydrophile-lipophile balance (HLB) surfactants. On the other hand, the type III formulations contain high HLB surfactants, which strongly affect the particle size of lipid droplets in an aqueous phase.

After the dispersion of an LBF in an aqueous buffer (GI fluid), the drug equilibrium between lipid droplets (solubilized concentration, C_sol) and bulk water (free concentration, C_free) is established by its partition coefficient.⁴⁾ Once the free drug is absorbed from the small intestine, the drug equilibration will be re-established, thereby resulting in quick drug absorption when compared with the solid form. In the meantime, the exogenous triglyceride, which is orally administered with a drug, is digested into diglycerides, monoglycerides, and fatty acids by pancreatic lipase.^2,4–7) These digestion products further interact with endogenous bile acids in the GI fluids.⁸⁾ This complexity of relation among lipid dispersion/digestion, drug (re-)equilibrium, and drug absorption makes it difficult to establish an in vitro–in vivo correlation (IVIVC) of LBFs.^2,9–12)

In order to establish an IVIVC of LBFs, Crum et al. have recently developed a new in vitro digestion model coupled with in situ single-pass perfusion to evaluate the mechanism of drug absorption from LBFs.⁴⁾ An in vitro digestion model has been developed and is commonly used to screen LBFs.^2,13–17) Using this in vitro–in situ model, they simultaneously monitored the dissolved (mostly solubilized) concentration during the digestion, and the flux into the mesenteric blood from type IIIA, type IIIB, and type IV fenofibrate formulations, quantified using a parameter called the supersaturation ratio (SR). The SR was determined by dividing the area under the dissolved concentration–time curve (AUC_solubilized) of LBF by the AUC_solubilized of the active pharmaceutical ingredient (API) that was added to the buffer containing drug-free formulation during digestion. The lipid-rich type IIIA and type IIIB formulations maintained a high solubilization capacity, giving a maximum SR of approximately 2 and 3, respectively. In contrast, the dissolved concentration of the type IV formulation, which comprised a hydrophilic surfactant and a co-solvent, rapidly decreased during dispersion due to high miscibility with the aqueous buffer, resulting in an SR below 1.0 (non-supersaturated). However, there was no significant difference in flux into the mesenteric blood from three formulations in rats. Anby et al. have also reported that lipid digestion acted as a trigger for drug supersaturation.⁷⁾ The dissolved concentration in an aqueous phase with or without a polymeric precipitation inhibitor, hydroxypropyl methylcellulose (HPMC) was measured during the digestion of two type IIIA (40 and 80% drug loaded) and a type IV (40% drug loaded) danazol formulations.⁷⁾ The dissolved concentration of danazol was supersaturated during the digestion of these LBFs, and the resulting maximum SR values were significantly increased in the presence of HPMC for three LBFs. However, the effect of HPMC to stabilize the supersaturation, resulting in enhanced oral absorption, was not observed in dog studies. The dissolved concentration that has been measured includes the C_sol in the lipids/digested products and the C_free in the bulk water, available for intestinal absorption. Because the C_sol is not a driving force for membrane permeation, it would not be able to predict the performance of LBFs. Many studies have also failed to establish IVIVCs of LBFs using the dissolved (mostly solubilized) concentrations.^{2,4,5,9–12)} Thus, understanding the change in C_free rather than the C_sol during the digestion of LBFs has recently received attention.^13,18,19) The previous study showed a good correlation between in vitro flux values (C_free profile) obtained from Franz cell chambers and in vivo oral AUC values in rats.¹⁸⁾ Therefore, in this study, we focused on determining the real-time change in the C_free released from LBFs using a side-by-side chamber system coupled with a digestion model as shown in Fig. 1.

Fig. 1. Experimental Setup for the in Vitro Side-by-Side Chamber System Coupled with a Lipid Digestion Model

Each formulation was added on the donor side filled with 7.0 mL of TM. Five minutes after applying the formulation, 1.0 mL of pancreatic lipase solution was added to the donor side to initiate lipid digestion. Samples (0.1 mL) were taken from the receiver side to determine the free concentration-time profile by calculating the ratio of permeation rates (LBF/suspension) at each sampling interval. During the experiment, the pH on the donor side was maintained at 6.5 using a pH-stat titration unit, which automatically neutralized the liberated fatty acids by adding 2.0 M NaOH.

The purpose of this study was to assess the in vivo absorption enhancement effects of LBFs through in vitro release studies. As model drugs, two Biopharmaceutics Classification System (BCS) Class II weakly basic drugs, dipyridamole and ketoconazole were selected because these drugs previously showed different supersaturated dissolution profiles.^20,21) The extent and the rate of absorption profile of the model drug formulated in the type IIIA-MC (medium-chain lipids) or the type IIIA-LC (long-chain lipids) formulation was first observed in rats. Then, the C_free profiles of both model drugs were determined in the in vitro release studies and compared with in vivo absorption rate profiles. The current data showed not only the simple rank order of LBFs but also the correlation between in vitro C_free profiles and the in vivo absorption rate profiles. This indicates that the release pattern of C_free from LBFs in the GI tract would be predicted using the current in vitro system.

