Data and Digitization – The key to better biology and more effective therapies

Data is the cornerstone of any scientific endeavor. But throughout history it’s been managed pretty poorly. Most data ever gathered on cells remain in hand-written lab notebooks, gathering dust on shelves after the experiment’s ended. Small snippets of data make it to the public eye through accepted publications, which shape future experiments and lines of investigation. But all that other data collected from hours of planning experiments and measuring endpoints that isn’t up to publication standards – like negative data – is essentially lost. Imagine how much more could be learned if all the data collected over the past century on cells was lifted, centralized, and analyzed.

Whilst this may not be feasible for the past, using digital systems could make this a reality in the future. Things like electronic lab notebooks are already used in the lab, helping to digitize note taking at the source making it easy to record, transfer, and store data. But the evolution of digital tools and equipment will:

• Automate monitoring of cells and the cell culture environment during and after experiments
• Process data to make it exploitable (e.g., images of cells into cell circularity, cell size, cell count, cell eccentricity)
• Organize data so it can be easily visualized
• Analyze data through machine learning algorithms to help uncover insights that isn’t currently possible

Before we can unlock the incredible power of digital systems, research labs and manufacturing sites need the right infrastructure and equipment. Communications standards are starting to come together to allow this, for example with OPC-UA, but there is an unmet need for hardware that can generate the actual data during cell culture.

At MFX, we build platforms that can manage up to 30 bioreactors in one device, with each bioreactor generating hundreds of thousands of data points thanks to integrated online sensors and mutualized analytical equipment. All the data is processed, organised, and presented to the user in a digestible format, and we partner with the best machine learning providers to generate insights.

The benefits of this are countless and include:

• Reducing process development time
• Reducing the number of experiments necessary to uncover insights
• Making it easier to transfer knowledge from one person/organization to another
• Discovering new treatments faster

What’s more, these benefits building exponentially as your database expands. Now, that’s an exciting future for biology.

Making cell therapies comPATible with large scale manufacturing

Process analytical technologies (PATs) cover a wide range of tools, playing an important role in initiatives such as quality by design (QbD), real-time release, and continuous manufacturing [1]. They are already widely utilized in pharma and biopharma manufacturing, and for the large-scale deployment of cell therapies, they are critical to ensure:

• Safety – Maintains a consistent process and product quality through in-line and on-line monitoring and analysis
• Speed – Allows insights into the process helping accelerate R&D and process development, especially when couple with data analytical tools
• Cost-effectiveness – Insights into culture throughout the process can help optimize reagent usage, reduce process duration, and lower batch failure rates

During process development, many parameters need to be measured across hundreds of experiments to identify the critical process parameters (CPPs) and critical quality attributes (CQAs) required for the desired cell product. Without PATs, this turns into an intractable (both costs and logistics-wise) amount of sampling and manipulation, and an enormous amount of Capex for analytical equipment.

Common parameters measured during process development include:

In-process

• pH
• Dissolved O2
• Temperature
• Dissolved CO2
• Metabolite concentration
• Amino-acid concentration

End of process

• Cell count
• Cell viability
• Phenotyping
• Karyotyping
• Potency assays

At manufacturing scale, CPPs and CQAs need to be measured constantly to demonstrate batch safety through quality control. With hundreds of thousands of patients per year who could benefit from cell therapies, maintaining the current practices of manually sampling and analysing swathes of data will become commercially and logistically unviable.

So how can you implement PATs into the cell therapy manufacturing process?

Automation

Online automation allows for multiple measurements to be taken in parallel, avoiding the need for sampling. For example, using specialist microscopes coupled with trained algorithms next to a bioreactor can supply increasingly accurate cell count and viability measurements, with fluid loops from the bioreactor maintaining a closed culture system. MicrofluidX has gone one step further, directly implementing microscopes on their bioreactor thanks to their bioreactor form factor.

If there isn’t capacity for online measurements, automation can be harnessed in other ways to save time and reduce manual handling, like automating sample prep for flow cytometry.

