Transcript Dustin Weaver Welcome to the Bullseye Breakdown, the podcast where CEOs, founders, and leaders of industry break down big decisions, big wins, and big lessons. No fluff, just real talk. Let's get into it. This is the Bullseye Breakdown. Welcome back, folks. Today, we're diving into a world where biotech, precision, and a whole lot of problem solving come together. I'm your host, Dustin Weaver, and with me is Brian Feth, the CEO and co-founder of Excel Biosciences, a company that's pushing the boundaries of cell therapy and AI-driven biotech. Brian's story is packed with real-world lessons, scaling a business in one of the toughest industries, turning groundbreaking science into real-world solutions, and building partnerships that move the needle. If you're an entrepreneur, founder, or just someone fascinated by the intersection of tech, medicine, and business, this episode is for you. We're talking biotech innovation, AI and drug development, and the challenges of scaling in a highly regulated industry, and what the future of medicine really looks like. No fluff, just sharp insights from someone who's in the trenches. So, let's take aim and let's get into it. Brian, welcome to the show. Brian Feth Thanks, Dustin. Appreciate you having me on. Dustin Weaver I appreciate you being here. I'm super excited to talk to you today. Doing the research and understanding kind of what Excel Biosciences is doing in the area of stem cell and T cell generation. It's amazing. You guys have such a lock on creating a process that is driving the way that we are treating diseases and cancers that are being affected in this role. Can you do me a favor? Can you just give us a bit of a background on what Excel Biosciences is and what you are doing in the market today? Brian Feth Yeah, we're a small piece of a much bigger picture. One of the things that I think people from outside the industry aren't necessarily aware of is that cancer care has really shifted from kind of this mindset from 50 years ago around using chemicals to try to just slow down the pace of cancer using radiation to try to mangle the DNA to slow down the pace of cancer. But over that last 50 years, it's shifted from kind of these broad sledgehammer-based approaches to more targeted approaches. A lot of those targeted approaches are just trying to identify one little protein on the surface of a cancer cell. It may not exist on all cancer cells, maybe just a little subdivision of a lung cancer or a colon cancer. And so we've been starting to chop off slivers of cancer where we can have an effect. And still, a lot of times, it can be a long delay, but the cancer can often come back. And we really over the last 10 years moved into a place where we're starting to think about the immune system and its capability to actually do a lot of the work for us. And that's something that's kind of new. I think people, and I certainly did when I was going through school, understood the immune system as really not being able to detect these cells because they often are so similar to your own cells. They're just, they are your cells. They just kind of break and start growing out of control. But what we've learned is that the immune system is actually really sensitive and can detect things that are awry in those cells and often clear it. And it's really whenever the cancer starts to cloak itself and figures out how to kind of evade that immune system that you get a problem and tumors start to form. And so, you know, there's been a class of drugs that's tried to interrupt that cloaking mechanism called checkpoint inhibitors. And they've had a lot of success. It's not effective in every cancer. It's, I think the stat is somewhere like 30% of the times it's used for the indications it's approved for it works. So, and when it does work, it's, I think it's, you know, pretty long-term duration. And that led to kind of this next class, which is where my company is focused, which is really thinking about using the tools that we have available to us now to actually modify the genomes of cells and to put a payload in, at least a signal for production of a payload. So you give it a certain signal to express something on its surface, and it kind of turns it into a heat-seeking missile. And that allows those cells to go after and find those cancer cells, those immune cells to go after and find those cancer cells. And it's been approved for a number of different indications within blood-based cancers like leukemia and lymphomas. And it's curing about half the patients. And cure is a very strong word, but we've got a lot of survivors that are 10 years plus out now from these therapies. And so usually in these therapies, almost everyone responds. It's like 80 or 90% get a response where the tumor clears and the patient feels good within two weeks. And then about half the time, it'll come back and the other half the time it doesn't. So my company is really focused on a big problem in this field, which is how do we get this to as many patients as we can? Because right now, a lot of the bottleneck is supply, supply side. There's a lot more patients that could be receiving this, and particularly because it's your cells, it's very safe, it's starting to move earlier lines of therapy. It's not kind of the last thing you do anymore. And so that is just broadening the base of people that can be using it. And, you know, it takes a lot of time to process this stuff right now. It can take weeks and weeks. Sometimes you lose a patient on that process, if pretty good it's the last line. You can also, it's just very difficult. It's kind of one process, one patient, and very manual. And so it's been challenging to get this to as many patients as needed. And it's very expensive. I think a lot of investors think about this as a cost issue. So our company focuses on getting tools to automate that workflow and really get these