MATERIALS AND METHODS

Materials

Dipyridamole, polyethylene glycol 400 (PEG 400), sodium chloride, glucose, N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES), sodium hydrogen carbonate (NaHCO₃), corn oil, hydrochloric acid (HCl), sodium hydroxide (NaOH), methanol (LC-MS grade), and acetonitrile (LC-MS grade) were procured from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). Ketoconazole was procured from LKT Laboratories, Inc. (St. Paul, MN, U.S.A.). COCONAD RK was obtained from Kao Chemicals (Tokyo, Japan). Capmul MCM was obtained from ABITEC Corp. (Janesville, WI, U.S.A.). Maisine 35-1 was supplied from Gattefossé (Saint-Priest, France). Cremophor EL, porcine pancreatin (8 x USP specifications), and 1-aminobenzotriazole (ABT) were procured from Sigma-Aldrich (St. Louis, MO, U.S.A.). Hanks’ balanced salt solution 10X (HBSS 10X) was obtained from Gibco Laboratories (Lenexa, KS, U.S.A.). Dialysis membrane (Spectra/Por^®, regenerated cellulose, MW cutoff = 1 kD) was procured from Spectrum Laboratories, Inc. (Rancho Dominguez, CA, U.S.A.). Water (18.2 MΩ·cm) was obtained from an in-house system (Simplicity UV, Merck) (Tokyo, Japan). All other reagents were of the highest purity.

Preparation of LBFs

The components of type IIIA-MC and IIIA-LC formulations are listed in Table 1.^18,21) The model drug was applied to each formulation at saturation solubility. Briefly, an excess amount of model drug was applied to each formulation and stirred with a magnetic stirrer at 200 rpm for at least 96 h in an incubator at 37 °C. After the sample was centrifuged at 20000 × g for 10 min at 37 °C, several drops of the supernatant were weighed in a 20 mL volumetric flask and dissolved in methanol to determine the saturation solubility in the formulation.

Table 1. Composition of LBFs and the Saturation Solubility of the Model Drugs in Each Formulation

	Composition		Proportion (w/w%)	Particle size (nm)	Saturation solubility (mg/g)
					Dipyridamole	Ketoconazole
IIIA-LCa)	Corn oil	Lipid (long-chain triglyceride)	32.5	45 ± 0.6	3.5 ± 0.3b)	10.8 ± 0.8b)
	Maisine 35-1	Lipid (blend of C16–20 mono- and di-glyceride)	32.5
	Cremophor EL	Surfactant (hydrophilic)	35.0
IIIA-MCa)	COCONAD RK	Lipid (medium-chain C8 triglyceride)	32.5	188 ± 4.6	4.8 ± 0.7b)	24.6 ± 1.7b)
	Capmul MCM C8	Lipid (blend of C8 mono- and di-glyceride)	32.5
	Cremophor EL	Surfactant (hydrophilic)	35.0

a) Ref. 19, 22). b) Ave ± standard deviation (S.D.) (n = 3–4).

In Vivo PK Study in Rats

All pharmacokinetic (PK) studies were approved by the Ethical Review Committee of Setsunan University. Sprague–Dawley rats (male, 8 weeks old, 250 ± 10 g) were purchased from Japan SLC, Inc. (Shizuoka, Japan). The rats were housed in a room maintained at a temperature of 25 °C and relative humidity of 55% with 12-h light/dark cycles and given free access to a commercial rat diet with tap water.²¹⁾ Rats were fasted overnight before starting the experiments. To observe the in vivo performance of LBFs without the confounding variables (pH shift-mediated supersaturation and first-pass metabolism), rats were pre-treated with omeprazole (for both drugs) and ABT (for ketoconazole only). Omeprazole was orally administered at 30 mg/kg 1 h before dosing model drugs.²³⁾ ABT was orally administered at 100 mg/kg 18 h before dosing ketoconazole.^24–28)

Oral Administration

The model drug formulated in each LBF was orally administered to rats at 1.0 mg/kg for dipyridamole or 4.0 mg/kg for ketoconazole. Water was immediately administered at 2.0 mL/kg to rinse the dosed LBFs. The dose concentration was set at 0.5 mg/mL for dipyridamole and 2.0 mg/mL for ketoconazole. For the control, the drug suspension in water was orally administered to rats. The area under the plasma concentration-time curve (AUC) was calculated using the trapezoid method. The mean C_max and T_max were calculated from the observed values.

Intravenous Administration

Each drug was dissolved in a saline/PEG400 (70 : 30) solution and injected into the jugular vein of rats with a syringe needle at 0.2 mg/kg. To calculate the bioavailability (BA) and absorption rate profile, the plasma concentration profile after intravenous injection was deconvoluted with the profile after oral administration.

