Some years ago, I came across a study that stopped me in my tracks. Cyclists who ate pistachios before a 75-kilometer time trial performed worse than those who didn't. Not marginally—4.8% slower. Statistically significant. Why would a healthy nut sabotage athletic performance?
The culprit: raffinose, a trisaccharide in pistachios. During intense exercise, your gut becomes more permeable—and raffinose from the colon leaked into the bloodstream. Once there, it correlated with oxidative stress and leukotoxic effects. Your body was literally fighting the fuel you just gave it.
This concept is discussed in greater detail in the Smart Biotech Scientist Podcast, hosted by David Brühlmann, founder of Brühlmann Consulting.
I thought: if raffinose can slip through a stressed gut barrier and trigger metabolic effects in humans, what does it do to CHO cells under bioprocess stress? That question led to one of the most surprising findings of my PhD work.
Before we dive into that story, let me give you the roadmap for today. First, I'll walk you through why glycosylation matters for biosimilars and what happens when it goes wrong. Then, I'll show you the biology of how raffinose changes glycan processing. After that, I'll take you through the experimental journey—what we tried, what failed, and what finally worked. And I'll close with the real-world impact this had on actual development programs.
But let me tell you what this episode is really about. It's about media design that increased high mannose glycans 2.8-fold in CHO cell culture runs. It works in multiple cell lines. It doesn't require cell line reengineering or expensive glycosidase inhibitors. It's not a universal fix—I'll tell you exactly when it works and when it doesn't—but if you're 6 to 12 months from filing, this could save your timeline.
Before we go further, a quick primer for anyone who's new to glycosylation or needs a refresher.
Glycosylation is the process where sugar chains get attached to your protein as it's being made. Think of it like decorating a cake as it comes out of the oven—except the decorations determine whether your drug works or gets cleared from the bloodstream in 10 minutes.
This happens in two cellular compartments. First, the endoplasmic reticulum adds a core oligosaccharide structure—14 sugars arranged in a specific tree-like pattern. That's the starting template. Then, the protein moves to the Golgi apparatus, where the real action happens. Enzymes in the Golgi trim off some sugars and add others, sculpting that core structure into the final glycan pattern.
For monoclonal antibodies, glycosylation affects a lot. Half-life in circulation. Immune activation through Fc receptors. Protein stability. And most importantly for biosimilar developers: whether regulators will approve your product.
Get it wrong, and you're looking at 6 to 12 months of rework. Get it right, and you're on track for approval.
Now, back to raffinose and why it matters.
Let me paint you a scenario that's all too common in biosimilar development.
You've spent 18 months developing a biosimilar mAb. Your cell line is stable. Titer is great—3 grams per liter. You're hitting your productivity targets. The team is feeling good.
Then analytical drops the bomb. Your high mannose levels are 1.4 percent. The reference product? 6 percent. You're out of spec.
Now, you might be thinking: why does this matter? It's just a small difference in sugar composition.
Here's why it matters. High mannose glycans affect antibody-dependent cellular cytotoxicity—ADCC. They affect receptor binding kinetics. They affect serum clearance rates. And most critically, they affect regulatory comparability. If your glycan profile doesn't match the reference product within an acceptable range, regulators will ask for additional studies. Or they'll reject your filing outright.
So what are your options?
1️⃣ Option one is a temperature downshift to ~32–33 °C. It can push the glycan profile toward higher high mannose, but the productivity tradeoff is clone- and process-dependent.
2️⃣Option two: kifunensine. It's a glycosidase inhibitor that blocks mannosidase enzymes in the Golgi. It's effective—you get high mannose, no question. But it's expensive, hard to scale, and regulatory agencies scrutinize any process that uses enzyme inhibitors. You'll spend months justifying it in your CMC package.
3️⃣Option three: re-engineer your cell line. Go back to the drawing board. Select a new clone with better glycosylation characteristics. This works, but it adds 12 to 18 months to your timeline. If you're racing to market against competitor biosimilars, that delay could cost you hundreds of millions in lost revenue.
What you really need is a media-based solution. Something that scales. Something that doesn't torpedo your productivity. Something that regulators understand because it's just a media component adjustment.
That's where raffinose comes in.
Let me explain the biology of what's happening when you add raffinose to your cell culture.
Raffinose is a trisaccharide—three sugars linked together: galactose, glucose, and fructose. CHO cells can't metabolize it efficiently, so it accumulates in the culture medium and in the cell.
It turns out that raffinose competes with a specific enzyme in the Golgi: N-acetylglucosaminyltransferase, or GlcNAc transferase. This enzyme is responsible for adding the first branch to the core glycan structure. Think of it as the fork in the road where a simple glycan becomes a complex, multi-antennary structure.
When raffinose is present, it binds to the active site of GlcNAc transferase. It doesn't shut the enzyme down completely—it just slows it down. The result? Your antibody ends up with Man5—mannose-5—instead of the complex, fully elaborated glycans you'd normally see.
This is different from kifunensine. Kifunensine blocks mannosidase I and gives you Man8 and Man9—much earlier intermediates. Raffinose arrests glycan processing at a later step, giving you predominantly Man5.
We confirmed this mechanism with transcriptomics. When we looked at gene expression in cells treated with raffinose, we saw downregulation of galactosyltransferase—GalT—and other late-stage Golgi enzymes. The cells were adapting to the metabolic stress by dialing back their glycan elaboration machinery.
The key thing to understand is this: raffinose doesn't shut down glycosylation entirely. It arrests it at a specific step. That's why you get Man5 enrichment, not a complete block. And that's why it's tunable—you can dial the concentration up or down to hit your target glycan profile.
Now, knowing the mechanism is one thing. Making it work in practice? That's where things got interesting.
Let me take you through what we actually tried in the lab. Because the path to the breakthrough was not straightforward.
1️⃣ Attempt one: standard supplements.
We started with the usual suspects. Manganese chloride—known to affect glycosyltransferase activity. Galactose and GlcNAc at various concentrations—trying to push or pull the glycan pathway. We tested these in 96-well plates across a range of concentrations.
Result? Minor glycan shifts. Nothing reproducible. Batch-to-batch variability was all over the place. No clear dose-response relationship. We were tweaking the system, but we weren't controlling it.
2️⃣ Attempt two: osmolality manipulation.
We knew that osmotic stress affects Golgi function. So we tried increasing osmolality by adding raffinose at high concentrations—50, 75, 100 millimolar. The idea was to stress the Golgi and slow down glycan processing.
Result? Growth crashed. At 50 millimolar raffinose, cell viability dropped below 70 percent by day 5. Titer tanked. We thought we'd hit a dead end. Raffinose looked like a non-starter.
But then we had a realization. We were confounding two variables: raffinose concentration and osmotic stress. High raffinose meant high osmolality, and we couldn't tell which factor was causing the growth inhibition.
3️⃣ Attempt three: the constant osmolality pivot.
This was the key experiment. We adjusted sodium chloride in the medium to keep osmolality constant at 315 milliosmoles per kilogram—regardless of raffinose concentration. So when we added 30 millimolar raffinose, we subtracted enough NaCl to maintain constant osmolality.
Result? Growth recovered. Viability stayed above 90 percent through day 14. And now we could test raffinose concentrations up to 100 millimolar without killing the cells.
Once we isolated raffinose's effect from osmotic stress, the data became crystal clear. We had a dose-response curve. We had reproducibility. We had scale-up potential.
It turns out that the breakthrough wasn't in finding a magic ingredient. It was in controlling the variables so we could see what raffinose was actually doing.
Here's what worked.
50 millimolar raffinose increased high mannose 2.8-fold compared to the control. The effect was robust. It worked in two different cell lines. It worked in two different basal media. It worked at every scale we tested: 96-well plates, shake tubes, benchtop bioreactors.
The surprise? The glycan profile was predominantly Man5. Not Man8 or Man9 like you'd see with kifunensine. This told us we were hitting a different step in the Golgi processing pathway.
Now, I was fortunate to apply this approach in actual development projects. I can't share specifics—those programs are confidential—but I can tell you this: when you're staring at a glycan mismatch 8 months before your IND filing, having a validated media lever in your back pocket is the difference between making your timeline and explaining to leadership why you need another year.
One more thing about this approach that's often overlooked: it's regulatory-friendly. Raffinose is a simple trisaccharide. It's available as a GMP-grade material from multiple suppliers. It's not an enzyme inhibitor. It's not a genetic modification. It's a media optimization—something process development teams do all the time.
If you're pre-IND, this is straightforward. You're optimizing your process, and media composition is part of that. Document it. Include it in your development report. Done.
So we proved raffinose works. But here's the hard part: how do you actually implement this in your process without spending 6 months on design-of-experiments studies?
In Part 2, I'll give you the exact decision tree. When to use raffinose. When it won't work. And the three experiments you need to de-risk it before committing to your manufacturing campaign.
If you want to dig into the full methods and data, the paper was published in the Journal of Biotechnology, 2017, volume 252, pages 32 to 42. DOI: 10.1016/j.jbiotec.2017.04.026.
Thanks for joining me in exploring the biology behind raffinose and glycan control in CHO cell culture.
If you found this episode valuable, I'd love your feedback. The best way to share it is by leaving a review. It helps other scientists discover these insights and lets me know what's resonating with you.
See you in Part 2.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
🧬 Stop second-guessing your CMC strategy. Our fast-track CMC roadmap assessment identifies critical gaps that could derail your timelines and gives you the clarity to build a submission package that regulators approve. Secure your assessment at https://stan.store/SmartBiotech/p/get-cmc-clarity-in-1-week--investor-ready
A new era is emerging in cancer diagnosis and therapy—one that extends beyond common indications like prostate cancer to address notoriously difficult-to-treat tumors that have long challenged oncologists. Radiopharmaceuticals, with their unique ability to both detect and destroy disease, are rapidly redefining what is possible in clinical research.
In this episode of the Smart Biotech Scientist Podcast, David Brühlmann speaks with Bryan Miller, Director of Scientific and Technical Operations at Crown Bioscience UK, about how advanced preclinical models are used to help clients advance innovative cancer therapies toward the clinic.
I think we are also seeing radiopharmaceuticals developed for a wider range of cancers. We do have therapies available for prostate cancer and neuroendocrine cancer, but we are seeing this expanding—and this will further expand over the coming years, which I think is a really exciting prospect. I think particularly the value that Crown Bioscience brings is what we touched on before in terms of the value of the models that Crown has in the radiopharmaceutical field. We can now offer our platforms of preclinical models to our clients to run radiopharmaceutical studies.
David Brühlmann [00:00:34]:
Welcome back. In Part One, Bryan Miller walked usthrough his journey from cancer research toradiopharmaceutical development and explained why this therapeutic class is capturing the attention of oncology researchers worldwide. Now we’re shifting gears to explore how Crown Bioscience is scaling its preclinical capabilities to meet exploding demand, the strategic partnerships shaping the field, and the technological innovations that could define radiopharmaceutical development over the next five years. Let’s dive back in.
Now I’d like to touch upon a different part of the technology. There’s a lot of innovation happening. Many are saying that we’re entering a new era for radiopharmaceutical innovation. What makes this moment or season different, and how are you as a company positioned to help shape that evolution as this field scales?
Bryan Miller [00:02:44]:
Yeah, I completely agree. I think we are entering a new era for it. It’s quite exciting what we’re starting to see, and one of the things that is most exciting to me—I touched on it a little before—is the development of diagnostics. We’re certainly seeing increasing interest and excitement about the potential of radiopharmaceuticals.
We’re also seeing increased interest in the theranostic use of radiopharmaceuticals—this idea that you can use the same material as both a diagnostic agent and a therapeutic agent. There’s huge potential here, particularly for patient stratification and rapid movement from diagnosis to treatment.
We’re also seeing radiopharmaceuticals developed for a wider range of cancers. We do have therapies available for prostate cancer and neuroendocrine cancer, but this is expanding and will continue to expand over the coming years, which is very exciting.
I think particularly the value that Crown Bioscience brings is in the preclinical models we offer for radiopharmaceutical studies. Our platforms give clients the opportunity to run studies using very advanced, well-characterized models, which opens up a lot of possibilities for research and development in this field.
David Brühlmann [00:04:14]:
What kind of limits do these models have, and where do you see the development going? What kinds of limitations are we likely to overcome very shortly?
Bryan Miller [00:04:26]:
I wouldn’t suggest there are strict limitations as such. As mentioned, we have a number of different platforms, so we can literally take a study from two-dimensional (2D) cell lines to three-dimensional (3D) organoids, then into CDX in vivo or PDX in vivo. That gives a very strong workflow in terms of maturing a study over time. By the time you’re running a study in a PDX model, you’re working with a model that recapitulates the heterogeneity of human tumors and provides highly clinically relevant information, which can generate very powerful data.
David Brühlmann [00:05:04]:
Perhaps give a definition once again of PDX and CDX. What does CDX mean?
Bryan Miller [00:05:11]:
CDX is a cell line-derived xenograft. These are xenograft models where a human cell line is implanted into a mouse to form a tumor, either subcutaneously or orthotopically.
For PDX, that’s a patient-derived xenograft. These models are set up from patient tumor samples—they’ve never been cultured in vitro. The tissue is grown in animals, then cryopreserved from the animal tumors. This approach retains many features of the original tumor, so it provides a more physiologically and clinically relevant output than a CDX, where the cells have undergone selection and adaptation to culture. PDX models retain much more of the tumor’s original characteristics.
David Brühlmann [00:05:51]:
The power of using models, as you highlighted before, is that you can move faster. But another purpose is to have more predictive models to get closer to your targets. What strategies do you use to achieve that?
Bryan Miller [00:06:09]:
I would say there are two main strategies, depending on whether you’re at the in vitro or in vivo stage.
Our organoid models are incredibly valuable. They form 3D structures that begin to recapitulate features you would see in an organ. That gives a much more clinically relevant output than a traditional 2D cell culture model.
With PDX models, the key is that they retain many features of the original tumor. Crucially, they maintain heterogeneity, unlike a tumor grown from a cell line. This intrinsic heterogeneity means that the model’s response to therapy better reflects what we would see in a clinical context.
David Brühlmann [00:07:04]:
I’m curious, Bryan. What I’m hearing is that there’s a lot happening in the space—innovation, changes, breakthroughs. What breakthroughs are you most excited about?
Bryan Miller [00:07:17]:
For me, it’s theranostics. That’s something I find very exciting, particularly their applicability to treating cancers. Take PDAC (pancreatic ductal adenocarcinoma)—very difficult to treat. Current treatment options are limited, and the five-year survival rate is still very low.
What excites me is the development of radiopharmaceuticals with theranostic applications for these hard-to-treat cancers. I see a lot of potential here, and we are already starting to see more development in this space. I expect it will continue to advance significantly in the coming years.
David Brühlmann [00:08:01]:
And how do you balance, in your role, moving fast and innovating quickly while also working in an industry where everything will eventually be regulated? Even at early stages, how do you factor in regulatory considerations and ensure that patients will ultimately receive what’s being developed? How do you manage these two worlds?
Bryan Miller [00:08:26]:
Well, what I would say is that robust quality standards are always absolutely crucial. At Crown Bioscience, we’re dedicated to delivering scientific studies of the highest possible quality. In setting up our strategic collaboration with Medicines Discovery Catapult (MDC), we partnered with an organization that shares our values and commitment to quality.
Having these robust quality systems in place allows our teams to perform work rapidly to meet essential timelines while maintaining extremely high standards. There’s no conflict between high-quality research and speed. If you conduct a low-quality study, that will set your timelines back—poor quality prevents you from getting the answers you need and delays the entire research program.
It’s crucial to work with partners who uphold very high standards of quality and QC, and this is ingrained in the culture at Crown. Our clients know the standards we work to, and we consistently deliver on timelines while maintaining the highest quality.
David Brühlmann [00:09:35]:
Let’s make this practical. What advice would you give a smart biotech scientist who’s curious about radiopharmaceutical development? Where should they start, and what can wait until later?
Bryan Miller [00:09:53]:
I would suggest working with the right partners. Developing a robust strategy for your study is critical—how you’re going to target the tumor, construct your therapeutic, and select preclinical models. Drawing on the knowledge of experienced people in the industry is crucial, and that’s a lot of the value a CRO can bring.
At Crown, we have extensive experience helping clients in the preclinical space. We can advise on study development, help design your study, guide model selection, and assist with isotope choice and labeling strategy for your test material. Partnering with the right experts early on ensures you have a strong strategy to deliver a successful study.
