David C. Sperry

Use of Artificial Stomach−Duodenum Model for Investigation of Dosing Fluid Effect on Clinical Trial Variability

Lilly Compound X (LCX) is an oncology drug that was tested in a phase I clinical study using starch blend capsules. The drug was given to a small patient population (4 patients) and showed large inter- and intra-patient variability. In order to evaluate the possible effect of stomach pH on exposure and ways to mitigate the variability issue, artificial stomach-duodenum (ASD) experiments were conducted to investigate the hypothesis that carefully selected dosing fluids would have an impact in minimizing exposure variability caused by the formulation, which could lead to more consistent evaluation of drug absorption in patients. The ASD data corroborates the observed variability, and was a good tool to investigate the effect of stomach pH and potential dosing solutions on duodenal concentrations. Administering capsules co-formulated with Captisol (10% drug load) along with Sprite was shown by the ASD to be an effective way to increase duodenal concentrations as well as to reduce the difference between duodenal concentrations for different gastric pH. The reduction in variability of duodenum AUC (in ASD) is expected to correlate well with a reduction of variability in patient exposure. The dosing regimen of Sprite/Captisol is therefore suggested for future clinical trials involving LCX. Furthermore, for design of early phase clinical trials, ASD technology can be used to assist in choosing the proper dosing solution to mitigate absorption and exposure variability issues.

Mechanism for enhanced absorption of a solid dispersion formulation of LY2300559 using the artificial stomach duodenum model.

An artificial stomach duodenum (ASD) model has been used to demonstrate the performance difference between two formulations of LY2300559, a low-solubility acidic developmental drug. The two formulations investigated were a conventional high-shear wet granulation (HSWG) formulation and a solid dispersion formulation. A pharmacokinetic study in humans demonstrated the enhanced performance of the solid dispersion formulation relative to the HSWG formulation. The Cmax and AUC of the solid dispersion was 2.6 and 1.9 times greater, respectively, compared to the HSWG formulation. In the ASD, the solid dispersion formulation performance was characterized by three main phases: (1) rapid release in the stomach, creating a supersaturated concentration of drug, (2) precipitation in the stomach, and (3) rapid redissolution of the precipitate in the duodenum to concentration levels that are supersaturated relative to crystalline drug. A series of complementary experiments were employed to describe this performance behavior mechanistically. Imaging experiments with a pH indicating dye showed that local pH gradients from meglumine in the solid dispersion formulation were responsible for creating a high initial supersaturation concentration in the stomach. Upon dissipation of meglumine, the drug precipitated in the stomach as an amorphous solid. Because the precipitated drug is in an amorphous form, it can then rapidly redissolve as it transits to the more neutral environment of the duodenum. This unexpected sequence of physical state changes gives a mechanistic explanation for the enhanced in vivo performance of the solid dispersion formulation relative to the HSWG formulation.

Dissolution Modeling of Bead Formulations and Predictions of Bioequivalence for a Highly Soluble, Highly Permeable Drug

The objective of this study was to assess the impact of observed in vitro dissolution rate differences on in vivo pharmacokinetics for two enteric-coated bead formulations of a highly soluble, highly permeable drug. A new bead dissolution model was developed to quantitatively simulate the dissolution profiles of the two formulations. The model is based on the boundary layer diffusion model and can be used to simulate dissolution profiles for bead formulations using physicochemical properties of the formulation. The model was applied to show that the observed differences in dissolution profiles can be attributed completely to the difference in surface area of the beads for the two formulations. An absorption/pharmacokinetic model (GastroPlus) was used to predict the in vivo plasma concentration time profiles for the formulations using their respective in vitro dissolution profiles as input. The simulation results showed that the plasma concentration-time profiles were not significantly impacted by slower dissolution rates. Additionally, a sensitivity analysis was performed with a range of dissolution rate profiles. The fastest dissolution rate reached 80% dissolved in 41 min, while the slowest reached 80% in 114 min. Over this range, the predicted C(max) decreased by 9% and the AUC decreased by 1%. An in vivo bioequivalence study on the two experimental formulations demonstrated the formulations were bioequivalent, consistent with predictions. The lack of sensitivity is attributable to the high permeability and long elimination half-life of the drug. The work presented in this article demonstrates the use of a bead dissolution model and an absorption/PK model to predict in vivo formulation performance.

In vitro monitoring of dissolution of an immediate release tablet by focused beam reflectance measurement.

Changes in in vitro drug release profiles of oral dosage forms are commonly observed due to storage of drug product at elevated temperature and humidity. An example is presented of an immediate release drug product which underwent changes to both release profile and crystal form on storage at elevated humidity. The dissolution rate for unstressed tablets was comparable regardless of the crystal form present. Decreased release rate was only observed for stressed tablets that exhibited crystal form conversion. The cause of the dissolution change was determined by evaluating tablets manufactured with three drug substance crystal forms by fiber optic ultraviolet detection and focused beam reflectance measurement (FBRM). Tablets were also analyzed by near-infrared spectroscopy for crystal form determination. The observed change in dissolution rate correlated with detection of a greater number of larger particles by FBRM. FBRM results indicate increased aggregation of the tablet material due to crystal form conversion, resulting in the presence of slowly disintegrating and dissolving granules during the dissolution process. The improved understanding of the dissolution process allows evaluation of the potential in vivo impact of the stability changes.