cells to be more potent. So you don't need as many of them. And potency is also really important for solid tumors, which, you know, it's been challenging to get the same strategy that has been effective in blood-based cancers to be effective in solid tumors. And there's been some hints of success over the past year, a couple of approvals in solid tumors. And really the key there is to just get those therapies, cell therapies more potent. So field is called immune cell therapy. Patients might've heard of CAR T therapy, which is kind of the main class. And so those are the, that's the nutshell of it. Dustin Weaver Let's move on to the section of the podcast. I like to call the pressure test. And this is where we talk about where are you at with your company, where you're moving and where you're headed down the road. Unfortunately, I know way more about this than I probably would have ever liked to. I lost my father almost 25 years ago to a form of cancer called multiple, excuse me. I lost my father almost 25 years ago to a form of cancer called multiple myeloma, which is in the classification of bloodborne cancers. And I know where your industry has been and to hear where you are now taking it is, I mean, I almost get emotional in thinking about it because knowing that if it was 25 years later, I may still be able to have my father who died in about three years from being diagnosed with the cancer from start to finish. And knowing that you're taking that out 10 years plus on a regular basis is amazing to me. It is something that, to your point, it's expensive. It's hard to replicate. And when you go into T-cell or stem cell transplants or therapy, you know, as it has transferred over the decades, it's very hard or it has been very hard on the patient. And I think that you're doing a lot to kind of adjust that and making it more palatable and a lot easier. To your point, it was, for my father, an end-of-line care. It was a last-ditch effort. to what you said previously, you're now moving this up because the results are so amazing that it's actually getting to a point where it's no longer the end goal or the last line of defense. It may be moving into second stage, right? And you're almost to the point where it might be first stage? Brian Feth Yeah. So multiple myeloma is one of those indications that has been really impacted by this field, this class of therapies. It has a BCMA, a mutation that can be targeted. There's a drug called Carvicti that's produced by Johnson & Johnson and a company called Legend. And it's been very successful. It's currently for relapsed or refractory multiple myeloma patients. So it has been later line. It got moved up to second line therapy this past, I think it's just this past year, last 12 months. And having attended conferences in this space over the last year, I think a lot of people think that we may be looking at a first line approval soon. And that That changes, that's huge for patients. It is a curative therapy for a lot of patients. The challenge has been that Legend has been making this stuff about 3,500 batches over the last year or two per year. They're talking about how to get to 10,000. That was part of the recent presentations and how they're thinking about strategy is really building more factories and more workflows. It's just one workflow, one patient. We need more workflows if we want to treat more patients. And their addressable population is like 100,000 or more right now. You know, 10,000 from 3,500 is a big deal. They're talking about tripling their capacity. But that's still one-tenth of what they could and should be able to treat if they had a workflow that was truly scalable. That's interesting, Dustin Weaver too, to note just in terms of what cancer research and cancer treatment is. You know, we talk – well, first of all, I'm going to say it. Cancer sucks. It's the worst. You know, it's affected almost every part of my family throughout my life. And, you know, it's not going anywhere. And so we talk about cancer, cancer research and, you know, curing cancer. You just identified a major problem in getting to that point. We might be able to find the cure for cancer, but the way to implement it to all the necessary people is so difficult and is such a scaling factor that it becomes almost impossible to reach. the amount of people that might be afforded the ability to have it or would be benefiting from it in the timeline that they can actually truly benefit from it before cancer takes its toll, correct? Brian Feth Yeah, that's right. I mean, you can think of it, so traditionally, small molecular chemicals, you know, you make it a VAT, you got huge economies of scale, and a lot of stuff is outsourced to other countries now because they just, they can make it super cheap. And they formulate it. Maybe they ship it overseas here as a raw ingredient. We formulate it, and we use it to treat a bunch of quality control around it. Which is kind of that next stage that's really evolved over the last 20, 30 years. Like antibodies, they figured out how to kind of scale and make mass production of that. It's not quite as scalable as chemicals, but you can get a lot of economies of scale. This field so far has really been challenged with economies of scale. It is a very manual process of one patient, one workflow. But it starts to look a little bit more like surgeries. A lot of people, if you can cut it out, that's your best chance of curing whatever your cancer is a lot of times. So that same process, we're not cutting it out. We're treating with your own immune cells. But those cells come out and they have to go back in and they need to be modified while they're out. And so it looks somewhat like surgery and somewhat like a process flow, like you think about for any process. Like how do you optimize your bank teller strategy? Where are the bottlenecks in that process? And how do you scale so