David Brühlmann [00:10:48]:
How do you make a partnership like this successful? Entering into a relationship with a CRO—or later, a CDMO—is sometimes like a marriage. It's very similar. You have to select very well your partner. What is your approach there? Because I'd say not every company fits every CRO. There are differences and it doesn't always make sense for a company to go to one because it's the best or whatever reason everybody recommends. Sometimes it makes sense to go to another. Right.
Bryan Miller [00:11:19]:
I think people choose CROs on different criteria. At Crown, we do work a very wide array of clients with different needs and different interests. And we can always tailor a study to the individual client. One of the things that we always try to do is we do like to see ourselves as an extension of that client's lab. We work in close collaboration in the design of the study. And there's very open communication during the progress of the study as well. I think the success is seen by the amount of repeat business that we get. And we do form these close collaborations with our clients because we are really dedicated to helping them achieve the objectives of their studies. And we've had the feedback time and again that they start to see us as part of their team. They almost see us as part of the research team and they collaborate very closely with us in terms of delivering the study successfully.
David Brühlmann [00:12:14]:
Now I would like to look forward into the future, since there’s a lot of innovation going on. What is your picture? Where do you see this field in five years?
Bryan Miller [00:12:26]:
It’s a dynamic field at the moment. There’s already a lot of exciting progress, and there’s more to come. We’re certainly seeing rapid growth in interest in radiopharmaceuticals, and I expect this to continue in the coming years.
What excites me is the prospect of increasing the number of cancer indications for which radiopharmaceuticals are a viable option—for diagnosis and therapy. This is particularly important for hard-to-treat cancers, where current treatment options are limited. I think radiopharmaceuticals offer the potential to really advance new treatments in these areas. So I anticipate we will see a significant impact of radiopharmaceuticals in treating these challenging cancers in the years ahead.
David Brühlmann [00:13:18]:
Bryan, what is the most important takeaway from our conversation?
Bryan Miller [00:13:24]:
I think the most important takeaway is the level of excitement and potential in radiopharmaceuticals, especially for oncology. Radiopharmaceuticals also have applications beyond oncology in other disease areas.
As I’ve mentioned, Crown Bioscience is a strong partner for these types of studies. We have a fantastic library of models, both in vitro and in vivo, available for radiopharmaceutical research. In partnership with Medicines Discovery Catapult (MDC), we have the expertise to help design and deliver highly successful radiopharmaceutical programs.
David Brühlmann [00:14:09]:
Where can people get a hold of you and further have a conversation about your work?
Bryan Miller [00:14:15]:
People can contact me through LinkedIn and I'd be delighted to give them any more information that they need on breeder pharmaceutical programs at Crown or indeed any other services that Crown offers.
David Brühlmann [00:14:25]:
Fantastic. I’ll leave the links in the show notes on Smart Biotech Scientist. Bryan, thank you so much for coming on the show today.
Bryan Miller [00:14:33]:
Thank you, David.
David Brühlmann [00:14:34]:
And that wraps up our conversation with Bryan Miller.
The radiopharmaceutical revolution is just getting started, and the preclinical strategies we discussed today could reshape how these therapies move from concept to clinic. If you are navigating the complexities of biotech development—whether in radiopharmaceuticals or beyond—I hope this conversation sparked some new ideas.
If it did, take 30 seconds to leave a review on Apple Podcasts or wherever you’re listening. Your feedback helps us bring more insights to the biotech community.
Thank you for tuning in. Until next time.
For additional bioprocessing tips, visit www.bruehlmann-consulting.com. Stay tuned for more inspiring biotech insights in our next episode. Until then, let’s continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Bryan Miller
Bryan Miller is Director of Scientific and Technical Operations at Crown Bioscience, with over a decade of experience in oncology research and preclinical drug development. He earned his PhD in Biochemistry from the University of Leicester and completed postdoctoral training at the University of Toronto and the Beatson Institute, specializing in in vivo and in vitro models of colorectal and pancreatic cancer.
After moving into the CRO field in 2015, Bryan has focused on translating strong science into effective preclinical strategies for oncology programs. Since joining Crown Bioscience UK in 2019, he has played a key role in guiding scientific operations and supporting clients across diverse cancer indications and therapeutic approaches.
Connect with Bryan Miller on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
Imagine treating cancer with something as precise as a guided missile—radioactive payloads delivered only where they're needed most. Radiopharmaceuticals are redefining both diagnostics and therapeutics, creating new momentum in oncology by pairing tumor-seeking molecules with potent radioisotopes. But what does it really take to develop these agents, and why are investors and scientists alike turning their attention to this field?
In this episode of Smart Biotech Scientist Podcast, David Brühlmann sits down with Bryan Miller, Director of Scientific and Technical Operations at Crown Bioscience UK, a Contract Research Organization (CRO) specializing in translational oncology and immuno-oncology drug discovery and development.
Radiopharmaceuticals are diagnostic and therapeutic agents. So they consist of a targeting agent—typically a small molecule or an antibody—and this will target the tumor. There’s also a radionuclide, and there’s a linker that connects the two. The radionuclide can act as either an imaging agent for diagnosis of cancer or as a therapeutic agent, where the radionuclide delivers a lethal dose of radiation to the tumor.
David Brühlmann [00:00:27]:
What if the next breakthrough in cancer treatment isn’t a pill or an infusion, but a precisely targeted radioactive payload that hunts down tumor cells like a guided missile? Today we’re diving into the radiopharmaceutical revolution with Bryan Miller, who is the Director of Scientific and Technical Operations at Crown Bioscience UK. From his early days studying colorectal cancer models to now leading preclinical development for one of oncology’s hottest therapeutic classes, Bryan is taking us inside science that’stransforming how we fight cancer.
Welcome, Bryan, to the Smart Biotech Scientist Podcast. It’s good to have you on today.
Bryan Miller [00:02:25]:
It’s great to be here, David.
David Brühlmann [00:02:27]:
Bryan, share something that you believe about early discovery research that most people disagree with.
Bryan Miller [00:02:34]:
So I’ve found that research progress is rarely linear or predictable. Over my career, I’ve seen the undruggable become druggable. I’ve seen false starts in some areas and surprisingly rapid advances in others. And I think that’s why I find it so dynamic and exciting—and why I love being a part of it.
David Brühlmann [00:02:51]:
Excellent. Draw us into your story. Tell us what sparked your interest in working on cancer models, and then what were some interesting pit stops along the way that got you to your current role?
Bryan Miller [00:03:03]:
Yeah, it’s quite interesting. So my PhD was actually in cardiac disease. So that’s where I started out. It was when I did postdoctoral fellowships—first at the University of Toronto and the second at the Beatson Institute (UK)—that really took me towards oncology models, both in vitro and in vivo. My postdocs were focused on models of colorectal and pancreatic cancer. And I think what motivated me at that point is exactly the same as what motivates me now. I was passionate about contributing to the development of new therapies for cancer and identifying new therapeutic targets. And really that’s what motivates me to go to work every day to this day. Being part of Crown Bioscience allows me to work across a large array of different cancer indications with a diverse range of clients. And it’s really motivating when you’re contributing to the development of new therapies. It’s fantastic when you’ve contributed to a research program that has led to a therapy going into the clinic.
David Brühlmann [00:03:58]:
Tell us a bit more about your current work. What does that look like when you’re developing these new therapies?
Bryan Miller [00:04:05]:
So we are a preclinical contract research organization, oncology focused, but we do work with a large array of different clients working on various therapeutic types. We have a range of advanced preclinical models available to our clients to run their research programs with. We work across most cancer indications and across most therapeutic types. So it’s a very diverse area of work that we conduct at Crown, which makes it really interesting. It’s very varied, it’s very interesting, and it’s absolutely fantastic to support this variety of client research projects.
David Brühlmann [00:04:42]:
And when you say it’s very diverse, that includes the various diseases you’re working on, and I imagine also very diverse companies with very diverse needs. Before we dive a bit further into radiopharmaceuticals, what is the main need companies have when they come to you? What is the main problem they want to solve?
Bryan Miller [00:05:04]:
They typically come to us with a target in mind and usually with a therapeutic that they wish to test. One of our main roles is to guide them towards the most appropriate preclinical model to address the question they’re asking. We have a range of different models available. We’ve got an excellent array of in vitro models, and we’re particularly well known for our organoid platform. So certainly at the in vitro phase, we have advanced models that can be very valuable. We’re also very well known for our PDX library. We have PDX models that cover most cancer indications, and we have genomic data associated with them. If you have targets in mind, we can guide you towards the most appropriate PDX model for your work. In addition, we have an array of TDX models, genetically engineered models, and humanized models. So really, while we’re cancer focused, for most questions around the development of cancer treatments—including neuro-oncology therapies—we will have appropriate models to support your research program and help you filter down to the most relevant ones.
David Brühlmann [00:06:08]:
When you say models, are these animal models, or are these 2D cultures, 3D cultures, or a combination of various models?
Bryan Miller [00:06:16]:
All of that, actually. So yes—2D cell culture, 3D cell lines, organoids, and various in vivo models. That includes syngeneic models, CDX models, PDX models, and humanized models. So it really is quite a wide array.
David Brühlmann [00:06:31]:
Excellent. So let’s dive into one specific area—radiopharmaceuticals. A lot of our listeners, I imagine, are not familiar with that. Can you start with the basics? What are radiopharmaceuticals, and how do they differ from traditional cancer therapies?
Bryan Miller [00:06:52]:Yeah, so radiopharmaceuticals are diagnostic and therapeutic agents. They consist of a targeting agent—typically a small molecule or an antibody—which targets the tumor. There’s also a radionuclide, and there’s a linker that connects the two. The radionuclide can act as either an imaging agent for the diagnosis of cancer or as a therapeutic agent, where the radionuclide delivers a lethal dose of radiation to the tumor.
David Brühlmann [00:07:18]:
Can you elaborate a bit more? Give us an example. How are they different from a traditional cancer therapy? Is it the morphology, the mechanism, or what exactly are the main differences?
Bryan Miller [00:07:32]:
I’d say there are probably two real advantages of radiopharmaceuticals. The first is safety, because you’re dealing with a therapeutic that can very specifically target a tumor. One of the early stages is to characterize the level of accumulation within the tumor and make sure you don’t have accumulation in other organs of the body. So they can have a greater safety profile than more traditional therapies. That’s one advantage.
The other thing that distinguishes them is the idea of theranostics. You can use an agent that combines diagnostics with therapy, using the same scaffold. You can use one radionuclide to diagnose a tumor and then switch to a second radionuclide to target and kill the tumor. Because you’re combining those within the same strategy, this allows for really good stratification of patients and very rapid progression from diagnosis through to therapy.
David Brühlmann [00:08:28]:
Okay, I see. So it’s faster, and there are a lot more things happening in parallel, giving you a much broader picture.
Bryan Miller [00:08:36]:
Yeah, and I think safer as well. A lot of them are much safer than traditional chemotherapies.
David Brühlmann [00:08:41]:
The radiopharmaceutical field has gone from a niche therapy to one of the hottest areas in oncology. What’s driving this rapid transformation, both on the scientific side and on the commercial and investor side?
Bryan Miller [00:09:04]:
That’s what we’re finding as well. There does seem to be growing interest in the area. It’s quite dynamic right now, and there are a lot of exciting things happening. I think it goes back to some of the advantages I just described, particularly the theranostic approach, which gives these therapies a considerable advantage.
I think the safety profile is another aspect that a lot of companies we work with find very valuable during development. And maybe the third aspect is the widening range of cancer indications that radiopharmaceuticals are being applied to. We’re starting to see more radiopharmaceuticals coming through that target a much more diverse range of cancers.
David Brühlmann [00:09:44]:
And the main purpose of all that is basically to make faster, more actionable, and more predictive decisions with all your models. I’m curious—on top of your models, do you add a layer of machine learning, AI, or advanced data-driven approaches? How does that work in your field.
Bryan Miller [00:10:05]:
In terms of model selection?
David Brühlmann [00:10:08]:
Yeah, and also in terms of interpretation of the data.
Bryan Miller [00:10:11]:
I would say that is certainly applicable to the selection of models. It’s absolutely crucial—you need to get your preclinical model right in order to have a successful research program. We have large datasets associated with all of our models. So certainly, if you’re looking for particular targets or particular gene profiles, we can help you with that. Our models are very well characterized, and that’s information we’re happy to share to guide you towards the most appropriate model.
We’re also very excited by the prospect of AI. I think that’s something that’s much more common now, and it’s something we’ll be incorporating into workflows as well.
David Brühlmann [00:10:51]:
Yeah, you’re making a very good point—you have to select the right preclinical model, because that’s the foundation, isn’t it?
Bryan Miller [00:10:59]:
Exactly. Yes. If you get the order wrong, you’re not going to have a successful program.
David Brühlmann [00:11:02]:
Yes. You’re going to answer the wrong question at the end of the day if you choose the wrong model. Besides that, what are other pitfalls when developing a new drug? What other choices or strategies really have to be right?
Bryan Miller [00:11:17]:
Yeah, particularly in the radiopharmaceutical field, it can be quite a challenging area. One early point is getting the targeting strategy for the tumor correct. That’s a very important aspect—you need high specificity in tumor targeting.
You also need to select an appropriate isotope and the correct labeling position and labeling strategy. Another important aspect is ensuring that you either have, or are working with someone who has, a reliable source of radionuclides. These materials have to be labeled fresh before dosing, so the supply needs to be reliable.
It’s also really crucial to have a robust labeling strategy and to ensure that proper and thorough quality control (QC) is performed on the labeled material. Nothing should be dosed unless it passes stringent QC tests. And, as we discussed, making sure that the model you select is the most appropriate one is critical to giving you the best chance of success.
These are all areas where we can assist—providing advice, guidance, and helping to design the most appropriate study design.
David Brühlmann [00:12:24]:
And compared to conventional drug development, what are the unique challenges associated with radiopharmaceutical development? You’ve mentioned quite a few things that need to be right. How different is radiopharmaceutical development compared to conventional drug development, and what are the unique challenges?
Bryan Miller [00:12:46]:I mean, obviously there are some commonalities. But I think the unique challenges are the need to prepare material fresh—you can’t make it in bulk and store it. You have to have fresh material that can be dosed almost immediately. There’s also the complexity of some of the strategies, particularly designing an appropriate radiolabeling strategy. Because this is quite a specialized area, you really need to work with people who have specialist expertise and can advise on the most appropriate way to label the molecule, the correct isotope to use, and, crucially, the linker. Those are all aspects of radiopharmaceutical design that are absolutely critical for success.
David Brühlmann [00:13:26]:
And how do you, as a company, manage all that and navigate these difficulties? What are some specific approaches you’ve developed?
Bryan Miller [00:13:35]:
One absolutely crucial element is the collaboration we’ve established with Medicines Discovery Catapult (MDC). We’ve recently launched this strategic collaboration, which brings together the range of preclinical models and preclinical expertise from Crown Bioscience with the radiolabeling expertise and radiopharmaceutical experience from MDC. It’s a really strong partnership, combining complementary strengths to deliver highly successful radiopharmaceutical studies for our clients.
David Brühlmann [00:14:10]:
And what kind of clients do you usually work with? Are they small companies, mid-sized companies, or large pharma organizations?
Bryan Miller [00:14:20]:
All three, actually. We work with a diverse range of clients—small, medium, and large. As we’ve alluded to earlier, there’s growing interest in the radiopharmaceutical field, so we’re seeing more and more organizations approaching us to discuss radiopharmaceutical development and appropriate strategies for further studies. That ranges from quite small companies through to larger pharma organizations.
David Brühlmann [00:14:39]:
I imagine their needs are quite diverse. Can you give us a sense of how those needs differ between small biotech and large pharma? What are the typical differences?
Bryan Miller [00:14:53]:
One of the main differences we see is the maturity of the program. Some clients approach us at a very early stage, without a defined radiolabeling strategy. In those cases, we may need to do significant optimization work—guiding isotope selection, developing the labeling strategy, and performing validation to ensure the material can be labeled effectively and is suitable for use.
Other clients come to us with more mature programs. They may already have an optimized labeling strategy and a defined protocol. In those cases, our role may be to perform the labeling through protocol transfer, rather than developing the process from scratch. So those represent two quite different types of client needs that we commonly see.
David Brühlmann [00:15:42]:
Since interest in radiopharmaceutical development is increasing on both the scientific and business sides, I imagine demand is growing rapidly. How do you ensure that you can scale your preclinical and translational capabilities to meet that demand?