Characterization of a novel cross-linked lipase: impact of cross-linking on solubility and release from drug product.

Liprotamase is a novel non-porcine pancreatic enzyme replacement therapy containing purified biotechnology-derived lipase, protease, and amylase together with excipients in a capsule formulation. To preserve the structural integrity and biological activity of lipase (the primary drug substance) through exposure of the drug product to the low-pH gastric environment, the enzyme was processed through the use of cross-linked enzyme crystal (CLEC) technology, making the lipase-CLEC drug substance insoluble under acidic conditions but fully soluble at neutral pH and in alkaline environments. In this report we characterize the degree of cross-linking for lipase-CLEC and demonstrate its impact on lipase-CLEC solubility and release from the drug product under relevant physiological pH conditions. Cross-linked lipase-CLEC was characterized via size exclusion chromatography (SEC) and capillary electrophoresis sodium dodecyl sulfate polyacrylamide gel electrophoresis (CE-SDS-PAGE). A combination of methodologies was developed to understand the impact of cross-linking on drug product release. Dissolution evaluation using USP Apparatus 2 at pH 5.0 with an enzyme activity-based end point demonstrated solubility discrimination based on degree of cross-linking, while full release was demonstrated at pH 6.5. The dissolution of the drug product was also evaluated using a dual-stage test employing a USP Apparatus 4 flow-through system to mimic the changing pH environments experienced in the stomach and intestine to understand the impact of cross-linking on drug product performance. Use of USP Apparatus 4 to characterize the pH-dependent release of lipase-CLEC represents a novel approach compared to the Apparatus 1 test employing an acid-challenge stage outlined in the USP for delayed-release pancrelipase, and the advantages of this approach may prove useful for understanding the pH-dependence of release for other drug products. Collectively, these studies confirmed that degree of cross-linking is a critical parameter that may impact in vivo release of lipase-CLEC, and also provided a risk assessment tool for understanding the potential impact of under- and over-cross-linked drug substance.

FIP Guidelines for Dissolution Testing of Solid Oral Products

Dissolution testing is an important physiochemical test for the development of solid oral dosage forms, tablets, and capsules. As a quality control test, the dissolution test is used for assessment of drug product quality and is specified for batch release and regulatory stability studies. In vitro dissolution test results can often be correlated with the biopharmaceutical behavior of a product.This article provides a summary of views from major global agencies (Europe, Japan, United States), pharmacopoeias, academia, and industry. Based on available guidance and literature, this article summarizes highlights for development and validation of a suitable dissolution method, setting appropriate specifications, in vitro-in vivo comparison, and how to obtain a biowaiver.

Development of an in vivo-relevant drug product performance method for an amorphous solid dispersion

HighlightsA clinically meaningful in vitro dissolution method was developed.The method was developed for an amorphous spray‐dried dispersion tablet.A risk assessment focused on those attributes that could have in vivo impact.The dissolution method was discriminating for crystalline drug substance content. Abstract The purpose of this work was to develop a meaningful in vitro dissolution method for evacetrapib spray‐dried dispersion (SDD) tablets that is discriminating for crystalline drug substance (DS) content. Justification of the method conditions included evaluation of dissolution media, rotation speed, surfactant selection and level of surfactant to achieve sink conditions. Discrimination was illustrated by testing SDD tablets spiked with 10%, 20%, and 30% crystalline DS. The results demonstrated a 13%, 22% and 32% drop in the dissolution end point, respectively, as compared to unspiked SDD tablets. Additionally, tablets containing crystalline DS and tablets containing SDD were tested in a relative bioavailability (RBA) study. Utilizing the proposed dissolution method, the dissolution end point of SDD tablets was determined to be approximately 4 fold higher than that of the tablets containing crystalline DS. These results compare favourably to the in vivo RBA study results where SDD tablets had a 4.6 fold increase in exposure compared to tablets containing crystalline DS.

In Vivo Predictive Dissolution and Simulation Workshop Report: Facilitating the Development of Oral Drug Formulation and the Prediction of Oral Bioperformance

In Vivo Predictive Dissolution and Simulation Workshop Report : Facilitating the Development of Oral Drug Formulation and the Prediction of Oral Bioperformance

Physiologically Based Absorption Modeling to Design Extended-Release Clinical Products for an Ester Prodrug

ABSTRACTAbsorption modeling has demonstrated its great value in modern drug product development due to its utility in understanding and predicting in vivo performance. In this case, we integrated physiologically based modeling in the development processes to effectively design extended-release (ER) clinical products for an ester prodrug LY545694. By simulating the trial results of immediate-release products, we delineated complex pharmacokinetics due to prodrug conversion and established an absorption model to describe the clinical observations. This model suggested the prodrug has optimal biopharmaceutical properties to warrant developing an ER product. Subsequently, we incorporated release profiles of prototype ER tablets into the absorption model to simulate the in vivo performance of these products observed in an exploratory trial. The models suggested that the absorption of these ER tablets was lower than the IR products because the extended release from the formulations prevented the drug from taking advantage of the optimal absorption window. Using these models, we formed a strategy to optimize the ER product to minimize the impact of the absorption window limitation. Accurate prediction of the performance of these optimized products by modeling was confirmed in a third clinical trial.