In Vitro, in Silico, and in Vivo Assessments of Intestinal Precipitation

Subscriber access provided by - Access paid by the | UCSB Libraries

In Vitro, in Silico, and in Vivo Assessments of Intestinal Precipitation and Its Impact on Bioavailability of a BCS Class 2 Basic Compound Dawen Kou, Chen Zhang, Hiuwing Yiu, Tania Ng, Joseph W. Lubach, Matthew Janson, Chen Mao, Matthew Durk, Leslie Chinn, Helen Winter, Larry Wigman, and Peter Yehl Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b01143 • Publication Date (Web): 09 Mar 2018 Downloaded from http://pubs.acs.org on March 16, 2018

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

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

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

Molecular Pharmaceutics

In Vitro, in Silico, and in Vivo Assessments of Intestinal Precipitation and Its Impact on Bioavailability of a BCS Class 2 Basic Compound Dawen Kou*†, Chen Zhang||, Hiuwing Yiu†, Tania Ng†, Joseph W. Lubach†, Matthew Janson†, Chen Mao†, Matthew Durk‡, Leslie Chinn§, Helen Winter§, Larry Wigman†, and Peter Yehl† †

Small Molecule Pharmaceutical Sciences, ‡Drug Metabolism and Pharmacokinetics, §Clinical Pharmacology, Genentech Inc., 1 DNA Way, South San Francisco, CA 94080, United States || Department of Chemistry, Michigan State University, 578 S Shaw Lane, East Lansing, MI 48824, United States ABSTRACT In this study, a multi-pronged approach of in vitro experiments, in silico simulations, and in vivo studies was developed to evaluate the dissolution, supersaturation, precipitation, and absorption of three formulations of Compound-A, a BCS Class 2 weak base with pH-dependent solubility. In in vitro 2-stage dissolution experiments, the solutions were highly supersaturated with no precipitation at the low dose but increasing precipitation at higher doses. No difference in precipitation was observed between the capsules and tablets. The in vitro precipitate was found to be non-crystalline with higher solubility than the crystalline API, and was readily soluble when the drug concentration was lowered by dilution. A gastric transit and biphasic dissolution (GTBD) model was developed to better mimic gastric transfer and intestinal absorption. Precipitation was also observed in GTBD but the precipitate re-dissolved and partitioned into the organic phase. In vivo data from the Phase 1 clinical trial showed linear and dose proportional PK for the formulations with no evidence of in vivo precipitation. While the in vitro precipitation observed in the 2-stage dissolution appeared to overestimate in vivo precipitation, the GTBD model provided absorption profiles consistent with in vivo data. In silico simulation of plasma concentrations by GastroPlus using bio-relevant in vitro dissolution data from the tablets and capsules and assuming negligible precipitation were in line with the observed in vivo profiles of the two formulations. The totality of data generated with Compound-A indicated that the bioavailability differences among the three formulations were better explained by the differences in gastric dissolution than intestinal precipitation. The lack of intestinal precipitation was consistent with several other BCS Class 2 basic compounds in the literature for which highly supersaturated concentrations and rapid absorption were also observed.

1 ACS Paragon Plus Environment

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

Page 2 of 23

KEYWORDS: Biopharmaceutics Classification System (BCS), weakly basic compounds, bio-relevant dissolution, biphasic, supersaturation, precipitation, absorption, physiologically based pharmacokinetic (PBPK) modeling