Bryan Miller [00:16:04]:
We have a lot of experience in this area. Both Crown Bioscience and Medicines Discovery Catapult have worked in our respective fields for many years, and we’re accustomed to running a wide array of projects simultaneously. Scalability is something we’re very experienced with.
You’re absolutely right—we’re seeing increasing demand for these types of studies. But we approach scalability in the same way we do for other programs: scaling our in vivo and preclinical capabilities on the Crown side, and scaling the radiolabeling and radiochemistry capabilities on the MDC side. Both partners bring significant experience in delivering complex studies at scale.
David Brühlmann [00:16:43]:
That’s where we’ll pause for today. In Part Two, we’ll explore the specific platforms accelerating radiopharmaceutical translation and get Bryan’s predictions on where this field is heading. If you’re finding value in these conversations, please leave us a review on Apple Podcasts or your favorite platform.
It helps other biotech scientists like you discover the show. See you in Part Two.
Alright, smart scientists—that’s all for today on the Smart Biotech Scientist Podcast. Thank you for tuning in and joining us on your journey to bioprocess mastery. If you enjoyed this episode, please leave a review on Apple Podcasts or your favorite podcast platform. By doing so, we can empower more scientists like you.
For additional bioprocessing tips, visit www.bruehlmann-consulting.com. Stay tuned for more inspiring biotech insights in our next episode. Until then, let’s continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Bryan Miller
Bryan Miller is Director of Scientific and Technical Operations at Crown Bioscience. He completed his PhD in Biochemistry at the University of Leicester, followed by two postdoctoral fellowships at the University of Toronto and the Beatson Institute. During this time, his research focused on the development and application of in vivo and in vitro models of colorectal and pancreatic cancer.
Since 2015, Bryan has worked in the contract research organization (CRO) sector, supporting oncology drug discovery and development programs across a range of therapeutic modalities. He joined Crown Bioscience UK in 2019, where he leads scientific and technical operations, leveraging advanced preclinical models to help clients progress innovative cancer therapies toward the clinic.
Connect with Bryan Miller on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
Our food systems face a monumental challenge: by 2050, the global population could reach 10 billion, demanding at least 60% more food than we produce today. This stark reality is one of the main drivers for innovation in agri-biotech. Traditional agriculture alone is unlikely to shoulder the burden, especially under the shadow of climate change, deforestation, and depleted resources.
In this episode of Smart Biotech Scientist Podcast, David Brühlmann steps into the lab with Steven Lang, Chief Technology Officer of California Culture. A veteran of both biopharma (and a former leader at Upside Foods), Steven is at the forefront of scaling plant cell culture for real-world impact.
Our future food systems have to incorporate new technologies because our population is growing. We're going to be 9 to 10 billion people on this globe relatively soon. And there are projections that we need to produce 60% more food than we're producing today by 2050 to support that population.
So what I like to say is that we need to continue with our conventional agriculture and supplement and add to it new technologies like what we're doing with plant cell culture and cellular agriculture, because there have to be multiple shots on goal to be able to feed our population. The alternative is malnutrition and starvation. To me, that's unacceptable.
David Brühlmann [00:00:46]:
Welcome back to Part Two of our conversation with Steven Lang from California Cultured. In Part One, we explored how plant cell culture works and the bioprocessing fundamentals behind cultured cacao. Now we're tackling the hard questions: Can this actually scale? What's the economic reality? And how do we move cellular agriculture from laboratory curiosity to commercial production? Whether you're in biopharma, considering a career pivot, or simply curious about the future of sustainable food, this conversation will change how you think about biomanufacturing.
On the taste part, because that's an important part for me. As a Swiss, I must say I'm pretty critical when it comes to chocolate. Producing cocoa that tastes well is quite a complex process because it's not just a plant. There is a complex fermentation process going on. How do you reproduce that in the lab to get as close as possible to natural cacao?
Steven Lang [00:03:03]:
That is a really important question. And I’d like to set the stage by basically giving a synopsis of how chocolate is made. The cacao pods are opened up on the forest floor at these small farms that grow cacao, and the beans are removed from the pod and then spread out on the floor to ferment. That develops some of the flavors. Then they dry those beans and go into the roasting process, followed by all the processing that goes into creating the cocoa nibs and extracting the cocoa butter to then recreate chocolate.
We’re starting with a different starting material because our cells are not analogous to the cocoa bean that's extracted from the pod, fermented, and roasted. So we have to take a food science approach and really think about how we can ferment our cocoa powder as well as roast it to achieve those flavors. The nice thing is that we can do that. In that process, we degrade some of the flavanols, so the astringent taste is decreased, and you can bring out some of the chocolate flavors.
Let me say to that developing great-tasting chocolate isn’t our primary goal right now, but it will be in the future—that’s where we’re heading toward the commodity market. Right now, we’re focused on scaling and commercializing, aiming to get onto the market in 2026 with our high-flavanol cocoa. Once we get there, we’ll have more time and bandwidth to build processes that allow us to pull levers to improve the sensory attributes of the cocoa powder to create really fantastic chocolate.
Once you understand the biology of those sensory attributes in the cell culture process, you can start building different varietals of tastes and flavors in chocolate. That’s really the exciting biology I’m looking forward to. And I think that will require, based on what we know about plant cell culture media optimization as well as improvements in bioreactors,…
David Brühlmann [00:05:05]:
What are the technologies you need to produce such high-quality cocoa products. Because you have the cell culture, and then you have the whole sensory area, which is highly complex—what do you need, and what kind of equipment do you need to use to make sure you get this high-quality product?
Steven Lang [00:05:24]:
That’s where I like to think that for cellular agriculture to be successful, we need to change people’s minds about it, not just overcome the technical challenges. We can take simpler processes like coffee or cacao cell culture to help explain the processes to people, which will create more consumer demand and actually pull the products from us.
With plant cell culture, we don’t use a lot of process analytical technologies. So we don’t control pH, there is no temperature control, and that’s essentially it. What we really need is macronutrient profiling at the end, flavanol concentrations at the end, and that’s essentially it.
Anything else relates to safety—microbial testing, for example. Speaking of safety, many people know there’s a heavy metal concern with most chocolates. Consumer Reports released a report showing that about one-third of chocolate products contain heavy metals above safe limits. The great thing about our cocoa powder is that because we can control all raw materials, we can produce products with essentially zero heavy metals. That’s a huge step forward when thinking about food safety and convincing consumers to try these products. Not only are they nutritious and healthy, but they’re also safer than conventional products.
David Brühlmann [00:06:50]:
This leads me to the bigger picture, because when we talk about cultivated meat or lab-grown coffee or chocolate, we hear a lot about climate change, deforestation, and environmental challenges, and you and other companies are trying to solve or at least alleviate these problems. How realistic is it that, maybe in five to ten years, we can produce enough in labs to reduce this burden? Or is this not realistic at all?
Steven Lang [00:07:22]:
No, I think this has to be the future. Our future food systems have to incorporate new technologies because our population is growing. We're going to be 9 to 10 billion people on this globe relatively soon. There are projections that we need to produce 60% more food than we produce today by 2050 to support that population.
What I like to say is that we need to continue with conventional farming and agriculture, and supplement it with new technologies like what we’re doing with plant cell culture and cellular agriculture. There have to be multiple shots on goal to feed our population. The alternative is malnutrition and starvation, which to me is unacceptable.
I am very motivated by using my cell culture background and everything I’ve learned in biopharma and at upstream biomanufacturing to really push the envelope. We need to ensure we’re improving human health, food security, and sustainability with the foods we eat in the future, rather than assuming that conventional industrialized farming will solve all our issues. Because frankly, that assumption distracts from the negative aspects of conventional agriculture.
David Brühlmann [00:08:40]:
And if we keep this future-focused lens, zooming out beyond coffee and cacao, what other lab-grown products do you see emerging? What are the hot technologies coming into this space?
Steven Lang [00:08:53]:
I think a lot of supplements as well as other food products could benefit. For example, one of my colleagues just started a saffron company using plant cell culture. Saffron is hugely expensive and difficult to produce, so if we can produce it in cell culture, it really reduces demand on the industry and also produces a higher-quality product. Other interesting products include ginseng, echinacea, and there’s even a startup trying to make cell-cultured wood. That one just blows my mind—it must be premium.
David Brühlmann [00:09:27]:
Wood must be very complex to make. I would imagine—you probably produce some cells in a bioreactor, and then you would need to assemble them to get your finished product at the end, right?
Steven Lang [00:09:43]:
Exactly. Very much like cultured meat. Wood probably has a lot of different components you need to recreate the structure. That’s where the complexity comes in. That’s why we need to start with simpler processes like cocoa or even cell-cultured wood—to build a foundation from which we can later be successful with cultivated meat and allow that technology to mature.
David Brühlmann [00:10:12]:
If one of our listeners thinks, “Well, this is interesting—cellular agriculture is great, you’re solving a big problem,” what advice would you give them as they transition into this exciting field?
Steven Lang [00:10:27]:
First and foremost, look for cell culture work outside of cultivated meat and biopharma. There are startups and companies using this technology—start thinking about where you can get involved. If you’re an entrepreneur, look for products that can be produced from plant cells, because those probably have the fastest route to market and are likely the most consumer-accepted.
The other thing to think about—and we discussed this back in 2023 for people coming from cell culture and biopharmaceutical backgrounds—is that you really need a safety and efficacy mindset. All the products we produce must be safe, and that’s paramount.
Next is efficacy. In drugs, we understand what an efficacious drug is. In foodstuffs or other products, it comes down to things scientists and engineers often overlook: texture, mouthfeel, aftertaste, packaging, and many other factors that are critical to consumers. If you build these considerations into your product intelligently, you can significantly increase your chances of success.
David Brühlmann [00:11:35]:
Before we wrap up, Steven, what burning question haven’t I asked that you’re eager to share with our biotech community?
Steven Lang [00:11:43]:
I’d like to highlight the collaborative aspect. One question I haven’t heard yet is: What does a team look like that can make this successful? I want to call out my technical team. They’ve done amazing work not only with cocoa, but also coffee and cocoa butter. With a small startup over a short period, they’ve developed cell lines capable of producing high-quality products.
We also recently brought in someone to help set up our data foundation. That’s near and dear to my heart. As a startup, setting up a data foundation and capturing all data in a way that can be used in the future is paramount. Many pharmaceutical or mature companies have siloed data systems, which prevents the use of machine learning or AI.
I really want to recognize my technical team, not only for their technical achievements, but also for how we’re building a foundation to implement advanced analytics—the kind of AI-driven insights that are rapidly transforming the industry.
David Brühlmann [00:13:02]:
What is the most important takeaway from all that we’ve discussed today?
Steven Lang [00:13:06]:
David, the most important takeaway is that cellular agriculture is much more than cultivated meat, and we need to start thinking about other processes using plant cell culture to produce products. I really encourage all of your listeners to explore opportunities where you’re doing non-traditional work and don’t be afraid, because cells and these technologies can be evolved to suit our purposes as we develop new products.
I really think about pressure-testing your systems and your assumptions about cells and what they’re capable of. I’ve been surprised throughout my 20-plus years working in cell culture at how far you can push a cell, and it’s amazing what they can do. So I encourage people to continue pushing and look for additional opportunities to help food security, human health, and sustainability.
David Brühlmann [00:14:02]:
This has been great, Steven. Thank you so much for helping us expand our horizons beyond traditional cell culture. It’s exciting what’s happening in that space. Where can people get a hold of you and also potentially taste this high-flavanol product or your coffee?
Steven Lang [00:14:22]:
Certainly reach out through LinkedIn. We’re hoping to be selling our product by the middle to late part of next year. We’re a B2B company, so we’ll be partnering with chocolatiers and hopefully co-branding, but that’s yet to be determined. I’d say definitely look for us in end-2026 or early 2027 in some chocolate products.
David Brühlmann [00:14:46]:
Fantastic. I’m also looking forward to that.
Steven Lang [00:14:51]:
As a Swiss, I’m sure you are.
David Brühlmann [00:14:52]:
Absolutely. Well, thank you, Steven, for these great insights, for expanding our vision, and thank you for being on the show today.
Steven Lang [00:15:02]:
David, I really appreciate what you do here on this podcast. It’s all about scientific communication and bringing people along with what we’re doing, which is very exciting and impactful. We just need more people to understand and get involved, so I truly appreciate the opportunity. Always good talking with you.
David Brühlmann [00:15:21]:
Steven Lang has given us a compelling vision of cellular agriculture’s potential to reshape food production while addressing real sustainability challenges. The bioprocessing fundamentals we discussed today apply far beyond chocolate—they’re principles you can use in your own CMC development work. If this conversation sparked new ideas, share it with a colleague and leave a review on Apple Podcasts or wherever you listen. Until next time, thank you so much for tuning in today and keep doing biotech the smart way. For additional bioprocessing tips, visit us at smartbiotechscientist.com. Stay tuned for more inspiring biotech insights in our next episode. Until then, let’s continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Steven Lang
Steven Lang is a biopharmaceutical executive with more than 20 years of experience across large pharma, CROs, and startups. As Head of R&D, Bioprocess, and Analytics at California Cultured, he leads CMC-driven development of plant-cell-derived cocoa and coffee products.
His expertise spans cell line development, process optimization, analytics, and regulatory strategy, with prior leadership roles at Genentech and Johnson & Johnson.
Connect with Steven Lang on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
What if your daily chocolate or coffee could be brewed without farms—or deforestation—but straight from a bioreactor? It sounds like a technological fantasy, but plant cell culture is quietly remaking some of the world’s most beloved food staples. This episode cracks open the world of cellular agriculture, moving beyond the hype of lab-grown meat to explore how plant cells are ushering in a new era of sustainable food manufacturing.
Joining Smart Biotech Scientist Podcast host David Brühlmann is Steven Lang, Head of R&D at California Cultured. Steven is no stranger to ambitious challenges. After nearly two decades in biopharma with industry giants like J&J and Genentech, he pivoted toward cultivated foods, determined to go beyond the narrow focus of animal cell-derived products.
Cellular agriculture is using cell culture to produce agricultural products. And the frustration that I've had over the last three or four years is that everything that is considered cellular agriculture is actually synonymous with cultivated meat. And there's so much more to cellular agriculture than just meat.
I really want, through this venue as well as other venues, to drive home the message that cellular agriculture includes any type of cell culture product that can be derived from animal cells like cultivated meat, plant cells like cacao, or precision fermentation with microbial cells. And that broader definition of cellular agriculture is important to me because I see there is a resistance from consumers to adopt or try cultivated meat.
David Brühlmann [00:00:55]:
Imagine a world where your favorite chocolate bar doesn't require a single cacao tree, where coffee comes from a bioreactor, not a plantation. Sounds like science fiction. It's happening right now in California. Today we're joined by Steven Lang, who is the head of R&D at California Cultured, who left two decades in biopharma to revolutionize how we produce food. He's cultivating plant cells to create real cacao and coffee. No farms, no deforestation, no compromise on flavor. Let's explore the future of what we eat.
Welcome back, Steven, to the Smart Biotech Scientist. It's a pleasure to have you on today.
Steven Lang [00:02:52]:
It's great to see you again, David. I'm really pleased to have this conversation again and catch up.
David Brühlmann [00:02:57]:
Sure, Steven. Share something that you believe about bioprocess development that most people disagree with.
Steven Lang [00:03:06]:
Interesting. Well, since we talked last, the thing that I've really come to learn is that we need to walk before we can run. And that is not necessarily around bioprocess development per se, but it's more on the biotech industry and how we have to have stage-appropriate models to really kind of push this industry forward. We can get into more of those details, but that's kind of a nutshell of what I think—that we need to really look at simpler models that can help us answer some of these fundamental questions, not only on the technical side, but also on the consumer acceptance side.
David Brühlmann [00:03:39]:
I like that. Learn to walk before you run.
Steven Lang [00:03:43]:
Exactly. Yeah.
David Brühlmann [00:03:45]:
And I'm excited to have this conversation today on cellular agriculture. Let's start with you, because not everyone listened to our first conversation. So I'd love to hear your origin story—what sparked your interest in biotech and what were some pivotal moments, because a lot of things happened actually since we last spoke.
Steven Lang [00:04:05]:
So my origin story we kind of went into in the last podcast, so I'd recommend listeners go back and listen to that one. But just as a synopsis, I've got almost 20 years of experience in biopharmaceuticals with large, mature pharmaceutical companies—Johnson & Johnson and Genentech—as well as small CROs.
About four years ago, I decided that I wanted to do something a little bit different outside of the pharmaceutical arena. The opportunity came along to work on cultivated meat, which blew my mind, because using these expensive bioprocessing technologies to produce a commodity just doesn't make any sense.
