Advances in Cell-based Therapeutic Process Development

The prospect of substituting damaged or diseased tissue by cell-based therapeutics has created much excitement in the last decades, since cell-based therapeutics open up new possibilities in patient treatment towards regenerative medicine. However, the first generation of cell-based therapeutics could not live up to market expectations and some were even withdrawn due to low reimbursement levels. While these products might be considered as market failures, identifying the obstacles early cell-based products were facing represents a major opportunity to improve the market performance of future products. In general, the inherent complexity of cell-based products in combination with limited experience represents a major obstacle translating a possible candidate from bench to bedside. A deeper insight was gained by the retrospective evaluation of the translational challenges where they could be further classified in three categories: (1) pre-market challenges, (2) manufacturing challenges and (3) post-market challenges. A fundamental step towards commercialization is the process development (industrial scale and grade) which affects all three translational challenge categories. Thus, looking at the process development associated translational challenges is important. Currently, process development for cell-based therapeutics is often done using a heuristic approach, however, a more directed and systematic approach would be highly beneficial in order to overcome the process development associated challenges. Such an approach would allow for gaining process knowledge and improve process performance and, thus, improve market performance. Effective process development for cell-based therapeutics is extremely important for product commercialization and could be achieved by implementing general bioprocess development tools and strategies. Hence, this doctoral thesis addresses the question if bioprocess development tools and strategies from the pharmaceutical industry can also be used for cell-based products. While process development requirements differ between unit operations and development stages, a general obstacle is to deal with limited material and time resources. As a consequence, most of the applied tools follow the principles of miniaturization, parallelization and automation. In order to demonstrate the power and versatility of bioprocess development tools, the objective of this thesis is the development and implementation of tools for different cell-based therapeutic manufacturing unit operations. The first part of the thesis focuses on the unit operation cell separation, more specific, the usage of cell partitioning in aqueous two-phase systems (ATPS) as cell separation method. Here, the usage of a high-throughput screening (HTS) platform is used to investigate the influence of the critical parameters polymer molecular weight and tie-line length on project. Here, batch data (cell partitioning in ATPS) is transferred into a microfluidic flow-through mode. The main scope of this project is the evaluation of 3D printing as a process development tool, since it allows for a fast and cheap realization of tailored process equipment. Whereas the second part of the thesis is focusing on cell formulation unit operations, namely, cell cryoperservation and bioprinting. A major clinical cell cryopreservation topic is the cryomedia formulation, since it needs to be toxin- and xenogen-free. Thus, the focus of the second part of this thesis is the development of a tool box to predict cryomedia performance to shorten development time. Additionally, an analytical strategy, allowing for evaluation of bioprinting process parameter on critical cell parameter, is developed. The evaluation of cell partitioning in ATPS as separation method in industrial scale cell downstream processing is the first central topic of this thesis. Cell partitioning in ATPS represents a promising alternative cell separation method to tackle the bottleneck condemned by limited availability of industrial scale methods, since it is scalable and facilitates a label-free separation. Up to now, the usage of cell partitioning in ATPS is limited due to its complex nature and not fully described partitioning theory. Thus, a time intensive and empirical screening is necessary in order to find optimal process parameter for each cell separation problem. First, a previously developed HTS platform (batch mode) is used for the investigation of critical ATPS parameter on cell partitioning behavior of five model cell lines aiming for a systematic and automated screening. The previously developed HTS platform was extended by a cell barcoding strategy allowing for a multiplexed high-throughput cell analysis in order to speed up the analytical strategy. Followed by a case study investigating the influence of the molecular weight and tie-line length on the resolution of the five model cell-lines in PEG-dextran ATPS. Both factors have an influence on cell partitioning behavior. The highest resolution of the five model cell lines could be achieved using low molecular weight PEG in combination with high molecular weight dextran. Additionally, a