(M1030-07-45) Effect of Drug Properties on the Formation of Amorphous Solid Dispersion

Purpose: Amorphous solid dispersions (ASDs) have been successfully commercialized to deliver poorly water-soluble drugs. Despite their increasing market prevalence, the mechanisms involved in ASD formation are not completely understood. While most of the current research is focused on understanding the factors affecting the physical stability and the dissolution behavior of the ASDs, the current study aims to understand the factors affecting the formation of ASDs. In this study, several physicochemical properties of the drugs, which were previously studied individually for their effects on the physical stability of ASDs, have been evaluated for their collective impact on the ASD formation of 15 drugs.

Methods: ASDs of 15 drugs at 15% w/w and 75% w/w were prepared with polyvinylpyrrolidone vinyl acetate (PVPVA) co-polymer. Dispersibility of each drug in PVPVA, defined as the formation of a single homogeneous amorphous phase persisting 1 day after preparation, was inferred from powder X-ray diffraction, differential scanning calorimetry, and polarized light hot stage microscopy data. The Flory-Huggins interaction parameter was determined using the dissolution endpoint method.¹ Crystallization tendency was determined using the DSC method described by Baird et. al.² The amorphous density of each molecule was taken from Bookwala et. al.³ R3m is a molecular descriptor previously found to be predictive of dispersion formation, and its values were taken from DeBoyace et. al.⁴. The R3m value of a molecule is the sum of R3m value of each atom at d (3; d_ij). The R3m values of atoms heavier than C are considered to calculate the percentage contribution of only electronegative atoms. The drug properties have been categorized for ease of comparison with dispersibility outcomes. The drug parameters were classified into two categories: A (unfavorable for dispersion formation) and B (favorable for dispersion formation) based on the cutoff values described in Table 1.

Results: The categorized drug properties and the observed dispersibility in PVPVA at 15% w/w and 75% w/w drug loadings are shown in Table 2. The first class of drugs cannot form ASDs with PVPVA, even at low drug loadings. These drugs have only 1 or 2 favorable drug properties. An example of this class of drugs is sulfanilamide. Although sulfanilamide has the highest amorphous density and contribution of electronegative atoms to R3m in the studied drug library, it still did not form a homogeneous ASD. This indicates that even though sulfanilamide has the atom types consistent with drug-polymer interactions, the drug molecules tend to interact with themselves and recrystallize. This behavior of sulfanilamide could be due to its high crystallization tendency and low molecular weight, which results in higher mobility, leading to recrystallization during ASD formation. The second class of drugs can form ASDs with PVPVA only at low drug loadings. These drugs have 2 or 3 favorable drug properties. An example is cimetidine, which has a low crystallization tendency and a high electronegative atom contribution to R3m. These factors might help cimetidine form ASD at lower drug loading due to high amorphous stability and higher drug-polymer interactions. The third class of drugs can form ASDs with PVPVA even at high drug loadings. These drugs have 3 or more favorable drug properties. An example of this class of drugs is ketoconazole. It has a high R3m value, contribution of electronegative atoms to R3m, molecular weight, and amorphous density, and the lowest crystallization tendency. Additionally, ketoconazole is a long linear molecule. All these drug properties of ketoconazole lower its mobility and aid its retention in the polymer matrix without recrystallization. If only one drug property is considered to predict ASD formation, it leads to multiple exceptions. For example, if we only consider the interaction parameter, although ketoconazole and itraconazole have the lowest interaction parameter of all 15 drugs, they still form ASDs even at higher drug loading. Therefore, in the current work, we utilized multiple drug properties to study their combined effect on ASD formation.

Conclusion: This study emphasizes that the formation of a homogeneous ASD depends not on a single property but on a combination of drug properties. Dispersibility is facilitated by the drug’s ability to interact with the polymer and the retention of heavier and denser molecules in the polymer matrix. It is possible to form homogeneous ASDs even if some drug properties do not facilitate drug-polymer interaction, if the drug loading is low.

References: 1. Tao J, Sun Y, Zhang GG, Yu L 2009. Solubility of small-molecule crystals in polymers: D-mannitol in PVP, indomethacin in PVP/VA, and nifedipine in PVP/VA. Pharmaceutical research 26:855-864.
2. Baird JA, Van Eerdenbrugh B, Taylor LS 2010. A classification system to assess the crystallization tendency of organic molecules from undercooled melts. Journal of pharmaceutical sciences 99(9):3787-3806.
3. Bookwala M, DeBoyace K, Buckner IS, Wildfong PLD 2020. Predicting density of amorphous solid materials using molecular dynamics simulation. AAPS PharmSciTech 21:1-11.
4. DeBoyace K, Buckner IS, Gong Y, Ju T-cR, Wildfong PLD 2018. Modeling and prediction of drug dispersability in polyvinylpyrrolidone-vinyl acetate copolymer using a molecular descriptor. Journal of Pharmaceutical Sciences 107(1):334-343.


Table 1: Categorization cutoff criteria for drug properties


Table 2: Categorized (A: unfavorable for dispersion formation and B: favorable for dispersion formation) drug properties and dispersibility outcomes of 15 drugs at 15% w/w and 75% w/w drug loading