As we spoke about back in 2023, that's really exciting for intelligent people—to have huge, audacious challenges to take on—and certainly cultivated meat presented that to me. That's what got me interested in it.
Throughout the biotech industry, there's been a lot of retraction and consolidation, especially in the food sector, which is more cost-conscious than the pharmaceutical sector. I was caught up in that and was laid off from UPSIDE Foods as part of a reduction in force, as they were consolidating their resources to extend their runway. I thought that was a little bit shortsighted, but I wasn't making those decisions.
I took about eight months off and did some introspection on what I wanted to do with the rest of my career. That brought me to wanting to have a larger impact on the world beyond just human health. That's where the opportunity came along to join California Cultured.
As we talk further, I hope to elaborate on how this cell culture technology can really have an impact on human health, food security, and sustainability—all of which are near and dear to my heart. Having the opportunity to work on something with that type of global impact was really compelling for me.
So I made the jump without much hesitation into California Cultured, where we're producing cultured chocolate (cacao powder) to begin with, as well as coffee. Those are the first products we're planning to get out the door.
This is a relatively small startup—about 18–19 people—founded in 2020, and we're building out not only plant cell culture products, but also the supporting biomanufacturing processes. We're actually developing new bioreactors for plant cells, which is fascinating work as well.
David Brühlmann[00:06:37]:
Yeah, I'd love to dive into more. It's fascinating, and I must say I'm a bit biased. As a Swiss, I love, love, love chocolate. And I ask some critical questions.
Steven Lang [00:06:49]:
Perfect, I'm ready for them.
David Brühlmann [00:06:57]:
Let's unpack it from the start because we have heard, or some of you listening have heard, the term cellular agriculture. But even to me, sometimes it's not clear what exactly is in that field. It's also evolving. So I would love to hear your definition. What is inside? Is even cultivated meat inside of that, or is it outside? Tell us more.
Steven Lang [00:07:16]:
No, that's great. Really good question. Because I think this definition has been evolving, and it's important for us as an industry to consolidate around a single definition. And I think that definition should be that cellular agriculture is using cell culture to produce agricultural products.
The frustration that I've had over the last three or four years is that everything that is considered cellular agriculture is actually synonymous with cultivated meat. And there's so much more to cellular agriculture than just meat. I'd also like to argue that the focus on cultivated meat has done us a disservice, because we have had some opposition to cultivated meat—political arenas, political grandstanding—where states and even countries have banned lab-grown meat even before it's available on the market.
So I really want, through this venue as well as other venues, to drive home the message that cellular agriculture includes any type of cell culture–derived products from animal cells like cultivated meat, plant cells like cacao, or precision fermentation with microbial cells. And that broader definition of cellular agriculture is important to me because I see resistance from consumers to adopt or try cultivated meat. So we need to get away from the singular focus on cultivated meat and move more toward these other products that people understand better and adopt more easily.
David Brühlmann [00:08:50]:
When you say precision fermentation is part of cellular agriculture, what comes to my mind—and this is a question—is beer production part of cellular agriculture or not?
Steven Lang [00:09:01]:
You could argue that, but it's traditionally been called fermentation, which is a stretch. You could call that precision fermentation.
David Brühlmann [00:09:08]:
Okay, fair enough.
Steven Lang [00:09:10]:
I think of precision fermentation as really using genetically modified microbes to produce a specific product, whereas beer and wine fermentation are essentially producing alcohol—which is important—but it's not as complicated as, say, producing lactoferrin from yeast.
David Brühlmann [00:09:26]:
Let's walk through the fundamentals of what you're doing. You're now growing cacao cells in a bioreactor. Tell us more. How does this work? Is it similar to a CHO cell culture? Is it very different, and more?
Steven Lang [00:09:40]:
I like to say cell culture is cell culture. What we do in plant cell culture is very analogous to a CHO cell line in cell line development or any type of cultivated meat, where you take a biopsy from the agricultural plant or animal, and you immortalize those cells and then expand them to produce biomass that is the product. So there are no differences there.
The differences are really about the terminology. With plants, what we do is take an explant instead of a biopsy. And plants are interesting because all of their cells are totipotent, so they can be differentiated into a full plant from any individual cell.
What we've done here at California Cultured, with our cell line development group, is take explants from a cacao pod. We aseptically open the pod, expose the beans inside, and take explants from those beans. Those explants are then put on solid agar media and induced to dedifferentiate. Those are called callus cells.
This technology has actually been around longer than a lot of animal cell culture, because it was historically used to propagate plants for the field. We can take those callus cells on the plate and use them as a cell bank, then screen and select the cell lines we want and adapt them to suspension culture, much like you would with animal cells, and then scale them up.
From there, once we go into scaled production, it's a very simple process—very much like cell culture. We're controlling the environment and the culture, running for a certain number of days, and then harvesting the bioreactor.
One big difference with the plant cell culture products we're working on is that there's relatively little downstream processing. Our downstream processing is dewatering, drying, milling, and packaging. Compare that to cultivated meat, where you take the biomass from the bioreactor and then have extensive downstream processing—such as scaffolding, maturation, or formulation with animal or plant-based proteins to create a meat analog.
With our plant cell culture products, downstream processing is very simple. And I think that alleviates some consumer concerns about cellular agriculture products because it's easier to understand. The products coming out of the bioreactor also really look like conventional cocoa powder or coffee. So it's easier for consumers to connect what we're producing with what they already think of as cocoa powder and coffee.
David Brühlmann [00:12:33]:
Regarding consumer acceptance, I think this cocoa product is much easier to sell because it's not a genetically modified cell, correct?
Steven Lang [00:12:43]:
Correct. We have not done any genetic modification to our cell lines or products. We're relying on the natural diversity in the genetics and epigenetics of cacao to select the cell lines we're interested in—not only based on growth parameters, but also the phenotypes we care most about, which are the sensory attributes and higher levels of bioactives that are important in cacao and cocoa powder.
David Brühlmann [00:13:11]:
Do you clone the cell or your cell lines as you would do in cell culture?
Steven Lang [00:13:16]:
Not necessarily. Since we're not doing any genetic modification, we don't need to ensure that our product originates from a single cell, like what the FDA requires for biopharmaceuticals. Essentially, we do perform clonal selection, but these cells tend to aggregate, so it's difficult to say whether the production unit is a single cell or an aggregate.
That said, we have strict quality control processes and specifications to ensure we're producing a safe, high-quality product.
David Brühlmann [00:13:46]:
And how do you manage the variability that comes from different cells? I imagine the taste will change, maybe the texture.
Steven Lang [00:13:55]:
There's not a lot of that. Right now, what we're finding is that if you control the environment very carefully for plant cells, you can control the consistency and quality of the product coming out. So we’ve been sticking with a very straightforward media and process, and we’ll talk more about that. The bioreactors themselves are pretty consistent, and that gives us stable production over time.
What we’ve demonstrated so far is stability over about six months at the lab scale. When we move to full scale, we’re hoping to translate that stability into a continuous manufacturing process.
David Brühlmann [00:14:30]:
I'm curious about the media cost, because that's a big challenge in the cultivated meat space. How does that work in cacao?
Steven Lang [00:14:38]:
Plants in the field just need fertilizer. And since we're growing them without sunlight, they also need a carbon source. Our media is relatively simple—about 18 components, compared to animal cell culture media, which typically has 40 to 60 components, including some very expensive growth factors.
Our components are mostly fertilizers—phosphorus, nitrogen—plus a carbon source and some relatively inexpensive growth factors that are already found in conventional food we eat today.
David Brühlmann [00:15:13]:
All of the components you're using are chemically defined, correct?
Steven Lang [00:15:16]:
Correct—chemically defined media. I should also add that all the components we’re using have gone through a GRAS (Generally Recognized As Safe) process and are totally acceptable for food production.
In addition to having fewer raw materials, the costs are dramatically different. Back when I was doing cell line development at Johnson & Johnson, we were paying around $40 per liter for media. I don’t know how much that’s changed, but in plant cell culture, we’re well below $10 per liter, and we expect to get below $1 per liter.
David Brühlmann [00:15:48]:
Wow. And what about the economics? Is it relatively easy to reach a point where the product is profitable?
Steven Lang [00:15:58]:
Yes, absolutely. And I should back up and describe our first product coming out the door, which is a high-flavanol cocoa powder. We’ve selected a cell line that produces higher levels of flavanols, which are bioactive compounds responsible for many of the health benefits of dark chocolate.
There was a very large clinical study called the COSMOS Trial, which included more than 21,000 participants, mostly elderly, and more than half had pre-existing conditions such as cardiovascular disease or diabetes. This long-term study supplemented diets with 500 mg of flavanols per day and found a 27% reduction in cardiovascular events. Additional analyses are still coming out of that study.
So we’re interested in improving not only flavor and taste, but also the health benefits of cocoa powder. Through our cell line selection, we believe we can do that. This allows us to enter the market with a premium product at a higher price point.
While we scale that product and generate revenue, we’ll continuously work on improving COGS and processes to reduce cost. That will allow us to move into commodity cocoa and eventually other products like coffee. So we start premium and then move toward commodity products.
David Brühlmann [00:17:38]:
That's an excellent strategy.
Steven Lang [00:17:40]:
Yeah. Other companies are doing this as well. I like to point out Gourmey, which has been very successful in France producing cultivated foie gras as a premium product. It really is the way to go—it’s more of the Tesla model, where you launch with a premium product first.
David Brühlmann [00:17:57]:
Yes, exactly. It’s a Tesla model.
Steven Lang [00:17:59]:
Exactly.
David Brühlmann [00:18:00]:
Before we go on to flavor, I just want to dive into the bioreactor side for a minute, because not everyone listening is familiar with plant cell culture. Can you describe how the bioreactor looks? What parameters are you controlling? How does a run typically work? Is it batch, fed-batch, or—you mentioned continuous—how does it work?
Steven Lang [00:18:23]:
With these cacao cell lines and our cocoa powder product, we’ve run the cells in both stirred-tank bioreactors and airlift bioreactors. They’re very robust and grow under a wide range of conditions, so we don’t have to control very much.
In one of your previous podcasts, you talked about Process Analytical Technology (PAT). For us, the main parameter we control is dissolved oxygen. These plant cells grow at ambient temperature, relatively low pH, and all they really need is oxygen and nutrients.
That’s not to say the processes are trivial, but they’re very flexible, and we can use different reactor modalities. Right now, we’re running batch processes, not fed-batch. A typical run starts with inoculation, and about seven days later, we harvest.
We use a higher inoculum than typical animal cell culture, but our cells are also much larger. We usually start with about 5–10% packed cell volume (PCV), and by harvest we reach 30–40% PCV. That yield is dramatically higher than what you typically see in even highly optimized animal cell culture processes.
So right out of the gate, the amount of biomass we can generate from plant cell culture is significantly higher than what animal cells can produce.
David Brühlmann [00:19:50]:
And how long does a run typically last?
Steven Lang [00:19:52]:
For seven days. And the seed train running up to that is your typical four- to five-day split ratio. Depending on the scale you're going up to, you'll need that time to get the inoculum ready for the final production vessel.
David Brühlmann [00:20:07]:
And what vessels are you using in the seed train? Are these shake flasks, spin tubes, the usual sub-specs, all that?
Steven Lang [00:20:14]:
Standard stuff—shakers and shake flasks, Fernbach flasks, the larger shakers—until we get into the bioreactors. At very small scale, starting at 5 and 10 liters, we keep them in a controlled environment, then scale from 5 to 10 liters to 100 liters, 500 liters, and then 2,000 liters.
David Brühlmann [00:20:34]:
How critical are these cultures with respect to contamination?
Steven Lang [00:20:39]:
The major problem with any scaled cell culture is contamination—how do you maintain a sterile envelope for the cells to really take off? The nice thing about plant cells, as I mentioned earlier, is that the culture media has a relatively low pH, which is not conducive to microbial growth. In addition, plant cells themselves produce antimicrobial compounds that help fend off contamination.
That said, we still deal with contamination, and that’s something we’re actively working through. As we talk about some of the bioreactor innovations we’re developing to create low-CAPEX, low-OPEX bioreactors, the sterilization strategy is a big part of that.
David Brühlmann [00:21:24]:
And you're using stainless steel tanks for the cell culture—it's not single-use, right?
Steven Lang [00:21:29]:
We’re not single-use, and we’re also not stainless steel, because we’re going after commodity products. One of the things that really attracted me to this company is that they’d clearly thought through the business model and understood they’re targeting commodities, so everything has to be interrogated for cost reduction.
Right now, we’re working with plastic bioreactors. The reason is simple: we can buy a 2,000-liter plastic tank for about USD 3,000. Try buying a 2,000-liter stainless steel bioreactor for less than USD 1 million. Right there, you can see we’re taking a very different approach—maybe even recreating the wheel in some cases.
But our cells are well suited to these low-cost bioreactors, which we can then scale out very easily. Another important part of the business model is that instead of scaling up beyond 2,000 liters, we scale out. That allows us to convert underutilized office space into laboratories where we can produce cocoa and coffee.
This is fantastic, because here in West Sacramento, where my labs are located, we’re using underutilized office space with very low cost per square foot, compared to what you’d pay for a GMP facility or a site designed for large-scale mammalian cell culture.
David Brühlmann [00:22:51]:
Yeah, that's interesting, because this model could also be very compelling for emerging markets. You could put this somewhere in Africa, Asia, or Southeast Asia. That works, and people there have the competencies to run it.
Steven Lang [00:23:10]:
Absolutely. And I’m really glad you brought that up. Not only can we co-localize these bioreactors because they’re so inexpensive, but we can also onshore production that historically hasn’t happened in Europe or the United States—particularly coffee and cocoa, which are restricted to regions around the equator.
In the U.S., the only places we can really grow coffee or cocoa are Puerto Rico and Hawaii, and that’s not enough. Our vision is to build not just the cell lines, but also the processes and infrastructure that allow us to deploy these production units anywhere in the world—co-localized with chocolatiers or manufacturers, right next to their facilities.
That also enables the creation of more biomanufacturing jobs. All of these are what I’d call positive externalities that come from focusing on low-CAPEX, low-OPEX biomanufacturing for products people genuinely want—and frankly, people are addicted to chocolate and coffee.
David Brühlmann [00:24:16]:
Yeah. And I think this is just the beginning, because you can produce all kinds of products in these cell culture systems.
Steven Lang [00:24:23]:
Yeah, absolutely. There are already plenty of precedents. Plant cell culture has been around for a long time. There are drugs being produced—at least four that I know of—from plant cell culture. There are also animal vaccines, as well as a wide range of nutraceuticals, cosmetic ingredients, and specialty compounds produced using plant cell culture.
David Brühlmann [00:24:44]:
That’s just the beginning of our conversation with Steven Lang. We've explored the foundations of cellular agriculture and the bioprocessing challenges of growing cacao in bioreactors. In part two, we'll dive into the economics, the scalability, and what this means for the future of food production.
If you’re finding value in this episode, please leave us a review on Apple Podcasts or your favorite platform. It helps other biotech scientists like you discover these conversations. We’ll see you in part two.
All right, smart scientists—that’s all for today on the Smart Biotech Scientist Podcast. Thank you for tuning in and joining us on your journey to bioprocess mastery. If you enjoyed this episode, please leave a review on Apple Podcasts or your favorite podcast platform. By doing so, we can empower more scientists like you.
For additional bioprocessing tips, visit us at smartbiotechscientist.com. Stay tuned for more inspiring biotech insights in our next episode. Until then, let’s continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Steven Lang
Steven Lang is the Head of R&D, Bioprocess, and Analytics at California Cultured, where he leads the development of sustainable cocoa and coffee produced directly from plant cells. He brings over 20 years of experience in cell culture and biomanufacturing from the biopharmaceutical industry, including roles at Genentech and Johnson & Johnson.
Steven is passionate about applying scientific rigor and scalable bioprocessing to build more resilient and sustainable food systems.
Connect with Steven Lang on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
For decades, biotechnology development relied on 2D cultures—cells grown as flat layers in petri dishes or flasks. While useful for many experiments, these models don't reflect the true complexity of living tissues.
In this episode of the Smart Biotech Scientist Podcast, host David Brühlmann speaks with Catarina Brito, Principal Investigator at ITQB NOVA and Head of the Advanced Cell Models Laboratory at iBET and ITQB NOVA in Portugal, about how 3D cell models are reshaping preclinical research and driving a fundamental shift in the field.
