Analysis of the moisture sorption behavior of amorphous drug–polymer blends

Authors

  • Alfred C. F. Rumondor,

    1. Department of Industrial and Physical Pharmacy, School of Pharmacy, Purdue University, West Lafayette, Indiana
    2. Pharmaceutical and Analytical Research Development, AstraZeneca Pharmaceuticals LP, Wilmington, Delaware
    Search for more papers by this author
  • Hajime Konno,

    1. Department of Industrial and Physical Pharmacy, School of Pharmacy, Purdue University, West Lafayette, Indiana
    2. Pharmaceutical Research and Technology Laboratories, Astellas Pharma, Inc., Yaizu, Shizuoka, Japan
    Search for more papers by this author
  • Patrick J. Marsac,

    1. Department of Industrial and Physical Pharmacy, School of Pharmacy, Purdue University, West Lafayette, Indiana
    2. Materials Characterization and Technology Assessment, Merck & Co., Inc., West Point, Pennsylvania
    Search for more papers by this author
  • Lynne S. Taylor

    Corresponding author
    1. Department of Industrial and Physical Pharmacy, School of Pharmacy, Purdue University, West Lafayette, Indiana
    • Department of Industrial and Physical Pharmacy, School of Pharmacy, Purdue University, West Lafayette, Indiana
    Search for more papers by this author

Abstract

Hydrophobic drugs are often formulated with hydrophilic polymers to form miscible blends called amorphous solid dispersions. The interaction of moisture with these blends is an important topic, both from stability as well as processing perspectives. In this study, the moisture sorption profiles of four different drug–polymer blends, [felodipine–poly(vinylpyrrolidone) (PVP), indomethacin–PVP, felodipine–hypromellose (HPMC), and felodipine–hypromellose acetate succinate (HPMCAS)] were experimentally determined at 25°C, and analyzed using various mathematical models. It was found that the moisture sorption profiles of the drug–polymer blends could not be reconstructed using the weight-averaged sum of the moisture sorbed by each of the components. Application of the Flory–Huggins model for ternary systems to extract drug–polymer interaction parameter (χ23) values using known values of water–drug and water–polymer interaction parameters led to ambiguous conclusions about the systems' thermodynamics. χ23 values extracted for felodipine–PVP and indomethacin–PVP using this model ranged from −9.6 to 26.9 and −20.4 to 22.0, respectively. It is thought that the presence of specific drug–polymer interactions changed the water–drug and the water–polymer interactions in the system. Combined with the mathematically small contribution from the term encompassing χ23 to the predicted amount of moisture sorbed by the drug–polymer blends, it was concluded that this method cannot be used to unambiguously determine drug–polymer interaction parameters in solid dispersions. Instead, a model with a mean interaction parameter (χ1,23) that considers the drug and the polymer in the blend as a single unit was found to better describe the changing affinity of water for the solid matrix with a change in composition or polymer type. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2010

Ancillary