that you don't have that bottleneck? and then you hit something else that's a rate-limiting step. And then there's also a lot of innovation. It's a new field. And a lot of the patients that were cured from treatments 10 years ago, those were very early technologies that were repurposed from other fields. Now we have kind of new best-in-class, and so you need to be able to kind of sub things in and out. And so it is a process-driven workflow with rate-limiting steps. And I think we're getting better. It's getting cheaper. We're creating automation at a lot of points that weren't there before. So I do feel like the field, you know, it's going to continue to be supply constrained for a while. But there's going to be a lot of patients over the next 10 years that are going to be cured that wouldn't have been last 10 years Dustin Weaver as a result. And that's by and large to what Excel Bioscience is doing with what your main goal is of limiting the amount of time it takes to create that batch of medicine and limiting or the batch of cells, excuse me, limiting the time and or the amount of those cells that you need to create. So thereby shrinking the process down to help to make the entirety of the process much faster and much more manageable on the large scale, correct? Brian Feth Yeah, that's right. You know, they're leading institutions that have been working on the University of Pennsylvania is kind of one of the main leaders in this. They kind of, you know, in some ways developed the early field and now they've been innovating on how to shrink it. We're a contributor to that. There's a couple of things you can do to get it to just a few million cells to be able to treat. a couple days instead of like, you know, a week or two. But there's other components. There's, you know, you need to do quality control, make sure that the batch isn't, you know, kind of contaminated with bacteria or mycoplasma. And those technologies need to catch up too, because right now you want to check for sterility, you got to let it grow. So that's like a week of just waiting to make sure your culture isn't contaminated. So there's a lot of contributions to getting that. But we, you know, we're proud to be, I think, a real key component of that, which is how do you metabolically shift these cells into a profile where they're much more effective and thereby you need fewer of them. And that allows you to have a timeframe for your manufacturing. And when that, remind that with the other elements, you know, you can kind of realize this goal of a short manufacturing workflow. And I think more importantly, you know, 90% of cancer is a solid tumor. You know, that's, we're kind of just entering this phase of cell therapy for solid tumors, like colorectal cancer, non-smoke cancer. There was approval this past year for a type of sarcoma. And then there was another approval for skin melanoma, which is obviously a big one. Yeah, that's a big location. And those were effective enough to get approved, but they're kind of curing like 5% of patients compared to like half. So there's a lot of work to do. And potency really matters. They're like treating with almost two orders of magnitude more cells than they are with these blood-based cancers. And that's because those cells just have a hard time getting the tumor. Once they're there, they have a hard time penetrating that really nasty environment that the tumor creates around it. And then they just don't have the right energy reserves to be able to actually kill effectively in this very hostile environment. So we're sending the cells to boot camp as part of it. Dustin Weaver Yeah. You're making super soldiers. Brian Feth That's right. Dustin Weaver And I mean, to put it into layman's terms, that's essentially what you're doing. You're taking these cells and you are amplifying them, turning them into superheroes, as it were, and thereby requiring less of them. One thing I want to make a note on, because I feel like you almost detracted to the fact that I can tell that it might be in your mind something that is hard to be working with right now, that you're not as effective as you are with tumor-based cancers as you are with blood-based. But coming from someone who has lost a parent to a blood-borne cancer, what I want our viewers to understand is prior to the implementation of T-cell stem cells and the advent that you guys are doing, there was no cancer to cut out. There was no clear path forward on how to attack that cancer and say, hey, we're going to give you an opportunity to have a successful remission or the ability to live another 10 years. So, you know, in looking at where you might not necessarily be, you know, front running and being as successful with how you are in bloodborne, I do not want to detract from the fact of what you have done with bloodborne cancers is unreal. It is amazing. And, you know, even though I lost my father to it, I've been a part of, of numerous charities that raise money for it in wholeheartedly, not as a podcast host, not as me, but just as a human being, I just, I want to take a second to say, thank you. Like it is, it is super amazing what you're doing with that. And I know that you're going to get there on the, on the actual tumor side of things. But to your point, it's a big glob of a fortress of cells that, you know, even if you're sending, I'm going to just carry this analogy forward, even if you're sending superheroes into the mix, you know, there's a lot of villains in there that they got to combat against. So I don't want to detract from what you're doing there and make sure that you understand that there are some people in the real world that appreciate what you're doing. Brian Feth Yeah. I mean, the real heroes of this, the patients that have been advocating for a broader awareness of this area, the Whitehead Foundation, Emily Whitehead Foundation has been a key one. they were