The Covid pandemic has made society aware of the importance of having appropriate concepts and strategies to prevent the spread of an infectious disease, to treat infected people and to provide preventive treatment. For the area of biotechnological production, the availability of a versatile toolbox comprising different expression systems and generic processing concepts, is the basis for fast and efficient development and availability of high-quality products like antigens or vaccines.

For the first time a single instrument combines single-cell cloning with a new imaging technology for 3D full well imaging to generate cell lines with probability of monoclonality of >99.99%. After single-cell cloning, the UP.SIGHT allows continues monitoring of cell proliferation and clone recovery. In combination with the C.BIRD  - a microbioreactor for early suspension culture in well plates - the timeline for generating high producer clones can be significantly reduced.

Our novel, automated high-throughput engineering platform enables the fast generation of large panels of multi-specific variants (up to 10.000) giving rise to large data sets (more than 100.000 data points). By combining data science and structure-based design workflows we leverage the potential of our unique data sets to guide the engineering of our next generation antibody therapeutics.

During the live cycle of a biopharmaceutical project the production needs to stay up to date with productivity and quality demands from early pre-clinical to market phase. Optimization can be done at different stages and on different levels with selecting the right cell line, selecting the right clone or optimizing the media/feed combination and / or optimizing process parameters. We provide case studies which address the different possibilities of optimization.
 

The challenge of processing more projects with lower budgets and more aggressive timelines (e.g., development of drugs like for COVID-19 treatment) force the pharma industry to apply their historical process knowledge in combination with novel technologies. Time needed for experimental testing shall be reduced to a minimum. Therefore the access to and easy evaluation of all historical knowledge of different departments (development, manufacturing…) is essential.

Initial data storage systems such as LIMS or MES enable an optimal storage of development and manufacturing data but also complicating the analysis of those with one software as data are not harmonized and no standards are defined.

This presentation shows a case study of how we can unlock the full potential of our data treasure by breaking through the data silos of different departments. The application of data lakes and harmonization of data between different departments enable scientists to get easy access to data and work on more value-adding activities such as integration of hybrid models and machine learning approaches. Furthermore, I will present how the connection of innovative web UI software with the data lake allows scientists to work highly efficient on topics like manufacturing deviations, development reports and submission documents.

Digital methods for process design can convert classical trial-and-error bioprocess development to a more systematized and targeted approach. The key component within such a framework is a digital twin interacting with its physical process counterpart. By interconnecting hardware, software, data and humans, currently untapped potential can be accessed. The underlying digital twin acts as a powerful knowledge management tool and a decisions support system along the full process development chain.

Changes in the biologics pipeline as well as changes in commercialization needs trigger changes in the technologies we use and work flows for cell line development. In this presentation, we will cover how Thermo Fisher Scientific leverages various innovative technologies to accelerate biologics development. The presentation will highlight the incorporation of the Berkeley Lights Beacon optofluidic system into our cell line development platform and the impact delivered on enabling an accelerated, more efficient, and regulatory ready cell line and cell culture development.

In this study, a Raman analyzer has been successfully used to implement a feedback control loop in a CHO cell culture based on glucose concentration. The feedback control loop implied a direct OPC UA connectivity between the analyzer and the bioreactor control system. The culture was fed with a complex feed containing glucose. As a result, glucose concentration was maintained steady for three days. The process performance remained similar to the ones of regular fed batch cultures and a noteworthy decrease in lactate concentration production was observed. The process was completely automated.

Platform bioprocesses offer standardized bioprocess development for a novel biologic with less experimentation than historically. Yet, the fine-tuning of the process conditions and scale-up to production consistently takes several month to years. By using an iterative learning approach we can reduce these times, efficiently transferring knowledge from project to project, scale to scale. Hybrid models constitute the core of our learning approach, efficiently integrating different knowledge sources. Our results suggest that significant acceleration can be achieved.

We will present the advantage of using human expressed cytokines and growth factors and GMP ancillary materials in bioprocessing.  Included with be a brief case study of Thermostable FGF-basic.

Scale down has lost none of its importance for faster bioprocess development and optimization. Miniaturization and parallelization, and the availability of better analytics, also in combination with tailored bioreactors, allow better possibilities to identify cellular stress responses to conditions on a large scale. Processes that differ from classical suspension cultures are also becoming the focus of scale-down approaches. The talk shall provide an overview of recent developments in the field.

Although perfusion processes gain interest in the biotech community, available perfusion media are scarce. Shifting from a fed-batch process to a competitive perfusion process moreover requires reduction in medium cost. We present case study data for a comparison of commercial perfusion media versus Glycotope's well developped own perfusion medium, show approaches to improve medium formulation, and present data for using a high-performing fed-batch medium in a perfusion process.

The major challenge of industrial production of mAb or vaccines is to reduce batch-to-batch variability. Initiatives such as the PAT, push the implementation of in-line sensors for real-time monitoring and control of the Critical Process Parameters and the Key Performance Indicators of those productions. To enable efficient automated control, process sensors must deliver reliable measurements: this is best addressed by the so-called “intelligent sensors”, which will be described in this contribution.

QUBICON^® stands for better product quality in pharmaceutical manufacturing. The software enables the close monitoring of processes and the immediate, advanced control response to deviations by collecting data and performing complex calculations in real-time. Through diverse examples in mammalian cell culture technology – such as feed-on-demand control and continuous mAb manufacturing – we demonstrate the versatile applicability of the software for better performing processes.

The development of bioprocesses is a complicated task due to the large number of design variables (media composition, feeding, control, etc.), the complex process behavior and the limited experimental resources. We will present methods for cultivations, mathematical modeling, design-of-experiments and process control which are applied at Bayer in order to be able to find optimal process designs in short development timelines.

Establishing a productive yet consistent mammalian bioprocess is an ongoing challenge throughout the industry. With case studies, we demonstrate how the application of predictive modelling can be applied across a range of parameters for the optimisation of process control within a cell culture bioreactor. Encompassing at-line analysis for predictive feeding, and machine learning controller tuning for DO and pH control, the two strategies demonstrate a novel approach for bioprocess control.

Deployment of automated sampling solutions in bioprocess development laboratories has demonstrated product quality attribute data can be generated in hours instead of weeks. Access to real-time product quality data allows development of dynamic experimental designs, reducing the overall cost of product development. Using an automated approach, scientists can save non-productive time spent manually sampling and running analyses, allowing focus on evaluation of the product quality data and rapid decision making.

Standard methods of monitoring cell growth in bioreactors are often by inefficient procedures of off-line cell staining, using methods that have been in use for decades. These can be replaced by continuous optical and dielectric methods that offer real-time analysis of not only cell density but the metabolic state of cells. This presentation will evaluate alternative methods of continuous monitoring cells and the information they provide during a bioprocess.

As recombinant protein drugs become more complex, post-translational and chemical modifications can have a significant influence on the structure function relationship. To ensure that the desired critical quality attributes are obtained, advanced analytics should be applied to understand how Process Development can influence the recombinant protein drug. This talk will focus on the challenges and approaches that can be employed to produce a protein drug with the desired CQA profile.

From cell line to process development, the tested conditions are intensified with the aim of automation and reduction of cell culture volume, reconsidering historical method for cell viability and viable cell count. The label-free cell counters NORMA allow the High Throughput cell culture monitoring from small volume of unprepared samples. Integrated to High Throughput microbioreactors, the NORMA is easy to use, accurate and provides a fully automated solution.