We have a breast cancer microenvironment model in which we include breast cancer cells alongside immune cell populations. We also incorporate stromal cells, namely fibroblasts, and we were studying antibody–drug conjugates (ADCs). These ADCs are very potent and can kill cancer cells quite quickly.
However, when we use a 3D model in which we reconstitute the tumor microenvironment, the response changes. The stromal cells—these fibroblasts—contribute to resistance to ADCs, and we can observe and study this effect directly in the 3D model. This is something you cannot—and would not—see in a flat, 2D culture flask.
David Brühlmann [00:00:46]:
Welcome back. In part one, Catarina Brito introduced us to the power of advanced 3D cell models that mirror the complexity of human tissues farbeyond traditional culture systems.
Now we’re going deeper into the immune microenvironment, the hidden factors that can make or break therapeutic success. From immunotherapy resistance in tumors to unexpected responses to gene therapy vectors, Catarina reveals how understanding innate immunity is reshaping preclinical strategy. Plus, she’ll share her vision for the future. Let’s continue.
Can you give us a concrete example of how modeling the tumor microenvironment in your system has revealed insights about specific interactions or mechanisms? How does that work?
Catarina Brito [00:02:53]:
There are different types of examples. One practical example comes from work we have published, where we developed a breast cancer microenvironment model that includes breast cancer cells, immune cell populations, and stromal cells, namely fibroblasts. In this study, we were evaluating antibody–drug conjugates (ADCs).
Typically, in 2D models, we culture HER2-positive breast cancer cells and test an anti-HER2 ADC. These ADCs are very potent and can kill cancer cells quite quickly. However, when we move to a 3D model, in which we reconstitute the tumor microenvironment, the response changes. The stromal cells—specifically fibroblasts—contribute to resistance to these ADCs. This effect can be observed and studied in the 3D model, and it is something you cannot—and would not—see in a flat 2D culture flask.
Another example comes from our work with neural models, which we use to study the early steps of immunogenicity to gene therapy vectors. In these models, we include neurons, glial cells, and microglia, which are the resident macrophages of the brain. Using this system, we were able to identify very transient but intense immune responses to viral vectors. These responses had not been detected in other preclinical models but were observed in patients. This illustrates how better reproducing human physiology can help identify factors that can later be used to improve therapies.
David Brühlmann [00:04:44]:
Where do you see these models evolving in the next few years? There’s a lot happening in the bioprocessing field—how we culture cells, how we maintain conditions, advances in analytical technologies, and now the integration of AI and machine learning. How will these technological advances affect the way you conduct your experiments?
Catarina Brito [00:05:16]:
I think advanced cell models combined with multi-omics data—multiple layers of omics information—and spatial information will be key. This is particularly relevant when we consider three-dimensionality and the multiple components of these environments.
This type of data will feed AI-based models, allowing a more accurate representation of human biology and pathophysiology. Until now, we have mainly been making observations and comparing them with human data. But to start making predictions, the integration with AI will be essential.
One starting point is thinking about digital twins of the models themselves, benchmarked against patient-derived data. From there, many things become possible: better tailoring treatments to biological patient profiles, optimizing dosing strategies, and designing drug combinations. Overall, this will be a major step forward for precision medicine and its meaningful implementation.
David Brühlmann [00:06:28]:
What is your advice to a smart scientist listening and wondering how they could implement these advanced models? We have many powerful tools available, but it can also feel overwhelming. Where do you start, and how simple can you keep these models while still gaining value from them?
Catarina Brito [00:06:53]:
I think the answer lies in focusing on the biological question you need to answer. I completely agree with what you suggested—we should aim to be as simple as possible to answer that question. The technology should not be the driving force; the question should lead us to define the biological endpoint, whether that is mechanism of action, potency, or safety.
From there, we decide on the level of complexity required. One risk today is overengineering systems too early, before fully understanding the cells and the question. That’s why I advocate for modular systems: starting simply with tumor or organ-specific spheroids, and then adding stromal or immune components only when the project demands it.
At the same time, we should avoid oversimplifying by treating these systems like regular 2D cultures. Factors such as mass transfer, cell source selection, and process control are critical. These are processes, and they require proper control to ensure reproducibility and standardization, which is often what feels daunting with these models.
David Brühlmann [00:08:17]:
At iBET, you collaborate extensively with pharmaceutical companies. I imagine you also guide them in how to use the models effectively. Can you describe the types of collaborations you have, how you run them, and how they shape model design and application?
Catarina Brito [00:08:42]:
That’s true—we collaborate extensively with industry. Over the years, we’ve been fortunate to work with several pharmaceutical partners who have helped shape how we operate. Pharma partners bring questions driven by translational needs: how to make models more predictive, more scalable, and more compatible with their development workflows.
These applied questions strongly influence model design and help us prioritize our research directions. They help ground the models in real-world use. A key factor for successful collaboration is alignment and transparency—being very clear about the biological question, the performance metrics, and the limitations of the system.
Open data sharing in both directions has been essential. This approach has helped move several of our platforms from proof-of-concept to tools that genuinely contribute to accelerating translation within companies.
David Brühlmann [00:09:49]:
What kind of companies do you generally work with?
Catarina Brito [00:09:53]:
I’ve worked mainly with pharmaceutical companies. Among our collaborations, we’ve worked with Boehringer Ingelheim, Merck Global Health—which is the social business of Merck KGaA, Germany—and AbbVie in the United States. These are large companies developing advanced and novel therapies.
David Brühlmann [00:10:14]:
Do you also work with smaller companies or startups?
Catarina Brito [00:10:18]:
Yes, we do. The collaborations I mentioned were long-term partnerships, often spanning several years, where we developed models from scratch to answer questions that were not addressable with existing systems. We’ve also worked with smaller companies and startups, typically on more focused questions related to their compounds, using models we already have established.
David Brühlmann [00:10:44]:
I have a question about potential pitfalls. As we mentioned earlier, there are many technologies available—so where do you start? What do you leverage? What is hype? What really makes a difference? Can you point out two or three pitfalls to watch out for?
Catarina Brito [00:11:03]:
As I mentioned earlier, one major pitfall is overengineering systems in a technology-driven way before truly understanding the cellular system, what you need the cells to do, and which cell types are actually required.
Another critical aspect is the cell source. Depending on your question, you may never get the answer you’re looking for from the cells you’re using—no matter how much you engineer them.
A third pitfall is insufficient validation and characterization. Moving too quickly to application without properly characterizing the system can be problematic. Thorough validation is essential before drawing conclusions or translating results.
David Brühlmann [00:11:45]:
Catarina, what burning question haven’t I asked that you’re eager to share with our biotech community?
Catarina Brito [00:11:52]:
I think we covered things quite thoroughly. It was a really good conversation—and you can probably tell how passionate I am about this topic.
David Brühlmann [00:12:05]:
Yes, absolutely.
Catarina Brito [00:12:07]:
And this also gives context to my work. My academic appointment is with NOVA University Lisbon, but my lab is also affiliated with iBET. It is through iBET that I collaborate with industry, because iBET is a partnering research organization that bridges engineering and biology to accelerate biopharmaceutical development.
This creates an ecosystem where we combine expertise ranging from cell line development, cell bioprocessing, and advanced analytics, and then close the loop with translational models under the same organizational framework. This setup gives us access to a highly collaborative environment and makes it easier to work with advanced therapeutics, from antibodies to gene and cell therapies.
David Brühlmann [00:12:59]:
This has been great. Catarina, what is the most important takeaway you want our listeners to walk away with from our conversation?
Catarina Brito [00:13:08]:
I think the key takeaway is that advanced cell models are here—and they are here to stay. They will play a major role in shaping translational research in the future, while at the same time driving transformation across the entire industry.
David Brühlmann [00:13:24]:
This was great. Catarina, thank you so much for letting us into the world of 3D cell models. Where can people get in touch with you?
Catarina Brito [00:13:33]:
I’m on LinkedIn, like most people, and also reachable through the iBET website and ITQB NOVA’s website. But I would say LinkedIn is the best way to contact me.
David Brühlmann [00:13:42]:
Excellent, smart biotech scientists—you’ll find all the links in the show notes. Please take this opportunity to reach out to Catarina. And Catarina, once again, thank you so much for being on the show today.
Catarina Brito [00:13:56]:
Thank you, David. It’s been a pleasure. Bye-bye.
David Brühlmann [00:13:59]:
That wraps up our conversation with Catarina Brito. Her work exemplifies the future of preclinical research, where complexity becomes predictability and where patient-specific insights move from aspiration to reality.
If this episode sparked new ideas for your own development programs, I’d love to hear about it. And please take a moment to leave a review on Apple Podcasts or wherever you listen—your support helps us bring more innovators like Catarina to the biotech community.
Thank you so much for tuning in today. Until next time.
For additional bioprocessing insights, visit us at www.bruehlmann-consulting.com. Stay tuned for more inspiring biotech conversations in our next episode. Until then—let’s continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Catarina Brito
Catarina Brito is Principal Investigator of the Advanced Cell Models Laboratory within the Animal Cell Technology Unit at iBET and ITQB-NOVA. Her research centers on the development of human cell models using induced pluripotent stem cells, patient-derived cells, and established cell lines, and on applying these models to investigate disease-associated alterations of the cellular microenvironment and their impact on therapeutic response.
Her key research interests include the innate immune microenvironment in cancer and neurological disorders, as well as gene and cell therapies and immunotherapies. Her laboratory is supported by funding from FCT (Portugal), the Innovative Medicines Initiative Joint Undertaking (IMI; EU & EFPIA), and the international pharmaceutical industry.
Connect with Catarina Brito on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
Ever wondered what actually happens to human cells when you scale up bioprocessing from a petri dish to a bioreactor? Most scientists see it as a matter of bigger equipment and higher volume. But that shift isn’t just technical—it's biological. The rules that govern cell behavior change, sometimes dramatically, reshaping everything we thought we knew about cell therapy development.
Smart Biotech ScientistPodcast host David Brühlmann is joined by Catarina Brito, Principal Investigator at ITQB NOVA and Head of the Advanced Cell Models Laboratory at iBET and ITQB NOVA , Portugal, to talk about how 3D cell models are driving a paradigm shift in preclinical research.
I really believe bioprocess development starts with understanding cell biology at scale. Cells are social entities, and they sense each other. They remodel their microenvironment, they rewire their signaling cascades when things such as cell density, mass transfer, and mechanical cues change. When we are moving from the T-flask to the bioreactor, I think we are not just increasing volume; we are changing the entire context in which the biology of the cells operates. That’s why I think we really need to respect biology as a primary design input.
David Brühlmann [00:00:39]:
Imagine a world where we could predict how your body would respond to a therapy before it ever enters clinical trials. Today’s guest, Catarina Brito from ITQB NOVA and iBET in Portugal, is making that vision tangible. She’s pioneering 3D human cell models that recreate the complex microenvironment of human tissues—from brain and liver to tumors—revolutionizing how we test biologics and gene therapies.
If you’re tired of the limitations of Petri dishes and animal models, this conversation will reshape how you think about preclinical research.
Catarina, welcome to the Smart Biotech Scientist. It’s great to have you on.
Catarina Brito [00:02:35]:
Hi David. Thanks for having me here. I’m glad to be here.
David Brühlmann [00:02:39]:
It’s a pleasure. Catarina, share something you believe about bioprocess development that most people disagree with.
Catarina Brito [00:02:47]:
I think people are starting to believe it in the bioprocess industry—maybe not as strongly as I do, or not from the beginning. It’s a strong conviction of mine because I truly believe bioprocess development starts with understanding cell biology at scale.
Cells are social entities, and they sense each other. They remodel their microenvironment, they rewire their signaling cascades when things such as cell density, mass transfer, and mechanical cues change. When we move from the T-flask to the bioreactor, we are not just increasing volume—we are changing the entire context in which the biology of the cells occurs.
That’s why I believe we really need to respect biology as a primary (PR) input—media composition, oxygenation, and nutrient delivery—to keep biology within the right operating window so that we can successfully scale up. This means thinking about more than just equipment and volume, but really starting with biology as the foundation.
David Brühlmann [00:03:49]:
Scale-up is much more than “scaling up.” And you said it very well—there are so many different parameters you have to consider when scaling a process.
Before we dive deeper into today’s topic, let’s talk about you. Catarina, draw us into your story. What sparked your interest in biotechnology, and how did you arrive at the exciting field you’re working in today?
Catarina Brito [00:04:19]:
It all started during my PhD and was driven by the questions I was trying to answer. I was studying mechanisms driven by glycan–protein interactions, and these mechanisms are very different between murine and human cells. That was an early wake-up call—it made me really think about human biology and the accuracy of the models we were using, as well as the need for models that truly reflect human physiology.
The cellular processes I was studying involved neural cells and axonal outgrowth—processes that are highly dependent on context, particularly the extracellular matrix. Yet we were growing murine neurons on plastic surfaces, which are extremely artificial compared to what actually happens in the brain.
This reinforced the need for better models of the physical context and how it shapes signaling, morphology, and neuronal connectivity. And then there’s the cellular context—neurons are not isolated; many critical cues come from neighboring cells, particularly glial cells. The models we were using were monocultures, which was quite frustrating.
All of this led me to look for a postdoc where I could tackle these questions. My motivation truly crystallized when I had the opportunity to join iBET for my postdoc. I worked under the supervision of Dr. Paula Alves, who was already doing pioneering work on 3D culture systems and demonstrating that biology can—and should—be a primary design input for experimental models.
It was also a very exciting time: it was the first time human pluripotent stem cells were being used in Portugal. That experience was incredibly important and really set the direction for my career—building models that integrate human, multicellular complexity.
David Brühlmann [00:06:24]:
That’s fascinating. Let’s unpack this a bit, because not everyone listening today is familiar with animal models, 2D culture, or 3D culture systems—and there’s a lot happening in this space.
Let’s start with the traditional tools: 2D cell culture and animal models, which have dominated preclinical research for decades. What are the critical limitations of these systems, and what advantages do newer models offer?
Catarina Brito [00:06:53]:
Both. I should start by saying that, because we are often on the defensive. I’m always advocating for advanced models and 3D models, but all models have value, right?
2D cultures and animal models have taught us a lot about biological questions and pathological aspects as well. But of course, they also have limitations. And I think that understanding the limitations of each model is extremely important, especially when we’re trying to develop advanced therapies.
2D cultures are cells grown on a flat surface. Typically, they involve immortalized cell lines that are easy to culture. They offer a lot of control and high throughput, but they lack the structural and functional reality of tissues. Tissues are not flat. Some tissues are layered, but most tissues are three-dimensional.
Cell polarity is altered; diffusion of nutrients—and also of compounds with therapeutic potential—is different. There is no physical confinement of cells and no proper neighborhood effects. As a result, cell–cell interactions change significantly.
If we think about cells under these conditions, their receptor localization, metabolism, and overall phenotype change when moving from 2D to 3D, or from 2D to native tissue. So if we’re thinking about the development of biologics, for example, these molecules depend on receptor engagement, transport, and access to the microenvironment. There is therefore a major difference in both biology and therapeutic response.
When we think about animal models, they are still invaluable because they provide systemic biology—they represent a complete organism. However, they miss many human-specific aspects, particularly in the immune system, glycosylation patterns (as I mentioned before), genetic variability, and even disease etiology.
This can result in false positives and false negatives due to interspecies differences that can distort the readout. Of course, animal models remain useful and are part of a toolbox that should be as comprehensive as possible—but they do have limitations.
David Brühlmann [00:09:09]:
There’s been quite a push from regulatory agencies—especially the FDA—to reduce the use of animal models lately. What’s your take on this? Do you think we’ll ever reach a point where animal models are no longer used at all, or is that unrealistic? Will we always need a hybrid approach?
Catarina Brito [00:09:30]:
I think we are on a path that may eventually lead to the replacement of animal models, although a lot of validation is still required. There is a strong effort to develop multi-organ systems in which pharmacodynamics and pharmacokinetics can be studied. So far, what regulators have mainly pushed for is the replacement of animal models in types of readouts where we already know that the systemic component is not required. But I would say we are clearly on a path that could take us there—we just need strong validation.
There are also efforts from the European Commission, including dedicated calls and roadmaps, to move us in this direction. With support from regulators and agencies, we may eventually get there.
David Brühlmann [00:10:17]:
This will definitely be a paradigm shift for the industry—a very different way of developing drugs.
And speaking of different approaches, let’s zoom in on 3D models and advanced models. Tell us more about that. What exactly is a 3D model, and why is it called “advanced”?
