This study investigated the effects of drug recrystallization on the in vitro performance of testosterone drug-in-adhesive transdermal delivery system (TDS). Six formulations were prepared with a range of dry drug loading in the adhesive matrix from 1% to 10% w/w with the aim of generating TDS with various levels of drug crystals. We visually quantified the amount of crystals in TDS by polarized light microscopy. The effect of drug recrystallization on adhesion, tackiness, cohesive strength, viscoelasticity, drug release, and drug permeation through human cadaver skin were evaluated for these TDS samples. The Optical images showed no crystals in 1% and 2% testosterone TDSs; however, the amount of crystals increased by increasing testosterone loading from 4 to 10%. A proportional and significant decrease (p < 0.05) in tack, peel, and shear strength of the adhesive matrix with increasing amount of crystals in TDS was observed. The drug crystals resulted in a proportional deterioration of the viscoelastic properties of the adhesive matrix. The 2% testosterone TDS showed faster drug release rate when compared to 1% testosterone TDS. The increase in drug loading from 2% to 4% w/w slightly increased the cumulative amount of testosterone released. Further increase in drug loading in TDS to 6, 8, and 10% was nonsignificant (p > 0.05) to affect the drug release and permeation. In conclusion, this study demonstrated that the extent of drug recrystallization can be quantitatively correlated with the deterioration of performance characteristics of TDS products.

OBJECTIVE: To test whether a side-by-side diffusion model is suitable for studying drug supersaturation in an absorptive environment.
METHODS: The µD/P model and the µFLUX model, using a Caco-2 cell monolayer/PAMPA membrane as the permeation barrier, respectively, were compared in terms of robustness and ease of handling, while studying the drug supersaturation-precipitation-permeation interplay. Continuing with the best model, the impact of the acceptor media and the importance of studying drug supersaturation in a combined dissolution-permeation model, as compared to a simple dissolution model, were evaluated.
RESULTS: The two models produced similar results in terms of supersaturation, precipitation and permeation. The µFLUX model was considered more robust and easier to handle based on its cell-free permeation system. Using the µFLUX model, it was found that an acceptor medium with a high surfactant concentration increased the amount of permeated drug. The effect of absorption on drug supersaturation was found to be dependent on the drug, and the tested level of supersaturation.
CONCLUSIONS: The tested models were comparable; however, Caco-2 cell monolayers were considered too sensitive to be used to study drug supersaturation. Further studies are needed to evaluate the observed drug-dependent effects of absorption on drug supersaturation.