Sample Collection, Process, and Analysis

Blood (0.25 mL) was routinely collected from a jugular vein with a heparinized syringe up to 8.0 h after oral administration, and 4.0 h after intravenous injection. The blood samples were immediately centrifuged at 5500 × g for 10 min at 4 °C to collect the plasma samples. The plasma sample (0.1 mL) was mixed with 0.9 mL acetonitrile and the mixture was centrifuged at 20000 × g for 20 min at 20 °C. The supernatant (0.8 mL) was dried and reconstituted with 0.1 mL of water/acetonitrile (50 : 50) containing 0.1% formic acid. The drug concentration was quantified by the previously established ultra performance liquid chromatography (UPLC)-MS/MS method.^21,29)

In Vitro Release Study

Preparation of Pancreatic Lipase Solution

Porcine pancreatin (0.3 g) was added to a microcentrifuge tube filled with 1.5 mL of pH 6.5 transport medium (TM)³⁰⁾ and vortexed well to extract pancreatic lipase. Approximately 5 µL of 5.0 M NaOH solution was added to the extract suspension to adjust the pH to 6.5. After the extract suspension was centrifuged at 20000 × g for 5 min at 37 °C, the supernatant containing 1000 tributyrin units/mL was collected and used as pancreatic lipase solution.^2,4,5)

In Vitro Assay

As illustrated in Fig. 1, a side-by-side chamber system mounted with a dialysis membrane coupled with a lipid digestion model was used to determine the C_free profile during digestion of the type IIIA-MC and IIIA-LC formulations. The dialysis membrane was mounted onto the interface between the donor and receiver chambers. The donor and receiver chambers were filled with 7.0 and 5.5 mL TM. Both sides were stirred continuously with magnetic stirrers at 800 rpm to minimize the effect of the unstirred water layer on membrane permeation.³¹⁾ Each formulation was added to the donor chamber at a final concentration of 0.5 mg/mL for dipyridamole or 2.0 mg/mL for ketoconazole. Five minutes after applying the formulation, 1.0 mL of pancreatic lipase solution was added to the donor chamber to initiate lipid digestion. The pH on the donor side was maintained at 6.5 using a pH-stat titration unit (Hiranuma Automatic Titrator COM-1700) (HIRANUMA, Ibaragi, Japan). Samples (0.1 mL) were collected from the receiver side at 5, 15, 30, 60, and 90 min. As a control, the drug suspension was added to the donor side. The drug suspension was prepared by adding the model drug to TM at 0.5 mg/mL for dipyridamole or 2.0 mg/mL for ketoconazole and then stirring for 24 h at 37 °C. Furthermore, to confirm the S_eq of the model drug was maintained throughout the experiments, the suspension samples (0.2 mL) were taken from the donor side at 0 and 90 min and immediately filtered through a polytetrafluoroethylene filter (Millex-LH^®, 0.45 µm) (Millipore, Billerica, MA, U.S.A.).

Determination of Free Drug Concentration Profile

The permeation rate (the slope of the time profiles of the permeated amount) across the dialysis membrane was proportional to the C_free on the donor side because the effective permeability (P_eff) and the effective surface area (SA) were constant during the experiment (Eq. 1). The C_free profile for LBF was determined based on the S_eq (100% free drug) and the ratio of the permeation rates (LBF/suspension) at every sampling time interval, i.e., 0 to 5, 5 to 15, 15 to 30, 30 to 60, and 60 to 90 min (Eq. 2). The median of each time point (i.e., 2.5, 10, 22.5, 45, and 75 min) is used on the X axis in Figs. 4B and 5B.

  

(1)

  

(2)

where P_eff is the effective permeability across the dialysis membrane, SA is the effective surface area of the dialysis membrane (1.77 cm²), C_free is the free drug concentration, and S_eq is the equilibrium solubility in TM (as 100% free drug).

Statistical Analysis

Statistical significance was assessed with Student’s t-test, and p-values of 0.05 or less were considered significant.

RESULTS

In Vivo PK Study in Rats

The plasma concentration and absorption rate after oral administration of the model drugs formulated in type IIIA-MC and type IIIA-LC formulations were investigated and compared to their suspensions in rats. The plasma concentration and absorption rate profiles are shown in Fig. 2 for dipyridamole and Fig. 3 for ketoconazole. The PK parameters are summarized in Table 2 for dipyridamole and Table 3 for ketoconazole.

Fig. 2. Plasma Concentration–Time (A) and Absorption Rate–Time (B) Profiles of Dipyridamole after Oral Administration to Rats at 1.0 mg/kg

The absorption rate-time profile was calculated by deconvoluting the plasma concentration-time profiles after oral and intravenous administration. Data represent mean ± standard deviation ( S.D.) (n = 3–4).

Fig. 3. Plasma Concentration–Time (A) and Absorption Rate–Time (B) Profiles of Ketoconazole after Oral Administration to Rats at 4.0 mg/kg

The absorption rate–time profile was calculated by deconvoluting the plasma concentration-time profiles after oral and intravenous administration. Data represent mean ± S.D. (n = 3–4).

Table 2. Pharmacokinetic Parameters of Dipyridamole after Oral Administration to Rats (Mean ± S.D., n = 3–4)

	AUC (ng·h/mL)	Cmax (ng/mL)	Tmax (h)	BA (%)
Suspension	7.1 ± 1.1	1.60 ± 0.14	2.0 ± 1.0	2.2 ± 0.3
IIIA-LC	6.6 ± 2.8	1.54 ± 0.69	3.4 ± 3.4	2.1 ± 0.9
IIIA-MC	11.9 ± 0.8a)	3.88 ± 0.33b)	0.5 ± 0.0c)	3.7 ± 0.2d,e)

a) p < 0.01 vs. AUC values of the suspension and the type IIIA-LC formulation. b) p < 0.001 vs. C_max values of the suspension and the type IIIA-LC formulation. c) p < 0.05 vs. T_max value of the suspension. d) p < 0.001 vs. BA of the suspension. e) p < 0.05 vs. BA of the type IIIA-LC formulation.