Catarina Brito [00:10:35]:
The notion of “advanced” goes beyond simply growing cells in three dimensions. These models need to recreate key aspects of the tissue they are meant to represent. The tissue microenvironment shapes how cells behave, so the goal is to have a bioactive model in which cells influence the system and are influenced by it as well—ideally in a way that achieves reproducibility and robustness.
There are several important aspects. First, architecture—the three-dimensional structure is essential for cell polarity, spatial organization, and physical confinement. Then there is the presence of diffusion gradients. These are not only gradients of oxygen and nutrients, but also of signaling molecules, which are distributed in tissues and strongly influence biology. They also affect drug penetration, drug binding, and clearance.
Another important component is mechanical cues and the extracellular matrix. Cells sense stiffness and tension, which are crucial for survival, migration, and—for example—immune evasion, which is particularly relevant in cancer therapeutics.
Then there is multicellular complexity. Tissues are composed of different cell types that interact with each other. If we think about tumors, they interact with stromal cells and immune cells. In neural tissue, neurons interact with glia and microglia. These interactions are essential.
Advanced models aim to recapitulate biology more closely to what happens in the body, allowing us not only to study disease mechanisms but also to better predict therapeutic responses.
David Brühlmann [00:12:26]:
In those 3D systems, do people need to visualize this in a particular way? For example, if we take a liver cell type—do we reconstruct a mini liver organ in a bioreactor? Or is it more like having a large number of individual cells, maybe floating?
Catarina Brito [00:12:47]:
We try to recapitulate tissues by capturing this multicellular complexity—having the different cell types that compose the tissue interacting with each other.
It also depends on the tissue. In the liver, for example, hepatocytes are tightly connected through junctions. You also have endothelial cells with fenestrations—essentially “windows” that allow molecular exchange. Then there are macrophages, immune cells that move within the tissue space.
So the goal is not necessarily to reproduce the exact anatomical architecture, but rather to recreate the relevant cell–cell interactions and how they contribute to tissue function.
David Brühlmann [00:13:34]:
And how long can you keep these cells in culture? If you want to start a new experiment, do you have to start from scratch every time? Can you keep something like a cell bank—maybe that’s not the right term—but how does that work?
Catarina Brito [00:13:49]:
It depends a lot on the cell of origin. With primary cells, they are typically used for a limited time. They may last several weeks in the model, but they cannot be propagated extensively.
If we’re talking about pluripotent stem cells, then we can differentiate progenitors and at least bank those progenitor populations. This allows us to work in a multi-step way that facilitates reproducibility.
The duration of the model itself also depends on the culture system. We use a lot of bioreactor technology under perfusion, precisely to prolong the lifespan of these models.
David Brühlmann [00:14:37]:
And how do you handle diversity? You’re working with liver cells, neural cells, stem cells—many different cell types with different requirements in a bioreactor setting. How do you manage that? Do you change conditions, media, bioreactor size or shape?
Catarina Brito [00:15:16]:
It really depends on the tissue. Each tissue has its own requirements.
For example, neural cells are extremely sensitive to oxygen tension and mechanical stress. We use bioreactors, but the design has to minimize shear stress, and oxygen levels must be kept low and very stable. Tight control of process parameters is critical.
In the liver, oxygen requirements are completely different. There, maintaining functionality is key, and perfusion flow is essential to support metabolic competence. It’s also crucial to optimize the ratios of different cell types—not only hepatocytes, but also the so-called non-parenchymal cells. These ratios must be tightly controlled.
In solid tumors, heterogeneity is central. We need systems that allow extracellular matrix remodeling, which is a key feature of tumors. This enables the formation of tumor niches, including hypoxic niches, which are highly relevant. On top of that, we aim to incorporate immune cells and study immune cell infiltration.
All of these requirements are driven by biology. For each model, we carefully choose design variables to meet those needs. Across all systems, robustness and scalability are essential. We aim to design models that are as modular and bioreactor-compatible as possible, ensuring reproducibility and throughput.
Finally, we place strong emphasis on validation, to confirm physiological relevance and ensure that it is not lost during scaling throughput.
David Brühlmann [00:17:22]:
I’m just curious, and I may have missed this earlier, but the purpose of all this, as you said, is better reproducibility and, ultimately, faster development. So how do these advanced systems compare, for instance, to animal models? Why can development be accelerated so significantly?
Catarina Brito [00:17:42]:
The throughput is completely different. When you develop models in scalable systems, you achieve much higher throughput, and these models can be applied much earlier in the drug development process.
They can be introduced earlier in pharmaceutical development than animal models, which are usually brought in quite late. This allows for much more selection and decision-making earlier on.
Additionally, these models are increasingly being adopted because much of the relevant biology—particularly human biology—is not captured in animal models. Even in cancer research, many of the newer therapeutic modalities, such as cell engagers and multispecific antibodies, rely on biological mechanisms that are not reproduced in animal systems.
So even when animal models are available, there is still a strong need to bring human-based models to the forefront.
David Brühlmann [00:18:28]:
We’ve just scratched the surface of how advanced 3D models are transforming drug development. In part two, Catarina will dive into the critical role of innate immunity in predicting therapeutic responses and share her vision for AI-powered personalized medicine platforms.
If you’re finding value in these conversations, please leave a review on Apple Podcasts or your favorite podcast platform—it helps other biotech scientists discover these insights. See you in part two.
All right, smart scientists, that’s all for today on the Smart Biotech Scientist Podcast. Thank you for tuning in and joining us on your journey to bioprocess mastery. If you enjoyed this episode, please leave a review on Apple Podcasts or your preferred podcast platform. By doing so, we can empower more scientists like you.
For additional bioprocessing insights, visit us at www.bruehlmann-consulting.com. Stay tuned for more inspiring biotech conversations in our next episode. Until then, let’s continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Catarina Brito
About Catarina BritoCatarina Brito is a Principal Investigator at ITQB NOVA and Head of the Advanced Cell Models Laboratory at iBET and ITQB NOVA in Portugal. Her research focuses on the development of complex human cell models to investigate disease microenvironments and therapeutic responses, particularly in cancer immunology and neuroinflammation.
By integrating fundamental cell biology with translational research, her work aims to accelerate the development of advanced therapies while reducing reliance on animal models. She has coordinated more than 19 research projects, authored 90 peer-reviewed publications, and works closely with pharmaceutical partners and clinicians to advance innovation in preclinical research.Connect with Catarina Brito on [LinkedIn](https://www.linkedin.com/in/catarina-brito-ibet/).
Connect with Catarina Brito on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
For years, protein engineering has struggled with a fundamental bottleneck: machine learning models, powerful as they are, have been starved of meaningful data. Most rely on just 10,000 antibody-antigen structures from the Protein Data Bank—a drop in the ocean compared to the billions of parameters powering modern large-scale AI models. The result? Promising molecules often get stuck in preclinical limbo, not because the models aren’t smart, but because they keep hallucinating with incomplete data.
In this episode from the Smart Biotech Scientist Podcast, David Brühlmann meets Troy Lionberger, Chief Business Officer at A-Alpha Bio, whose team has quietly shattered the data ceiling by measuring and curating more than 1.8 billion protein interactions..
Data is always going to be, I think, the constraint in industry—whether it's training your models or validating them. I think there's an increased appreciation that we need differentiated sources of data. A-Alpha Bio is hopefully going to play some part in that ecosystem, but we need more than just a handful of companies.
And I think the second thing I hope your audience walks away with is that, with the right data, AI can be an incredibly powerful tool in accelerating your work. That said, everyone would be forgiven for being confused by the claims of many, and so I would just encourage diligence as we watch this space unfold.
David Brühlmann [00:00:40]:
Welcome back to Part Two with Troy Lionberger from A-Alpha Bio. In the first half, we explored why protein–protein interactions are so fundamental yet historically difficult to characterize, the limitations of traditional antibody discovery methods, and where de novo design fits into the picture. Now we dive into the solution: how A-Alpha Bio’s platform generates massive datasets at unprecedented scale, what happens when you feed 1.7 billion measurements into machine learning models, and why this data revolution is transforming therapeutic development. Let’s jump back in.
Now I would like to move on to machine learning and data in general, because AI and machine learning are powerful, but they depend on good data. I always hear my former boss—he was quite skeptical about technologies—telling us, garbage in, garbage out. And that’s very true for modern technology. For AI, you need good data, and currently there’s a limitation because engineering models today are constrained by public-domain data. So can you paint us a picture of what that looks like and what the impacts are today on drug discovery?
Troy Lionberger [00:03:11]:
I think when we look at modern protein engineering models, I mean, many of them have been trained, as you said, on public data. In that case, it's typically the PDB, so the Protein Data Bank that is being referenced in many of these cases. And just to put a perspective on that, there are a little more than 10,000 antibody structures of antibodies bound to their target antigen in the PDB today. So when you compare 10,000—the number 10,000—to how much data has been generated to train ChatGPT, for example, or other large language models, I mean, there's a 10-order-of-magnitude difference between the amount of data that has gone into generalizable machine learning models in large language models compared to what data is available for brilliant protein engineering teams today to develop their next model.
The constraint is really felt when the biggest failure mode in many of these models is what they call hallucinating. So if you have one example of a solid structure that you're training your model on, it's actually quite difficult to then ask the model, if I make a mutation in one of these residues, will my molecule still bind? And if so, how tightly? Oftentimes these models are hallucinating because they've seen the structure before, there's a minor modification to it, so the tendency—the bias, if you will—is to revert to the solution that they've seen before.
So what's missing is an ability to generate data that can give these models many, many, many examples of both positive and negative data. So being able to definitively tell this model, no, that aspartic acid was changed to an alanine, I know from my experiments it did not bind anymore. Now correct yourself so that you stop hallucinating, right? I mean, that's basically what we're trying to do with AlphaSeq—is generate large amounts of data that introduce minor variations to these molecules and directly tells these models not only do they still bind, but what is the strength of that binding interaction.
And one thing for your audience to, I think, appreciate is when we talk about making generalizable models, I think the grand ambition in the industry is that we can one day teach a machine learning model physics. If you want to teach a machine learning model physics, the missing ingredient is not just the structures, but physical parameters like Gibbs free energy (ΔG). And it just so happens that binding affinity is directly related to ΔG. So what the AlphaSeq platform is giving us is not just structural information, but also energetic information that is really going to be, I think, transformative in how these models ultimately become predictive. And the predictive nature is going to increase the better they get at understanding the underlying physics.
David Brühlmann [00:06:07]:
Wow, that's powerful. That's powerful. Even physics, yeah. What can you do today with the technology, and what will be possible as you build your models? If I'm not mistaken, you have measured 1.7 billion protein interactions, or have—I’m not sure if this is by your company or in general. Where is this going to lead?
Troy Lionberger [00:06:30]:
I think there are multiple layers to that right now. I think the latest count is 1.8 billion, but what's 100 million between friends, I guess. So that data has been used—and we generated that data—in large part to train our internal models and also to do partnerships with therapeutic developers as well.
To answer the first part of your question, what are we doing today? Today we're offering access to our platform under fee-for-service arrangements. Someone comes to us with an antibody that they want to diversify or increase the affinity of or make more developable. We have a very standard process now that three months later these customers are getting what they want, and it's very easy for us to do this work, so we're happy to do it.
Where I think things get more challenging are applications where you require multiple iterations of the AlphaSeq platform. So what I just described as a kind of fee-for-service—someone comes with a molecule, we’ll improve it for them—that requires one iteration, if you will, of our experimental platform.
For more challenging examples, I'll give you two. One would be work we did with Gilead, where we were able to essentially optimize biologics specific for HIV. So to make a neutralizing biologic for HIV, the challenge that was put in front of us is: can you make a molecule that can hit not just one or two versions of HIV variants that are present in the population, but over 800 variants? If you had asked me before I joined this company three months ago if it would have ever been possible for someone to optimize a molecule against more than 800 different variants of a target, I would tell you there's just no way. You might get one or two, but not 800. And the fact that they were able to simultaneously improve the efficacy of this molecule for virtually every variant of HIV available is really exciting.
And I think that, for clarity, took three turns of the crank. So that's how powerful I think things get when you iterate a few times on this.
Another example that I'll give you that I didn't think possible before three months ago is reliably and reproducibly making biparatopic molecules. So what I mean by that is in most cases, the parts of antibodies called the variable domains that ultimately interact with your target antigens typically will bind to a single antigen—so one variable domain, one target antigen.
What are called dual-specific antibodies are those that actually have the ability to have that same domain bind to two different antigens. I've seen a few examples of this in the past where it was serendipitously discovered. And I think a lot of therapeutics historically has been kind of serendipity—it's bespoke and artisanal. We're getting to the point where now these things are tractable and systematic.
And this is one example where we can now almost arbitrarily introduce two different paratopes into this antibody and have them interact with two different antigens. And this opens up all kinds of unique modes of action and safety profiles for therapeutics that really haven't been possible before. But this is where this ability of AlphaSeq and AlphaBind to reproducibly engineer these molecules is, I think, quite exciting.
David Brühlmann [00:09:54]:
Yeah, definitely—that's quite exciting. And it's powerful to have all these tools at our disposal. And now this leads me to the next question. Because if you're developing your company, especially if it's a smaller company or a mid-sized company, you always face the question: Well, we do have very knowledgeable and successful CROs. We have service providers. We have all kinds of technologies. So what do we do in-house? What is the capability we're building in-house? What is the knowledge or the understanding we want to generate in-house versus sourcing out? So as you are working with different companies, what is your observation or your recommendation?
Troy Lionberger [00:10:33]:
David, I think you're right that many companies want to keep their core intellectual property and contributions to these molecules in-house. I absolutely understand the tendency to want to do that.
I think my response is: there's a reason that everyone decided to move to Microsoft Word instead of typewriters. I mean, at some point we realized that there are more efficient ways to move forward. And I think it just comes down to dollars and cents.
The reality is that these therapeutic developers, they don't want to be spending their time engineering molecules—if we're honest with ourselves. I mean, what they want to be spending their time doing is de-risking molecules, not making them. And so the quicker that they can get to de-risked molecules, the better for everybody.
And that is really where I would encourage those in the field to ask the question: Could you accomplish more, faster, and cheaper than service providers? If the answer is no, then everyone should use a service provider. That is really how we're operating now in many of our applications.
We're simply trying to accelerate these teams by making their work easier, faster, and cheaper—and more importantly, de-risking that entire process, because we're quite confident that we will be able to ultimately deliver a molecule. There isn't as much of a question anymore in terms of risk. I mean, we haven't yet had a project that we haven't been successful with. Another aspect of this is just mitigating the overall risk.
David Brühlmann [00:12:01]:
Troy, before we wrap up, what burning question haven't I asked that you are eager to share with our biotech community?
Troy Lionberger [00:12:09]:
If I'm honest, there’s something that we’re really excited about—it’s coming, and we’ve been talking about it publicly, so I’m happy to share it here.
But back to this question about being ultimately limited to the entirety of the PDB and only having about 10,000 structures to train your models on—what can we do about that? Crystallography is not going to solve that problem. We have around 10,000 structures, and only a few hundred are added each year. That’s not going to give us the scalability that we ideally need for these models.
And if I were to predict where the field is going—at least on the machine learning and protein engineering side—I think there’s going to be an appreciation across the industry that synthetic epitopes will become an indispensable tool for training these models.
This took me a bit to get my head around, so I’ll try to distill it. The basic idea is that if you have one antibody, it will bind to one target—but that’s not necessarily always true by physics. You could design other antigens—synthetic antigens—specifically to bind that same antibody.
So we’ve been working on something called the Synthetic Epitope Atlas. This is a program that generates predictions about synthetic partners for antibodies, for the purpose of validating that we can actually see a structure and measure the ΔG for that interaction. And even though these data are purely synthetic—these binders do not exist in nature—we’re showing that you can train a model almost just as well as with the PDB, but using things that don’t exist in nature.
You’re teaching these models structure, epitopic diversity, and feeding them ΔG and affinity measurements. That, for me, is really exciting to be a part of at these early stages. And increasingly, we’re starting to hear across the industry that more and more groups are pursuing synthetic epitope development specifically for machine learning and model training.
Just to put this into perspective: I mentioned that there are about 10,000 antibody structures available today to train models on. In just one of our measurements, we’ve shown that you can generate about 1,000 viable structures that are all validated, along with 29,000 confirmed non-binders—molecules with simple mutations that break affinity. That’s roughly 30,000 distinct measurements from a single experiment that can be used to train models. That’s pretty exciting when you think about how an alternative to the Protein Data Bank could ultimately grow.
