Scaling Manufacturing For Personalized Cell Therapies With Cellino's Marinna Madrid, Ph.D.

On this week's episode of the Business of Biotech, Marinna Madrid, Ph.D., Cofounder, Board Member, Chief Product and Regulatory Officer at Cellino, talks about scaling up an automated manufacturing process she co-invented for personalized autologous induced pluripotent stem cell (iPSC) therapies using laser-activated substrates and AI. Marinna describes her work with industry groups to improve regulatory guidelines, Cellino's partnership strategy, the implications of receiving FDA's Advanced Manufacturing Technologies designation, and the future of iPSC therapies.  

Access this and hundreds of episodes of the Business of Biotech videocast under the Business of Biotech tab at lifescienceleader.com.  

Subscribe to our monthly Business of Biotech newsletter.

Get in touch with guest and topic suggestions: ben.comer@lifescienceleader.com

Find Ben Comer on LinkedIn: https://www.linkedin.com/in/bencomer/

Transcript

Welcome And Guest Overview

Ben Comer 0:03

Welcome back to the Business of Biotech. I'm your host, Ben Comer, Chief Editor at Life Science Leader, and today I'm speaking with Marinna Madrid, PhD, co-founder, board member, and chief product and regulatory officer at Cellino, a company developing AI-powered autonomous manufacturing systems used to create personalized regenerative therapies, such as autologous-induced pluripotent stem cell therapies. Marinna co-invented laser-based intracellular delivery techniques during her PhD program at Harvard and Applied Physics. And I'm excited to speak with her today about deploying those inventions at Cellino and getting to personalize cell therapies at scale. We'll talk about Cellino's partnership with Karis Bio, getting an advanced manufacturing designation from the FDA and the regulatory work that Marinna is doing and what's next for the company. Thank you so much for coming on the show, Marinna.

From Journalism To Biophysics

Marinna Madrid, Ph.D. 1:02

Thank you for the invitation. I really appreciate it.

Ben Comer 1:05

I'm pleased to have you. I want to start out as we do with your background. And I'm curious, um, what interested you initially about biophysics? Was physics something that you know you you were always found compelling and were interested in? Or how did you come to that field?

Marinna Madrid, Ph.D. 1:25

Yeah, so it's it's interesting. I actually didn't start college thinking about physics or even science at all. I started college. So I grew up in San San Bernardino, Southern California, and I started college at NYU studying journalism. Um super, super different. And it ended up being a terrible fit for me. I think, you know, I was young, I didn't like the city, I didn't like the school, I didn't like the major. I actually ended up dropping out in the middle of my first year and moving back home and starting over at community college. And community college is really where I fell in love with science. I um the science classes, I will say, like they came to me more easily and they were more interesting to me. And so for my biology, my math, my physics courses, I ended up um kind of teaching mini lessons outside of class to help the students who didn't get the material the first time around. And my teachers noticed this. They got me a job as a supplemental instructor at my community college. So I was getting paid for it. They were just very, very supportive of these interests. Um, and ultimately they encouraged me to transfer to a university to study, to study science. And so I ended up transferring to UCLA. I did major in biophysics. At the time, UCLA didn't have a very um established biophysics program. It was basically a physics degree plus like one or two bio and chem classes added to it. Um, so it was essentially a physics degree. Um, but I was studying biophysics at the time with the goal to be a full-time community college teacher. So to teach physics um at a community college full-time. That was my goal. And you don't really need a PhD for that, you just need a master's, but a PhD is paid, you a PhD is free, whereas a master's, you have to pay for it. So I applied to PhD programs, um, got into Harvard. The PI that I went to um work with at Harvard is unique, Eric Mazur, in that he does, he's in the applied physics program, he has a femtosecond laser lab, but he's also very passionate about physics education research. And so that's actually what drew me to that lab is I thought I'm gonna be a community college teacher, I'm gonna teach physics full-time, and so I want to learn how to teach physics really, really well. Um, and so I so I went to that lab. I was studying physics education research, I was teaching physics way more than like the typical grad student because that's what I was interested in. And Nabiha Saklayen at the time was also in the lab, and she was working on this biophotonics project that involved using lasers to manipulate cells. And she presented um for her in preparation for her qualifying exam, and I just fell in love with the work, and that's what really drew me into it. So um didn't initially focus on science, but was drawn into it in community college, and also didn't initially focus on entrepreneurship as a career, but kind of ended up serendipitously taking that path later in grad school.

Ben Comer 4:14

So Nabiha Saklayen, uh, she is the CEO at Cellino, I'm sorry. Um, you so did you guys meet uh in the the PhD program at Harvard?

Marinna Madrid, Ph.D. 4:26

Yep, exactly. Yeah. So she was in the PhD program. We started working together because I was drawn to her project, and we've just worked together ever since. So over a decade now.

Ben Comer 4:36

Wow. Okay, so the uh the the physics curriculum in community college you know really clicked with you. You you enjoyed doing it. Did you experience that same thing with with bio, with biology? I mean, um uh was that was that similar similarly something that the the actual coursework really brought you into?

Laser Substrates And Bubble Control

Marinna Madrid, Ph.D. 4:55

Yeah, definitely I would say so. I would say my physics training is more traditional. I've taken a lot of physics courses and like gone through the you know typical physics textbooks. A lot of my biology learnings have been more on the go. Biology is such a vast field that typically what's happened is I've been on some sort of project, whether it was like a research project in undergrad or a PhD thesis project or the work we're doing at Cellino, and then I've had to learn the relevant biology on the go. Um, and that's worked really, really well for me. I mean, I think it's um a fairly accessible field. You can learn the same way you learn anything by reading a lot and then talking to a lot of people who have expertise in that area. Uh, but yeah, I would say the biology, what I love about biology is that uh there's still a lot of mystery. With physics, physics is a much slower moving field. You get a major discovery once every 30 to 40 years. You know, in the 40s, it was the atomic bomb, and like, was it the 70s or 80s? It was lasers, and then more recently, gravitational waves. You don't get major discoveries in physics as often anymore. It's a slower moving field. With biology, there's still so much unknown that we have a major discovery that's published almost every other day. Um, so I think biology, biology, I just I love the momentum that there is in this field, and I love that there are still so many unknowns. We don't have a fully operational mathematical model of even the cell, like let alone the human body yet.

