The pharmaceutical industry has in recent years been faced with the challenge of rethinking their business and Research and Development (R&D) models due to expiring patents competition from generic products, tighter regulations and a decrease in profits from their own products. The result-ing decline in profitability has led to reduced reinvestment in R&D. One way of rethinking the R&D model is to speed up the pharmaceutical development process to reduce energy costs, operating costs and waste. In this PhD thesis, the focus is on the crystallization process. Crystallization is a significant separation and purification technique widely used in the chemical, agrochemical, cosmetic and pharmaceutical industries, where more than 90% of active pharma-ceutical ingredients (APIs) are produced by this process step. Crystallization serves as the link be-tween synthesis and downstream processes such as filtration and tableting, where performance and processability depend on properties of the crystal product such as size and shape. This PhD thesis presents a comprehensive study of process analytical technology (PAT)-enabled strategies leading to systematic approaches. The goal is to provide insight into the mechanisms involved in the crystallization process, identify critical operation parameters and implement the novel meth-odology for crystallization process optimization and scale-up. The focus of this thesis is the crystallization of model APIs from solutions. The first part of the the-sis investigates crystallization of a biopharmaceutical. The use of crystallization as a purification and separation technique in biopharmaceutical processing is uncommon. Hence, developing strat-egies to optimize the process is essential for the successful implementation of the crystallization process in biopharmaceutical production. Batch-cooling crystallization was performed using the biopharmaceutical cephradine monohydrate as the model compound. The relative thermodynamic stability between the anhydrate and monohydrate of the biopharmaceutical was established at different water and methanol contents at temperatures ranging from 5°-50 °C. Five different crys-tallization strategies were applied to improve the filtration performance; a seed load of 2.5% and a slow cooling rate were shown to perform best. The second part of this thesis investigates the principles of Quality-by-Control (QbC) and direct design. These principles were implemented, and different PAT tools, such as Particle Vision Meas-urement (PVM) and focused beam reflectance measurement (FBRM), were applied. A novel scale-up strategy was developed, ensuring the desired polymorph and particulate properties of the crys-talline product were obtained throughout the process and across scales. Three different strategies were developed using the model-free approach: supersaturation control (SSC), direct nucleation control (DNC) and a sequential SSC and DNC approach. For the two first strategies, the effect of different process parameters, including seed load, seed size and supersaturation setpoint, on pro-cess time, polymorphic form and particle size distribution (PSD) were investigated. The addition of antisolvent/solvent and temperature profiles obtained from the different feedback control strate-gies were implemented directly in open-loop control and investigated for reproducibility. The op-erating profiles were successfully scaled-up by one order of magnitude, and this newly proposed scale-up strategy was shown to be successful at reproducing similar particulate properties across scales. The polymorphic compound indomethacin was used as the model compound, and a ternary solvent system comprised of methanol (33.5 wt%), and acetone (66.5 wt%) was used as the solvent and water as the antisolvent. This solvent system has been shown to mitigate the acetone solvate formation and increase the yield, compared with single-solvent cooling crystallization. The last part of this thesis focuses on the paradigm shift from batch to continuous crystallization in the pharmaceutical industry. For a successful change towards the continuous processing of APIs, it is essential to identify the potential risks associated with the process. Many mechanisms are in-volved in the continuous antisolvent crystallization of a polymorphic compound. Hence, many po-tential risks may cause process failure. To address these, a risk assessment of a single-stage mixed-suspension-mixed-product-removal (MSMPR) was conducted for the antisolvent continuous crystallization of indomethacin using the ternary solvent system. To monitor the effect of different process parameters, including initial seed load and solvent/antisolvent inlet feed ratio, on poly-morphic form and time to reach steady state, various PAT tools were applied. It was shown that seeding was essential to obtain the desired polymorph, and that the inlet feed ratio decreased the time required to reach steady state.
|Date of defence||13. Nov 2020|
|Place of Publication||Odense|
|Publication status||Published - 2020|