(W0930-01-04) Spray Drying Versus Coprecipitation for Processing Amorphous Solid Dispersions

Purpose: Around 40% of active pharmaceutical ingredients (APIs) on the market and 90% under development have poor aqueous solubility¹. The poor solubility of compounds may be associated with poor bioavailability and insufficient dose uptake². Formulation of an amorphous solid dispersion (ASD) consisting of API dispersed in a hydrophilic polymer may increase its solubility³. The aim of this project is to increase the solubility of compounds by formulation of ASDs produced by different manufacturing techniques. The impact of different manufacturing methods on the final critical quality attributes of the obtained powders is investigated, focusing on their physical characteristics and stability.

Methods: Two solvent based methods, namely coprecipitation (CP) and spray-drying (SD), were employed for processing combinations of the model API, hydrochlorothiazide (HCTZ) with one of two different polymers, PVP VA 64 or Soluplus®. Drug:polymer ratios of 3:7 (sample CP 1, SD 1 with PVP VA 64 / CP 3, SD 3 with Soluplus®) and 4:6 (sample CP 2, SD 2 with PVP VA 64 / CP 4, SD 4 with Soluplus®) were used. The spray dried samples of API and polymer were prepared by dissolving in a solvent composed of 95% (v/v) ethanol and 5% (v/v) deionized water. Solutions were spray-dried using a Büchi B-290 Mini spray dryer (BÜCHI Labortechnik AG, Flawil, Switzerland). Coprecipitated samples were prepared by dissolving API and polymer in ethanol at a temperature of 74°C and stirring at a rate of 400 rpm. The solution was added to antisolvent (hexane) at 10 °C in an EasyMax^TM (Mettler Toledo, Columbus, Ohio, United States) vessel in a 1:4 solvent: antisolvent ratio at a rate of 5 ml/min with continuous overhead stirring at 600 rpm. The aging time was 5 minutes in situ. The obtained CP was isolated by filtration and dried in a vacuum oven. The solid-state characteristics of materials obtained by the two methods, CP and SD, were analysed by mDSC, pXRD and FTIR. Samples were stored at 25^oC/75% RH and solid state characteristics analysed again after one month’s storage. Drug content was determined by HPLC, which was also used to assay samples from dynamic solubility studies, conducted in 0.1 M HCl (pH 1.2) over three hours.

Results: The experimental drug content did not differ significantly from theoretical values (Table 1). Freshly prepared samples and those kept for one month at 25^oC/75% RH all remained pXRD amorphous despite the high drug loading of the samples. The Tg of each of the obtained powders was higher than the Tg of their single components. The Tg values of the processed samples were compared to the theoretical values based on the weight fraction of components and their individual Tgs, using the Gordon-Taylor equation (Fig. 1). The positive deviation from the theoretical values suggests the presence of intermolecular interactions between drug and polymer. These interactions were also confirmed by FTIR spectra, where shifts signifying hydrogen bond formation were observed (Table 2). In dynamic solubility studies, all HCTZ-polymer systems demonstrated supersaturation relative to crystalline HCTZ. The yields obtained for each of the systems were: 65.14 (± 8.3)%, 66.20 (± 6.23)%, 57.00 (± 6.8)%, 62.5 (± 7.89)%, 71.00 (± 2.00)%, 72.67 (± 2.52)%, 65.33 (± 3.36)%, 67.67 (± 3.51)%, for systems CP 1, CP 2, CP 3, CP 4, SD 1, SD 2, SD 3, SD 4 respectively.

Conclusion: The solid-state characteristics and drug content of samples prepared by SD and CP methods are similar. Yields for both processes were also similar and in the range 57-75%, although CP yields were more variable. SD and CP methods both produced amorphous samples in the systems with PVP VA 64 or Soluplus®, which stayed amorphous after 4 weeks of a physical stability study at 25^oC and 75% RH. The physical stability of the ASDs prepared may be attributed to hydrogen bonding between the API and polymer, which stabilizes the amorphous form of the API.

References: 1. Loftsson, T.; Brewster, M. E., J Pharm Pharmacol 2010, 62 (11), 1607-21.
2. Abuhelwa, A. Y.; Williams, D. B.; Upton, R. N.; Foster, D. J., Eur J Pharm Biopharm 2017, 112, 234-248.
3. Iyer, R.; Petrovska Jovanovska, V.; Berginc, K.; Jaklic, M.; Fabiani, F.; Harlacher, C.; Huzjak, T.; Sanchez-Felix, M. V., Pharmaceutics 2021, 13 , 1682.

Acknowledgements: This work was supported by Science Foundation Ireland (SFI) 18/EPSRC-CDT/3587 and the Engineering and Physical Sciences Research Council EP/S023054/1. It was part funded by SFI and co-funded under the European Regional Development Fund 12/RC/2275.


Figure 1. Glass transition temperature (Tg) predicted by Gordon-Taylor equation (red and yellow lines) and experimentally obtained values (dots).


Table 1 . Theoretical and actual API and polymer content of SD and CP systems.


Table 2. Infrared peak assignments for physical mixtures (PM) and CP, SD samples.