Podcast Transcript

David Brühlmann [00:00:45]:
Picture this: a biologics program moves through early development without a hitch, and then at scale-up, everything stalls. The root cause? Decisions made years earlier in the cell line and process development stage.

Today, we’re digging into manufacturability, what it really means, and why it’s so much harder to achieve than it sounds. Joining me is Sigma Mostafa, who is the Chief Scientific and Technology Officer at KBI Biopharma, and who has spent her career turning early-stage science into commercially manufacturable reality.

Welcome, Sigma. It’s good to have you on today.

Sigma Mostafa [00:02:40]:
Thank you, David. So great to be here.

David Brühlmann [00:02:43]:
Absolutely a pleasure. And it’s good to have you back on. So we did an interview a couple of years ago. Really excited to talk about very important topics.
So to start us out, Sigma, share something that you believe about bioprocess development that most people disagree with.

Sigma Mostafa [00:03:01]:
I do think that when companies are at the candidate selection stage, being able to do some level of assessment of the molecules and bring in manufacturability even at that stage is important.

I think most companies look at candidate selection more based on PK/PD data. Even small improvements in performance in terms of activity are the focus, but not manufacturability.

Some of it is cost-related. Some of it is just not realizing how big a disadvantage you might have by choosing the one candidate that is not particularly manufacturable.
So that’s one area where I don’t see a lot of traction, where people are willing to think about manufacturability at the candidate selection stage.

David Brühlmann [00:03:51]:
Yes, manufacturability is essential, and I’m very much looking forward to diving deeper into that topic and unpacking it. Before we do that, tell us your story. I’m sure some of the listeners haven’t listened to your previous interview, but if you haven’t, please go back and listen to that one.

But tell us your story. What drew you into science, and what were some interesting pit stops along your way?

Sigma Mostafa [00:04:18]:
My dad was a mathematician who named me after a Greek alphabet, Sigma. So I was, in some ways, predestined to be in STEM, I guess. But I really loved math, and I always knew I wanted to be in a field where I could bring math and biology together, where I saw having a true impact and feeling fulfilled by working in a field where you are bringing life-saving drugs to market.

So early on, I didn’t quite know how I’d bring biology and math together. I got into chemical engineering at the University of Texas at Austin and had the opportunity to work with world-renowned scientists in their labs, with George Guo and Jeff Hubbell. I got to work on biochemical and biomedical-type research as an undergrad and knew that I wanted to be more on the biochemical engineering side, this kind of process development aspect.

And then my PhD work was on stem cells, but essentially growing stem cells into types of blood cells so they can be put back into cancer patients. But it was all about engineering, like how to grow these cells, what’s the impact of pH, oxygen, etc. It was a natural transition into biotech.

What really still excites me after 25-plus years is that, unlike the chemical industry and some of the other industries, there is still a bit of art to it, in the sense that we are working with biological organisms to generate these proteins.

And when I started in industry in 2000, a lot of it was art. A lot of things were not understood as to what genes are impacting what in CHO cells or E. coli to get to better productivity, product quality, etc., and how to just manipulate the systems. But we still don’t fully understand. So there is this component of magic to it, and there is still a lot of opportunity to optimize, which is what makes it exciting.

There is still a big space in terms of truly optimizing the system. So we are not just doing the same thing again and again. And at the same time, the molecule complexity is going up. So now that we are looking at all these conjugates, we are not just talking about the biological part, but the chemical components that are getting conjugated. We need to understand both sides of it. So I think it continues to be very exciting for me and others.

David Brühlmann [00:06:50]:
No doubt about that. And I think innovation is happening faster than ever. So there’s new modalities, there’s new technologies, there’s a lot of things happening.

Sigma Mostafa [00:07:00]:
Exactly.

David Brühlmann [00:07:01]:
Let’s dive into manufacturability, because I’ve seen cases where research comes with an amazing molecule, great affinity, perhaps even great stability in some cases—the perfect properties.

But then you have the discussion, and you just realize it’s not manufacturable. So why is it actually much harder than it sounds? It’s actually pretty logical. You need to be able to make the molecule. But why is that conversation so much harder? And what does it actually really mean for a molecule to be manufacturable?

Sigma Mostafa [00:07:32]:
I’ll start with an example. When I was at Eli Lilly, my first job out of university, the typical modality at that time was to work with HEK cells—for R&D. And you have lower productivity; you are really not looking at product quality. So R&D developed—they had worked on the molecule—and then it gets transferred to development in our shop. And I was responsible for all of CMC for this one program.

And when we moved to CHO cell lines, we started seeing that there was a truncated product, which was never realized early on. The reason is that the productivity was so low, the impurity level was so low, no one was really even looking into it. As we looked at the higher quantity of protein, and we started looking at why we were having low yield in downstream, there was this realization that there was a secondary product.

So that’s an extreme example where, at the protein engineering level, there could be a truncated product, but that sort of gives an example of the disconnect. If you use a different kind of cell line or different concentration of products, you may not see all the problems that will show up at scale. And those problems, if you don’t study the manufacturability early enough, can truly become a blocker to the point that you cannot get to market.

And I don’t think, because there are different groups working on the molecule at different stages, there is still full understanding of how much of a detriment this can be.

A big portion is definitely: is the protein you are working on in the early phase, does it have the right product quality?