Mutualization

All these PATs can add up when running multiple bioreactors in terms of costs, space, validation requirements, logistics and maintenance. This is why mutualization is key, and it can be done in a few ways. Robotics can manage the movement of bioreactors to and from pieces of analytical equipment in closed systems, while autosamplers can manage the movement of samples from different bioreactors to one piece of analytical equipment. This drastically decreases Capex, increases productivity of your equipment, and removes variability from manual handling.

At MFX we combine all these approaches to deliver a cost-effective, analytically driven bioreactor platform. Our Cyto Engine™ can be deployed in both process development and large-scale manufacturing. 

[1] Clegg, I. 2020. Specification of Drug Substances and Products, Second Edition. Elsevier, pp 149-173

How do you solve a problem like scaling up microfluidics?

Culturing cells using microfluidics offers several advantages over traditional cell culture techniques – the main one being the ability to precisely control the microenvironment. This means the availability of nutrients, oxygen levels, and pH can be finely tuned to the cells’ needs, resulting in more accurate and reproducible results with optimized reagent consumption (1).

Microfluidic cell culture is well-suited for online analytics. Microfluidic devices can be equipped with sensors to allow continuous, real-time monitoring of cells as they grow and respond to various stimuli (2). This enables detailed data collection on cell behaviour and responses, which can provide valuable insights and facilitate the development of new therapies. Microfluidic devices can also be easily automated, making them the perfect next generation bioreactor platform for cell research – more precise and controlled processes, automatable, low usage of reagents, and compatible with online analytics.

But what about scale? Getting 100,000 perfect cells doesn’t help patients. Millions, often billions of cells are needed to make a dose. And that’s were microfluidics stopped being relevant… until now.

At MFX, we have parallelized dozens of microfluidic bioreactors onto a larger array that forms our manufacturing bioreactor, producing up to 6 billion cells. Our parallelization strategy is inspired by the semiconductor industry and adapted to incorporate fluid and cells, ensuring homogeneous distribution to each bioreactor. The benefit of this is that everything remains the same from the point of view of the cells – densities, nutrient and metabolite concentrations, shear stress, gassing, and even analytics. The cells experience the same controlled environment, and the throughput of the system is clinically relevant. Voilà! 

(1) Ram, K. R., Pang, K. K., & Thian, E. S. (2013). Microfluidic platforms for in vitro cell culture and analysis. Lab on a Chip, 13(2), 252-268.

(2) Lai, C., Kim, J., & Wang, L. (2016). Microfluidic cell culture platforms for studying cell-cell interactions. Lab on a Chip, 16(2), 199-212.

Make your cells feel at home with microfluidics

It can be easy to forget just how alien an environment in vitro culture is when culturing cells. In vivo, cells exist in a multi-faceted micro-environment made up of a carefully regulated biochemical milieu combined with cell-to-cell/matrix contacts and specific physiochemical properties. Yet culture vessels have evolved over the years, often with little consideration of what is actually best for cells. Historically, cell culture vessels looked like they had been taken straight out of the kitchen (and they probably were): simple bottles of reagent with rudimentary agitation systems. When research needed something smaller, basic plates and flasks were utilized for adherent culture.

More recently, innovative cell culture vessels such as stirred tank bioreactors have been developed – optimized for things such as gas transfer and mixing. 3D cell culture techniques have enabled cells to be grown in three dimensions, replicating some in vivo interactions, whilst the introduction of cell culture cartridges have taken this approach one step further. However, most of these innovations have been focused automation and cell culture for non-mammalian cells. A truly cell-friendly vessel for mammalian cells is yet to be developed.

We asked MFXs VP of Microfluidics, Césaré Cejas, PhD what makes microfluidic cell culture so very different; “Microfluidics  offers a new approach, allowing for an unprecedented level of control of the cell micro-environment. Microfluidic cell culture has been around for less than 30 years but is already revolutionizing the field of mammalian cell culture. In addition to higher cell densities, a significant improvement in transduction efficiency and better and more persistent phenotypes can be achieved.”

The fact that microfluidics is highly automatable, reduces manufacturing costs, and allows the flexibility to easily fit channels with sensors is an added bonus!

MFX’s Cyto Engine™ is leveraging microfluidics to create the best possible environment for mammalian cells. By putting the cells first, we deliver incomparable process efficiency and control.