Ben Comer 6:20

Yeah. Yeah. Well, did lasers enter the picture uh with Dr. Eric Mazur? Is that where you first started working with lasers? Yeah.

Marinna Madrid, Ph.D. 6:29

Yes, exactly. So his lab is a femtosecond laser lab, and there are a bunch of different, there are a lot of different work streams in that lab. We were the we were part of the biophotonics group. So what we were doing is we were using lasers to um using lasers and laser activated substrates to basically poke holes in cells for cargo delivery. That's what we were doing in our PhD. That's not what we're doing at Cellino now. We're not using lasers for intracellular delivery. Um so the the, you know, the work and the focus has evolved a lot since then, as it should, once you actually kind of get out into the real world and start engaging customers. But the common theme has always been lasers and cells.

Ben Comer 7:07

Well, I came across this phrase, laser-activated thermoplasmonic substrates. And I would love for you to tell me and probably a lot of the audience uh what those are and what they can be used to do.

Marinna Madrid, Ph.D. 7:20

Yeah, so they're very efficient at absorbing laser energy and then transferring the energy to the surroundings in a way that can be used to manipulate cells. So in the past, in the early days of laser manipulation of cells, lasers would be focused directly onto the cell membrane. And that works really well. You can get the effect you want, you can poke a hole in the cell membrane to deliver cargo, but it requires very precise focus of the laser on a cell membrane. And so, as you can imagine, it's not very scalable because if you were wanting to deliver to a million cells, you would have to laser focus onto each and every single cell membrane one at a time. Um, so what we wanted to do was to make that whole process more scalable. So instead of laser focusing on each and every single cell membrane one at a time, we uh developed these laser-activated substrates that you can culture cells on. So you could culture millions of cells on them at a time. And then it's very fast to just scan the laser across the substrate. And what the substrate does is it absorbs the energy in the laser photons and then transfers it to the aqueous surroundings in a way that creates a bubble. And it's actually the bubble then that expands and collapses and interacts with the cell. And so you can precisely control the size of that bubble based on the energy of the laser pulse. So less energy is going to give you a smaller bubble, more energy is gonna give you a larger bubble, a smaller bubble might just poke a hole in the cell membrane to deliver cargo, a larger bubble might completely kill or lice a cell or lift a cell off a surface.

Ben Comer 8:48

What are the substrates actually made of?

Marinna Madrid, Ph.D. 8:52

So it's, you know, it varies. I've worked with a lot of different laser-activated substrates over the years. Initially, you know, our first publications were on um gold pyramids and titanium nanocavities. Um, and so initially we were focused on some combination of nanostructure and metallic coating. But the truth is that there are quite a lot of quite a lot of structures that will work. And people have published on non-metallic structures as well. Um, but the nanostructures are efficient at interacting with the laser energy, essentially.

Ben Comer 9:25

And the the way that lasers were formerly used to puncture the cell membrane and create a place for delivery. Um, what is the delivery mechanism? What's used to actually deliver things into the cell in that?

Marinna Madrid, Ph.D. 9:38

Yeah. So it's just diffusion. So like you have the delivery cargo in the cell media, it's already surrounding the cells, and you poke a hole in the cell membrane, then the delivery cargo can diffuse in. Um, and then the cell membrane heals on its own. It just rearranges and closes up that hole that was formed.

Ben Comer 9:54

Well, you are a co-inventor in this uh in the laser active uh thermoplasmic substrate space. Can you describe what your uh invention was?

Marinna Madrid, Ph.D. 10:05

Yeah, so there's, I mean, I'm a um I'm an inventor on a number of patents, but essentially it's covering the space of if you have a nanostructured surface, can you use it to absorb laser energy? And then can you use that to do a number of things to cells? Um and it's basically via this bubble-mediated mechanism.

Ben Comer 10:26

And this were this was uh um discoveries that you were making during your PhD program uh at Harvard. At what point did you start looking for ways um to apply those technologies to treating patients?

Marinna Madrid, Ph.D. 10:40

So we were always very application focused. But what I would say is in your PhD, you don't have a very strong understanding of what the actual market need is. Like you think you understand what the need is, and it's mostly based on literature review, but that's a fairly weak understanding of the field compared to actually going out into the world and talking to a bunch of people. So when we started Cellino, um, we went out and talked to, let's say, like 100 folks in the biotech and big pharma space and talked about this technology. And um, basically we're talking about like, hey, if you have this technology, what would you want to use this for? And our assumption had always been that intracellular delivery would be the most interesting. But what we learned was that in the in the context of manufacturing a cell therapy, very important that every single cell is high quality. And if you have any unwanted cells in that cell therapy product, you're not going to want to transplant it into a patient. That could be a safety risk or it could be a less efficacious product. And so just being able to get rid of the unwanted cells is a really important problem to solve in cell therapy manufacturing. And it's interesting because from a laser perspective, it's also a much easier problem. It's way easier to use a laser to just kill a cell than to create a pore in a cell that's not too big to kill it, but still simultaneously big enough to allow cargo to diffuse into. So it's just interesting because, you know, like if you're doing your PhD in a laser physics program, you're not going to go for the easiest, simplest thing because you're going to go for the thing that seems a little bit more, you know, scientifically interesting or complex, but that's not necessarily where the need is in the market.