If you’re going to have aggregation issues or fragmentation issues, or at the high concentration you have to go to because your competitor is already at high concentration and you need every week or every month kind of injection, therefore you need high-concentration drug substance, all of that is not taken into account. And by the time you realize that, you don’t have a manufacturable product.

David Brühlmann [00:09:41]:
How early should you start thinking about manufacturability?

Sigma Mostafa [00:09:45]:
As I started by saying, I think almost at the candidate selection stage, because we do see that in some cases the molecule is not manufacturable, or there’s going to be a lot of extra studies to come to kind of a very bespoke solution.

Then we find out, “Oh, they had another candidate, the second one, which did not have this issue at all,” but there was no realization. I think the biggest things that absolutely we need to understand early on are: is there any kind of property of the molecule that leads to, let’s say, thermal stability versus aggregation, that will pull you in different directions in how to solve it?

And finding a common solution that gets you to the level of long-term stability of your drug product versus the type of concentration you’re looking for. Can you find a solution that makes it viable to be in the market?

David Brühlmann [00:10:42]:
What are the main reasons that some companies tend to look into manufacturability too late? Is that a lack of awareness early on? Is that perhaps a time pressure because there’s increasing pressure on being faster in development? Is that something else? What are you seeing?

Sigma Mostafa [00:11:01]:
I think the time factor is a big one. There is such a need to show clinical data. There used to be a time when you had just filed an IND, and that was a good enough benchmark and you could get funding, but you now have to show that in clinic you are doing well to continue to get funding and all of that.

So time is a factor, associated cost is a factor, and of course everyone thinks their molecule is in good shape. If they knew that there might be certain issues with this particular molecule, they would be more amenable.

But because the studies haven’t been done, the deep characterization hasn’t been done, that is not understood. And they just think this is extra cost, extra time, and something that can be done in a later phase.

David Brühlmann [00:11:53]:
Something I had to say multiple times when I was running my consultancy, especially to founders who come from different backgrounds, is that CMC should not be an afterthought. It should be built from the very beginning.

Sigma Mostafa [00:12:08]:
Exactly. Exactly. I mean, a lot of folks see that and think, “It will just come. That’s just scaling, and that’s not really science.” But the reality is, it is very hardcore science, and there are a lot of things that can go wrong with scalability or with getting to the right concentration of drug product, all that.

David Brühlmann [00:12:28]:
When we talk about CMC, one aspect that comes to my mind is cell line development. That’s the other part. Once we have a molecule that is manufacturable, we then have to develop the cell line. What are important considerations on that end in order to make sure that you make the right decisions early on?

Sigma Mostafa [00:12:47]:
So, cell line—that decision is probably the most important one you are making because, in most cases, you are not going to change your cell line. You can change your downstream process, make it more manufacturing-friendly and all that. But if you do not have a robust cell line, it will haunt you for the lifecycle of the molecule.

We, of course, have our Selexis platform, but we work with all kinds of cell lines and probably 20-some CHO cell line vendors. We have worked with all kinds of cell lines, from standalone cell line providers to integrated programs that got tech transferred into us.

So we see from other CMC demos as well. And what we see is, if your cell line is not robust enough—and what I mean by robustness is, let’s say, pH sensitivity: if your process goes off by, in your cell culture, 0.1 pH unit, the cell line growth rate gets impacted, or that level of sensitivity—it should be a red flag. We do see that now that we are onshoring a lot of programs in the current environment. We do see cell lines where we are surprised by the sensitivity to certain things that, in this day and age, we don’t expect.

So one other thing around cell lines is, once you have made your master cell bank, you are kind of locked. And what media you made that cell line in is important. For example, what we should not have is, after making the master cell bank, you’re like, “Oh no, I’m going to change my media. I’m going to make some changes on my seed train.“

Because what that means is, when you do VLT for your GMP runs, during seed train, you are causing adaptation to the cell line, which is not a good idea because every time you may see—you’re creating stochastic scenarios where the cells may adapt differently. Ideally, you’d be in the right media at least, and you have the right cell line going into the master cell bank.

David Brühlmann [00:14:51]:
Absolutely. And perhaps another important aspect in cell line development is all the regulatory aspects and that you’re compliant with raw materials and so on. What are some common pitfalls you see on that end?

Sigma Mostafa [00:15:03]:
In terms of the regulatory side, you have to have your clonality data. And then, in terms of raw materials, being completely animal-source-material-free. And not just in terms of what goes into your culture, but what reagents were you using for selection, any kind of selection process you are doing.

We have seen situations where realization of animal-source material led to going back and redoing the cell line. Extremely important, especially in Europe. That can be a big factor. There’s very little tolerance for animal-source material. But other regulatory considerations, I’d say more around: did you show your clonality properly, and do you have the right documentation?

David Brühlmann [00:15:52]:
That’s a lot to sit with. The idea that today’s tech transfer headaches so often trace back to choices made long before anyone thought about scale-up. Sigma Mostafa’s perspective is a reminder that manufacturability isn’t an afterthought; it’s a mindset from day one.

We’re just getting started, so stick around. If this conversation is giving you something to think about, please leave a review on Apple Podcasts or your favorite platform. It will help scientists like you discover the show.

Thank you so much for tuning in, and I’ll see you next time for Part 2.