Evaluation of SinglePass Tangential Flow Filtration

Monoclonal antibodies (mAbs) are an important class of biopharmaceuticals used to treat various chronic and infectious diseases. The downstream processing i.e., the recovery and purification, is typically considered the bottleneck in the manufacturing of mAbs. Reasons for the low productivity of downstream processes are the use of suboptimal standardized
platform processes and the negligence of superior process alternatives as a result of cost and time pressure during early development stages. Often, large inprocess volumes and low product concentrations limit the productivity of mAb processes due to increased processing times and facility footprints, which also drive up the manufacturing costs. Throughout the past several years, singlepass tangential flow filtration (SPTFF) has found use in many manufacturing processes as an inline concentration technology to diminish the negative effects of large volumes. SPTFF is a relatively new ultrafiltration (UF) technology in biopharmaceutical manufacturing that provides the opportunity for volume reduction in a single pass i.e., without recirculating the product. High concentration factors are achieved through a characteristic multistage design of the filters, which, however, increases the complexity of the filtration technology and makes it difficult to understand how the filter design, the operating conditions, or the feed properties affect the filtration performance. The unknown relation between these design aspects and the process outcome gives rise to questions such as where in the process to reduce volumes to obtain the highest gain in productivity, how to design and operate a filter to realize a target product concentration, or which membrane to use to achieve a robust volume reduction process for a specific product.

This thesis aims to provide answers to these questions by performing a comprehensive evaluation of SPTFF. The primary objective is to enhance the current understanding of singlepass filtration to achieve efficient use of the filtration technology and increase productivity in mAb manufacturing. To accomplish this objective, a multiscale approach incorporating both modeland experimentbased approaches is applied, which divides the overall problem into three subproblems. (1) First, SPTFF is evaluated at the process scale to identify suitable placements for volume reduction in a process. For this, a framework for the analysis of flowsheet alternatives is presented and applied to a case study with the objective of increasing the capacity of a mAb downstream process through volume reduction with SPTFF. The framework uses a recipedriven flowsheet model and discreteevent simulation (DES) to virtually represent the process, simulate alternative designs, and select the best alternative(s). The results drive the evaluation of SPTFF at the equipment scale to determine how the concentration targets that were set by the evaluation from the flowsheet perspective can be realized from the unit operation perspective. (2) Therefore, the second subproblem is defined as the evaluation of SPTFF at the equipment scale. Here, computational fluid dynamics (CFD) is applied to study the flow of fluid and particles in a UF membrane cassette to understand how the operating conditions, the cassette design, and the filter configuration affect the filtration process. A threedimensional CFD model of an entire UF membrane cassette is presented, which provides the opportunity to simulate the spatial distribution of important physical quantities in the cassette. Subsequently, the model is used to simulate and compare different multistage filter configurations concerning the achieved concentration factors and generated pressures to support the selection of a suitable filter configuration. (3) Lastly, an experimental evaluation of different UF membrane cassettes used for SPTFF is presented. The objective is to study the impact of the membrane and product properties on the filtration performance. Pressureflux excursions using different cassettes and mAbs, as well as cleaning experiments, are performed to evaluate the cassettes regarding their usability in singlepass filtration and to assist in the selection of suitable cassettes. In conclusion, the presented multiscale approach provides new insights into SPTFF at different scales that were not previously available and are of use to answer the questions posed above to efficiently implement SPTFF in a mAb process.