Ben Comer 12:10

Right. Okay. Well, um, did you you you've just said you were always focused on application. Um, did you know that you wanted to start a life sciences company though? Or um how did you, how did you come to that decision or or where did that, I guess, opportunity first first spark with you?

Marinna Madrid, Ph.D. 12:27

Yeah, I would say definitely did not know that. Um, it was never the career path that I would have, you know, initially chosen. I we were focused on applications for the work we were doing because we wanted the work we were doing to be important and have impact. And that was always kind of even just how we talked about it. For our qualifying exam, the way we set up that slide deck was first talking about like, what's the problem that we're trying to solve and how does the work that we're doing fit into that solution development space. Um, but that wasn't what I was focused on for my career. I did want to be a community college teacher and teach physics. But what was interesting about Harvard and just this Boston area in general, it's so collaborative. We were in the physics department, but we were literally across the street from the Harvard Stem Cell Institute. We were collaborating with some biologists who are um, you know, very, very entrepreneurial. So George Church is known for being very entrepreneurial. Derek Rossi is known for being very entrepreneurial, uh, founder of Moderna. And so we were, we were collaborating with these biologists who encouraged us to consider commercializing the technology. And that's really what pushed us. And it's not something that Nabiha or I would have considered before. But Nabiha asked me, she was like, Hey, would you be interested in trying a startup? I wouldn't want to do it without you. And I agreed. And at the time I felt very low stakes. I was like, yeah, let's do it. If we don't like it, we can quit after a year. Like it'll be a good experiment to run. And then a year in, and every year since then, there's just been so much momentum that it's kept going. But I don't think we would have ever envisioned it getting to this place. This was um, yeah, not anything that we would have like anticipated for our careers at the beginning of this.

Ben Comer 14:03

And the company was actually Cellino was actually founded in 2017, is that right?

Marinna Madrid, Ph.D. 14:08

Incorporated in 2017, yeah.

Ben Comer 14:10

Incorporated, okay. Okay. Um, I'm curious about, you know, given uh what you just described about coming to, you know, this idea of forming a company, what would you say were some of the biggest surprises about that process uh and and just during the first couple of years of the organization, uh coming from, you know, a less exposure to the the kind of entrepreneurial field, more focused on the science, you know, what what did you what surprised you? What did you learn in those first couple of years?

Marinna Madrid, Ph.D. 14:39

Yeah, you know, I always tell people that um doing a startup out of a PhD is a little bit like doing things backwards. Like maybe the right forwards way to build a company would be to identify a problem and then figure out the best solution to that problem. When you're coming out of a PhD and you're connected to a certain type of technology, it's a little bit backwards. You a little bit have a solution that you're trying to backfit to the right problem and you're trying to validate market demand. And so that's like a very interesting way of going about things. And so customer discovery and market research become really, really important. That's just not something you ever really do in your PhD. Um, the other piece is um that like ultimately the most important thing about a company to ensure that a company is successful is the quality of the team. And that's not necessarily something that you have control over in your PhD lab, you know, who you're working with. Like you're working with your lab mates. Um, but when you're building a company, that becomes the most important thing. And hiring initially is very difficult because you're not well established, you're not well known. Now it's much easier for us to hire. But in the early days, that was a that was like a difficult thing to do, and it was a very new thing for us. Um, so those are all very new. And then the the other, and this is like so long ago now, we don't, you know, this isn't something that it even occurs to us now. But when I'm talking to people who are like very first building a startup out of their PhD, what I tend to notice is that their pitch decks and the way they talk about the work is all very focused on the science and how the science works and how the technology works. And that's like an important part of the story, but a relatively small part of the story, the much, much more important, much bigger part of the story is what problem are you solving and what is the demand for that solution?

Ben Comer 16:19

Right. Yeah. And um, you talked about kind of being almost a hammer looking for a nail coming out of uh of you know your doctorate program, your familiarity with this this very specific technology, linking that to a real problem out in the world that patients have that you could potentially treat. So did you, when you actually started up uh Cellino, was there a specific product at that point, you know, that you were already working to develop? Or did you have to have all you know all of the conversations that you've talked about, uh do the market research, et cetera, to then land on, you know, where you thought you could make a difference?

Marinna Madrid, Ph.D. 16:59

Yeah, we we definitely had to do a ton of research and a ton of talking to everyone in the industry to land where we landed. So our focus now is on autologous-induced pluripotent stem cells. And that's something that Nabiha and I are um super, super passionate about. Just we want to enable a world where everyone has access to replacement cells, tissues, organs, limbs on demand. And autologous-induced pluripotent stem cells is a way to enable that without requiring immune suppression, without requiring donor matching. Um, so like truly the path towards accessible regenerative medicine. But that is not something that we were really thinking about or focused on in that way in the first year of Cellino. We knew about IPSCs and we are working with IPSCs, but it hadn't become so clear to us that autologous IPSCs was what we were going to focus on until honestly, at least two, maybe even three years in.

Why Autologous iPSCs Stay Artisanal

Ben Comer 17:52

Okay. Yeah. All right. I wanna, we're gonna uh talk about manufacturing um in a little bit, but I I wanted to just maybe set the table with this question. You know, why are induced pluripotent stem cells so challenging to manufacture at scale? It uh it seems like it's almost an artisanal process, particularly when you're talking about autologous, you know, personalized cell therapies. Um, why is that?

