(T1330-07-37) Coexistent API-Polymer Halogen and Hydrogen Bonds in Amorphous Solid Dispersions Affect Drug Solubility and Miscibility

Purpose: Poorly water-soluble drugs can potentially be formulated as amorphous solid dispersions (ASDs), allowing improved API apparent aqueous solubility. In glass solutions, the API can be stabilized by non-covalent interactions with the polymer, which reduces the potential for recrystallization and/or restricts diffusion in the highly viscous carrier polymer. API-polymer solubility and miscibility are strongly associated with ASD physical stability and are directly influenced by non-covalent drug-polymer interactions. In this research, we investigated the simultaneous formation of drug-polymer halogen and hydrogen bonds in a chlorpropamide-PVPva glass solution and evaluated how these coexistent interactions affected the solubility and miscibility of the API in the co-polymer. Results were compared with tolbutamide-PVPva, which can only form hydrogen bonds with the co-polymer.

Methods: Chlorpropamide and tolbutamide were prepared as ASDs in polyvinylpyrrolidone-vinyl acetate co-polymer (PVPva) via spray drying at 10, 20, 30, 50, and 60 or 70% w/w drug-loading. Confirmation of stable dispersion formation was done using powder x-ray diffraction, differential scanning calorimetry, and polarized light microscopy after 24 h of ambient temperature vacuum drying of samples. Raman and Fourier transform infrared (FTIR) spectroscopy were used to characterize ASDs, physical mixtures of crystalline drug and amorphous polymer, and pure amorphous drugs. The API-polymer solubility and the Flory-Huggins interaction parameter (χ) were determined using well-established dissolution endpoint¹ and melting point depression² methods, respectively.

Results: Chlorpropamide and tolbutamide were selected, owing to their close structural similarity. The C-Cl bond in chlorpropamide was identified in the Raman spectrum and was wider for amorphous chlorpropamide relative to the crystalline solid, with a peak maximum for both forms at 756 cm^-1 (Figure 1). When the same spectral region was probed for ASDs made using five different concentrations of chlorpropamide, this maximum shifted to lower wavenumbers (Figure 1). The largest peak maximum shift to lower Raman values was observed for the ASD containing 10% drug (753 cm^-1); with increasing drug-loadings resulting in peak maximum shifts to increasing values (i.e., closer to the values for the pure amorphous drug). No peak maximum shifts were observed for any of the physical mixtures, reinforcing that the chemical environment around the Cl was changed in the ASD by formation of halogen bonds between chlorpropamide and PVPva. These results are contrasted for tolbutamide ASD, where the peripheral Cl atom (chlorpropamide) is replaced with a CH₃ group in the same position. Since tolbutamide does not contain a halogen atom, Raman peak in the halogen region was not observed highlighting the absence of halogen bond with PVPva.

Both the chlorpropamide and tolbutamide structures contain two hydrogen bond donors at the same relative intramolecular positions. The electronic environment of the secondary amide region of ASDs for both drugs was assessed using FTIR, showing strong, concentration-dependent peak maximum shifts in ASDs relative to pure amorphous components, confirming drug-polymer hydrogen bond formation.

A temperature-composition phase diagram for both chlorpropamide and tolbutamide (Figure 2) is shown along with the Couchmann-Karasz-predicted T_g for each ASD. The solubility of chlorpropamide in PVPva was consistently higher across multiple measurements relative to its non-halogenated tolbutamide analogue. The melting point depression plot was used to identify χ, where linear relationships for PVPva concentration ranging from 0-50% w/w for chlorpropamide (R²=0.97) and tolbutamide (R²=0.99), resulted in respective values of χ=-2.08±0.1 and χ=-1.78±0.03. An independent t-test demonstrated that χ for each ASD was statistically different (p = 0.001). The increased solubility and χ for chlorpropamide was attributed to the halogen bonding interactions that it can form with PVPva in addition to hydrogen bonds, relative to tolbutamide ASDs, where only hydrogen bonds are formed between drug and co-polymer.

Conclusion: Non-covalent adhesive interactions play a key role in increasing ASD shelf-life. Hydrogen bonding between drug and polymer has been the most extensively studied specific interaction, and their formation is frequently associated with stabilizing dispersions. This work not only confirms the coexistent formation of hydrogen and halogen bonds in chlorpropamide-PVPva ASD, it further demonstrates the implications of these on solubility, and miscibility of the API in PVPva co-polymer. Formation of both halogen and hydrogen bonds between chlorpropamide and PVPva increased both the solubility in the polymer, and –χ by approximately 17% relative to a structural analogue (tolbutamide), which could only form hydrogen bonds with PVPva. These findings open a new avenue for better predicting the stabilization of ASDs.

References: 1. Tao, J.; Sun, Y.; Zhang, G. G.; Yu, L., Solubility of Small-molecule Crystals in Polymers: D-mannitol in PVP, Indomethacin in PVP/VA, and Nifedipine in PVP/VA. Pharmaceutical Research 2009, 26 (4), 855-64.
2. Marsac, P. J.; Li, T.; Taylor, L. S., Estimation of Drug-polymer Miscibility and Solubility in Amorphous Solid Dispersions using Experimentally Determined Interaction Parameters. Pharmaceutical Research 2009, 26 (1), 139-51.


Figure 1: Raman spectra showcasing region specific to the C-Cl bond in chlorpropamide. In ASD, concentration dependent peak shifts are observed indicating changes in the electronic environment of the halogen.


Figure 2: Temperature-composition phase diagram shows that chlorpropamide has consistently higher solubility in PVPva at all temperatures relative to a structural analogue tolbutamide.