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


INTRODUCTION The dissolution and absorption of an orally dosed drug in the GI tract can be a complex process that affects its bioavailability and pharmacokinetics (PK), depending on its physicochemical and biopharmaceutics properties. The Biopharmaceutics Classification System (BCS) has been widely used to characterize drugs on the basis of their permeability and solubility 1. The BCS classes can be further divided into several subclasses based on acidity or basicity in addition to solubility and permeability 2. Each subclass has its own dissolution and absorption characteristics 2. Many drugs in the BCS Class 2 (high permeability and low solubility) are weak bases, with higher solubility at acidic pH in the stomach but lower solubility at near neutral pH in the small intestine under fasting conditions. Although dissolution of such drugs can be fast at acidic gastric pH facilitating ionization, drug absorption generally is minimal in the stomach. In the small intestine where most drugs are absorbed, the higher pH and lower solubility could affect the amount of drug in solution. During and after gastric transfer, several processes take place simultaneously. The drug and gastric content gradually enter the small intestine, bile and pancreatic juice are secreted to neutralize the gastric content, and the drug gets absorbed across the intestinal membrane and removed from the intraluminal space. With the pH increase, the drug may remain in solution at a higher concentration than its intrinsic solubility at the pH, a phenomenon known as supersaturation. The drug may also precipitate out of the supersaturated solution. At the same time, the dilution effect of mixing gastric and intestinal fluids and rapid drug absorption by the small intestine can reduce drug concentration, supersaturation, and potential precipitation. With multiple factors in play, the interactions can be complex and the net effect is not known a priori. Therefore, it is critically important to study whether and how intestinal precipitation occurs in order to better understand and assess its impact on bioavailability, preferably prior to dosing in clinical studies. The phenomena of drug supersaturation and precipitation due to pH and solubility changes have been studied extensively in in vitro studies under various mixing and fluid transfer conditions 3-7. In vitro precipitation was found to be affected by the starting concentration, degree of supersaturation, transfer rate, hydrodynamics, and presence of particles 3-7. Several multi-compartmental systems that include a gastric compartment and at least one intestinal compartment have been developed, such as Artificial Stomach and Duodenum (ASD) 8-14, Gastric Intestinal Simulator (GIS) 15-17, and TNO Gastro-Intestinal Model (TIM-1) 18. ASD and GIS measure drug concentration in the intestinal compartment where precipitation may occur. The TIM-1 system utilizes membrane filtration to remove the dissolved drug from the intestinal compartments mimicking absorption. Physiologically based pharmacokinetic (PBPK) modeling and simulation for in vivo drug absorption can be a valuable tool to predict bioavailability. Certain software packages provide the capability to adjust the dissolution rate and precipitation rate as variables to simulate their impact on drug absorption and generate corresponding drug plasma concentration profiles. In silico modeling has been used either on its own or in combination with in vitro data to simulate the impact of precipitation on absorption 7, 19-20. Interestingly, evidence of in vivo precipitation of BCS Class 2 weak bases in the literature is limited, and mixed results have been reported. In one study 21, dipyridamole and ketoconazole were dosed in healthy volunteers in the fasted state using oral solutions of each compound. Samples from the upper small intestine contents were aspirated post dosing up to 70 min at 10 min internals. Only minimal or limited precipitation was found in the samples (≤7% for dipyridamole and ≤16% for ketoconazole). The in vivo precipitation observed was substantially less than that in in vitro studies with dipyridamole and ketoconazole 7, 19. In another study 22, two formulations of a BCS Class 2 basic compound, AZD0865, 3 ACS Paragon Plus Environment


were administered in three clinical studies at a wide range of doses, and studied in in vitro mixing and stirring experiments. The bioavailability and PK (Cmax, AUC, and Tmax) of the two formulations were linear and dose proportional across the dose range with no evidence of in vivo precipitation that would have produced non-linear PK. However, significant precipitation was observed in the in vitro mixing and stirring experiments. In a third study 23, solutions of mebendazole were dosed in dogs. Samples taken from the jejunum showed little precipitation at the low dose and limited precipitation (~11%) at the high dose. In contrast, in vitro experiments produced significant precipitation. Yet, in another study 24, significant in vivo precipitation (up to 92%) was reported for posaconazole, a BCS Class 2 basic compound and an antifungal agent similar to ketoconazole. Given the apparent disconnect between in vivo and in vitro precipitation and differences in in vivo precipitation among BCS Class 2 weak bases reported in the literature, there is a need to study in vivo and in vitro precipitation of other compounds in the same class, and to evaluate the utilities and limitations of various approaches in estimating precipitation. Genentech Compound-A is a weakly basic BCS Class 2 compound with high permeability and pH dependent solubility. Figure 1 shows the pH-solubility profile of Compound-A with a steep, exponential drop in solubility from >35 mg/mL at pH 2 to 0.001 mg/mL at pH 5, which could lead to supersaturation and possibly precipitation in the small intestine. Therefore, Compound-A offers a good opportunity to study intestinal precipitation and its impact on bioavailability. Due to its lower solubility at pH>4, capsules containing neat API of Compound-A showed poor bioavailability in healthy volunteers with hypo-/achlorhydria (elevated gastric pH), when co-dosed with acid reducing agents (ARA) such as proton pump inhibitors (PPI) in a drug-drug interaction study. A significant portion of the target population of Compound-A take ARA as co-medication. Therefore, it was critical to develop formulations to overcome the reduced dissolution and absorption caused by hypo/achlorhydria due to ARA as co-medication. The rationale and strategy for developing such a formulation and bio-relevant dissolution models for formulation screening were discussed in a recent publication 25. 40 35 Concentration (mg/mL)

