(605h) Simulation Aided Pharmaceutical HME Scale-up

Josip Matić^*, Abdelhamid Mostafa*, Carolina Alva*, Johannes Khinast^*,**

*Research Centre Pharmaceutical Engineering GmbH, Inffeldgasse 13/III, 8010 Graz, Austria

**Institute for Process and Particle Engineering, Graz University of Technology, Inffeldgasse 13/III, 8010 Graz, Austria

Email for correspondence: khinast@tugraz.at

Introduction

Hot melt extrusion (HME) is a continuous manufacturing process primarily facilitated by co-rotating intermeshing twin-screw extruders (TSE). The process is mostly used in order to produce amorphous solid dispersions of poorly soluble active pharmaceutical ingredients (APIs) by dispersing them in polymer carriers. The process setup is modular, allowing the process to be tailor-made for the processed formulation. This might however be problematic in the sense of choosing the appropriate and efficient setup for an unknown formulation. Besides the process setup challenge the process scale-up is a considerable issue during the different product development stages (from formulation development to pilot plant/clinical study to production scale). The current state-of-the-art is based on utilizing different extruder similarity based approaches (simple 0D models), developed mostly in the polymer industry and heavily coupled to extensive experimental efforts. The immediate restriction in the pharmaceutical industry is the lack and the expense of APIs in the early development stages in addition to the connected costs and efforts in handling the process in a GMP environment. Moreover, the 0D models do not provide any insights into the process, as they are not predictive and do not guaranty the most important aspect in the process transfer and scale-up: the equivalence of product quality at different scales. Hence, understanding why certain process settings may lead to a certain product quality is the key for ensuring an easy process setup and a reliable subsequent process scale-up.

Methodology

As a response to the challenges, simulations have been increasingly used as means for process understanding. In general, HME simulations have been used for understanding individual extruder screw elements, the HME process as a whole and to relate HME to other unit operations in a continuous manufacturing line. The most notable among them is the individual extruder screw investigation using the Lagrangian based Smoothed Particle Hydrodynamics (SPH) approach for understanding the melt flow and mixing in fully and/or partially filled extruder screw elements, which applies for both Newtonian and non-Newtonian fluids [1]–[5]. The knowledge gained is applied in a less complex and faster 1D HME code, allowing for the accurate representation of the process as a whole[6], [7]. These approaches have a focus on understanding the process, either by investigating the melt flow in detail and/or by investigating the influence of the process settings on the process state variables like melt temperature and RTD. Based on the knowledge gained from process simulations, experiments and product investigations, the next challenge is the development of tools for the in-silico prediction of the product performance. Hence, relating the process state variables (melt temperature, SMEC, RTD...) with the product performance is the next step towards understanding the HME process.

Results

In our last talk we looked at the degradation of Famotidine-Eudragit RL formulation as a function of different HME process parameters of the 12mm TSE from Leistritz. The API degradation was taken as a measure of the product quality after extrusion and the process states that lead to the degradation were analyzed in detail using the 1D HME simulation code developed in-house. As a next step, a number of scale-up approaches were tested experimentally, for a process transfer from the Leistritz 12mm to 18mm TSE. The goal was reaching the same API degradation state formed in the original process. The different resulting HME processes were analyzed using the 1D HME software in order to explain variations of the product quality in comparison to the original process, keeping in mind the temperature peaks and mean residence times archived in the different processing zones. In silico based process settings were then obtained and tested with the goal of achieving the same product quality. This lays the foundation for automated process setup and scale-up with a prescribed product quality in mind.

Literature

[1] J. J. Monaghan, “Smoothed particle hydrodynamics,” Reports Prog. Phys., vol. 68, no. 8, pp. 1703–1759, Aug. 2005.

[2] K. Kohlgrüber and W. Wiedmann, Co-Rotating Twin-Screw Extruders. Munich: Carl Hanser Verlag, 2008.

[3] A. Eitzlmayr and J. Khinast, “Co-rotating twin-screw extruders: Detailed analysis of conveying elements based on smoothed particle hydrodynamics. Part 1: Hydrodynamics,” Chem. Eng. Sci., vol. 134, pp. 880–886, 2015.

[4] A. Eitzlmayr, J. Matić, and J. Khinast, “Analysis of Flow and Mixing in Screw Elements of Corotating Twin-Screw Extruders via SPH,” AIChE J., 2017.

[5] R. Baumgartner, J. Matić, S. Schrank, S. Laske, J. G. Khinast, and E. Roblegg, “NANEX: Process Design and Optimization,” Internaitonal J. Pharm., vol. 506, pp. 35–45, 2016.

[6] A. Eitzlmayr et al., “Mechanistic modeling of modular co-rotating twin-screw extruders,” Int. J. Pharm., vol. 474, no. 1–2, pp. 157–176, 2014.

[7] J. Matić, A. Witschnigg, M. Zagler, S. Eder, and J. Khinast, “A novel in silico scale-up approach for hot melt extrusion processes,” Chem. Eng. Sci.