Continuous Chromatography of Biopharmaceuticals: Next Generation Process Development

The biopharmaceutical industry is moving from a batch to a continuous mode of manufacturing. This shift promises to reduce costs and manufacturing footprint while improving productivity and consistency of the product. This thesis implements miniaturized and automated high-throughput screening techniques alongside a mathematical chromatography model for the development of an integrated continuous chromatography process. The model is used for in-silico optimization of a capture and polishing step of a monoclonal antibody (mAb). The optimization focusses on chromatographic processes that would have to deal with higher titer solutions.
The transition to Integrated Continuous Biomanufacturing (ICB) is welcomed by industry and regulatory agencies, which are working together to accomplish this shift. Process development plays a crucial role in defining new processes or adapting existing processes to different modes of operation. High-Throughput Process Development (HTPD) has been used in the biopharmaceutical industry to accelerate and reduce costs of process development, by using miniaturized assays and performing computer-aided studies. However, the industry experiences gaps and sees opportunities for improvement in the HTPD tools that can help the transition to ICB. These gaps, together with a state-of-the-art of HTPD for ICB are presented in Chapter 2. Experts in the field identified microfluidics and modeling to be the most promising technologies to fill in the gaps in process development for ICB.
Subsequently, an overview on the state-of-the-art of automation and miniaturization for biopharmaceutical process development is given in Chapter 3. The focus is on different degrees of miniaturization and automation of the technologies for process development, for both Upstream and Downstream processing (USP and DSP, respectively). Liquid-Handling Stations (LHS) are the epitome of automation for process development, and have seen great adoption for the past decades. Examples of the use of this tool for USP and DSP process development are provided. A greater emphasis is placed on the often overlooked microfluidics and how it can also be used as a screening tool, and a SWOT analysis on LHS and microfluidics as potential process development tools is provided.
Further comparison between HTS tools for chromatographic process development is needed, since process development efforts for chromatography mostly rely on LHS-based experiments. Three methodologies are selected for this comparison: LHS, microfluidics, and Eppendorf tubes (Chapter 4). To achieve this, protein equilibrium adsorption isotherms are determined with each of the aforementioned methodologies. The microfluidics chip produced in-house provides a platform for resin screening that achieves liquid and resin volume reductions of 15- and up-to 200-fold, respectively. Accurate resin volume determination is ensured with an image analysis software, and resin consumption is as high as 200 nl in the microfluidics system. After validating the HTS methodologies, a cost consideration study aims at fairly comparing the three methodologies for their chromatographic process development potential. Although at a lower Technology Readiness Level, microfluidics can be a viable alternative tool when the protein to be studied is very expensive or scarce (such as in early stages of process development), due to the high degree of miniaturization. Furthermore, it is discussed what would be the possible applications of the different methodologies in chromatographic process development.
The HTS methodologies developed paved the way for the implementation of a HTPD approach for the study and optimization of continuous chromatography (Chapters 5 and 6). A large database on the adsorption equilibrium isotherms of mAbs to different protein A (ProA) and Cation-Exchange (CEX) resins is generated from experiments with a LHS. This database is then used to further reduce resin candidates to be used in subsequent experiments. Four resin candidates are used to study the equilibrium adsorption isotherms of mAb to ProA ligands with a clarified cell culture supernatant (harvest). It is shown that pure mAb experiments reflect the same adsorption behavior as harvest experiments for all resin candidates, reducing the need to duplicate experiments in the future. The parameters determined are further used in a mechanistic Lumped Kinetic Model (LKM), used for the in-silico study of column chromatography (Chapter 5). The LKM uses a lumped overall mass transfer parameter that is linearly dependent on feed concentration, in line with mass transfer theory. The hybrid approach to HTPD emphasizes the importance of computational, experimental, and decision-making stages in chromatographic process development.
The LKM model described is further developed for the study of continuous chromatography. The continuous model is used for the in-silico optimization of a 3-Column Periodic Counter-current Chromatography (3C-PCC) capture and polishing step, for the purification of mAbs from high-titer solutions (Chapter 6). The model maximizes Productivity and Capacity Utilization (CU) keeping the yield high (99%) and having the flow rate and the percentage of breakthrough achieved in the interconnected phase as constraints. The shape of the breakthrough curve plays an important role in the optimization of continuous chromatography. The optimization results are validated for three different ProA resins, from which the best resin candidate is selected to continuously capture mAb from a harvest solution. The eluates of this operation are pooled and used as input for the continuous CEX step. The experimental results show very good agreement with model’s predictions (lower than 7% deviation) and the proposed methodology helps to develop and optimize a continuous chromatography process in a short amount of time.
In summary, this thesis presents the exciting journey of process development for continuous chromatography, from conceptualization and selection of screening techniques until the end result of performing an optimized continuous chromatographic step for the successful capture and polishing of a mAb.