Table 3. Pharmacokinetic Parameters of Ketoconazole after Oral Administration to Rats (Mean ± S.D., n = 3–4)

	AUC (ng·h/mL)	Cmax (ng/mL)	Tmax (h)	BA (%)
Suspension	5940 ± 1590	1108 ± 205	4.25 ± 1.26	61 ± 12
IIIA-LC	5167 ± 182	873 ± 7.07	3.33 ± 1.15	51 ± 2.0
IIIA-MC	6817 ± 2498	1106 ± 336	4.50 ± 1.73	66 ± 24

Dipyridamole

The type IIIA-MC formulation showed a higher initial plasma concentration, and a faster initial absorption rate than the suspension (p < 0.05, 0.003/h for suspension vs. 0.019/h for IIIA-MC at 0.25 h) (Fig. 2). As a result, the type IIIA-MC formulation gave a 1.7-fold higher BA than the suspension (p < 0.001, BA 2.2% for suspension vs. 3.7% for IIIA-MC) (Table 3). In contrast, the type IIIA-LC formulation had plasma concentration and absorption rate profiles similar to those of the suspension. After all, the type IIIA-LC formulation did not show any improvements in the BA. The rank order of the BA and the initial absorption rate for dipyridamole was suspension ≤ IIIA-LC < IIIA-MC.

Ketoconazole

Both type IIIA-MC and type IIIA-LC formulations had plasma concentration profiles similar to that of the suspension even though the type IIIA-MC formulation showed a slightly higher plasma concentration until 2.0 h (Fig. 3A). For all treatments, plasma concentrations were sustained over 8 h due to ABT-treatment (Fig. 3A), while the absorption was almost complete by approximately 3 h (Fig. 3B). As with dipyridamole, the type IIIA-MC formulation showed a faster initial absorption rate than the suspension (p < 0.05, 0.111/h for suspension vs. 0.628/h for IIIA-MC at 0.25 h) (Fig. 3B). The rank order of the initial absorption rate for ketoconazole was suspension = IIIA-LC < IIIA-MC, which was very similar to that for dipyridamole.

In Vitro Release Study: Determination of Free Drug Concentration Profiles

To elucidate the mechanism that the type IIIA-MC formulation increased the initial absorption rates and that the type IIIA-LC formulation did not show any improvements for both drugs in rats, the profiles of free drugs that were released during the digestion of both LBFs were determined in the in vitro system depicted in Fig. 1. The permeated amount and the free drug concentration profiles are shown in Fig. 4 for dipyridamole and Fig. 5 for ketoconazole. The initiation time of lipid digestion is shown as 0 min in Figs. 4 and 5. The permeation rates and C_frees are summarized in Tables 4 and 5.

Fig. 4. Permeated Amount–Time (A) and Free Concentration–Time (B) Profiles of Dipyridamole

In Fig. 4B, the dotted line indicates the equilibrium solubility. The free drug concentration-time profile was determined based on the equilibrium solubility and the ratio of the permeation rates (LBF/suspension) at every sampling time interval. Data represent mean ± S.D. (n = 3–4).

Fig. 5. Permeated Amount–Time (A) and Free Concentration–Time (B) Profiles of Ketoconazole

In Fig. 5B, the dotted line indicates the equilibrium solubility. The free drug concentration-time profile was determined based on the equilibrium solubility and the ratio of the permeation rates (LBF/suspension) at every sampling time interval. Data represent mean ± S.D. (n = 3–4).

Table 4. Permeation Rate and Free Concentration of Dipyridamole for the Type IIIA-MC and IIIA-LC Formulations

Time (min)	Suspension	IIIA-MC			IIIA-LC
	Mean permeation rate (ng/h)	Permeation rate (ng/h)	Ratio of permeation rate (vs. suspension)	Free concentration (µg/mL)a)	Permeation rate (ng/h)	Ratio of permeation rate (vs. suspension)	Free concentration (µg/mL)a)
2.5 (0–5)	2.12	10.38 ± 0.92	4.88 ± 0.43	18.65 ± 1.65	5.19 ± 1.65	2.44 ± 0.77	8.17 ± 2.26
10 (5–15)	1.82	1.92 ± 0.54	1.05 ± 0.29	4.02 ± 1.12	1.62 ± 0.48	0.89 ± 0.26	3.68 ± 1.03
22.5 (15–30)	2.36	3.78 ± 0.27	1.60 ± 0.12	6.11 ± 0.44	1.21 ± 1.36	0.51 ± 0.58	2.82 ± 1.66
45 (30–60)	1.54	2.87 ± 0.30	1.87 ± 0.19	7.14 ± 0.74	1.05 ± 0.75	0.68 ± 0.49	1.83 ± 1.24
75 (60–90)	4.64	2.12 ± 1.14	0.46 ± 0.25	1.75 ± 0.94	1.96 ± 1.13	0.42 ± 0.24	1.77 ± 1.07

a) Free concentration = Ratio of permeation rate × Equilibrium solubility (3.8 µg/mL).