just up on Capitol Hill talking about the need for continued funding of science. Carl June, Bruce Levine, Michelle Sadeline, these are some of the early pioneers that took a risk when everybody else thought they were crazy for modifying and putting these cells back in patients. And this was 12 years ago or so whenever the first patients were treated. And I think we're going to see some Nobel Prizes for these folks pretty soon. Steven Rosenberg is another one out of the NIH. He pioneered basically just mashing up solid, there's a solid tumor field I'm speaking about now, but mashing up the tumor, pulling the immune cells out, effectively waking them up. And they're familiar because they've been in the tumor and putting it back in the patient. That's also been frequently a curative approach. And he's been doing that for several decades. So I think the field is still early, but we're going to be hearing a lot more about it. And it's been tough. Investors got really excited about this, maybe kind of peaked three to five years ago. But given the challenges in scale, they just started shifting attention elsewhere. So I think we're in a bit of a lull in this field in terms of attention. But the more patients that are cured and the more people that are aware of this, the more that there's going to be attention back into the space. Dustin Weaver Well, let's get into the business of it. I just want to ask you how you got to where you are today. I know you got your undergrad at Purdue and went and got a master's at Berkeley, both in science fields. Had you always known that you wanted to get into some form of cancer research? Is this something that developed over time? Brian Feth Yeah, you know, I've always been interested. When I was early, I guess, late high school, early college, I lost three of my four grandparents to different cancers, and I was old enough to spend time with them and really grieve that loss. And while they were still living, the process of grieving to some extent starts, you just know that this is a process that's unlikely to go well. And I think I was really motivated coming into undergrad at Purdue to kind of focus on a meaningful impact to get an education. And science kind of quickly became the passion. And then I ended up doing a lot of research in cancer while I was at school. I managed to kind of like TA'd a class, I guess. It was called The Nature of Cancer. So I invested a good amount of time during that educational period of my life to that. And then started building a career, went into consulting, worked in finance. And I think I was always interested in being involved in this field. I ended up on the business side of things fairly quickly out of school and focused in strategy and finance. And so when I decided to go back and get my MBA, it was really, I'd already had a business kind of science-oriented degree at one of the Claremont schools called the Keck Graduate Institute of Applied Arts. Nice, okay. Brian Feth And that was right after undergrad. So I didn't really feel like I needed the MBA from a business skills practical standpoint. But I went back because I'd been kind of soul searching and thinking after consulting and kind of thinking about what I wanted to be doing next. I spent a year in Africa and really got to see entrepreneurship at its grassroots. You know, just finding opportunities to kind of make money and solve a problem. And it didn't always look like the venture capital funded businesses. You know, kind of can be very basic. Some of the people that we worked with would go home on the weekends and they would buy 10 gallons of rice and then break it up into buckets and sell it into their hometown. So, you know, you just saw that it's not as hard as it sounds and that, you know, you just need to do it. And so that was kind of the thought process coming into Berkeley was I'm really here to start a company. And I knew that Berkeley, you know, Berkeley chose specifically because of the wealth of technical talent. And I went back, I'd actually done a fellowship at Berkeley in my undergrad and went back and talked to that PI about the work that they'd been doing and started this process of kind of looking at all the different science that was happening and evaluating what labs I might want to pull something out of. And then I kind of, you know, the other, that was kind of the thoughtful way of like, how do I systematically find an idea that I want to launch? But you have the problem of, you know, is there a technical co-founder that that's going to come out of that lab as well? Dustin Weaver Right. Brian Feth And then the other side is founder dating effectively through entrepreneurship organizations where technical people are coming to explore what it means to start a company. And you know, I had some background working with angel groups, understood kind of the process of pitching and raising money. And so when I met my co-founder, it was through that process. It was, he had some ideas out of his research that he was thinking about might, might be useful as a business. He and I kind of spoke about what that might look like and where the opportunities might be. And then we, we ran it through one of the business school classes, which is a class called the Lean Launchpad. It's a famous business guru called Steve Blank, who kind of created this concept of kind of this lean process of building a company. And one of the big points of dogma is that startups are not just small versions of big companies. They really have to run this process of seeking out how to optimize your business model. And a lot of that comes through just getting out of the building and having conversations with people. And so we met, did like 100 phone calls with different researchers and doctors to try to hone our business model. And then at the end of that, we decided we would do it for real. And we called back a lot of those