(M1530-03-18) Impact of Manufacturing Processes of Supersaturation Optimized Ternary Amorphous Solid Dispersions on Solid State and Dissolution Performance

Purpose: The preparation of amorphous solid dispersions (ASDs) is a well-established technique to supersaturate poorly soluble drugs in aqueous media and to overcome low bioavailability. As supersaturation generation and stabilization depend on different specific drug-polymer interactions, it has been demonstrated that combining two polymers (ternary ASDs) can optimize the supersaturation profile of a drug [1, 2].
The aim of this study was to investigate the potential superiority of ternary ASDs compared to the respective binary formulations in terms of solubility enhancement and supersaturation stabilization of the poorly soluble model drug efavirenz (EFV). As different manufacturing methods can be used for ASD preparation, we investigated the impact of ternary ASD processing on the solid state and on the dissolution performance by preparing ternary ASDs of the same composition using vacuum compression molding (VCM), hot-melt extrusion (HME) and spray-drying (SD).

Methods: Manufacturing: Binary ASDs of EFV: Eudragit L 100‑55 (EL 100-55) and EFV: Hydroxypropyl cellulose (HPC)‑SSL and a ternary ASD with 50:50 (w/w%) polymer ratio were prepared using VCM (MeltPrep GmbH, Graz, Austria). Approx. 500 mg of the physical mixtures were loaded into the VCM device with 20 mm disc geometry and molded under vacuum at 160 °C for 15 min. Extrusion of the ternary ASD (EFV: EL 100‑55: HPC‑SSL 10:45:45) was carried out using a 12 mm co-rotating twin screw extruder ZE 12 (Three‑Tec GmbH, Seon, Switzerland), equipped with a 2 mm die at temperatures of 40:75:150:150:150 °C. Feed rate and screw speed were kept constant at 2 g/min and 100 rpm, respectively. Both the VCM ASDs and the extrudate were milled and passed through a 355 µm sieve. For the SD ternary ASD (EFV: EL 100‑55: HPC‑SSL 10:45:45), EFV and the polymers were dissolved in ethanol (96%) and spray-dried at an inlet temperature of 90 °C utilizing a B290 mini spray dryer (BÜCHI, Essen, Germany). After processing, the spray-dried ASD was dried under vacuum for 24 h at 40 °C. Analytics: For solid state analyzes of neat EFV and ASDs, X‑Ray powder diffraction (XRPD) was performed in reflection mode utilizing an X’Pert MRD Pro (PANalytical, Almelo, Netherlands). Differential scanning calorimetry (DSC) was conducted using a DSC 2 instrument (Mettler, Gießen, Germany) in a temperature-modulated program (TOPEM-mode) with temperature rising of 2 °C/min from 0 to 150 °C. Non‑sink dissolution testing was performed utilizing a miniaturized USP dissolution apparatus II (MiniDissolution apparatus) in 20 ml 0.05 M phosphate buffer (pH 6.8) at 37 °C and a paddle speed of 75 rpm.

Results: Sharp reflection peaks showed the crystallinity of neat EFV (Fig. 1). The absence of reflection peaks in all ASDs indicated complete amorphous character of processed EFV. For the EFV: EL 100‑55 ASD a glass transition temperature (T_g) of 107.3 ± 0.5 °C was detected, while our DSC method could not determine the T_g of the EFV: HPC‑SSL ASD (Fig. 2). Dependent on the preparation method, solid state miscibility of the ternary systems was observed. While the HME and SD ASDs showed a T_g of 79.2 ± 1.0 °C and 80.1 ± 0.7 °C, indicating single-phased ASDs, VCM did not lead to a single-phased ternary system, as a similar T_g compared to the EFV: EL 100‑55 ASD was detected. In the dissolution testing, crystalline EFV showed low solubility for the entire observation period (max. 5.2% EFV) and only poor improvement was received dissolving the HPC-SSL ASD (max. 13.9% EFV) (Fig. 3A). Up to 50.8% of free EFV was measured by using the EL 100‑55 ASD. However, the EFV concentration could not be stabilized, as precipitation occurred after 24 min and the concentration decreased to 27.0%. The combination of both polymers (VCM) led to slightly slower dissolution rate, but to a constant increase of free EFV to almost 60% after 80 min. This concentration was maintained for the entire observation period of 180 min. The alternative manufacturing procedures HME and SD of the ternary ASDs led to even superior dissolution profiles of EFV (Fig. 3B). The HME ASD showed fast dissolution rate and a max. EFV concentration of 78.1%. The SD ASD showed slower release of EFV compared to the HME ASD; however the same extent of supersaturation was achieved after 40 min. The concentration was maintained for the entire period with a small decrease of free EFV, ending in a final concentration of 63.1%.

Conclusion: The combination of EL 100‑55 and HPC‑SSL for ternary ASD processing with EFV showed superior dissolutions compared to the corresponding binary formulations, independent on the manufacturing technique. Although VCM did not lead to a single-phased ASD, the polymers demonstrated a beneficial interplay, as EFV precipitation was prevented for the entire observation period. Using HME and SD, single-phased ternary ASDs were processed, resulting in synergistic interactions between the polymers and EFV and to even increased supersaturation of EFV compared to the ternary VCM ASD.

References: [1] Six, K. et al. Increased Physical Stability and Improved Dissolution Properties of Itraconazole , a Class II Drug, by Solid Dispersions that Combine Fast‐ and Slow‐Dissolving Polymers. Journal of Pharmaceutical Sciences 93, 124–131 (2004).
[2] Xie, T. & Taylor, L. S. Dissolution Performance of High Drug Loading Celecoxib Amorphous Solid Dispersions Formulated with Polymer Combinations. Pharm Res 33, 739–750 (2016).

Acknowledgments: The authors would like to thank Nisso Chemical Europe GmbH for funding and supporting this research project.
Edmont Stoyanov is an employee of Nisso Chemical Europe GmbH and does not own stocks on Nippon Soda Ltd. Florian Poestges is a PhD student and Karl G. Wagner a professor at the department of Pharmaceutical Technology and Biopharmaceutics at the University of Bonn, Germany. They have no additional conflicts of interest to report.

Fig. 1: X-ray powder diffraction (XRPD) diffractograms of neat EFV and ASDs.



Fig. 2: Differential scanning calorimetry (DSC) thermograms (exo up) of ASDs.



Fig. 3: Non-sink dissolution profiles of binary and ternary EFV ASDs processed by vacuum compression molding (VCM) (A) and of ternary EFV ASDs processed by VCM, hot-melt extrusion (HME) and spray-drying (SD) (B).