Marinna Madrid, Ph.D. 18:18

Yeah, so that's that's a great way of describing it. It's very much an artisanal process. And and IPSC technology has been around for 20 years now. Like this is the 20th year anniversary of discovering IPSCs. Um, and it's still a very difficult process. They're very sensitive cells. Generating them takes a long time. You're starting with adult cells, maybe fibroblasts, maybe blood cells, and then you're delivering some reprogramming factors to epigenetically reset them to a younger age. And they're very finicky during that entire time. And so what you end up getting is you end up getting these like stem cell scientists who are very, very talented and have magic eyes and magic hands, and they just always produce the highest quality cells. And you might have a team of 10 stem cell scientists, and all 10 scientists might have a decade of experience, but there's still one person that just happens to have magic eyes and magic hands. And that's okay for the early days of IPSC research. You know, that's okay for a PhD program, that's okay even for early stages of clinical trials, where maybe you're only dosing five to seven patients. You can have a relatively small team of IPSC scientists that are literally just working around the clock and going in seven days a week and managing these cells, but it obviously doesn't scale. Um, and so what we've done is we've looked at the parts of that manufacturing process that are difficult to scale and figured out where we can use technology to scale this. So, where you have a really talented stem cell scientist looking at these cells by eye and thinking, based on their own judgment and expertise, these cells look good, these cells don't look good. Instead, we have AI-based image analysis so that we can automatically image cells and then characterize the cells in process and characterize them in a very consistent way. Um, and instead of a um stem cell scientist that has just like very fine motor control and is very precise with their hand movements and can very carefully scrape away the unwanted cells without harming the cells that they want to keep. Instead of doing that, we use the laser mediated bubbles to lift off and remove unwanted cells. So we're essentially taking the parts of that process that are very manual, that are very artisanal, and dependent on like very deep talent and expertise and automating it. And we still have stem cell scientists in the loop. They're a very, very important part of the process, but they're not going into the lab every day and carrying out these processes by hand anymore.

Ben Comer 20:34

When did the uh the AI technology become available that was able to, you know, to actually do this?

Marinna Madrid, Ph.D. 20:42

So it's been available. Some version of AI-based image analysis has been available the entire time that Cellino has been around. You know, it's like a really cool question, actually. It's a really interesting question that you ask because everyone has heard about AI kind of exploding in the past couple of years. It's not that there were huge advancements or developments in AI models or algorithms in the past few years. Like these models and algorithms have existed. Um, what has changed in the past few years, at least in the case of the LLMs, is access to like a massive training data set. You know, with LLMs specifically and natural language models, every day on the internet, like trillions and trillions of words of pieces of data are being generated. And so there's been this huge, huge, huge explosion of like generative AI models that are being able to be trained on this data. But the type of models that are used for AI-based image analysis, those have been around. The Broad Institute has done a lot of work here. Um, we are applying them specifically to IPSC manufacturing, but they have been around. They haven't been introduced into manufacturing processes for cell therapies. And that's kind of a shame because I think they go a long way in making the cell therapy manufacturing more scalable and more consistent.

Nebula Closed Cassettes For Scale

Ben Comer 21:54

Yeah. Well, um, Cellino is gonna potentially uh prove that. Um, can you and uh I want To just get a maybe a high-level uh description of the nebula platform technology, which is what Cellino is using for manufacturing, I believe.

Marinna Madrid, Ph.D. 22:09

Yeah, so we are building Nebula, which is um based on a closed cassette. So it's essentially a closed manufacturing system. The reason that's important, if you're doing itologous manufacturing, you need to be able to manufacture for many, many patients in parallel. Otherwise, it would not be scalable. Um, but you want to minimize cross-contamination. So you don't want to be openly processing multiple patient samples in the same facility or the same space because you could have cross-contamination between the samples. So having a closed manufacturing process is really important. And the way that we're processing these cells, using AI-based image analysis to characterize, using a laser to manipulate cells, those are both compatible with closed manufacturing. Um there's the optical bioprocess, which is how we refer to the cycle of imaging, AI-based image analysis, laser-based cell manipulation. And that optical bioprocess is compatible with a range of manufacturing formats. You could use open wall plates, you could use a closed system, but Nebula is the closed system that we're building because that is what's necessary to parallelize clinical manufacturing. Um and so that is funded by our ARPA H grant. So we earned a government grant to fund that work. And that is something that is, yeah, very important ongoing development work at Cellino.

Ben Comer 23:19

And that is uh actually a mobile unit that could be moved around from place to place, or am I wrong about that?

Marinna Madrid, Ph.D. 23:25

That's the idea. Right now they are being built at Cellino. Um, but the idea is to place one on site at Mass General Hospital to support cell therapy manufacturing work that's happening there. So you could have um manufacturing being done, point of care, like down the hallway from where the surgical transplantation is taking place.

Ben Comer 23:46

Yeah, which seems like the real end goal for autologous cell therapy writ large, because then you you know completely eliminate the the logistic, the logistical aspect, which I think has held that therap those therapies back uh in the past, which is sending them out, having the cells engineered, sending them back just in time, you know, uh making sure the cells are healthy, everything stays intact. Whereas if you can do it, you know, right alongside in the hospital, uh, that that seems like an ideal situation. But is it is it possible to to scale that in the way that you want to scale it? And is it would the manufacturing be decentralized, you know, across the nation where you would have essentially these cassettes at at every available care center where they're treating these patients, or do you get to a centralized manufacturing um where you're you're you are still maybe handling and sending and receiving cells? Um how how would how would that look?

Marinna Madrid, Ph.D. 24:44

So I think it's exactly what you're saying, which is like you can go for a decentralized manufacturing model and have manufacturing be done on site, basically across the world. And I think that's really important for um making sure that the therapy is truly scalable and geographically accessible for a range of patients. So you can imagine having these nebulas being run in little clean room trailers near a very rural hospital. You know, it's um we don't want these therapies to only be be available on the coasts, for example, or in major cities.