people that we had talked to that were very supportive of us as students and ended up getting a source of samples, be able to test out the technology, found a lab space and kind of built it out of that. Interesting. So, you know, I think I resonated with my co-founder because of the concept that he had and the focus of being focused on cancer. It was a different business than it is now and kind of a different concept then. Dustin Weaver Yeah, sure. Brian Feth But I knew it wasn't going to come from me. The concept wasn't going to come from me. The technology really needed an external tech inventor. It's interesting. Dustin Weaver So you went about this in a way that, you know, we talk a lot about bootstrapping. We talk a lot about, you know, building up into mid-cap space on here. And you went the route of investment and startup in a true sense of going and finding investment into a product that you had a vision for. But you still had to pivot that across. And I guess one of my questions there that is maybe more about those early days is when you had to make the iterations on the company and as you were building out, had you already received funding? And had those investors already understood that this is where we were going? And then when you decided, no, we're going to shift this a little bit, what's that conversation like with those investors to get them behind you to say, oh, yeah, this is a good idea? Brian Feth Yeah, it's a messy affair. So when we started the company, we did Bootstrap for really almost first two years. Very supportive partners at the time that helped kind of cover the bills. We did everything as cheap as we could, bought lab supplies off of eBay, mostly expired. And, you know, you just did your best kid. We hustled to get free patient samples, which a lot of times these places make you pay for, or at least there's some strings attached. So, you know, we hustled, did things as cheap as we could. And the original business concept was, we called it the technology avatar, which is still its name because the idea was it was a version of yourself in a petri dish. And what we wanted to do was the early technology was very much the same it is now. It was an environmental chamber that is hyperbaric and hypoxic and allows a really nice environment for cells to grow and thrive. And it does create an environment that those cells are much more in vivo-like. It's a low-oxygen environment. Tumors are particularly low-oxygen in how they thrive. And that's a different way of processing sugar that, you know, we talked about metabolic shift. For those that had that in school, it's basically how you use sugar to generate energy. If you have oxygen, you can make a lot more energy, a lot more ATP. But if you don't, the process looks different. And it's a much more kind of lean process in generating energy. But that's what you need to do whenever you're in these literally hypoxic tissues. So the original idea was take a patient's tumor and grow it and then be able to test drugs on it. So it was a diagnostic company. We started setting up kind of this diagnostics workflow, ClioLab. And then we started getting pharmaceutical companies saying, well, that's pretty neat. I would actually like to test my drugs on it. And so they started paying us. But, you know, the type of arrangement, like we thought it was customer was the patient, the doctor. And then pharmaceutical companies said, well, we'd like to use that. And then at some point they just said, this is working really well for us. Can we just buy the workflow from you? And we had actually built it to look nice. It had some reliability issues early on. And this was part of the mistakes you make as a startup. We thought we were hiring an engineering group to just kind of build a breadboard, like one of these. And then we accidentally, hindsight is ridiculous, but we hired an industrial design firm to do this. And so they made it look really pretty, but not very manufacturable. It looked great. And so it did feel like something that we were confident to actually sell. And the group that built it was willing to make more. So we ended up kind of selling some of these early instruments. And I think we're a research tools company now rather than a diagnostics company. Can we do both was kind of part of the thought process. And I actually built two different pitch decks. And we went out to two different investment groups and said, I think this is the business. High content, lots of data, better diagnostic decision making. And the other was, we're going to sell a lot of these boxes. And what we found was that we got really far on this big data concept. There's a lot of tech investors that like big data. and biological data is something that often has barriers around it. So this was kind of interesting to them, but couldn't get it over the finish line. And then on the other side, we did have some smaller investors that were willing to write smaller checks, but appropriate for that business model that basically said, I think there's something bigger here than like, I see a path. You guys are going to generate some revenue. You're going to get a lot of voice at customer. But if you stop here, I'll be disappointed. Like you really, there's a clinical opportunity here to do something more important for patients, not just make better research products for researchers. And we wholeheartedly agreed with that. That was a good match for us. And so we went and kind of started off building more of these boxes and really thinking about who are the key users and what's the problems they're solving. And that led us to really this group of researchers that were doing drug screening. And in particular, they cared a lot about these environments for the tumor because immune cells are dysfunctional there. They just, metabolically, they're not functioning well. And they get exhausted and they're repressed. And