Increasing capacity utilization and lowering manufacturing costs are critical for pharmaceutical companies to improve their competitiveness in a challenging environment. Development of next generation cell lines, improved media formulations, application of mature technologies and innovative operational strategies have been deployed to improve yields and capacity utilization. This article describes a large-scale perfusion strategy for the N-1 seed train bioreactor that was successfully applied to achieve higher inoculation cell densities in the production culture. The N-1 perfusion at 3,000-L scale, utilizing a inclined settler, achieved cell densities of up to 158 × 10(5) cell mL(-1) at perfusion rates of 2950 L day(-1) and a retention efficiency of >85%. This approach increased inoculation cell densities and decreased cultivation times by ~20% in a CHO-based, fed-batch antibody manufacturing process while providing comparable culture performance, productivity, and product quality. The strategy therefore yielded significant increase in capacity utilization and concomitant cost improvement in a large scale cGMP facility. Details of the strategy, the cell retention device, and the cell culture performance are described in this article.

Most discussions about stirred tank bioreactors for cell cultures focus on liquid-phase motions and neglect the importance of the gas phase for mixing, power input and especially CO(2) stripping. Particularly in large production reactors, CO(2) removal from the culture is known to be a major problem. Here, we show that stripping is mainly affected by the change of the gas composition during the movement of the gas phase through the bioreactor from the sparger system towards the headspace. A mathematical model for CO(2)-stripping and O(2)-mass transfer is presented taking gas-residence times into account. The gas phase is not moving through the reactor in form of a plug flow as often assumed. The model is validated by measurement data. Further measurement results are presented that show how the gas is partly recirculated by the impellers, thus increasing the gas-residence time. The gas-residence times can be measured easily with stimulus-response techniques. The results offer further insights on the gas-residence time distributions in stirred tank reactors.

The production of therapeutic proteins by mammalian cell culture is complex and sets high requirements for process, facility, and equipment design, as well as rigorous regulatory and quality standards. One particular point of concern and significant risk to supply chain is the susceptibility to contamination such as bacteria, fungi, mycoplasma, and viruses. Several technologies have been developed to create barriers for these agents to enter the process, e.g. filtration, UV inactivation, and temperature inactivation. However, if not implemented during development of the manufacturing process, these types of process changes can have significant impact on process performance if not managed appropriately. This article describes the implementation of the high-temperature short-time (HTST) treatment of cell culture media as an additional safety barrier against adventitious agents during the transfer of a large-scale commercial cell culture manufacturing process. The necessary steps and experiments, as well as subsequent results during qualification runs and routine manufacturing, are shown.

The monitoring and control of biotech processes in different phases of the product lifecycle from early development to commercial production is key for accelerated development and stringent process controls. Effective methods of monitoring are required to develop, optimize, and maintain processes at a maximum efficiency and desired product quality. In the last decade more and more research has been devoted to developing specifically designed sensors, sampling strategies and integrated data management systems to allow better and more detailed process monitoring. Especially with the Process Analytical Technology (PAT) framework published by the Food and Drug Administration (FDA) in 2004 the measurement, monitoring, modeling and control of biotech processes has become more important. This article will describe general operational aspects of sensors, sampling technologies and methods of process monitoring, advanced applications like soft sensors and metabolic controls including integrated data management and analysis. Practical examples and case studies are used to illustrate the potential of soft sensors, models and advanced sensors.

The production of monoclonal antibodies by mammalian cell culture in bioreactors up to 25,000 L is state of the art technology in the biotech industry. During the lifecycle of a product, several scale up activities and technology transfers are typically executed to enable the supply chain strategy of a global pharmaceutical company. Given the sensitivity of mammalian cells to physicochemical culture conditions, process and equipment knowledge are critical to avoid impacts on timelines, product quantity and quality. Especially, the fluid dynamics of large scale bioreactors versus small scale models need to be described, and similarity demonstrated, in light of the Quality by Design approach promoted by the FDA. This approach comprises an associated design space which is established during process characterization and validation in bench scale bioreactors. Therefore the establishment of predictive models and simulation tools for major operating conditions of stirred vessels (mixing, mass transfer, and shear force.), based on fundamental engineering principles, have experienced a renaissance in the recent years. This work illustrates the systematic characterization of a large variety of bioreactor designs deployed in a global manufacturing network ranging from small bench scale equipment to large scale production equipment (25,000 L). Several traditional methods to determine power input, mixing, mass transfer and shear force have been used to create a data base and identify differences for various impeller types and configurations in operating ranges typically applied in cell culture processes at manufacturing scale. In addition, extrapolation of different empirical models, e.g. Cooke et al. (Paper presented at the proceedings of the 2nd international conference of bioreactor fluid dynamics, Cranfield, UK, 1988), have been assessed for their validity in these operational ranges. Results for selected designs are shown and serve as examples of structured characterization to enable fast and agile process transfers, scale up and troubleshooting.