Ben Comer 25:13

Right. Would you start out potentially with, you know, like academic medical centers and then expand out into community practices and or or I guess follow wherever the patients are that are getting this this treatment? Do you have you thought about that yet?

Marinna Madrid, Ph.D. 25:28

Yeah, I mean, that's the idea. You know, we're starting with MGH. That's one of our first collaborations that we've publicly announced. And they are they're a great partner. Um, and they're also like down the street from us. So it's very convenient. So it's a really ideal first sight. But um the idea would be to also be able to expand to more rural areas as well.

Ben Comer 25:47

Right. So the the idea with with this program um would be to kind of create a case study to show uh other care delivery centers that, you know, this is what you could potentially do.

Karis Bio And Partner Model

Marinna Madrid, Ph.D. 25:58

Exactly. Yeah, yeah. This can be copy-pasted to other parts of the world.

Ben Comer 26:02

Um, Cellino announced a partnership with Karis Bio a little over a year ago. I wonder if you could talk about that partnership a little, maybe in terms of how it came together and um and what role each company plays.

Marinna Madrid, Ph.D. 26:16

Yeah. So Karis Bio um is developing this very amazing autologous IPSC-based cell therapy for peripheral artery disease. Um so they have been, they started in Korea and they've also been operating in the US as well. And it's a disease that affects, um, should look this up, but I want to say around 12 million patients in the US.

Ben Comer 26:39

Yeah, it's a large population for sure.

Marinna Madrid, Ph.D. 26:41

It's a large, large patient population. Um, and so they are also interested in the these, you know, scalably manufactured IPSCs because they do want to go forward with an autologous version of that program.

Ben Comer 26:54

And so what uh I guess what piece of the therapy is Karis supplying, and then what is is Cellino supplying?

Marinna Madrid, Ph.D. 27:02

So right now, Karis is developing the entire therapy. We're in the early stages of collaborating with them, but the idea would be to leverage Cellino's IPSC manufacturing technology.

Ben Comer 27:12

Oh, okay. So you would essentially be like the manufacturing partner for this therapy that they did they discover that, do you know, uh at Karis? Yeah.

Marinna Madrid, Ph.D. 27:20

Yeah, yeah. And they've been in this space for quite a while. And a lot of the cell therapy developers, so a lot of the cell therapy developers that are out there are experts in a specific differentiation process or a specific differentiated cell type. And then the IPSC generation for many of them is something that they've kind of had to learn to be able to um develop this as an autologous therapy. But yeah.

Ben Comer 27:43

Got it. Okay. Yeah. And uh I think, you know, one of the really interesting and incredible things about pluripotent stem cells is that they can transform into essentially any cell type in the body. I wanted to ask you about indication. I mean, you're with this partnership with Karis, your your focus on cardiovascular disease, PAD, you mentioned uh as maybe a kind of lead indication at Karis, but what do you see as the potential for um the therapies that that Cellino could help create uh across therapeutic areas? You know, what what's exciting or or what what areas are you thinking about?

Marinna Madrid, Ph.D. 28:19

So it's really any cell type in the human body. So we actually um, you know, my co-founder generated this Claude model that has like kind of scraped all of the literature. And there are differentiation processes from IPSCs for or over 200 cell types in the human body right now. And that's like almost as many cell types as there are in the human body. So, really, any disease that is characterized by a loss of cells or a loss of functionality of cells, you could imagine coming up with a treatment for based on auto IPSC-based therapy. All of the early therapies were pretty focused on CNS. So, um, like dopaminergic neurons for Parkinson's, retinal pigma epithelial cells for age-related macular degeneration. And that's mostly for historical reasons, not because, you know, they're very important diseases, but it wasn't that people thought, okay, these are the most important diseases to treat with IPSCs. It's that these were some of the first differentiation protocols to be developed with IPCs.

Ben Comer 29:12

Oh, that's I was wondering about that. Well, you know, given that they can change into any cell type, you know, why do you see these kind of repeating uh uh target areas? Yeah.

Marinna Madrid, Ph.D. 29:20

Yeah, it's it's very much it's kind of historical. It's um, it's that these differentiation processes were some of the first to be developed. So retinal pigment epithelial cells, you know, the differentiation process is quite long. I don't want to say it's easy, but one of the first differentiation processes for retinal pigment epithelial cells were discovered because the scientists literally left IPSCs in the back of their incubator for too long over Christmas break and came back and like made into RPEs. Like that is not something that you would see for beta cells, for example. Beta cells are like very, very difficult to differentiate. Um, but so these therapies, uh, you see like more prevalent therapies in the retinal space and in the dopaminergic neuron space because the differentiation protocols were developed earlier. But also, what I'll say, what's very nice about those therapies is that they require relatively small therapeutic doses. Um, in the case of RPEs, a therapeutic dose might be tens of thousands or maybe low hundreds of thousands of cells. In the case of dopaminergic neurons, a therapeutic dose might be ones or low tens of millions of dopa neurons. Um, so that's in contrast to, you know, a cardiovascular treatment, which might require many, many more cells, or a muscular dystrophy, which might require billions of cells if you're trying to replace lost muscle. Um, so that's why those programs have advanced so far. Also, a lot of the a lot of the early work is in the allogeneic space because autologous is just so much more difficult to manufacture for. And these are all therapies in relatively immune-privileged sites. So that's not to say that you don't need immune suppression. The patients still do need to go under immune suppression, and it is a serious risk. There has been a patient death due to an opportunistic infection that arose during an immune suppression course for one of these therapies. But um, but it is more immune-privileged than other sites of the body. And so um typically the immune suppression that they undergo is a short course.

