Pheroid® technology in the transdermal delivery of selected non-steroidal anti-inflammatory drugs
Drug delivery through the skin still remains an area that enjoys active research with much of the research focusing on physical or chemical methods to reversibly alter the skin’s permeability to compounds because of the skin’s very low permeability. Vesicular carriers are one of the many chemical approaches used in order to deliver drugs into and sometimes through the skin. A review article offers an overview of various vesicles that have been investigated during dermal and transdermal drug delivery research, with special emphasis on a relatively new carrier, namely the Pheroid™. Based on the review of previous work that was done it was found that it is still unclear whether an API requires certain physicochemical characteristics in order for the Pheroid™ to enhance its permeation or not. It was also observed that the type of formulation had a significant impact on the API’s transdermal delivery behaviour. To aid in this investigation of the optimal physicochemical properties for the transdermal delivery of selected non-steroidal anti-inflammatory drugs (NSAIDs) dispersed in the Pheroid™ delivery system, a novel flux-independent mathematical model was derived and tested. The model is based on restrictions made to Lipinski’s rule of five. The results suggested that the same molecular size and log P (octanol-water partition coefficient) ranges that had determined the skin permeability of an API having been dissolved in PBS, also determined its permeability when dissolved in Pheroid™, and that these ranges were consistent with the previously described restrictions that should be placed on Lipinski’s rule of five, to account for the formidable resistance offered by the skin’s stratum corneum. The model gave reasonable approximations of the experimentally observed concentrations. The results suggested that certain restrictions placed on Lipinski’s rule of five could be used to accurately model transdermal delivery. The ability of Bayesian networks to determine the correct probabilistic dependencies between skin permeability and the physicochemical properties of selected drugs dissolved in PBS at pH 7.4 and dispersed in a lipid-based drug delivery system was also investigated. The networks identified a probabilistic dependence between pKa and skin permeability for drugs dissolved in PBS (pH 7.4), which was not observed for the same drugs dispersed in the lipid micelles of the Pheroid™ drug delivery system. For both dissolved and dispersed drugs the same probabilistic dependencies existed between topological polar surface area (TPSA), melting point (MP) and cumulative amount of drug dissolved in PBS (pH 7.4) (CPBS), which permeated the skin after 12 h, in comparison to the cumulative amount of drug dispersed in Pheroid™ (CPher). Although both networks shared a causal relationship between permeability and molecular size descriptors, the network determined from the dissolved drugs displayed a probabilistic dependence between molecular weight (MW) and permeability, while the dispersed drugs displayed dependence between molecular volume (MV) and permeability. Both networks identified similar physicochemical regions for optimal transdermal delivery, i.e. low MP, small TPSA to molecular size ratio and, in the case of the dissolved drugs, a pKa value close to the pH of the buffer solution. Regions for poor transdermal delivery were also identified, i.e. high MP and a TPSA to molecular size ratio close to 0.5. The results suggest that Bayesian networks determined from online bioinformatics and cheminformatics databases can be viable classification tools in early drug discovery and development, and can aid in the identification of suitable drug candidates and formulation strategies. Exploratory data analysis of the dependencies between skin permeability, MW and log P was also investigated as they remain the most frequently used physicochemical properties in models that predict skin permeability. The results suggest that, in general, MW and log P are poorly correlated to log Kp (permeability coefficient). However, after employing several exploratory data analysis techniques, regions within the dataset of statistically significant dependencies were identified. As an example of the possible applicability of the information extracted from the exploratory data analyses, a multiple linear regression model was constructed, bounded by the ranges of dependence. This model gave reasonable approximations to log Kp values obtained from skin permeability studies of selected NSAIDs administered from a buffer solution and a lipid-based drug delivery system. Knowing the ranges within which molecular weight and log P are statistically related to log Kp can supplement existing methods of screening, risk analysis or early drug development decision making to add confidence to predictions made regarding skin permeability.
- Health Sciences