countercurrent distribution model was applied to calculate the theoretical purity and yield for the described separation issue. By applying the screening conditions to the countercurrent distribution model, an isolation of four of the five model cell lines is theoretically possible with high purity (> 99.9 %) and yield. A technology transfer of batch HTS cell partitioning data into a microfluidic flow-through mode is described in a subsequent project. Such a technology transfer is normally associated with high investment costs. Aiming for a flexible process development approach with low financial risk, the usage of 3D printing technologies as process development tool is proposed in this study. In general, 3D printing technologies allow for the layer-by-layer manufacturing of 3D objects. The advantage of those technologies is the relatively cheap and fast realization of complex geometries. Thus, they represent a promising and versatile tool for process development, since they enable a rapid manufacturing and implementation of tailored equipment. In this project, 3D printing was used to manufacture tailored devices complementing standard laboratory equipment for the realization of a cheap and flexible microfluidic experimental set up. The following tailored equipment was designed and 3D printed for the realization of the experimental set up: • replication master for microfluidic device manufacturing • camera mount, using the existing camera shaft, allowing for implementation of a small high performance camera • manual tailor-made fractionator with optimal installation space enabling the integration on the microscope table • frame for the microfluidic device: the frame consists of three parts, the bottom plate, an outlet- and inlet bridge. The in- and outlet tubing were connected with the microfluidic device and stabilized via the bridges using fittings. Implementing these tailored equipment parts, analysis of cell partitioning in ATPS in a microfluidic flow-through set up was possible. Cell suspension, top and bottom phase are pumped individually into the microfluidic device using three pumps. The cell suspension is focused at the interphase of the top and bottom phase allowing for cell partitioning. Simultaneously, a video of the microfluidic channel near the outlet is recorded. Using the developed snap shot analytic of the video material, an automatic estimation of cells in top and bottom phase is performed. Additionally, live cell rate was determined via flow cytometry by analyzing pooled samples. In order to provide a systematic process development and performance evaluation, the technology transfer case study was performed using the DMAIC (Define, Measure, Analyse, Improve and Control) framework. Using this DMAIC framework a successful implementation of the developed microfluidic process set up was performed. Additionally, two main factors influencing cell partitioning in the flow-through mode were identified, namely, cell load and fluid velocity. Process development of different cell formulation unit operations is the focus of the second part of this thesis. Depending on the manufacturing strategy of cell-based therapeutics, long-time storage of the cell product needs to be performed. Usually, the unit operation cryopreservation is performed for this purpose, since it allows for stocking of living material while maintaining critical cell characteristics (no genetic or metabolic alterations). While cell cryopreservation is a well described topic in literature, cell cryopreservation process development of cell-based therapeutics is currently rather time consuming and follows an empirical approach. Aiming for a streamlined and more systematic cell cryopreservation process development, a video-based tool for the characterization of the freezing and thawing behavior was developed in the first formulation project. Freezing and thawing profiles are critical process parameter influenced by many factors including e.g. cryomedia formulation and scale up (working volume and container geometry). In order to enable rapid process development for cryopreservation despite the high number of process variable, fast and directed analytical tools are required. To evaluate the performance and flexibility of the video-based tool, a cryopreservation case study was performed with the _-model cell line INS-1E. Here, the freezing and thawing behavior of two working volumes (1 mL and 2 mL) were analyzed. Additionally, cell recovery and proliferation were evaluated. As expected, a delay in freezing and thawing behavior due to scale up was detected, resulting in a decrease of process performance determined by live cell recovery. While a high live cell recovery (0.94 (±0.14) %) was achieved with 1 mL working volume, a decrease to 0.61 (±0.05) % could be observed for the 2 mL working volume. These findings are in good agreement with expectations, confirming the tool performance. The objective of the consecutive second cryopreservation project is the development of a tool box allowing for the prediction of cryomedia process performance by characterization of media properties, freeze/thaw behavior (using the previously developed video-based tool) and media