Table 5. Permeation Rate and Free Concentration of Ketoconazole for the Type IIIA-MC and IIIA-LC Formulations

Time (min)	Suspension	IIIA-MC			IIIA-LC
	Mean permeation rate (ng/h)	Permeation rate (ng/h)	Ratio of permeation rate (vs. suspension)	Free concentration (µg/mL)a)	Permeation rate (ng/h)	Ratio of permeation rate (vs. suspension)	Free concentration (µg/mL)a)
2.5 (0–5)	26.85	44.04 ± 18.68	1.64 ± 0.70	6.85 ± 2.91	27.05 ± 2.63	1.01 ± 0.10	4.21 ± 0.41
10 (5–15)	12.55	23.50 ± 15.61	1.87 ± 1.24	7.82 ± 5.20	12.28 ± 4.02	0.98 ± 0.32	4.09 ± 1.34
22.5 (15–30)	17.95	24.40 ± 10.85	1.36 ± 0.60	5.67 ± 2.52	10.82 ± 3.01	0.60 ± 0.17	2.52 ± 0.70
45 (30–60)	11.52	26.92 ± 3.40	2.34 ± 0.29	9.75 ± 1.23	9.00 ± 1.86	0.78 ± 0.16	3.26 ± 0.67
75 (60–90)	13.71	18.78 ± 3.10	1.37 ± 0.23	5.72 ± 0.95	7.96 ± 1.87	0.58 ± 0.14	2.43 ± 0.57

a) Free concentration = Ratio of permeation rate × Equilibrium solubility (4.2 µg/mL).

Dipyridamole

The type IIIA-MC formulation showed a significantly higher permeation rate than the suspension until 5 min (Fig. 4A). After 5 min, the type IIIA-MC formulation showed a permeation rate similar to that of the suspension, on the other hand, the type IIIA-LC formulation showed a lower permeation rate than the suspension. The dissolved concentration after applying the suspension remained constant (3.8 µg/mL) throughout the experiments and was used as the S_eq (Fig. 4B). The C_free of the type IIIA-MC formulation at 2.5 min was 4.9-fold higher (p < 0.01, 18.7 µg/mL) than the S_eq and decreased to approximately the S_eq by 10 min. In contrast, the C_free of the type IIIA-LC formulation became lower than the S_eq over time although this concentration at 2.5 min was 2.1-fold higher (p < 0.01, 8.2 µg/mL) than the S_eq.

Ketoconazole

The type IIIA-MC formulation showed a higher permeation rate than the others throughout the experiment (Fig. 5A). In contrast, the type IIIA-LC formulation showed a lower permeation rate than the suspension. The dissolved concentration after applying the suspension remained constant (4.2 µg/mL) and was used as the S_eq (Fig. 5B). The average C_free of the type IIIA-MC formulation was 1.7-fold higher (7.2 µg/mL) than the S_eq. This low degree of supersaturation was maintained throughout the experiment. In contrast, the C_free from the type IIIA-LC formulation became lower than the S_eq.

For both drugs and both LBFs, the profiles of in vitro C_frees were consistent with those of in vivo absorption rates. The rank order of the performance was suspension ≤ IIIA-LC < IIIA-MC for both drugs (Figs. 2–5).

DISCUSSION

In this study, we have assessed the in vivo enhancement effects of LBFs containing the model drug through the in vitro system depicted in Fig. 1. The in vivo rat study showed that the type IIIA-MC formulation, but not the type IIIA-LC formulation, significantly increased the initial absorption rate for both drugs. The in vitro release study revealed that the C_free of both drugs was immediately supersaturated after initiating digestion of the type IIIA-MC formulation. Hence, the observed enhancement of absorption rates by the type IIIA-MC formulation would be attributed to the supersaturation of the free drug. Overall, the profiles of in vitro C_free profiles were consistent with in vivo absorption rate profiles for both drugs and both LBFs. Also, the in vitro side-by-side chamber system coupled with a lipid digestion model helps predict the in vivo performance of oral LBFs.

Both dipyridamole and ketoconazole previously showed different supersaturated dissolution profiles,^20,21) and thus were selected as model drugs in this study. These two drugs are classified as BCS Class II weakly basic drugs and have the potential to supersaturate when moving from the acidic milieu of the stomach (where the drugs are more soluble) to the neutral pH of the proximal small intestine (where the drugs are less soluble).^20,21) In addition, ketoconazole, known as a substrate and potent inhibitor of CYP3A, often causes drug-drug interactions.^32,33) Our previous studies also demonstrated that a supersaturated suspension of ketoconazole strongly inhibited its first-pass metabolism via CYP3A, resulting in much higher systemic exposure than predicted from an in vitro dissolution study.^21,29) Thus, to assess the in vivo performance of LBFs without the confounding variables of pH and first-pass metabolism, rats were pretreated with omeprazole (for both drugs) and ABT (for ketoconazole only) before the in vivo PK studies. The BA of ketoconazole was 51–66% for all treatments in pretreated rats (Table 3). ABT strongly inhibits intestinal and hepatic first-pass metabolism, suggesting that F_g and F_h values were close to 100% and that the BA of ketoconazole was almost equal to the F_a value. In the current case, the maximum effect of LBFs on enhanced oral absorption is expected to be relatively small (1.6-fold = 100%/61%) due to its high F_a value, although neither of the LBFs improved the BA of ketoconazole.