Ben Comer 31:17

Do you think the immune issue with allogeneic cells is something that um the you know the industry will overcome in in the next several years? Uh I that's I remember, you know, when I first started covering cell therapies um probably eight or 10 years ago, there was a real hope that allogen, allogeneic was seen as kind of the future. And if we can get this off the shelf, cell therapies without, you know, donors, without having to take cells from individual patients. But I think the the immune suppression piece is, you know, um really, really makes it more difficult. Uh, but do you do you think that that, do you think that uh that that that's an issue? Like, do you think you'll be working on allogeneic or or autologous cells, you know, like I don't know, let's say five or ten years from now? What do you think?

Marinna Madrid, Ph.D. 32:04

Yeah, I think it's, you know, I think there are some there are some subtleties. I think aloe and auto will continue to coexist. I think it's widely accepted that auto is what would be safest and best for the patient in most cases. Right now for sure, yeah. Yeah, especially for chronic degenerative diseases, and that if you could manufacture auto in a scalable and cost-effective manner, that's what would be best for the patient. The challenge is just manufacturing and cost and scale. Um, but I think, you know, the scenarios where aloe might win out, even if um manufacturing cost and scale were equivalent, is when you need a really, really fast therapeutic dose. So if you're treating so for chronic degenerative diseases, you typically have time. You can manufacture autologous dose, but let's say something like cancer, if you're developing IPSC derived NK cells to treat cancer, then speed becomes more important and NK cells are actually like it's just kind of a weird thing of the biology. MK cells actually work best if it's um if they're allogeneic. So I do think there's always going to be case, some cases where allo is better and some cases where auto is better. But for chronic degenerative diseases, auto would be safest for the patient.

Ben Comer 33:11

Um just one last question about the the Karis Bio partnership. Is that something that you that that Cellino hopes to replicate with other companies?

Marinna Madrid, Ph.D. 33:21

Um Yeah, we have other collaborations um um that have been announced. So uh Matricelf uh is a company that's developing an autologous IPSC derived um therapy for spinal cord injury. And so that's a company that we're collaborating with as well. It's essentially there are a lot of these companies that are looking for a solution to scalably manufacture IPSCs, and and that's who we're working with.

Ben Comer 33:48

Yeah, I'm my my colleague who uh is the chief editor at Cell Engine uh writes and talks a lot about the manufacturing whole, you know, being the kind of bottleneck for cell therapies becoming more more widely used. So that that's exactly what it sounds like you're trying to address. Um will Cellino develop its own internal development pipeline, or do you see the company more as a a partner, um a manufacturing partner?

Marinna Madrid, Ph.D. 34:17

Right now we're trying to partner with the cell therapy developers that are doing this very important work. Um, I think, you know, we really want to make sure that the manufacturing technology uh can be used to enable an entire industry. And so if we were focused on just commercializing one specific therapeutic application, I think that would be less likely to happen. So we're really focused on on building the foundations to enable the entire auto-IPSC cell therapy industry.

Ben Comer 34:46

Yeah. Do you have a sense uh, and I I'm I'm pretty sure the FDA has never uh approved uh an induced pluripotent stem cell therapy. Um correct me if I'm wrong on that, but do you do you have a I guess a ballpark time frame of when you expect uh such a therapy to enter human clinical trials? Maybe there's there may be clinical trial. Actually, I know there's uh a couple of clinical trial programs um ongoing, but maybe a timing for your manufactured cells uh entering human clinical trials and then how far out we are from, and I'm asking you to gaze into a crystal ball here. I apologize, but um, maybe your sense of when the FDA might approve the first uh IPCSC therapy.

Marinna Madrid, Ph.D. 35:32

Yeah, okay. This is um so you're right, the the U.S. has not approved any IPSC-based therapies yet. There aren't any IPSC-based therapies on the market in the U.S. Japan has approved two. So one for heart failure and one for Parkinson's. And that was very recent, and they were conditionally approved. Japan has these accelerated pathways for regenerative medicine. The U.S. um doesn't have pathways like this for these specific types of therapies. And so the IPSC-derived cell therapy development is a little farther behind in the U.S. But there are many active ongoing clinical trials for both auto and allo IPSC derived cell therapies in the U.S. And so I definitely expect that within the next decade we'll see some of the first approvals. Um, even, you know, there are some programs that are starting to enter their pivotal trial. So even within the next five years. Yeah. For Cellino, we're targeting clinical manufacturing within the next couple of years.

Ben Comer 36:25

Okay. Do you do you know by chance what uh what companies those were in Japan that got those approvals?

Marinna Madrid, Ph.D. 36:30

Yeah, Sumitomo. Um, Sumitomo and Okay, the other one is QORIPS, C-U-O-R-I-P-S. So Sumitomo developed an allogeneic IPSC derived dopaminergic neuron cell therapy for Parkinson's, and that uh received conditional approval from Japan's regulatory body. And so that's super exciting. That's on the market now. Patients can access that therapy. And then the other therapy is um an auto-IPSC derived cardiovascular tissue therapy for heart failure.

Ben Comer 37:02

Oh, okay. Interesting. Yeah. Yeah, I hadn't heard about that. I'm gonna have to, I'm gonna have to look those up. Um, but we're on the subject of regulatory, which is an area I wanted to ask you about. You became chief product uh and regulatory officer this year. Um, how did you, I guess, get up to speed on the various regulations that govern uh this particular space and the work that you're doing in Cellino? I mean, some of this I think is is uh you know, white space. It's brand new. And some, I'm sure some aspects of the cassette you're developing are are new. How do you how do you navigate the regulatory environment um and just you know understand uh what what you need to understand to to progress the the manufacturing?

