Predictive Model to Determine the Aqueous Solubility of BCS Class 4 Drugs in Amorphous Solid Dispersions
Poster Number
19B
Introduction/Abstract
Solubility is one of the key factors to achieve desired drug absorption after oral administration. Therefore, poor aqueous solubility is the major challenge for a drug to be delivered orally[1]. To improve the oral absorption, various solubility enhancement techniques are used to improve solubility of drugs with limited aqueous solubility. The poor aqueous soluble drugs as amorphous solid dispersions (ASDs) is being widely studied and successfully used in several products due to higher solubility of amorphous form of drug. A molecular dispersion of an amorphous drug in desired polymer can alter the dissolution rate and solubility leading to improved drug absorption. Experimentally determining the enhanced solubility is time consuming, laborious and might end up having no significant results. Predictive models improve the efficiency of formulation development or provide a direction for further formulation studies [2]. The present work focuses on the development of an empirical model to predict solubility of physically modified BCS (Biopharmaceutical classification system) class 4 drugs.
Purpose
Experimentally determining the enhanced solubility is time consuming, laborious and might end up having no significant results. Predictive models improve the efficiency of formulation development or provide a direction for further formulation studies.
Method
The ASDs of the five BCS class 4 drugs (acetazolamide, chlorothiazide, furosemide, hydrochlorothiazide and sulfamethoxazole) and four water-soluble polymers (polyvinylpyrrolidone, copovidone, hydroxypropyl methyl cellulose and Soluplus) were prepared by hotmelt process. The equilibrium aqueous solubility of ASDs was determined by shake flask method. The amorphous nature of drug was analyzed by differential scanning calorimetry using SII EXSTAR 6000 DSC and drug polymer interactions were studied by FTIR analysis using Thermo Scientific Nicolet Summit Pro. Multiple linear regression was used to develop solubility predictive model, using statistical software SPSS 28.0.1.1(15) version, and validated by leave one out method.
Results
ASDs of all compounds have exhibited significant (P
Regression analysis of molecular descriptors of the drug and polymer as predictor variables and experimental solubility of ASDs as response variable resulted in the following equation and was validated by leave one out method.
Log Solubility = 5.094 + (0.024*polymer ratio) + (0.001*polymer HBA) + (0.017*Δ solubility parameter) – (0.006*mol.wt drug) – (0.004*melting point drug) – (0.196* drug rotatable bonds)
The present model has an R2 of 0.833 and P value less than 0.001 and the log experimental vs predicted solubility plot shows an R value 0.9.
Significance
The empirical model developed serves as a predicting tool for the BCS class 4 drugs solubility in ASDs with an accuracy greater than 80%.
Location
Library and Learning Center, 3601 Pacific Ave., Stockton, CA 95211
Format
Poster Presentation
Predictive Model to Determine the Aqueous Solubility of BCS Class 4 Drugs in Amorphous Solid Dispersions
Library and Learning Center, 3601 Pacific Ave., Stockton, CA 95211
Solubility is one of the key factors to achieve desired drug absorption after oral administration. Therefore, poor aqueous solubility is the major challenge for a drug to be delivered orally[1]. To improve the oral absorption, various solubility enhancement techniques are used to improve solubility of drugs with limited aqueous solubility. The poor aqueous soluble drugs as amorphous solid dispersions (ASDs) is being widely studied and successfully used in several products due to higher solubility of amorphous form of drug. A molecular dispersion of an amorphous drug in desired polymer can alter the dissolution rate and solubility leading to improved drug absorption. Experimentally determining the enhanced solubility is time consuming, laborious and might end up having no significant results. Predictive models improve the efficiency of formulation development or provide a direction for further formulation studies [2]. The present work focuses on the development of an empirical model to predict solubility of physically modified BCS (Biopharmaceutical classification system) class 4 drugs.