The type IIIA-MC formulation mainly comprised COCONAD RK (C₈ triglyceride) and Capmul MCM (a blend of C₈ mono and di-glyceride). Kataoka et al. have reported that approximately 80% of MC lipids, including COCONAD RK, were digested by pancreatic extract within 10 min.³⁴⁾ In addition, Anby et al. have reported that the solubilizing capacity of the digested monoglycerides and liberated fatty acids from MC lipids was lower than those from LC lipids.^2,7) Therefore, it was indicated that the digestion of the type IIIA-MC formulation rapidly decreased its solubilization capacity and that the preconcentrated drug in the type IIIA-MC formulation was released in an aqueous phase. These processes resulted in an initially high absorption rate for both drugs (Figs. 2B, 3B). In addition, it was confirmed using ketoconazole that the initial supersaturation was not induced in the non-digestion condition (data not shown). Previous studies have shown that the supersaturation profiles are drug- and dose-dependent.^21,29) In general, the drug in the supersaturated state precipitates such as dipyridamole (Fig. 4B). In contrast, the supersaturation of ketoconazole was maintained throughout the in vitro study when the type IIIA-MC formulation was digested (Fig. 5B). This long duration of supersaturation (but low degree, only 1.7-fold) was due to a lack of precipitation energy.²¹⁾ In vivo, regardless of the in vitro long duration of supersaturation, a rapid decrease in the absorption rate was observed (Figs. 3B, 5B). This was probably explained by the difference in absorption/permeation clearance (P_eff ×SA) between in vitro (small clearance) and in vivo (high clearance) conditions.³⁵⁾ Previously, the effect of the SA/V ratio on drug transport from ketoconazole supersaturated solutions with and without HPMC was investigated through in vitro permeation and in vivo rat absorption studies. As a result, the effect of HPMC (duration of supersaturation) on enhanced absorption became small with the increase of the SA/V ratio (absorption clearance).³⁵⁾ Therefore, it was indicated that in the current study, the free (supersaturated) ketoconazole caused by lipid digestion was quickly absorbed due to the high absorption clearance despite the long duration of supersaturation (Fig. 3B).

In contrast to the type IIIA-MC formulation, the type IIIA-LC formulation did not improve the BA for both drugs (Figs. 2–5). In general, LC lipids are digested more slowly than MC lipids. However, the extent of digestion for the type IIIA-LC formulation used in this study has been reported to be 67–95%.³⁶⁾ The reason for the high extent of digestion is explained by the excellent dispersibility of the type III-LC formulation with an aqueous buffer, as compared with the type I-LC and II-LC formulations.³⁶⁾ Again, the solubilization capacity of LC lipids/digested products is usually high compared with MC lipids/digested products.^14,17,37) Therefore, one possible explanation for the lack of absorption enhancement effect for the type IIIA-LC formulation is that a large amount of the drug was distributed in LC lipids/digested products and that the free concentration decreased below the S_eq for both drugs (Figs. 4B, 5B). The same phenomenon was previously observed using the LC formulation.¹⁸⁾

For dipyridamole only, supersaturation of free concentration (with short duration and low degree, 2.1-fold) was observed at 5.0 min after initiating digestion of the type IIIA-LC formulation, suggesting a drug-specific mechanism (Fig. 4B). However, this had no effect on increasing the plasma concentration of dipyridamole (Fig. 2). Because 35% of the type IIIA-LC formulation was composed of a hydrophilic surfactant (Cremophor EL), its miscibility with the aqueous buffer could decrease the solubilization capacity, thus generating a low degree and short duration of supersaturation (Fig. 4B).^2,17,38)