David Brühlmann [00:14:51]:
Yeah, that’s exciting. With everything we’ve covered today, what is the most important takeaway?
Troy Lionberger [00:14:57]:
The most important takeaway I hope your audience walks away with is, first of all, that data is always going to be the constraint in this industry—whether it’s training models or validating them. There’s an increased appreciation that we need differentiated sources of data.
A-Alpha Bio is hopefully going to play some part in that ecosystem, but we need more than just a handful of companies. And the second thing I hope your audience takes away is that, with the right data, AI can be an incredibly powerful tool for accelerating your work.
That said, everyone would be forgiven for being confused by the claims being made today, and so I would just encourage diligence as we watch this space unfold.
David Brühlmann [00:15:41]:
Troy, this was fantastic. Thank you so much for sharing your insights and your perspective on where drug discovery is heading. Definitely exciting times ahead. Where can people get in touch with you and learn more about your technology?
Troy Lionberger [00:15:56]:
Yeah, our website is www.aalphabio.com—there’s a lot of information there. You’re also welcome to connect with me on LinkedIn. I’d be happy to connect with your audience.
David Brühlmann [00:16:10]:
Excellent. I’ll put the links in the show notes. Smart biotech scientists, please reach out to Troy. And Troy, once again, thank you so much for being on the show today.
Troy Lionberger [00:16:20]:
Thank you, David. It was great speaking with you.
David Brühlmann [00:16:22]:
What a conversation. From measuring billions of protein interactions to predicting the next generation of therapeutics, Troy Lionberger has given us a glimpse into the future of drug discovery. The integration of high-throughput biology and machine learning isn’t just incremental progress—it’s transformational.
If this episode sparked new ideas for your own work, share it with a colleague and leave a review on Apple Podcasts or wherever you listen. Until next time, keep innovating. Thank you so much for tuning in. Until next time, keep innovating.
For additional bioprocessing tips, visit us at www.bruehlmann-consulting.com. Stay tuned for more inspiring biotech insights in our next episode. Until then, let's continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Troy Lionberger
Troy oversees Business Development and Alliance Management at A-Alpha Bio. He started his career in the lab after earning a PhD from the University of Michigan and completing postdoctoral training at UC Berkeley. During nearly a decade at Berkeley Lights, Troy held senior leadership roles in R&D, product strategy, and business development, driving commercial initiatives that secured over $200M in deals across biopharma, agriculture, and industrial biotech.
Before joining A-Alpha Bio, he was Chief Business Officer at Abbratech, guiding the company from stealth-mode antibody discovery to a partner-focused biotech. Troy combines deep technical expertise with a track record of scaling platform companies through strategic, transformative partnerships.
Connect with Troy Lionberger on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
Antibody therapeutics have transformed modern medicine, but for many scientists, developing new candidates still feels like searching for a needle in a haystack—a slow, expensive, and unpredictable process. Structural biology and high-throughput data generation are now collapsing that haystack, offering unprecedented visibility into the molecular handshake that drives life: protein-protein interactions.
In this episode from the Smart Biotech Scientist Podcast, David Brühlmann meets Troy Lionberger, Chief Business Officer at A-Alpha Bio, a biotechnology company harnessing synthetic biology and machine learning to measure, predict, and engineer protein-protein interactions.
As a biologist, I would tell you that ultimately life doesn't exist if proteins aren't on some level interacting with other proteins. So whether it's catalyzing force in your muscles or replicating DNA, proteins have to interact with other proteins to carry out all of the cell functions that are necessary for life. If there's ever cell dysfunction, it's oftentimes in some way, shape, or form tied back to some sort of protein–protein interaction that's both the origin of many disease states, but also the opportunity for therapeutic intervention.
David Brühlmann [00:00:35]:
Protein–protein interactions govern almost every biological process and hold the key to treating cancer, infectious diseases, and neurological disorders. Yet, with only 10,000 antibody–antigen structures in public databases, we're building tomorrow’s medicines on yesterday's data. Today, Troy Lionberger, Chief Business Officer at A-Alpha Bio, reveals how measuring millions of interactions simultaneously changes everything. By generating unprecedented quantities of high-quality data, they are accelerating the discovery of rare antibodies, engineering better protein therapeutics, and training AI models that predict what works before you ever touch the lab bench.
Let's explore how. Welcome, Troy, to the Smart Biotech Scientist. It’s good to have you on today.
Troy Lionberger [00:02:42]:
Thanks, David. It's a pleasure to be here.
David Brühlmann [00:02:43]:
Troy, share something that you believe about biotherapeutic development that most people disagree with.
Troy Lionberger [00:02:51]:
It's an interesting question. I think the most controversial view I harbor right now is—given my background—there is an overwhelming historical acceptance that antibody therapeutic development is artisanal and bespoke, that you're really hunting for needles in a haystack, if you will, as is often the common analogy.
I think the controversial statement I would make is that it’s far more systematic today than I ever imagined. For example, most people are surprised when I tell them that there are tractable and reproducible ways to make antibodies that have the same affinity for their human therapeutic target as animal targets. I mean, not just cross-reactive, but the same quantitative affinity, which could help streamline preclinical development, for example, of therapeutic antibodies. Most people I’ve spoken with seem to think that is a flight of fancy. Fundamentally, there are processes that make this happen today. That is the most surprising thing I share with people on a day-to-day basis.
David Brühlmann [00:03:52]:
And this will open avenues to novel therapeutics and also more efficacious drugs.
Troy Lionberger [00:03:59]:
That's right. I mean, preclinical development of antibodies is fundamentally constrained by the challenges in developing these therapeutic molecules. In large part, getting them to work with the animal models required in your studies is problematic. Oftentimes, the affinities of your molecules for those animal targets are far worse than for your human targets. So while you may have a drug that works quite well in humans, you can’t get it to the clinic because the animal might have toxicity issues, simply because you had to administer so much drug into it.
David Brühlmann [00:04:33]:
Before we dive deeper into today's topic, take us back to the beginning. What first sparked your interest in biotech, and how did that journey lead you to A-Alpha Bio?
Troy Lionberger [00:04:43]:
The origin for me was really in college, when a faculty member teaching structural biology started describing proteins as nanomachines. That visual has always stayed with me. It got me interested in science and wanting to understand how these fascinating machines, which operate with very different materials and properties than anything we’ve created as human beings, function and work.
That naturally led to understanding that these proteins, which we barely understand, are ultimately at the root of all human disease—leading to cell dysfunction, which we describe as disease states. It’s ultimately the understanding of these basic building blocks of life that drives biotechnology: figuring out how these proteins can be manipulated and controlled to elicit therapeutic effects.
To answer your question about how I ended up at A-Alpha Bio, my career in biotechnology started at a life science tools company called Berkeley Lights, where I helped invent an exciting technology to discover therapeutic antibodies. That experience naturally led to working with many teams in the industry to support their discovery efforts, and I became increasingly aware of the next major constraint: the preclinical development of those drugs. That is, in large part, the problem we are trying to solve at A-Alpha Bio right now.
David Brühlmann [00:06:05]:
When I looked at your website, what struck me is that your company, A-Alpha Bio, describes itself as a protein–protein interaction company. Why are these interactions so fundamental to drug discovery, and why are they also so difficult to characterize at scale?
Troy Lionberger [00:06:22]:
It's a great question. My background is in biology. As a biologist, I would tell you that ultimately life doesn't exist if proteins aren't on some level interacting with other proteins. So whether it's catalyzing force in your muscles or replicating DNA, proteins have to interact with other proteins to carry out all of the cell functions necessary for life. If there's ever cell dysfunction, it's oftentimes in some way, shape, or form tied back to some sort of protein–protein interaction.
That's both the origin of many disease states, but also the opportunity for therapeutic development.
Being able to characterize these protein–protein interactions—there are many technologies that have come forth to help do this. Surface plasmon resonance (SPR) is an industry-standard way to study protein–protein interactions, and we call this affinity. Understanding the strength of those interactions, or the affinity of those interactions, is ultimately how biophysical characterization describes these protein–protein interactions.
The problem historically is that, despite how advanced these technologies are, they are also quite costly and difficult to use. And when I say difficult, I don’t mean it’s impossible—people do this every day in labs all around the world. It’s just that if your goal is to make millions and millions of those measurements, it’s not a scalable technology.
A great example: to generate the amount of affinity measurements that come off just one of our experiments using SPR, it would take a few weeks. At A-Alpha Bio, the equivalent amount of affinity data would cost you between $1 and $500 million if done at a CRO. We do a million measurements at a time. That math illustrates the fundamental constraint in the industry. Despite increasing awareness that this volume of data is transformative for understanding biology, no one is going to pay $100 million for a weeks-long experiment.
So the constraint we’re trying to solve is making these data—otherwise far too expensive and too hard to generate—easier, more affordable, and more economical.
David Brühlmann [00:08:40]:
That's exciting, and that's definitely the way to go—to be able to screen a lot more and then find, quote-unquote, “needles in the haystack,” but for a much smaller, modest budget. If we just look at the general picture—because drug discovery has been done for decades—how do companies do this traditionally? What are the traditional workflows and methods? Let’s start with the basics.
Troy Lionberger [00:09:06]:
The branch of therapeutic discovery that I come from is called in vivo discovery. In in vivo discovery, you are typically relying on an animal model whose competent immune system is ultimately responding to the presence of an antigen that’s presented, raising an immune response against that antigen. On the discovery side, scientists will access those antibody-producing cells, identify the ones producing an antibody very specific to your disease target, and then sequence those antibodies to move forward in developing them into a drug.
There’s a complementary approach called in vitro discovery, where you use what are literally called panning methods. You can imagine gold miners panning for gold, which gives you an appreciation for the basic philosophy behind current discovery: needles in a haystack, mining for gold. In phage panning, you use bacteriophage to express a version of your therapeutic and access very large diversities—many different combinations of molecules. You expose these to your therapeutic targets, find those that bind, sequence them, and move them forward in the development process.
David Brühlmann [00:10:20]:
I imagine there are a lot of advantages to this traditional approach. Can you highlight what those advantages are, and also what the limitations are?
Troy Lionberger [00:10:27]:
The advantages of the in vivo approaches using animals are that you're taking advantage of really one of the world's most sophisticated ways of generating diverse sequences of antibodies, which is a competent immune system. To date, while there is promise in AI, AI has not been able to generate the diversity of functional antibodies that a competent immune system can. That is the promise for the future of in silico methods. But to date, hands down, one of the finest ways of generating a diverse antibody response is using an immune system.
The advantages are the diversity. The disadvantages are that, in many cases, you're not able to get human antibodies because it would be unethical to immunize human beings for the purpose of generating antibodies for therapeutics. So we’re limited to animal models that produce antibodies that then have to be further engineered to be compatible with human biology. We have developed humanized animal models to solve that problem, but these are expensive and not commonplace. That is the challenge there.
On the in vitro side, using phage panning, it's much faster. The downside is there are often biophysical characterization issues with those molecules. For example, we’re phage panning at room temperature, but antibodies have to survive body temperature. If they melt or denature at body temperature, that's a problem. So there are other liabilities with the in vitro technologies.
David Brühlmann [00:11:58]:
With the new technologies advancing very rapidly, what is the picture you're seeing? Are we going to have a side-by-side approach, or eventually will AI, machine learning, and so on take over?
Troy Lionberger [00:12:11]:
It's definitely top of mind for me personally, and I should be upfront and say this is the first time I've worked on the machine learning and AI side of the industry. I'm definitely new to the game. So with that caveat, I’ll just say I’ve mentioned in vivo, in vitro, and now, as you mentioned, in silico approaches, which are now complementing the first two antibody discovery approaches.
De novo antibody design is the name of the field that is essentially trying to predict sequences of antibodies that will bind efficaciously to a therapeutic target. Right now, I see all of these as complementing one another. As I said, there are advantages to in vivo and in vitro technologies. In silico approaches often take the output of those approaches as inputs to their models. They’re absolutely interlinked today.
I think the promise of in silico methods is to eventually amass enough data that you can generalize these models so that you don’t actually need a wet bench. But I would argue the constraint is always—even if you didn’t need data to train your models, you will need data to validate your outputs. There’s no escaping the data cycle in this space.
There’s a lot of talk in AI about AI models taking the lead. I think there are advantages to de novo design in terms of epitopic accessibility and creating next-generation molecules. But as it stands today, I would describe them as complementary.
David Brühlmann [00:13:43]:
And I imagine that with AI and machine learning, you’ll be able to accelerate the workflows. It's not one or the other, as I’m hearing, but it’s definitely making certain steps of the process faster and more efficient.
Troy Lionberger [00:13:58]:
That's absolutely right. And I’ll give you a great example—an area that A-Alpha Bio is heavily involved in right now: optimizing antibodies to be lead candidate molecules for preclinical teams. Typically, after the discovery of an antibody, you want to optimize that candidate. That could involve making it more human so that it doesn’t interact negatively with the human immune system when injected into patients. It could involve improving developability, like reducing the propensity for aggregation or increasing how much you can produce from cells at scale. But you’re also optimizing affinities.
Historically, this has been a very complicated process. You’d focus on driving affinities to where you want them, then worry about secondary characteristics of molecules that may affect manufacturability downstream. After making those changes, you’d have to go back to ensure your affinity hasn’t veered off course. It’s a slow, recursive, iterative process that’s expensive and time-consuming at each phase, often ping-ponging back and forth through various parts of the value chain. This could take over a year of hard work and significant investment.
What we’re doing now, leveraging disruptive data generation to inform machine learning models in a bespoke way, is transformative. We can generate datasets that train models to predict higher-affinity antibodies while simultaneously optimizing other characteristics. For example, we can insert up to 21 mutations into an antibody and still achieve greater than 50% accuracy in maintaining higher affinity than the parental antibody. That shows the flexibility now possible—where previously you could only add two or three mutations to achieve desired characteristics.
We can now predict sequences that are more human, more developable, higher expressing, higher affinity, and more stable—all at the same time—in a three-month process. This condenses what used to take potentially years of effort into just a few months.
David Brühlmann [00:16:25]:
So traditionally, drug discovery would take years—like two years, three years? What is the benchmark?
Troy Lionberger [00:16:33]:
It really depends. Having been in this space, it depends on the approaches taken and the complexity of the target. But in terms of lead optimization, the part of the process I’m referring to, that could traditionally take one to three years, which we are now compressing into a three-month process.
David Brühlmann [00:16:54]:
Wow, that’s massive. Yeah, that’s a big change.
Troy Lionberger [00:16:56]:
Yeah. And after that three-month process, we’ve also done things that historically couldn’t be done. For example, we can drive the affinities of preclinical animal model antibodies to match human target affinities. That’s where it gets really exciting—asking how this changes the dynamics of the overall ecosystem if everyone’s projects could be accelerated through preclinical development. These molecules are essentially tailor-made for the experiments planned in preclinical studies.
David Brühlmann [00:17:30]:
At the end of the day, we need to find effective antibodies—the purpose of the whole drug discovery process is to find the most efficacious antibody. Now what I’m hearing is that we have technologies to accelerate the workflow. My question is: do these technologies also enable finding that very effective antibody, or do we need additional technologies on top of that?
Troy Lionberger [00:17:59]:
No, I would argue that the problem I just described is not just about making a faster, more efficient process. We are also hitting the target product profiles required for these therapeutics. There’s no sacrificing quality by accelerating the timeline, which is, I think, a rare example. In this three-month process, you’re actually starting to guarantee results—something I never thought would be possible with a service provider.
David Brühlmann [00:18:28]:
Can you lead us into how you’re achieving that? What is the technology used, and what are the major steps?
Troy Lionberger [00:18:34]:
At a high level, here’s how it works: we generate tens of thousands of mutations of a parental antibody. We randomly insert arbitrary point mutations, making mutants with one, two, or three mutations at a time. We then measure the affinity of those tens of thousands of antibodies against a panel of different targets—not just the human target, but maybe mouse and cynomolgus targets, as well as point mutations of targets or comparable family members. This helps avoid early readouts of nonspecific binding.
We gather data not only on binding affinity and cross-reactivity, but also specificity from each of those tens of thousands of molecules. We then use all that data to fine-tune AlphaBind, our computational platform. The AlphaBind model has been trained on close to a billion different affinity measurements generated by the company. Fine-tuning the model with data from a specific parental antibody trains it to predict mutations that can be introduced into the original molecule.
The machine learning picks up on synergistic and compensatory mutations that might not be obvious to the human eye but are clear in the data. As a result, we can be greater than 90% confident in generating antibodies with higher affinity, even with up to 15 point mutations.