Microfluidics- the perfect meeting-place for viruses and cells

Viruses get a bad press. SARS-CoV-2 did little to improve matters, but those of us working in gene and cell therapy are only too aware of the critical role lentiviruses, adeno- and adeno-associated viruses play in the manufacture of these life-saving medicines.

Viral vectors are super-effective at inducing gene expression in cells and conferring T-cells with therapeutic properties such as tumor detection; but the production of these biomolecular tools is complex and expensive, and yields are often incredibly low. A typical process requires culturing CHO cells to produce plasmids, which are used to transfect HEK293 cells, which in turn produce the viral particles, all under GMP conditions. A few microliters of virus can cost several thousand dollars to manufacture, so when using large volume conventional bioreactors things quickly become prohibitively expensive.

Infecting target cells with the viral vector presents further challenges. The objective is to achieve optimal interaction between cell and virus particle so that as many cells as possible are infected within the bioreactor, whilst minimizing multiple infection of cells (regulators typically specify <5 infections per cell). Conventional bioreactors are not well-designed for homogeneous interaction between virus particle and cell: some cells may get infected multiple times, others not at all.

But there is another way. Both of these problems (non-uniform cell infection and low viral yield) can be overcome by the use of microfluidics. As Césaré Cejas, PhD, MFX’s VP of Microfluidics explains; “Microfluidics require significantly less virus, and co-localizing target cells and virus particles within microfluidic channels enables fine control of the interaction between cells and the virus. The cells are homogeneously distributed across a thin film of fluid ensuring that every cell is in contact with the fluid – not buried under other cells. The viral particles, moving by Brownian motion, do not have to travel far to encounter a cell and all the cells are equally likely to be infected.”

Microfluidics requires around 10x less virus, shortens the viral transduction process by several days, and ensures a homogeneous number of infections per cell. Our proprietary Cyto Engine™ platform is harnessing the power of microfluidics to transform viral transduction processes for cell and gene therapies.

Scale up, look sharp – why analytics in large-scale cell therapy manufacture is so difficult

CAR-T and other cell and gene therapies represent an exciting new paradigm in how we approach previously untreatable diseases. But as ‘living drugs’ where the cell is the product, QC / analytical testing is more important than ever to ensure safety and efficacy. Analytics can be thought of in terms of:

• Process analytics – collecting data on variables that affect the quality of the cells, like temperature, pH, dissolved oxygen, and metabolite concentration, throughout the process to analyse and refine the process
• Product analytics – collecting data through or at the end of the process to analyse and characterize the product (in this case the cells), including cell count, cell viability, and flow panels

As per GMP guidelines, analytics that are critical to the manufacturing process and not measured automatically, ‘Critical Quality Attributes’ and ‘Critical Process Parameters’, are measured by the QC team. While batched QC analysis is achievable in traditional pharma manufacture, it isn’t optimal for production of complex advanced therapies. In order to scale production of cell and gene therapies and reduce the all important vein-to-vein time developers need to:

• Limit people-intensive manual measurements and recording which are laborious and error-prone
• Minimise manual sample and data exchanges between manufacturing and QC teams

The obvious solution would then be to have online measurements and sample analysis. But how can this be achieved across multiple bioreactors without significant capital outlay? Fitting an array of different analytical technologies all measuring different things on each and every bioreactor is expensive (usually being restricted to just a few bioreactors at a time), and the unfortunate truth is that current technologies can barely automate pH and dissolved oxygen measurement reliably. Automated cell count is often only an estimate, only a handful of early-stage technologies are looking at measuring online cell viability, and no one can do label-free flow cytometry at scale.

“From an engineering perspective, we knew we had to think about things differently to solve the challenge of making online analytics a reality as we designed the Cyto Engine platform. So we came up with the concept of mutualising the analytics we integrated, i.e. one set of analytical tools that can make measurements on multiple bioreactors. We also looked to build in flexibility on what we can integrate  – that allows us to add to the system as analytical technology improves as well as allowing customization on what analytics come built in,” says James Davies, VP of Engineering at MFX.

What’s more, the mutualisation contributes negligible CapEx, making manufacturing cost-effective in the long-term.