About 30% of commercial and developmental drugs exhibit poor solubility and thus poorbioavailability. Strategies that enhance solubility of such compounds have become morepopular. Cocrystallisation is one of these strategies, so characterisation of in vitroperformance of cocrystals is essential. Conventional dissolution systems (USP apparatus 1and USP apparatus 2) are often not suitable for testing poorly soluble drugs due to failure inproviding sink conditions and inability to change the media during the experiment. Thisproject involves designing appropriate dissolution methods that will help understand themechanism of dissolution of cocrystals. Successful dissolution methods were used to testmarketed formulations of Carbamazepine (CBZ) (IR and PR Tegretol® tablets) andIndomethacin (IND) (IR Indocid® capsules). Flow-through cell (USP 4) apparatus proved tobe more suitable to test poorly soluble formulations than basket (USP 1) apparatus.Formulations and their cocrystals were tested in four combinations of media: compendial(SGF/SIF), modified I (MGM/MIM-I), modified (MGM/MIM-II) and biorelevant media(FaSSGF/FaSSIF-V2). USP apparatus 4 allowed capturing and quantifying the precipitationof IND samples upon the media change. All Indomethacin formulations exhibitedprecipitation; however, this occurred into the smaller extent in biorelevant media. The greatestenhancement in CBZ dissolution was observed for Saccharin cocrystals of CBZ. On the otherhand, Nicotinamide cocrystal of IND improved dissolution of IND greater than Saccharincocrystal. Dissolution profiles with physicochemical and pharmacokinetic parameters wereused to develop a Physiologically Based Pharmacokinetic (PBPK) model using in silicoprogram Simcyp®. Successful models were then used to predict in vivo performance ofcocrystals. Successful PBPK models were developed for IR and PR formulations of Tegretol®tablets and IR Indocid® capsules using dissolution data tested in biorelevant media usingdissolution USP apparatus 4. It was found that in vivo absorption of CBZ from cocrystals didnot increase dramatically. An overall 2 % increment was observed when compared against theCBZsp sample. However, the rate of absorption for CBZ-SACss samples was significantlyfaster than the rate of CBZsp absorption. API and cocrystals of IND completely absorbedwithin 3 hours and IND-NICss cocrystal were found to absorb slightly faster than the othercocrystals. Absorption was 8% higher at 1 h in comparison to IND. However, overall theimprovement was not statistically significant. In summary, this research demonstrates thatselection of appropriate medium and apparatus is essential to build a successful PBPK model.
|Date of Award
|31 Dec 2013
|Nikoletta Fotaki (Supervisor) & Robert Price (Supervisor)