Impact of dissolution medium pH and ionization state of the drug on the release performance of amorphous solid dispersions.

Abstract

Amorphous solid dispersions (ASDs) are widely employed as a strategy to improve oral bioavailability of poorly water soluble compounds. Typically, optimal dissolution performance from a polyvinylpyrrolidone vinyl acetate (PVPVA) based ASD is observed at relatively low drug loading limit. Above a certain drug load, termed limit of congruency (LoC), the release from ASDs significantly decreases. So far, the majority of the dissolution behavior has been tested in conditions where the drug primarily exists in unionized form. In this work, the impact of pH of the dissolution environment on the release performance of ASDs of an ionizable drug was studied. Atazanavir (ATZ), a weakly basic drug with a pK_a of 4.5 was used as a model compound and PVPVA was used as a non-ionizable matrix polymer. Dissolution rate was measured using Wood's apparatus which normalizes the surface area of the dissolving tablet. The pH of the dissolution media was varied between 1 and 6.8, cover a range where ATZ exists as >99% ionized or unionized species. At pH 6.8, near complete release was observed only when the drug load was ≤ 6%. Unlike typically observed drastic decline in release behavior for PVPVA based ASDs above LoC, ATZ ASDs underwent gradual decline in dissolution behavior when the DL was increased to 8%. This was attributed to potential formation of an ATZ-PVPVA associated phase with dissolution rate slower than neat PVPVA. However, the 10% DL ASD showed negligible ATZ release. On another extreme (pH 1) where ATZ is ∼100% ionized, the dissolution rate of ATZ was faster than that of PVPVA. ASD dissolution rate was found to be slower than that of the neat drug but faster than PVPVA and interestingly, did not change with DL. This can be attributed to formation of an ionized ATZ-PVPVA phase which controls the dissolution rate of the ASD. At pH 3, where the drug is ∼97% ionized, near complete release was observed for drug loads ≤ 8%. To observe significant increase in drug loading with near complete release, >98% ionization of ATZ was required. At pH 2 where ATZ is ∼99.7% ionized, near complete release was observed for drug loads up to 30%. Furthermore, the deterioration in dissolution performance with an increase in drug load continued to be gradual at pH 2. The enhancement in dissolution performance did not correlate with solubility enhancement of ATZ due to ionization. We theorize that the enhancement in the dissolution performance due to ionization is the result of formation of an ionized ATZ-PVPVA phase which increases the hydrophilicity and the miscibility of the ASD. This can help resist water induced phase separation during ASD dissolution and therefore, result in continuous, and congruent dissolution of the drug and polymer.