so there's a lot there at this time, a lot of researchers that were focused on cell therapy and about how to make it more active at the tumor site. So particularly I'm talking about solid tumors now. And and so that that, you know, we listened and we built another technology that was really focused on that specific application, just a generational version of the one we'd already built, you know, kind of allowed us to measure killing really efficiently. And then we found out something in one further step was that researchers were exploring how exposing these cells to conditions that would shift them metabolically, kind of prime them, would allow them to be more effective when they actually went in to do the fighting. And they saw some really outstanding results. Like these cells were just really effective and persistent in killing. And it was such a big difference that we're like, I don't think we can be an instrument like research company anymore. we need to make this, get this in doctor's hands, because if it's making these therapies that much more potent, that's going to just cure that many more patients. And so we decided we're just going to shift the whole company. We're going to stop any ancillary activities that are not focused on this idea of like, how do we get this into a system that is able to treat patients, which is not an easy task. It has to be in this very sterile environment. Everything has to be kind of closed. The materials have to be a certain type of material. And then on top of that, like there's all these like user requirements for the doctors. Like it can't be here. It has to be like this. And so we kind of had enough information to paint the picture for investors. And we went out shortly after the pandemic to investors and said, this is the data we've got. This is what we've been hearing. This is where we think the opportunity is. And we had talked to a number of strategics in space that were interested in this. And they said, if you build that, you know, we're going to buy you. And so investors love, you know, love hearing that. And voice of strategic, like if we build this, they will come. Dustin Weaver Right. Brian Feth So that led to our last round of financing and ultimately what we've built now, which is, and, you know, hired this amazing team that's really skilled at building these types of tools. Although the first of its kind in a lot of ways, like their collective skill sets kind of perfectly built for building this technology. And really in just a little over a year, we went from like napkin to fully functional GMP instrument, which is ridiculous in terms of speed. Dustin Weaver Was the fully functional GMP instrument created by an industrial designer? Brian Feth No, this time we did it right. Yeah. No, I, hopefully you learned from your mistakes. We actually talk Dustin Weaver about that a lot when it goes into, I have a manufacturing background too. And so, you know, understanding that, you know, when you're making one of things, it's easy to be pretty, but you have to figure out how to make 10,000 of that effectively or else things are not going to work right. So I laugh at when you said that because I knew exactly what you were saying. Brian Feth Yeah. And, you know, coming out of, we had gotten a lot of money from that round of financing, and I kind of built the budget. I just said, how many people do we need to hire to pull this off? We need, like, a couple of software people and systems integration engineer and double E. And, like, you know, I kind of built the model that way. It was going to be, like, a team of 40. And I knew you've got these cycles of product development, and, like, we're going to be slow at some point once we build this thing. You know, are we going to be able to? That was, and I don't, nobody wants to lay people off. And like, so you want to create a sustainable organization that's going to continue to be functional. I was kind of trying to figure this out. And then we hired some of the leadership first, hired our CTO and another individual who is now our VP of engineering. And both of them just said, look, we've spent our entire careers using outsourced engineering groups. We know how to do this really efficiently. Like we know the difference between industrial design and mechanical engineering firms. And they said, you know, and this kind of solves for the issue that you've got, which is like, what do I do with a team of 40 once the product's built? They said it's going to be a little more expensive, but on your timelines, it will get you there in the time that you're looking for. And so they said, the team we've got here already is really good. We're going to make a couple additions to it. We'll do all of the alpha testing, and then we'll hand them the designs that they need. We'll make those subsystem decisions, like it should be like this versus this. And then they're going to go through and iterate further on that sub, and we'll test those, and then they're going to lock it in. That's ultimately what we did. We did this intensive several hundred page bid document where we went through with all the requirements of the system that we'd found through, you know, first thing actually the team did when we hired him as voice a customer. Called a bunch of people and tried to understand what their user requirements are. Dustin Weaver Yeah, sure. Went Brian Feth through, put that in the product requirements as best we could, several hundred pages, and then sent it out to 12 groups and all 12 put bids in. It was an exciting project. A lot of them wanted to develop the experience in this area. And so we had our choice. We went and interviewed a bunch and picked ultimately the one we felt like was best equipped to do it. They had a really talented software group, hardware group, had some experience in GMP platforms, had built something upstream of our system already. So it