CONCLUSION

In this study, we have assessed the in vivo absorption enhancement effects of LBFs through in vitro release studies. The in vitro and in vivo investigations demonstrated that the profiles of in vitro free concentrations were consistent with those of the in vivo absorption rates for both drugs and all LBFs. Also, the in vitro release study successfully detected the supersaturation of free concentration after initiating digestion of the type IIIA-MC formulation and showed the importance of measurement of free concentration to predict the performance of LBFs. These findings indicate that the in vitro system used in this study helps predict the in vivo performance of oral LBFs and establish an IVIVC.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Pouton CW. Formulation of poorly water-soluble drugs for oral administration: physicochemical and physiological issues and the lipid formulation classification system. Eur. J. Pharm. Sci., 29, 278–287 (2006).
• 2) Feeney OM, Crum MF, McEvoy CL, Trevaskis NL, Williams HD, Pouton CW, Charman WN, Bergström CAS, Porter CJH. 50 years of oral lipid-based formulations: provenance, progress and future perspectives. Adv. Drug Deliv. Rev., 101, 167–194 (2016).
• 3) Pouton CW, Porter CJ. Formulation of lipid-based delivery systems for oral administration: materials, methods and strategies. Adv. Drug Deliv. Rev., 60, 625–637 (2008).
• 4) Crum MF, Trevaskis NL, Williams HD, Pouton CW, Porter CJH. A new in vitro lipid digestion – in vivo absorption model to evaluate the mechanisms of drug absorption from lipid-based formulations. Pharm. Res., 33, 970–982 (2016).
• 5) Williams HD, Trevaskis NL, Yeap YY, Anby MU, Pouton CW, Porter CJ. Lipid-based formulations and drug supersaturation: harnessing the unique benefits of the lipid digestion/absorption pathway. Pharm. Res., 30, 2976–2992 (2013).
• 6) Alskär LC, Keemink J, Johannesson J, Porter CJH, Bergström CAS. Impact of drug physicochemical properties on lipolysis-triggered drug supersaturation and precipitation from lipid-based formulations. Mol. Pharm., 15, 4733–4744 (2018).
• 7) Anby MU, Williams HD, McIntosh M, Benameur H, Edwards GA, Pouton CW, Porter CJ. Lipid digestion as a trigger for supersaturation: evaluation of the impact of supersaturation stabilization on the in vitro and in vivo performance of self-emulsifying drug delivery systems. Mol. Pharm., 9, 2063–2079 (2012).
• 8) Yeap YY, Trevaskis NL, Porter CJ. The potential for drug supersaturation during intestinal processing of lipid-based formulations may be enhanced for basic drugs. Mol. Pharm., 10, 2601–2615 (2013).
• 9) Do TT, Van Speybroeck M, Mols R, Annaert P, Martens J, Van Humbeeck J, Vermant J, Augustijns P, Van den Mooter G. The conflict between in vitro release studies in human biorelevant media and the in vivo exposure in rats of the lipophilic compound fenofibrate. Int. J. Pharm., 414, 118–124 (2011).
• 10) Griffin BT, Kuentz M, Vertzoni M, Kostewicz ES, Fei Y, Faisal W, Stillhart C, O’Driscoll CM, Reppas C, Dressman JB. Comparison of in vitro tests at various levels of complexity for the prediction of in vivo performance of lipid-based formulations: case studies with fenofibrate. Eur. J. Pharm. Biopharm., 86, 427–437 (2014).
• 11) Heshmati N, Cheng X, Dapat E, Sassene P, Eisenbrand G, Fricker G, Müllertz A. In vitro and in vivo evaluations of the performance of an indirubin derivative, formulated in four different self-emulsifying drug delivery systems. J. Pharm. Pharmacol., 66, 1567–1575 (2014).
• 12) Dahan A, Hoffman A. The effect of different lipid based formulations on the oral absorption of lipophilic drugs: the ability of in vitro lipolysis and consecutive ex vivo intestinal permeability data to predict in vivo bioavailability in rats. Eur. J. Pharm. Biopharm., 67, 96–105 (2007).
• 13) Tanaka Y, Nguyen TH, Suys EJA, Porter CJH. Digestion of lipid-based formulations not only mediates changes to absorption of poorly soluble drugs due to differences in solubilization but also reflects changes to thermodynamic activity and permeability. Mol. Pharm., 18, 1768–1778 (2021).
• 14) Porter CJH, Kaukonen AM, Boyd BJ, Edwards GA, Charman WN. Susceptibility to lipase-mediated digestion reduces the oral bioavailability of danazol after administration as a medium-chain lipid-based microemulsion formulation. Pharm. Res., 21, 1405–1412 (2004).
• 15) Huang Y, Yu Q, Chen Z, Wu W, Zhu Q, Lu Y. In vitro and in vivo correlation for lipid-based formulations: current status and future perspectives. Acta Pharm. Sin. B, 11, 2469–2487 (2021).
• 16) Thomas N, Richter K, Pedersen TB, Holm R, Müllertz A, Rades T. In vitro lipolysis data does not adequately predict the in vivo performance of lipid-based drug delivery systems containing fenofibrate. AAPS J., 16, 539–549 (2014).
• 17) Williams HD, Anby MU, Sassene P, Kleberg K, Bakala-N’Goma JC, Calderone M, Jannin V, Igonin A, Partheil A, Marchaud D, Jule E, Vertommen J, Maio M, Blundell R, Benameur H, Carrière F, Müllertz A, Pouton CW, Porter CJH. Toward the establishment of standardized in vitro tests for lipid-based formulations. 2. the effect of bile salt concentration and drug loading on the performance of type I, II, IIIA, IIIB, and IV formulations during in vitro digestion. Mol. Pharm., 9, 3286–3300 (2012).
