Overcoming Bottlenecks in Organoid Production

One of the biggest challenges with traditional cell culture is that the workflow is still rooted in manual techniques that haven’t meaningfully changed since the 1990s. Researchers still spend a large percentage of their time every day in the cell culture suite; feeding cultures, passaging cells, and visually assessing growth using manual methods.

That level of hands-on care is both time-consuming and exhausting. Biology-driven timelines mean that working late, weekend shifts, and constant pressure to monitor something that is inherently variable from day to day is the norm for cell culture scientists.

When you layer organoids onto this already demanding process, the complexity only increases. Organoids contain multiple functional cell types and self-organize into tissue-like structures, which makes them powerful models of human biology—but also highly sensitive. Small differences in how or when media is changed, how cells are handled, or how growth is judged can introduce variability that affects both biological behavior and experimental outcomes.

These manual constraints also connect to a broader issue the scientific community has been grappling with: reproducibility. Results can vary across researchers, labs, or even across batches generated by the same person simply because the process depends so heavily on individual technique, judgment, and timing. That lack of consistency limits what scientists feel comfortable attempting.

Many researchers tell us they don’t want to spend their time growing organoids—they want to focus on planning experiments, designing screens, and analyzing data, where their scientific skills and knowledge make the most impact. 

When organoid production becomes the bottleneck, it restricts experimentation and makes it difficult to scale studies or compare results. That’s why we’ve focused on producing large, reproducible batches of organoids: to give researchers access to enough consistent material that they can explore ideas more freely, including experiments they might not have attempted when months of manual culture were on the line.

Our patented method introduces a controlled bioreactor-based process that fundamentally changes the scale and uniformity of organoid production. Rather than manually seeding or passaging organoids in small batches by hand, we dissociate them into a cell suspension, sieve the suspension to select for a defined size range, and then seed them into a bioreactor with an extracellular support matrix.

This gives us tight control over the starting cell cluster size, which directly contributes to more homogeneous organoids in terms of size and surface–area–to–volume ratio. That consistency helps reduce variability—a key source of irreproducibility in manual organoid culture.

Using a bioreactor creates an environment in which culture conditions (nutrients, growth factors, oxygen) can be more precisely regulated. This scalable, standardized system moves away from labor-intensive, variable manual workflows and toward a process that supports high-throughput applications: large, reproducible batches that are assay-ready. That scalability means researchers can run larger screens or replicate experiments more reliably, without being constrained by the bottlenecks of traditional manual handling.

The sieving step is important for reproducibility in the final organoid product. By narrowing the size distribution of the clusters that go into the bioreactor, we ensure the organoids that grow out are more uniform. Uniformity in starting size, together with the controlled growth environment in the bioreactor, translates into less variance in growth dynamics, and ultimately organoids that are more consistent in size and morphology.

For the end user, this means that when they use the organoid product, they are working with a group of organoids that are roughly the same size and growing at approximately the same rate, which reduces a major source of experimental noise. Consistent size means more consistent diffusion of nutrients, oxygen, and growth factors into the organoid, more consistent exposure to any therapeutic agents being tested, and more reproducible responses across the batch.

That improves the quality of the models and gives scientists greater confidence that differences in downstream assays reflect real biological effects, not random variation arising from uneven starting materials.

The use of ECM for organoid culture is usually necessary in order to maintain stable cultures long term; not only does it provide important structural support, but it also supplies critical signaling cues to ensure that organoids grow and develop correctly.

However, the use of ECM is also a pain-point for researchers. It is usually animal-derived, with significant variability from batch to batch, it is costly, and its unusual physical properties make it difficult to pipette and handle reproducibly without cell loss.

While there are multiple practical and scientific benefits to removing or substantially lowering the use of ECM—such as improved sustainability, reduced cost, and easier handling—the criticality of the ECM for organoid maintenance and development means that the approach to this needs to be very cautious, ensuring that the biology of the model system is not inadvertently negatively impacted.

Our bioreactor-based process is compatible with organoids growing embedded in ECM, ensuring they can be maintained in their optimal growth environment. However, we can also optimize the bioprocess operating parameters on a line-by-line basis to ensure maximum yield from the bioreactor for each specific organoid line. This means we can maximize the efficiency of ECM use compared to manual culture, and tailor conditions to the needs and characteristics of the line at hand. 

There’s growing recognition across the scientific community that 2D models alone aren’t enough to support the kinds of data packages researchers are now expected to deliver. Funding bodies, peer reviewers, and regulators are increasingly asking for more human-relevant systems—3D models, organ-on-chip approaches, and even more complex multi-tissue constructs. For many researchers, especially those who haven’t worked with these systems before, that can feel overwhelming.

Moving deeper into complex biology depends on having models that are well-characterized and validated, and that can be used reliably across applications ranging from relatively simple co-culture assays to more sophisticated 3D tissue structures. Those models also need to be robust enough to be used consistently over time, spanning a research and development project timeline that can often reach 10 years of work.

A major focus for us has been lowering the barrier to entry by making these models as straightforward as possible to handle—so researchers can adopt them without needing extensive training or specialized expertise.

As adoption grows, the next challenge is standardization. If scientists are going to depend on more complex, human-relevant models, those models need to be consistent, defined, and validated for specific contexts of use. This is particularly important in areas such as toxicity testing, where the US Food and Drug Administration is increasingly encouraging the development and validation of complex models to support drug applications or guide decisions about which patients may benefit from particular therapies.

Establishing standard go-to assays and well-understood models is a large undertaking, but it’s an important one, and something the field will need to tackle together. 

At the same time, the future is being shaped by the creativity coming out of academia—new model systems, new engineering approaches, and entirely new ways of shaping and structuring tissues to influence cell behavior. It’s an exciting convergence of disciplines: tissue engineers, materials scientists, and automation and process engineers are all contributing advances in scaffolds, physical microenvironments, fluid flow, and scalable manufacturing. That interdisciplinary momentum is what will continue driving organoid research forward, even as we collectively work through the challenges of making these models more accessible, standardized, and widely usable.

The introduction to this interview includes text that has been created with the assistance of generative AI and has undergone editorial review before publishing. Technology Networks' AI policy can be found here.