gave us a lot of confidence. And so that's how we did it. Dustin Weaver I think that's awesome. And it's a really smart way of doing things. just from a CEO perspective of knowing that there's going to be an end date on where you're going to have all those people and not necessarily knowing that there are contract groups personally and not understanding that there's contract groups out there currently that would go in and filter that out. It makes complete sense. And it's so great because just from an efficiency standpoint, you don't have anybody worried about the deadline and trying to lag it out. They want to reach that finish line as quickly as possible because that's what they've been brought on to do. And so thereby you're pushing. And as you said, it may cost a little bit more because you're hiring an outside group. But long term, grand scheme of things, it's probably cheaper because, again, you don't have any of that uncertainty as you're coming to the project close of who's going to stay on and where it's going to go. And so you probably finished it in a magnitude of months prior than it would have been if you would have done it internally and move forward with it. I think it's genius. That's an amazing thought process. Brian Feth Yeah. You know, I got some bad advice. Talk to an investor around this time prior to hiring the team. And I just said, you know, how do I determine whether I should build a team or outsource? And so he introduced me to another hardware company. and the CEO was like really kind of there's a lot of them out there that are a little bit like you know the kind of jerks and he had a lot of swagger and said well we hire our own engineers I want the control, I would never trust anyone else to build something and it was said with confidence making me feel small because we had outsourced our last product and maybe that was his goal but it was advice, I took it and then I had another conversation with an engineer at a happy hour or a business plan competition or something who had lost his job at a company. I think it was a company called Interra. They built the workflow and then they all got laid off. And he was like, yeah, that's kind of normal. That happens a lot and it sucks. And just between those two things, once we had a couple of people that had kind of gone through this process, it was very obvious to not take the advice from Dustin Weaver the CEO. Yeah. By making that decision, even if you were to make that decision and decide a month into the process, that, hey, this is maybe the wrong decision and we do need to go internally. You can walk that back and get out of that. Of course, it's going to cost you a lot, but then you can go and hire internally. The reverse of that is darn near impossible to hire a bunch of people and then fire them all. Not only would it make you feel bad as a person, because I think a lot of people forget that as CEOs, we have feelings. We're real people. You know, it's not like we're just, well, apparently there's some that are just jerks. They're not invited on this. But, you know, like the reality is, is that, you know, we do everything that we do on a day-to-day basis, trying to make these decisions on what we think is the best path forward. And a lot of the times we, there may not be an answer, nor do we have somebody that we can go to and ask. Or in this scenario that you just described, you go and ask somebody and he's going to tell you something that maybe is not the information that you wanted to choose from. But I think that that is an A way of making a decision on a B decision. You know, and I talk about A decisions and B decisions, you know, things that you can walk back easily and things that you really have a harder time of walking back on. Brian Feth Yeah. And to be fair, you know, at the time we had a team of software engineers, maybe five or so. We were trying to figure out how to keep them and kind of use them as an adjunct to this team that we were using externally. In the end, they didn't have the right skill sets. That's the other issue. When you're outsourcing some of this stuff, they can kind of pull people, particularly if it's a larger organization, to have that exact type of electrical engineering experience you need for this. If you're using somebody in-house, you're limited by what their skills are, and you end up having to pay more anyway for them to manage somebody externally that can support them. So we did do a layoff to kind of adjust the team to be able to be the right size and complement to the external team. But that was probably one of the worst days I've had because it was, for us, it was a big number for our small team. Dustin Weaver How many people were you at that time? Brian Feth I think we were like 28, and Dustin Weaver there Brian Feth was eight people that we let Dustin Weaver go that day. That's tough. Brian Feth It was tough. It was really hard. Dustin Weaver But in those situations when you have to cut to bone like that, You hope that you're doing it in a manner that they're going to understand. And I would venture to say that whilst they were probably upset, they understood where you were coming from. And, you know, going back to what we were just talking about, as a CEO, you want to make the decisions that you're making for the best interest of not only the company, but for all your team and everybody that's involved on it. So I can only imagine. Brian Feth I mean, we made a lot of phone calls to try to find jobs. All of them ended up getting employed, and a lot of them wanted us to know when we were ready to hire him back. Yeah, that's great. They understood it was just a pragmatic decision. Dustin Weaver Yeah, yeah, yeah. It doesn't cut any less on you, but, you know, it's – Brian Feth yeah. Dustin Weaver No, that's intriguing. So you built it out with the contract group, and you were able to get it to a point where now