• 18) Higashino H, Develin C, Higashino C, Lim T, Martin A, Zhou F, Strab R, Patel R, Bhoopathy S, Hidalgo I. In vitro digestion–diffusion model for predicting in vivo performance of lipid-based formulations. J. Drug Deliv. Sci. Technol., 83, 104439 (2023).
• 19) Higashino H, Minami K, Kataoka M, Tomimori N, Rogi T, Shibata H, Yamashita S. Control of oral absorption of nutritional supplement using lipid-based formulations (LBFs): application to the poorly water-soluble ingredient. J. Drug Deliv. Sci. Technol., 57, 101675 (2020).
• 20) Tsume Y, Matsui K, Searls AL, Takeuchi S, Amidon GE, Sun D, Amidon GL. The impact of supersaturation level for oral absorption of BCS class IIb drugs, dipyridamole and ketoconazole, using in vivo predictive dissolution system: Gastrointestinal Simulator (GIS). Eur. J. Pharm. Sci., 102, 126–139 (2017).
• 21) Higashino H, Minami K, Takagi T, Kataoka M, Yamashita S. The effects of degree and duration of supersaturation on in vivo absorption profiles for highly permeable drugs, dipyridamole and ketoconazole. Eur. J. Pharm. Biopharm., 189, 48–55 (2023).
• 22) Abbasi S, Higashino H, Sato Y, Minami K, Kataoka M, Yamashita S, Harashima H. Maximizing the oral bioavailability of poorly water-soluble drugs using novel oil-like materials in lipid-based formulations. Mol. Pharm., 18, 3281–3289 (2021).
• 23) Wada I, Otaka M, Jin M, Odashima M, Komatsu K, Konishi N, Matsuhashi T, Horikawa Y, Ohba R, Itoh H, Watanabe S. Expression of HSP72 in the gastric mucosa is regulated by gastric acid in rats-Correlation of HSP72 expression with mucosal protection. Biochem. Biophys. Res. Commun., 349, 611–618 (2006).
• 24) Balani SK, Zhu T, Yang TJ, Liu Z, He B, Lee FW. Effective dosing regimen of 1-aminobenzotriazole for inhibition of antipyrine clearance in rats, dogs, and monkeys. Drug Metab. Dispos., 30, 1059–1062 (2002).
• 25) Emoto C, Murase S, Sawada Y, Jones BC, Iwasaki K. In vitro inhibitory effect of 1-aminobenzotriazole on drug oxidations catalyzed by human cytochrome P450 enzymes: a comparison with SKF-525A and ketoconazole. Drug Metab. Pharmacokinet., 18, 287–295 (2003).
• 26) El-Kattan AF, Poe J, Buchholz L, Thomas HV, Brodfuehrer J, Clark A. The use of 1-aminobenzotriazole in differentiating the role of CYP-mediated first pass metabolism and absorption in limiting drug oral bioavailability: a case study. Drug Metab. Lett., 2, 120–124 (2008).
• 27) Strelevitz TJ, Foti RS, Fisher MB. In vivo use of the P450 inactivator 1-aminobenzotriazole in the rat: varied dosing route to elucidate gut and liver contributions to first-pass and systemic clearance. J. Pharm. Sci., 95, 1334–1341 (2006).
• 28) Takahashi M, Washio T, Suzuki N, Igeta K, Yamashita S. Investigation of the intestinal permeability and first-pass metabolism of drugs in cynomolgus monkeys using single-pass intestinal perfusion. Biol. Pharm. Bull., 33, 111–116 (2010).
• 29) Higashino H, Hasegawa T, Yamamoto M, Matsui R, Masaoka Y, Kataoka M, Sakuma S, Yamashita S. In vitro–in vivo correlation in the effect of supersaturation on intestinal absorption of BCS class 2 drugs. Mol. Pharm., 11, 746–754 (2014).
• 30) Kataoka M, Yokoyama T, Masaoka Y, Sakuma S, Yamashita S. Estimation of P-glycoprotein-mediated efflux in the oral absorption of P-gp substrate drugs from simultaneous analysis of drug dissolution and permeation. Eur. J. Pharm. Sci., 44, 544–551 (2011).
• 31) Yano K, Masaoka Y, Kataoka M, Sakuma S, Yamashita S. Mechanisms of membrane transport of poorly soluble drugs: role of micelles in oral absorption processes. J. Pharm. Sci., 99, 1336–1345 (2010).
• 32) Daneshmend TK, Warnock DW. Clinical pharmacokinetics of ketoconazole. Clin. Pharmacokinet., 14, 13–34 (1988).
• 33) Remmel RP, Amoh K, Abdel-Monem MM. The disposition and pharmacokinetics of ketoconazole in the rat. Drug Metab. Dispos., 15, 735–739 (1987).
• 34) Kataoka M, Ohi Y, Sakanoue K, Minami K, Higashino H, Yamashita S. Impact of dietary intake of medium-chain triacylglycerides on the intestinal absorption of poorly permeable compounds. Mol. Pharm., 17, 212–218 (2020).
• 35) Higashino H, Develin C, Higashino C, Lim T, Miezeiewski B, Strab R, Patel R, Bhoopathy S, Hidalgo I. The impact of surface area per volume (SA/V) ratio on drug transport from supersaturated solutions of ketoconazole. Eur. J. Pharm. Biopharm., 188, 147–152 (2023).
• 36) Williams HD, Sassene P, Kleberg K, Bakala-N’Goma JC, Calderone M, Jannin V, Igonin A, Partheil A, Marchaud D, Jule E, Vertommen J, Maio M, Blundell R, Benameur H, Carrière F, Müllertz A, Porter CJH, Pouton CW. Toward the establishment of standardized in vitro tests for lipid-based formulations, part 1: method parameterization and comparison of in vitro digestion profiles across a range of representative formulations. J. Pharm. Sci., 101, 3360–3380 (2012).
• 37) Sek L, Porter CJ, Kaukonen AM, Charman WN. Evaluation of the in-vitro digestion profiles of long and medium chain glycerides and the phase behaviour of their lipolytic products. J. Pharm. Pharmacol., 54, 29–41 (2002).
• 38) Cuiné JF, Charman WN, Pouton CW, Edwards GA, Porter CJH. Increasing the proportional content of surfactant (Cremophor EL) relative to lipid in self-emulsifying lipid-based formulations of danazol reduces oral bioavailability in beagle dogs. Pharm. Res., 24, 748–757 (2007).