Statistical Evaluation of the Factors Impacting In Vitro Drug Dissolution
Results from the PLS-R analyses, conducted to understand the vehicle impact on the dissolution of the two drugs, are shown in Fig. 10.
Fig. 10 Standardised coefficients corresponding to the variables studied for dissolution of montelukast, mesalazine and both drugs. Colour denotes coefficients with a moderate (lighter colour) and significant (darker colour) impact on the response (VIP > 0.7 and 1, respectively). (B.C., buffer capacity; ST, surface tension)
For the dissolution of montelukast from the granules and crushed chewable tablets, the PLS-R model developed was defined by three components and showed a good fit to the experimental values (R2 = 0.83) and a good predictive power (Q2 = 0.79). The statistical analysis revealed that vehicle pH and percentage of fat and drug solubility in each vehicle were the factors with the most significant positive impact on drug dissolution from the two montelukast formulations tested, with a moderate negative impact from vehicle osmolality, viscosity, percentage of sugar and of protein.
For dissolution of mesalazine, the model constructed was defined by two components and showed a good fit to the experimental values (R2 = 0.70) and a good prediction power (Q2 = 0.64). PLS-R analysis revealed that the type of formulation was the factor with the most significant positive impact on drug dissolution, whilst significant negative effects from vehicle viscosity, percentage of protein and buffer capacity were observed. Moderate negative effects from vehicle osmolality, percentage of sugars and drug solubility were also observed.
For dissolution of both drugs (all formulations), the PLS-R model built was defined by six compartments, had good predictive power and showed a good fit to the experimental values (Q2 = 0.62 and R2 = 0.70, respectively). PLS-R analysis showed that the drug characteristics (logP and ionisation percentage) and the pH of the vehicles were the factors with a significant negative effect on drug dissolution. A moderate negative effect from vehicle buffer capacity was also observed. In contrast, significant positive effects from the type of formulation, drug solubility in each vehicle and percentage of fat of the vehicle were observed. Moderate positive effects from vehicle osmolality, viscosity and percentage of protein were also observed.
Overall, PLS-R results showed that knowledge of the physicochemical properties and macronutrient composition of the food and drinks and drug/formulation properties can help understand the potential vehicle-impact on drug dissolution. This impact should be taken into consideration during compatibility assessments of the vehicle-drug product and could be used to predict potential alterations on drug product behaviour.