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

Page 4 of 23

30 25 20 15 10 5 0 2.0

3.0

4.0

5.0 pH

6.0

7.0

8.0

Figure 1. pH-solubility profile of Genentech Compound-A


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


The primary objective of this study was to develop a comprehensive approach to investigate the interrelationship between dissolution, supersaturation, precipitation, and absorption of Compound-A as a case study of a BCS Class 2 basic drug, including in vitro dissolution and precipitation experiments, in silico simulations, and in vivo studies. The in vitro experiments studied precipitation behaviors in simulated intestinal fluid after gastric transfer, with and without an organic phase to simulate absorption. The in silico simulations used different models based on dissolution rate or solubility coupled with different precipitation rates. The in vivo data were from clinical studies with three formulations and a dog study using the tablet formulation. The overall data were used to assess the impact of potential precipitation on bioavailability of the three formulations. The second goal was to determine whether the different in vivo performances of the formulations were due to incomplete dissolution or precipitation in the GI tract. The third objective was to compare Compound-A with other BCS Class 2 basic compounds in the literature to identify factors that may explain their similarities and differences in in vivo precipitation. EXPERIMENTAL SECTION Materials. The drug substance or active pharmaceutical ingredient (API) of Compound-A was a crystalline free base from Genentech (South San Francisco, CA). Fumaric acid (FA) was from Bartek (Stoney Creek, ON, Canada). Ammonium formate, formic acid, HCl, and 1-nonanol were from Sigma/Aldrich (St. Louis, MO), and acetonitrile from J.T. Baker (Avantor, Center Valley, PA). All chemicals and reagents were ACS or HPLC/GC grade. Purified water was obtained from a Millipore Milli-Q unit. FaSSIF powder (V1) was purchased from biorelevant.com (London, UK). Three formulations of Compound-A were used in clinical trials: a solution formulation and APIfilled capsules (50 mg dose strength) for Phase 1, and a tablet formulation (50 mg dose strength) for Phase 2. The solution formulation contained 0.2 mg/mL of API in 25 mM phosphate buffer at pH 2.3. The capsule contained 50 mg of neat crystalline free base API in size 0 hard gelatin capsule shells (Capsugel, Morristown, NJ). The Ph2 tablet formulation was developed to improve the dissolution and bioavailability of Compound-A in hypochlorhydric and achlorhydric patients (low or no hydrochloric acid in the stomach) who take acid reducing agents (ARA) as co-medication. The tablet contained 15% (50 mg) crystalline free base API , 15% FA (50 mg, to acidify transient gastric pH and increase dissolution), and other excipients including lactose, croscarmellose sodium, magnesium stearate, and microcrystalline cellulose (MCC). When doses higher than 50 mg were needed in in vitro or in vivo studies, multiple units of capsules or tablets were used. During Ph2 formulation development, several other tablet formulations were also made and compared with the 15% FA tablet in in vitro and pre-clinical studies, including formulations containing 0%, 5%, or 10% FA. In these formulations, part or all of FA was replaced by MCC, otherwise they had the same composition and weight as the 15% FA tablet. Detailed development of the FA based formulations and comparison with formulations containing other acidic modifiers and a formulation using a salt form of Compound-A had been discussed in the earlier paper 25 and are not repeated here. In this study, the no-FA tablet was further compared with the Ph2 tablet in in vitro experiments. Two-stage Dissolution. In the first stage, 500 mL of acidic dissolution media, FaSSGF at pH 1.6 or dilute HCl at pH 2.4, at 37ºC was used to mimic the normal gastric condition in the fasted state. At 30 min, 500 mL of FaSSIF (2x concentration) at 37ºC was added to the same vessel to start the second stage mimicking the small intestine environment after gastric transfer. A USP Dissolution Apparatus 2 was used with the paddle speed at 75 rpm. Samples were taken at predetermined time points up to 240 min, 5 ACS Paragon Plus Environment