Marinna Madrid, Ph.D. 37:47

Yeah, you know, my learning process for every new field is pretty similar, which is just read as much as I possibly can and then talk to as many people as I possibly can. And like that's how I got up to speed on the IPSC industry, that's how I got up to speed on the regulations. Um, and then, you know, of course, doing is very effective for learning. And so I led the um I led Cellino's advanced manufacturing technology designation submission. So there was a lot of like regular regulatory learning that came in that process. I'm on a number of IPSC committees. So I'm on an um, I'm on the Japanese Society for Regenerative Medicine and International Society for Cell Therapy's joint IPSC committee. I'm also on the Alliance for Regenerative Medicine Cell Therapy Advisory Group. And so these are groups that meet on a fairly regular basis, every couple months, maybe once a quarter. And it's a lot of the top experts in the IPSC field that are actively developing therapies. So there's also a lot of learning that comes from engaging with them and, you know, just talking to the experts. I would say I also very heavily leverage AI now for my learning process. So Claude is very helpful because I can use Claude to kind of um ask questions about like what are all of the relevant regulations for this body of work that we're doing. Um, and I would say the FDA has been surprisingly collaborative and surprisingly open. I mean, maybe I shouldn't say surprisingly, they've been like very pleasantly collaborative and very pleasantly open. And there are a number of guidance documents out there that are very useful for learning this space because you're right, a lot of these regulations aren't set in stone yet. There's a lot of white space for the IPSC field, but also for things like using AI and cell therapy manufacturing. So there was a very useful guidance that the FDA put out on how they intend to regulate AI in the context of cell therapy manufacturing. And that just came out in, I want to say January of 2025. And so, you know, like I interacted with that guidance, provided comments on it. Since then, I've engaged with different regulatory bodies around how to how to regulate AI in the context of cell therapy manufacturing. So all of the learning has been a very active process of like reading and engaging and talking to people.

Ben Comer 39:55

Yeah, I wanted to ask you about that if you were actually advising uh regulatory bodies at the FDA or globally. Are you and maybe as part of the groups that you've just mentioned or or uh you know as a as an independent uh party? I mean, are you are you speaking with global regulators about this field as it as it continues to progress?

Marinna Madrid, Ph.D. 40:15

Absolutely, yeah. Um, and through both mechanisms, partially through some of these groups that have, you know, regular meetings with different regulatory agencies around the world, and then partially just via Cellino. And what's really cool is that a lot of the regulatory bodies are they're all aligned on wanting to regulate AI in a risk-based manner. Um, and they're all very interested in learning and understanding what all the different types of use cases are right now. So they're very open to engaging with folks like us who are using AI so they can kind of understand how we're approaching it, how we're training the algorithms, how we're assessing performance of the algorithms.

AMT Designation And FDA Access

Ben Comer 40:51

Yeah, well, that's very positive to hear. I'm I'm glad to hear you say that. I I did want to ask a follow-up on the advanced manufacturing technology designation. That was something that that you led. Can you um maybe just talk about that process a little bit, uh, how it happened and and the the key benefits of that designation?

Marinna Madrid, Ph.D. 41:12

Yeah, so the advanced manufacturing technology designation, it's still quite new. When we applied for it, the final guidance hadn't even come out yet. But we knew about it ahead of time. We knew about it even before the initial draft guidance came out because of some of these committees and um groups that I'm a part of. So that was really valuable for us to be able to learn about that. And it's it's been a really important designation and kind of eye-opening because when I first started, when we first built Cellino, what we consistently heard was that the FDA cares about safety and efficacy and nothing else. They don't care about manufacturing scale. They don't care about cost. That's what we were told initially. And the fact that they put out this advanced manufacturing technology designation program shows that they actually do really care about manufacturing scale and cost and patient access. And I think it also speaks to how big of a challenge manufacturing and access has been in the self-therapy space, that even the regulators are kind of taking it upon themselves to try to solve this problem in collaboration with manufacturers and self-therapy developers. So very meaningful. Um, and so we, you know, we uh submitted um our application for the advanced manufacturing technology designation, and that was a really great process. The FDA is, you know, they're very smart, very collaborative. They asked great questions, and it was um a much smoother, more efficient process than I would have guessed it would be. And the advantage of having that designation is that any man any therapy that gets manufactured using the technology that has that designation. So it doesn't have to be like a Cellino-owned therapy, a cell therapy developer who is So Karis Bio would would benefit from this. Exactly, exactly. Karis Bio, any of our cell therapy developer partners would benefit from this. What it gives you is earlier and more frequent access with the FDA to be able to collaborate to figure out how to bring this therapy to market faster. So it accelerates the um development timelines and the commercialization timelines.

Ben Comer 43:11

Um as everyone at this point knows, um, FDA Commissioner Martin Makary is leaving the agency. Um CBER has has yet to appoint uh a new director. Uh, the I I have a sense that there's gonna be acting directors uh both at the very top and at CBER, probably for some time, just given the the way that the Congress functions and with midterms coming and number of other reasons. Um, but I I am curious, uh, and you've had you've just talked about it having a very positive experience in terms of engaging uh with the FDA. Um uh if you were in charge though, uh Marinna, you know what what kind of person would you like to see come into the those top roles, maybe in terms of past experience and and skills? What do you think about that?

