Pluripotent stem cell therapies need high cell doses exceeding traditional 2D capabilities. We present a closed, scalable, semi-automated process to expand pluripotent stem cells as high-density aggregates in stirred tank reactors. It achieves a 22-fold expansion over four days while retaining pluripotency markers. Acoustic perfusion enables automated medium exchange with improved control over aggregate quality. Harvested aggregates were differentiated to natural killer cells in 3D.

Laboratory automation in biotechnology has made significant progress in recent years, revolutionising the way experiments and research are conducted. The automation of entire workflows from strain engineering to bioprocess optimisation is currently the focus of attention in industry and academia, and various biofoundries are emerging around the world. I will introduce the Jülich Biofoundry and discuss challenges and solutions in the field of automated microbial bioprocess development. In particular, the standardisation and automation of molecular cloning procedures are crucial for exploring, utilising, and shaping the broad genetic parameter space of different model organisms.

Mammalian cell-based mAb production is a complex process involving many factors that determine process performance and product quality. To leverage model-based decisions in development, we have established a versatile modelling toolbox with fit-for-purpose models addressing different aspects of the overall process. We will show selected applications of mechanistic, hybrid and machine learning models, and how these digital approaches can help improving the efficiency and timelines in robust process development.

The integration of a PAT-based viable cell density (VCD) soft sensor facilitates model-based process monitoring, automation, and control within the biopharmaceutical industry. This innovative technology harnesses real-time data to estimate VCD, enabling biopharmaceutical manufacturers to make informed decisions swiftly such as proceeding to the next process step or controlling the feed flow. It enhances product consistency and reduces production costs, underscoring the vital role of data-driven strategies in advancing production of pharmaceuticals and biologics.

Overcoming the inherent heterogeneity of cells grown in vitro is essential to deliver effective, uniform and safe biological medicines. This talk will focus on the application of single cell omics to understand the CHO cell biological system and how the technology can be used to understand the emergence of phenotypic instability. We will provide examples of studies DNA, RNA as well as at the protein level.

The use of modelling tools for bioprocess development and manufacturing has gained significant attention. But what does it take to create digital bioprocess-twins? Is it only about advanced algorithms, or do we need more to accurately represent and transfer knowledge from development to manufacturing? Key discussion points include the purpose of process modelling, experimental design, tailored modelling approaches, accelerated development, seamless scale-up, and real-time model use for monitoring and control.

Acceleration of commercialization of biologics including the filing of a robust control strategy is of utmost importance for biosimilars up to new modalities. Digital twins capture CMC knowledge and allow multiple deployments. We will show 

• How end-to-end digital twins can help saving 50% of experimental effort by incorporating drug substance specification when designing unit operations and
• How real-time application allows for prediction and control on process performance?
• How we can know screws to compensate for deviations in real time, and have a clear OOS probability perspective for real-time release and there with clearly differs from classical SPC approaches
  

Raman spectroscopy plays an important role in bioprocessing by providing in situ measurements and enabling real-time process control. In this study, we demonstrate the use of Raman spectroscopy as a key solution for inline and real-time monitoring of CPPs and CQAs along the upstream process.

In industrial bioprocess control, conventional PI controllers, while effective for high-frequency measurements, often overlook complex variable interactions. They mainly focus on single input-single output scenarios. However, many situations require holistic consideration of multiple factors. Model Predictive Control (MPC) offers a versatile multi-input, multi-output framework, managing variables and constraints simultaneously. Challenges include model development, computational demands, expertise, and real-time implementation. We'll illustrate MPC's practical efficacy with industrial case studies across various processes. This highlights MPC's advantages and its ability to address complex bioprocess control needs, from batch to continuous operations.

• Critical evaluation of the current industrial evolution
• AI vs mechanistic modelling: what to choose?
• Workflow vs data: where to invest?
• Outlook how AI will change the way of bioprocessing in the future​

Monoclonal antibodies are increasingly dominating the market for therapeutic and diagnostic agents. For this reason, continuous methods—such as perfusion processes—are being explored in an ongoing effort to enhance product yields. In this work, we demonstrate a 3D-printed microfluidic spiral separator for cell retention integrated in a small-scale bioreactor system. This device achieves a separation efficiency of up to 100% and can readily be adapted according to process conditions (plug-and-play system) due to its fast and flexible 3D printing. In addition, we also improved production conditions by successfully separating dead cells and increasing cell viability in bioreactors.

Single-use technologies are well established in biopharma production. In this context, the question arises to what extent sustainability aspects play a role in the choice between reusable and disposable equipment. How can model processes and procedures be established in order to be able to carry out meaningful comparative studies and provide standardizable decision-making tools in an open access format? An overview is given of what has been explored so far, with respect to life-cycle assessments in the field of single-use technologies in bio-based applications and what still needs to be achieved.

Bioprocessing has made significant progress in the last decades, with advancements in cell-line engineering, media platforms, and process control. Delivering product with an increased space-time-yield, intensified processes seem to replace classical cell culture fed-batch processes step-by-step. Still, it is worth to discuss at which point even the combined use of process intensification and good producer cell lines has its limitations.