toxicity. Using good manufacturing practice (GMP) compliant cryomedia is an essential requirement for the cryopreservation of cell-based therapeutics, meaning it needs to be chemically defined without the addition of animal-derived or toxic compounds. For this reason, using the well described and often used cryoprotectants dimethyl sulfoxide (DMSO) and fetal bovine serum (FBS) is not an option. Supplementing a cryomedia with cryoprotectant agents is, however, crucial for cryopreservation process performance since they protect cells from severe stress, e.g. non-physiological osmotic pressure conditions, during freezing and thawing. Hence, development of an effective and GMP compliant cryomedia formulation is necessary for cell-based therapeutics when long-term storage is part of the manufacturing strategy. In this project, the usage of the developed predictive tool box for cryomedia formulation screening is proposed. The project comprised a case study evaluating the developed predictive cryomedia screening tool box including two commercial, DMSO-free cryomedia, one negative (without any cryoprotectant) and one positive (supplemented with DMSO and FBS as cryoprotectants) cryomedia example. The case study was performed with the _-model cell line INS-1E. Additionally, a conventional cryomedia performance assessment was done in order to evaluate the predictive tool box. Using the tool box the commercial Biofreeze® media was classified as unsuitable for the beta-model cell line INS-1E due to media toxicity, while the second commercial cryomedia CryoSOfreeTM was classified as possible candidate. These findings were confirmed by the conventional screening proving the power and usefulness of the tool box. Thus, implementing the predictive tool box will facilitates a fast pre-selection of possible candidates with low sample volume and cell number, shortening the overall process development time for the development of cryomedia formulations. The focus of the third cell formulation project within this thesis is bioprinting. More precisely the development of an analytical strategy for cell-based bioprinting applications. Bioprinting is a relatively new field, however, it is believed to have a huge impact on regenerative medicine and tissue engineering, since it enables the manufacturing of artificial 3D tissue. Great efforts are being made in this field to move from academic research towards pharmaceutical industry or clinical applications. A critical hurdle to overcome this transition process is the development of a robust and well-known process, while maintaining critical cell characteristics (CCC). In order to gain process knowledge, a systematic and more directed process development approach for 3D bioprinting applications is required which includes the monitoring of CCC. Depending on the application, the critical cell characteristics may vary. However, universal CCCs are cell viability and proliferation. Flow cytometry represents a highly flexible, powerful and often used cell analysis method capable of analyzing multiplex CCC issues in parallel and might also be useful in the field of 3D bioprinting. However, flow cytometry analysis can only be performed with cell suspensions and needs, therefore, a destruction of the cell-laden 3D-structure prior analysis. Hence, the objective of this project is the development of a flow cytometry-based analytical strategy as a tool for 3D bioprinting research. The development of this anayltical strategy was conducted using a model process with the _-model cell line INS-1E, a commercially available alginate-based bioink and a direct dispensing system as 3D bioprinting method. The destructive strategy enables the evaluation of cell viability and proliferation. By using a flow cytometer set up including an autosampler, the strategy enables an automated high-throughput screening. Following the development of the strategy, an evaluation of the process steps was performed including: suspension of cells in bioink, 3D printing and cross-linking of the alginate scaffold after printing. The evaluation showed that each individual process step (using the selected process parameters) had a negative influence on cell viability and must therefore be carefully monitored. This highlights the importance of process optimization in 3D bioprinting and the usefulness of the flow cytometry-based analytical strategy. The presented strategy has a great potential as a cell characterization tool for 3D bioprinting and can even be extended to a multiplex CCC analysis. Thus, the developed tool may contribute to a more directed process development in the field of 3D bioprinting. In summary, several process development tools and strategies for different process units were developed within this doctoral thesis, demonstrating the applicability of general bioprocess development tools for cell-based products. The presented methods facilitating a more directed and systematic process development approach for cell-based therapeutics. These powerful tools represent an opportunity to streamline process development time, tackle translational hurdles and, ultimately, improve process understanding aiming for a QbD approach in the cell therapy industry