Marinna Madrid, Ph.D. 43:57

Oh, that's a really interesting question. Yeah, you know, I I would say. There's been a lot of change at the FDA in the past couple of years, and we've been just really pleasantly surprised by how communicative they've remained with us throughout all of it. So that has been just folks that we're in contact with are, you know, still very responsive and still very much present. Um, I think what we need for the cell therapy field to really be able to advance is um a willingness to be flexible on establishing more fit-for-purpose regulatory frameworks. You know, the way that these therapies are being developed is not the same way that small molecules were developed. The regulatory frameworks might not be the same fit. Um, and also for some of the diseases that are addressed by regenerative medicine. So, for example, rare diseases, um, a flexibility in clinical trial design is going to become very important too. So, I'm on the uh Alliance for Regenerative Medicine Cell Therapy Advisory Committee. And this is some of the work that they do in engaging with the FDA, is like collaboratively working with the FDA to try to understand how do we advance and shape regulatory frameworks so that they are fit for purpose for some of these newer therapeutic modalities that no one could have ever envisioned before when these regulatory frameworks were first getting established.

Ben Comer 45:14

Yeah, and uh we don't know, like we we just got this this news about um um Commissioner Makary leaving yesterday, but my I I I believe that a lot of the programs, for example, you know, the the clinical trial real-time monitoring program, for example, are are likely to stick around just based on um, you know, what I know about previous uh uh commissioners' administrations and the way that the FDA works. Um, so I mean, maybe I'll just use that one as an example. Is that is that something that would potentially benefit the therapies that that you're contributing to or developing? Um, or are there any other programs that that are are new that you like or that that you would like to see?

Marinna Madrid, Ph.D. 45:55

Yeah, I mean, I think that one's very helpful. The RMAT designation. We we see that benefit a lot of the therapies that are in our field, um, breakthrough designation as well. I think RMAT and breakthrough are two of the two of the designations that I see a lot of the cell therapy developers in our field going for.

Near-Term Priorities Funding And Advice

Ben Comer 46:12

Excellent. Um, I want to ask you about your uh your key priorities right now and future plans uh at Cellino. Um what uh what still needs to happen before uh Cellino's uh autologous uh IPSCs uh can enter human clinical trials? Uh are there any kind of final issues or or or challenges uh that that you're focused on right now to make that happen?

Marinna Madrid, Ph.D. 46:38

Yeah, we are in process development and engineering and building mode. So we're in building nebula mode. We are in the process of like optimizing our IPSC process so that it is fit for clinical manufacturing. We previously developed um an optical bioprocess to generate research grade IPSCs, and those are the IPSCs that we've shipped around the world to cell therapy developer partners, and those have been successfully differentiated to a number of different cell types: random pigment epithelial cells, spinal cord tissue, cortical neurons, um, um, dopaminergic neurons, but we need to transition that process so that it's appropriate for the clinic. And that's the bulk of the work that we're doing right now.

Ben Comer 47:19

Does uh Cellino have enough uh funding capital at this point, do you think, to get to uh human clinical trials? Are you gonna have to do some fundraising uh during that process? I mean, most biotechs are are kind of in a constant fundraising mode, but so maybe it's not as a good question. But um, I I guess, you know, I'm I'm curious about, you know, the resources that you have at your disposal now, you know, compared with what you're trying to do.

Marinna Madrid, Ph.D. 47:47

Yeah, it's a good question because I do think a lot of biotechs are in constant fundraising mode. We're actually not actively fundraising right now. Um, the, you know, it's no secret to anyone that the biotech industry has kind of been through the ringer. It's been a past couple of years. We've been very grateful to be uh well resourced these past couple of years.

Ben Comer 48:09

Is that primarily through uh, you know, series that through through fundraising or through the partnerships or combination of both?

Marinna Madrid, Ph.D. 48:17

We have the ARPA H grant. Yeah.

Ben Comer 48:19

The ARPA H grant, yeah. I I didn't ask, is that 25 million? Is that what that grant was? Yeah. Yes, yes. Awesome. Okay, what about key priorities for you right now, Marinna? You know, what what are you most focused on kind of in the short term, like for the rest of 2026?

Marinna Madrid, Ph.D. 48:36

For the rest of 2026, advancing this um MGH partnership, that's for us really important to get clinical validation of the nebula platform, um, supporting the team in terms of advancing bioprocess development. I think that's something that is really critical and underlying underlies all of our business goals. Um, and essentially getting Cellino to the point where, you know, we continue to build this strong relationship with the FDA because we have the AMT designation, but that's definitely not where our regulatory strategy stops. So there's more that we need to do to be um manufacturing clinical grade IPSCs to support all of these different cell therapy developers.

Ben Comer 49:18

You've uh you've touched on this already a little bit at the beginning of our conversation, but I'm curious about what advice you might give to other researchers who are thinking about starting up a biotech company, maybe coming out of a PhD program.

Marinna Madrid, Ph.D. 49:33

Yeah, I mean, I think the biggest piece of advice is to really understand the problem that you're trying to solve and do the market demand, do the market research to fully understand that. Um, understand the voice of the customer. And then the other advice I would give, you know, I think when you're when you're a scientist and you haven't really been trained in all of these other entrepreneurial skills, it's easy to assume that you wouldn't be able to do it. Um, but founder-led companies are very successful and I think very unique and special in that the founders do have a strong connection to the technology, and that's important for understanding um how to build the business.

Ben Comer 50:09

Well, uh Marinna, thank you so much for coming on the show. I really appreciate it.

Marinna Madrid, Ph.D. 50:13

Yeah, no problem. Thank you so much for the invite.

Ben Comer 50:15

Yeah, we've been speaking with Marinna Madrid, PhD co-founder, board member, and chief product and regulatory officer at Cellino. I'm Ben Comer, and you've just listened to the Business of Biotech. Find us and subscribe anywhere you listen to podcasts, and be sure to check out our weekly video casts of these conversations every Monday under the Business of Biotech tab at life scienceleader.com. We'll see you next week, and thanks as always for listening.