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

Page 6 of 23

and analyzed by HPLC. The samples from the first stage were analyzed as is. The samples from the second stage were diluted in acetonitrile (total volume of 5 mL after dilution) to prevent precipitation in the sample solutions. The capsules and Ph2 tablets at 100, 200 and 400 mg were compared in 2-stage dissolution using FaSSGF at pH 1.6 in Stage 1. The ending pH in Stage 2 for the 400 mg dose was measured to be 4.6 for the capsules and 3.5 for the tablets. In another experiment comparing capsules and tablets, pH 2.4 HCl was used as the dissolution medium in Stage 1. The capsules at doses of 100-400 mg were pre-dissolved in pH 2.4 HCl with sonication and the resulting solutions were used to start Stage 1. The tablets at 400 mg dissolved quickly in pH 2.4 HCl within 30 min. The ending pH in Stage 2 was 6.5 for all capsule doses and 6.1 for 400 mg tablets. Gastric Transit and Biphasic Dissolution (GTBD). Figure 2 shows a schematic diagram of this dissolution model, constructed to mimic gradual fluid transit from the stomach to small intestine and drug absorption in the small intestine. The system consisted of a USP Apparatus 2 and two dissolution vessels. The first vessel was filled with 250 mL of pH 4.5 HCl solution as dissolution medium to mimic the hypochlorhydric stomach. The rationale for using pH 4.5 HCl had been extensively discussed 25. The second vessel contained 250 mL of FaSSIF as the aqueous phase and 250 mL of 1-nonanol as the organic phase on top. The temperature of the contents in both vessels were maintained at 37ºC. In Vessel 1, 200 mg dose of the Ph2 15% FA tablets were pre-dissolved for 60 min, and then the solution was transferred to Vessel 2 over 30 min with 10 µm in-line filters to keep the undissolved particles from entering the pump and Vessel 2. Two additional paddles were attached to the middle of the paddle shaft in Vessel 2 so that at least one paddle was submerged in each layer at any time during and after fluid transfer. The paddle speed for both vessels was 75 rpm. Samples were taken from the organic phase at predetermined time points up to 120 minutes, diluted 1/1 with acetonitrile, and analyzed by HPLC. The ending pH in the aqueous phase in Vessel 2 after GTBD was 4.0, likely due to the FA remaining in the aqueous solution while the API free base had partitioned into the organic phase.

Figure 2. Schematic diagram of the Gastric Transit and Biphasic Dissolution Model


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


As a comparison with the Ph2 tablet formulation, tablets without FA were also tested. The no-FA tablets (200 mg total dose) were added directly into Vessel 2, since little drug would dissolve without FA in Vessel 1 due to its low solubility in pH 4.5 HCl. The ending pH of the aqueous phase was 6.7. Simulation of Plasma Concentrations Based on Absorption Profile from GTBD. The absorption profile from GTBD was used to simulate the plasma concentration profile using the following equation and Microsoft Excel software.

Cp () =

( )

( − )

(1)

where Cp is the plasma concentration, D is the dose, F is the fraction absorbed, Vd is the volume of distribution, Ke is the elimination rate constant, and Ka is the absorption rate constant. The percentage of drug absorbed in the organic phase in GTBD was used as F. Dissolution Equipment. Multiple units of USP apparatus 2 were used, including Agilent model 708-DS with 8000 dissolution sampling station (Agilent Technologies, Santa Clara, CA) and Distek models Symphony 7100 with Evolution 4300 autosamplers (Distek, Inc. North Brunswick, NJ). Distek ezfill 4500 was used for dissolution media preheating, deaeration, and automated dispensing. Samples of 1.5 mL were taken at various time points with automated sampling, with a 10 µm filter for each sampling line. A pH meter with pH/ATC Accument AB150 electrode (Fisher Scientific, Hampton, NH) was used to measure the pH of dissolution media before and after dissolution. HPLC Instrument and Method. Agilent 1200 HPLC systems were used for sample analysis. Empower 3 chromatographic data system (Waters Corporation, Milford, MA) was used for instrument control, data acquisition and processing. The HPLC method used an Agilent Poroshell EC-C18 column (150 mm x 3.0 mm, 2.7 µm particle size), an isocratic mobile phase of 35:65=10 mM ammonium formate at pH 3.7: acetonitrile at a flow rate of 0.6 mL/min, UV detection at 245 nm, and an injection volume of 1 µL. The column temperature and autosampler temperature were 40ºC and ambient, respectively. Quantitative analysis of the dissolution samples was performed against duplicate standard solutions of Compound-A, with the second standard as a QC control. Filter compatibility with API was evaluated and found to have no impact on recovery. Powder X-Ray Diffraction (PXRD). Precipitate from the 2-stage dissolution was collected by filtration on Whatman® qualitative filter paper, Grade 602 h,

In Vitro, in Silico, and in Vivo Assessments of Intestinal Precipitation

Recommend Documents