Track: Formulation and Delivery - Chemical - Biopharmaceutics - Predictive Modeling
Category: Poster Abstract
Understanding In Vivo Drug Dissolution of Indomethacin Immediate Release Drug Products to Guide Biopredictive In Vitro Dissolution Method Development
Purpose: Indomethacin, a nonsteroidal anti-inflammatory drug (NSAID) indicated for treatment of mild to moderate acute pain, is a biopharmaceutical classification system (BCS) class 2 (i.e. low solubility and high permeability) weak acid drug with pKa 4.5. The solubility of indomethacin is dependent on the pH in the gastrointestinal tract (GIT), which ranges from a low of pH of 1.2 to high of pH of 7.4. As the permeability is high, indicated by systemic bioavailability of oral drug products (~100%), dissolution is the rate-limiting step for the absorption of indomethacin immediate release (IR) solid oral drug products. Currently implemented in vitro dissolution tests for indomethacin IR products employ compendial methods using the paddle or basket apparatus. The in vitro dissolution data of indomethacin IR products exhibits a rapid dissolution, i.e., 85% (mean) dissolved in 30 minutes. The discriminating ability of the compendial methods towards formulation/manufacturing and establishment of an in vivo-in vitro relationship are often inadequate. This research is aimed to understand in vivo drug absorption/dissolution of indomethacin IR products for providing insight for development of bio-predictive in vitro dissolution methods. Methods: The drug plasma concentration-time profiles after administration of intravenous indomethacin was extracted from literature, and pharmacokinetics profile of the selected IR indomethacin drug products in healthy volunteers were collected from Drugs@FDA. Numerical deconvolution was conducted based on the mean concentration-plasma profile with Phoenix 64 IVIVC module (Certara, L.P.). A unit impulse response (UIR) function was generated using indomethacin in vivo pharmacokinetic profile after intravenous administration. The fraction of absorbed indomethacin as a function of time was obtained for each IR drug product. Tested formulations include indomethacin capsules as the reference drug product (product 1) with the strength of 50 mg, a submicron indomethacin drug product (product 2) with doses of 20 mg and 40 mg. The in vivo drug absorption profiles of IR indomethacin under fasting and fed conditions were also obtained. Results: A UIR function of Cδ(t) = 75.95e12.95t + 89.52e1.08t was generated with the intravenous PK data in the healthy volunteers at a dose of 50 mg. The fraction of absorbed indomethacin-time profiles of IR drug products showed that it took approximately 3 hours for 85% of absorption. As the permeability of indomethacin is high, the in vivo absorption profiles can represent the in vivo drug dissolution. The absorption fraction figure showed a biphasic in vivo dissolution curve with lag time of 15-20 minutes. The total absorbed fraction showed no significant differences in formulations with different doses, however, the micronized indomethacin drug product showed a faster drug absorption than the reference indomethacin capsule product. With food, product 2 showed a slower in vivo absorption and a lower total fraction of absorption compared to that of administered in fasting condition. Conclusion: The in vivo fraction of absorption-time profile of indomethacin from IR products indicates a slow drug dissolution in the gastrointestinal tract, which is not consistent with the results of the in vitro dissolution testing. Further, in vivo dissolution profile differentiates the submicron indomethacin product and the reference indomethacin capsule. The in vivo dissolution of the submicron indomethacin product in fed condition exhibits a slower absorption with a longer lag time compared to that in fasting condition. In vitro dissolution testing fails to show such differences. The discrepancy of in vitro dissolution studies and in vivo performance might be due to the medium of in vitro dissolution testing which provides higher solubility than the in vivo GIT environment as well as the differences of hydrodynamics between in vitro and in vivo. Exploration of biorelevant dissolution medium and/or in vitro testing conditions which can generate in vitro profiles resembling in vivo dissolution profiles are suggested for future biopredictive dissolution method development.