In parallel, we use off-the-shelf developability models to downselect molecules. While optimizing affinity, we simultaneously evaluate expression, solubility, and melting temperature to ensure the molecules are manufacturable.
David Brühlmann [00:20:33]:
So you generate millions of affinity measurements in a week. Are you using yeast display and genomics to do that, or what’s the trick behind it?
Troy Lionberger [00:20:43]:
Yeah, let me go a bit under the hood. Our wet-lab technology is called AlphaSeq. It’s a yeast display system that takes advantage of yeast mating pathways in nature. In yeast genetics, the A and α (alpha) strains have mating receptors that engage to form a diploid cell containing the genetics of both parent cells.
What our co-founder and CEO, David Younger, proved while in David Baker’s lab is that there’s nothing extremely specific about these mating receptors. You can attach molecules of interest to the outside of these cells, and if those molecules have measurable affinity, they will increase the rate of diploid formation—or yeast mating.
Here’s how it works in practice: you have a culture with a library of A strains expressing an antibody library and α strains expressing targets of interest. Under optimized culture conditions, mating occurs. After some time, you sequence everything. The number of times you see a specific gene pair correlates with binding affinity.
The diploid cells contain the genetics of both “mom” and “dad,” so each readout provides a genomic barcode corresponding to a specific antibody–antigen pair. You can quantitatively relate that to affinity in molarity, which correlates well (about 0.85) with standard measurements like SPR (surface plasmon resonance) or BLI (biolayer interferometry).
The trick here is extracting a biophysical measurement from a genomic readout. Next-generation sequencing provides the scalable data we need, and we can reliably relate it to binding affinities for millions of molecules simultaneously.
David Brühlmann [00:23:18]:
You have a good correlation. I’m curious—what is the affinity range you’re able to measure?
Troy Lionberger [00:23:25]:
Great question. The affinity range we usually see—though we can tune our culture conditions to adjust sensitivity—is typically from hundreds of picomolar up to tens of micromolar. So it’s a very large dynamic range that covers the therapeutic range most people are interested in.
David Brühlmann [00:23:53]:
That wraps up part one of our conversation with Troy Lionberger on the data revolution in antibody discovery. We’ve explored the limitations of traditional methods and how A-Alpha Bio’s AlphaSeq platform is changing the game.
In part two, we’ll discover how machine learning transforms this massive dataset into predictive power. If you’re finding value in this episode, please leave us a review on Apple Podcasts or your favorite platform—it helps other scientists like you discover the show.
All right, smart scientists, that’s all for today on the Smart Biotech Scientist Podcast. Thank you for tuning in and joining us on your journey to bioprocess mastery. If you enjoyed this episode, please leave a review on Apple Podcasts or your preferred platform. By doing so, we can empower more scientists like you.
For additional bioprocessing tips, visit us at www.bruehlmann-consulting.com. Stay tuned for more inspiring biotech insights in our next episode. Until then, let’s continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Troy Lionberger
Troy Lionberger serves as Chief Business Officer at A-Alpha Bio. He started his career as a research scientist after earning his PhD from the University of Michigan and completing postdoctoral training at UC Berkeley. During nearly a decade at Berkeley Lights, Troy held senior leadership positions spanning R&D, product strategy, and business development.
Before joining A-Alpha Bio, he was Chief Business Officer at Abbratech, where he guided the company from stealth into a partner-focused antibody discovery biotech. At A-Alpha Bio, Troy applies his strong technical background and experience scaling platform companies through strategic partnerships to help position the company as a key enabler in the industry.
Connect with Troy Lionberger on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
For generations, silkworm pupae were simply a byproduct of silk spinning. Today, the biotech spotlight is shifting to their dormant power: transforming these “waste” organisms into natural protein factories. Turns out, when silkworm pupae are harnessed as living bioreactors, they can produce complex recombinant proteins and vaccine antigens at a scale—and cost—that makes mammalian cell culture systems look cumbersome.
In this episode of the Smart Biotech Scientist Podcast, host David Brühlmann speaks with Masafumi Osawa, a global strategy leader at KAICO with an unconventional path into biotechnology. Originally trained in cultural anthropology, Masafumi’s career was shaped by hands-on experience in pharmaceutical business development across diverse markets.
Silkworms were once used purely for producing silk, and their pupae were often considered waste. Today, those same silkworm pupae have the potential to address major global health challenges and offer new modalities for vaccines and therapeutic proteins. If any researchers listening today are struggling with difficult protein expression—whether it's a VLP, membrane protein, or complex antigen—I would be very happy to explore how we can support your R&D.
David Brühlmann [00:00:36]:
Welcome back to Part Two with Masafumi Osawa from KAICO. In Part One, we explored how silkworm pupae function as natural bioreactors expressing complex proteins using a baculovirus expression system. Now we’re moving from platform science to product reality. KAICO isn’t just offering contract services—they’re developing injectable and oral vaccines for both human and animal health. We’ll examine their development pipeline, discuss the unique regulatory considerations when your bioreactor is alive, and explore where silkworm-based manufacturing fits into the future of biologics production. Let’s continue our conversation.
Let’s shift gear, Masa. Let’s focus on your product pipeline. What are the kind of molecules you're developing? How well advanced are you in these various molecules? I'm curious to see what you're working on right now.
Masafumi Osawa [00:02:50]:
Our pipeline is designed around a stepwise regulatory strategy, starting with segments that allow for rapid entry and progressing toward human applications after livestock validation. Our most advanced program is the PCV2 oral immunization product for pigs, registered in Vietnam as a functional feed additive. In real farm environments, it has shown performance comparable to injectable vaccines while substantially reducing labor and stress costs. This provides strong validation of oral antigen delivery using silkworm-derived proteins in livestock.
We are also developing oral vaccines for cats, including FPV (feline panleukopenia virus), FCV (feline calicivirus), and FHV-1 (feline herpesvirus type 1), and producing purified CPV (canine parvovirus) antigens for dogs. Our aim is to reduce vaccination stress in animals and offer alternatives to in-clinic injections, with implications for human health as well.
Our recombinant injectable human norovirus vaccine is preparing for Phase I clinical trials next summer under Japan’s AMED SCARDA program. This marks a major step toward establishing insect-based platforms in human pharmaceuticals. Beyond internal programs, we collaborate with partners to express complex proteins and antibodies, leveraging the unique capabilities of the silkworm system. Overall, our pipeline reflects a long-term progression from livestock to companion animals to human injectables and eventually oral medical vaccines for humans.
David Brühlmann [00:04:39]:
Oral vaccines are an exciting delivery approach because they reduce distress during administration. Are there specific antigen characteristics that work better—or worse—for oral vaccine applications, and where are the limits?
Masafumi Osawa [00:04:57]:
Silkworm-derived antigens offer significant advantages for both injectable and oral vaccines, although the reasons differ by modality. For injectable vaccines, the key strength lies in the ability to express antigens that are difficult or sometimes impossible to produce in other systems. This includes complex structural proteins and virus-like particles (VLPs), which often do not require mammalian-type N-linked glycosylation and therefore assemble particularly well in insect-based platforms.
The silkworm pupal environment provides a dense, multicellular physiological setting that naturally supports proper folding, multimerization, and high-yield expression. This is why we can obtain enough purified antigen from a single pupa to immunize several hundred pigs. When manufacturing injectable swine vaccines, this level of efficiency is extremely difficult to match with conventional cell culture systems.
For oral vaccines, the advantages are even more distinct. Previous plant-based oral vaccine approaches—such as rice or other edible crops—have struggled due to low expression levels or sharply increasing production costs at industrial scale. As a result, despite scientific interest, few plant-derived oral vaccines have reached commercial feasibility.
Silkworm pupae, however, function as naturally concentrated bioreactors, delivering expression levels far higher than those typically achieved in plants. At the same time, silkworms can be mass-produced at low cost, making them well suited for oral vaccine applications where dosage volumes are much larger than injectables. In our PCV2 oral program, for example, we formulate approximately 1.5 g per pig, accounting for variation in feed intake and ensuring sufficient mucosal exposure. By contrast, injectable formulations—with higher purity and potency—allow a single pupa to cover hundreds of animals.
Additionally, silkworm-based production avoids large bioreactors, extensive culture media, sterile water systems, and intensive cleaning operations. These factors significantly reduce manufacturing costs and environmental burden, giving our platform a strong economic advantage—particularly in vaccine applications where global accessibility and scalability are critical.
David Brühlmann [00:07:49]:
During the COVID pandemic, we heard over and over how important it is to be able to develop vaccines very quickly. I’m curious—using the silkworm platform, how fast can you develop a new vaccine? Is this comparable to, for instance, mRNA, or how does it differ?
Masafumi Osawa [00:08:08]:
As long as the DNA sequence is available, we can produce any kind of recombinant protein. That’s the first point. One example is that during COVID we also produced SARS-CoV-2 recombinant spike protein. It took just three months after the outbreak of COVID-19. I think that is one remarkable aspect of our platform.
David Brühlmann [00:08:32]:
That’s remarkable—very fast. And the advantage I see with your platform is that because you’re scaling out, it’s very easy to expand production rapidly and produce massive amounts of vaccines in a short time. On our podcast we usually focus on human medicine, so I’d like to take a quick deep dive into the animal side of things because that’s also very important. We often forget that there’s a huge market for animal health. What are the main regulatory differences between the two? Can you give us the two-minute version? Obviously, I'm sure there's a lot of more details, but what is the two minute version of the differences between human and animal health?
Masafumi Osawa [00:09:14]:
Regulatory pathways for silkworm-derived products differ significantly between human and animal health. Human vaccines follow globally harmonized standards, but because silkworm-derived antigens are unprecedented, we must work closely with Japan’s PMDA (Pharmaceuticals and Medical Devices Agency) to define raw material controls, GMP, CMC expectations, and quality control frameworks.
In the animal health sector, pathways differ by country. In Vietnam, our PCV2 product was registered as a functional feed additive rather than as a pharmaceutical, enabling rapid market entry. Companion animal vaccines follow pharmaceutical regulatory frameworks, but typically with shorter timelines than human vaccines. These differences allow us to pursue a staged development strategy—starting with faster, more accessible applications, generating real-world validation, and gradually advancing toward more tightly regulated markets.
David Brühlmann [00:10:18]:
Okay, so it’s country-dependent. And as I recall, there are also major differences between human health and animal health regulations?
Masafumi Osawa [00:10:27]:
Yes. However, for companion animals, regulatory standards are often harmonized through VICH (International Cooperation on Harmonisation of Technical Requirements for Registration of Veterinary Medicinal Products). For products like our PCV2 immune-enhancing feed additive, since there is no comparable product globally, the regulatory classification depends on how local authorities decide to position it.
David Brühlmann [00:10:55]:
Let’s look ahead. You’re still a relatively young company. What does the future hold? If we look at therapeutic areas, product types, or services, what are your next steps?
Masafumi Osawa [00:11:10]:
While vaccines represent our core focus, the silkworm system has broader therapeutic potential. Many antibodies and complex recombinant proteins are difficult to express in standard systems due to folding challenges or instability. Silkworm pupae, with their diverse cell types and chaperone-rich environment, offer an alternative solution for these difficult targets. Sustainability is another strong feature of silkworm biomanufacturing. Because the system requires minimal water, no bioreactors, and low energy inputs, the overall environmental footprint is significantly lower than conventional platforms. This opens possibilities for distributed manufacturing models where production can occur closer to the end user, including in emerging regions with limited infrastructure. In the long term, the flexibility of small-batch production means that silkworms may contribute to personalized biologics or rare disease therapeutics.
David Brühlmann [00:12:16]:
When do you think we’ll see the first biologic approved that was produced in silkworms? Do you have a sense of timing—five years, ten years?
Masafumi Osawa [00:12:27]:
That’s very difficult to answer because we are about to enter a Phase I clinical study next year—next summer, to be precise. I don’t know how many more years it will take, but it’s becoming very real. In the past, no one believed that a live silkworm body could be a source of APIs or vaccine antigens, but now it’s becoming a reality. Entering a Phase I clinical study means that products derived directly from silkworms are about to be administered to humans. So I cannot give a timeline, but this is a major step.
David Brühlmann [00:13:10]:
That’s a major milestone and shows that you’ve done the homework and achieved initial regulatory acceptance. Obviously, there’s still a lot ahead, but entering Phase I trials shows that regulatory bodies see the potential and trust the technology.
Masafumi Osawa [00:13:32]:
Yes. Regulatory authorities now accept our quality control strategy and how we manage consistency and safety, which opens the door to further pharmaceutical development opportunities.
David Brühlmann [00:13:45]:
Absolutely. You’ve demonstrated proof of concept, and once that foundation is laid, you can build on it. That’s wonderful. Before we wrap up, Masa, is there any burning question I haven’t asked that you’d like to share with our biotech community?
Masafumi Osawa [00:14:04]:
One thing I may not have mentioned is the production volume of a single silkworm pupa. Productivity is another strong advantage. One silkworm pupa, about 2 to 3 centimeters in size, can express 10 to 20 milligrams of norovirus virus-like particles (VLPs). After purification, this typically yields 1 to 2 milligrams per pupa, which is still a substantial amount. This is why scaling out is such a strong advantage of our platform. If we need 100 milligrams of product, we simply require 100 silkworm pupae. The total space needed is about the size of a laptop. Compared to large-scale manufacturing equipment, this is extremely compact, making it a key benefit of our platform.
David Brühlmann [00:15:09]:
As we wrap up, Masa, what is the most important takeaway from our conversation?
Masafumi Osawa [00:15:15]:
Silkworms were once used purely for silk production, and their pupae were often considered waste. Today, those same pupae have the potential to address major global health challenges and offer new modalities for vaccines and therapeutic proteins. If any researchers listening are struggling with difficult protein expression—whether VLPs, membrane proteins, or complex antigens—I would be very happy to explore how we can support your R&D.
David Brühlmann [00:15:49]:
This has been great, Masa. Thank you for sharing your work and for helping democratize life-saving therapies by pushing boundaries beyond what many people think is possible. Where can people connect with you?
Masafumi Osawa [00:16:08]:
Please feel free to connect with me on LinkedIn: Masafumi Osawa.
David Brühlmann [00:16:17]:
There you have it, Smart Biotech Scientists. Please reach out to Masa and learn more about the technology. Once again, thank you very much for being on the show today.
Masafumi Osawa [00:16:28]:
Thank you very much for having me, David. It was my pleasure.
David Brühlmann [00:16:33]:
Thank you for joining us for this deep dive into silkworm-based biomanufacturing with Masafumi Osawa. From ancient silk production to modern vaccine development, KAICO is proving that nature still has lessons to teach us about efficient bioprocessing. If this episode expanded your thinking about alternative expression platforms, please leave a review on Apple Podcasts or your favorite platform. Your feedback helps us bring you more cutting-edge biotech insights. Thank you so much for tuning in today and I'll see you next time.
All right, smart scientists, that's all for today on the Smart Biotech Scientist Podcast. Thank you for tuning in and joining us on your journey to bioprocess mastery. If you enjoyed this episode, please leave a review on Apple Podcasts or your favorite podcast platform. By doing so, we can empower more scientists like you. For additional bioprocessing tips, Visit us at www.bruehlmann-consulting.com. Stay tuned for more inspiring biotech insights in our next episode. Until then, let's continue to smarten up biotech.
Disclaimer: This transcript was generated with the assistance of artificial intelligence. While efforts have been made to ensure accuracy, it may contain errors, omissions, or misinterpretations. The text has been lightly edited and optimized for readability and flow. Please do not rely on it as a verbatim record.
Book a free consultation to help you get started on any questions you may have about bioprocess development: https://bruehlmann-consulting.com/call
About Masafumi Osawa
Masafumi Osawa brings more than a decade of experience in the pharmaceutical sector, with a strong focus on driving innovation to address global health needs. He is the Business Development Lead at KAICO Ltd., a Japan-based biotechnology start-up specializing in recombinant protein production using silkworms as biological reactors.
At KAICO, he leads strategic partnership development, represents the company at industry events and technical forums, and applies his strengths in market analysis, CRM, and communications to clearly articulate KAICO’s vision and promote its distinctive technologies, including oral vaccine platforms for both human and veterinary use.
Connect with Masafumi Osawa on LinkedIn.
David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.
He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.
Hear It From The Horse’s Mouth
Want to listen to the full interview? Go to Smart Biotech Scientist Podcast.
Want to hear more? Do visit the podcast page and check out other episodes.
Do you wish to simplify your biologics drug development project? Contact Us