you're in a position. Is this kind of where we are at in this iteration of where you are with Avatar currently? Brian Feth Yeah, we've had, you know, we generated 14 beta systems for this foundry, this GMP cell therapy manufacturing system. And we have that out at beta sites over the last year. A lot of those are tracking toward treating patients either at the end of this year or beginning of next year. There's this process of doing process development that usually precedes clinical use to make sure that you've got, you know, good controls, good understanding of how to create a successful therapy. and that when you go into actually using inpatients, you've got that process really locked down and you can measure whether you're doing it correctly or not. And so that takes some time. And there's also a period of time that required, then once you have that data package, you've got to give it to the FDA and say, we're going to treat patients, you know, check my work. They kind of review that package and they hopefully give you the thumbs up. So a lot of those processes are happening this year where the data package has already been submitted or being submitted. then the patients will be treated. And it covers a whole, really a whole range of different types of cancer treatments. A lot of them look like the traditional kind of CAR-T that I mentioned, which is that classic modified therapy. We have a few other partnerships that are on something called TCRTs, which are kind of a similar approach. And then we have a couple of partnerships in this TIL workflow, Tumor Infotrating Lift Site, which is that mash-dept tumor I was talking about. And those are pretty exciting too. I mean, those have just really broad. If you're to kind of a silver bullet in this space, it would be teletherapy. They just have the efficacy across a whole range of different cancers. Just the workflow really is terrible. So it's kind of a perfect problem for us to be looking at. Dustin Weaver Do you consider those three situations, those three different scenarios as verticals? Like are you building towards selling to those individual spaces? Or is it all the same product and it's just a difference of the workflow inside? Brian Feth Yeah, that's a great question. We typically think about the verticals as being the indications, but a lot of times at the stage, early stage, you're kind of thinking about indications just broadly. So I would say that, you know, the reason I labeled those three, I think, is because the workflow varies slightly. And a lot of that is driven by the scale. So I mentioned CAR-T therapy for multiple myeloma is maybe currently being dosed at one to two billion cells. Right now, it takes 10 days plus to get to that number. I mentioned there are groups at University of Pennsylvania, they're trying to get that to like 10 million, maybe 100 million cells. So we're talking about kind of two orders of magnitude less or so, maybe an order or two. Dustin Weaver What does that do to the time? Brian Feth What's that? What does Dustin Weaver that do to the timeline if you take those two Brian Feth orders Dustin Weaver of magnitude down? Easy for me to say. What does that do for the timeline there? Brian Feth Yeah, it's like one to three days because at that point, you're not really growing the cells. You're trying to modify them to get as much of the payload in as possible as a percentage of the starting population. That becomes really important. And then you're trying to get those cells in a fit fighting condition as quickly as you can. And you're only getting a little bit of growth. They're not turning over much at that point. Dustin Weaver So it's Brian Feth quite a lot shorter. It's a tenth of a ton. And the TCRT, that's a different, because that's like 10 to 25 billion cells that they're dosing with. Another order of magnitude bigger. In the till therapy, they're dosing at 100 billion cells. So another order of magnitude on top of that. And you can imagine 100 billion cells, that requires liters and liters and liters of fluid to grow this stuff. And so those workflows are just big plastic containers and incubators right now. very manual. And so they're kind of two different, you know, when I think about the world, I think about a lot of the blood-based cancers like multiple animal where we've got existing therapies and we're just trying to shorten those workflows. And those don't require big volumes, but they do require very high quality potent cells. So in terms of how we're thinking about choosing sites, those are kind of key partnerships, short, rapid manufacturing. And then on the other side, these things that are predominantly focused on solid tumors tend to be really large numbers because they're not very potent. And so we're also thinking about how to shrink those down. Can we reduce it by an order of magnitude or two to just get them more potent? It's probably going to still take two weeks of time, maybe more, three weeks. But you're not trying to get to $100 billion anymore. You're trying to get to $10 billion, for instance. Dustin Weaver I think that sets the picture up pretty well for us to move into the next section of the podcast. And this is the portion that I like to call the scaling secrets. You've done a really good job of explaining where your company sits and how you've gotten to the space that you're at. In biotech and specifically your version of biotech, it seems like time is a very large portion of what you're dealing with. How do you effectively set yourself up to scale to move the company forward and know that the next steps you're taking down the road are going to grow the business for yourself and your investors? Brian Feth Yeah. Yeah. Biotech is tricky because it does have longer timelines and higher capital requirements. You know, every instrument that we build to sell, you know, requires the preceding time and money to build it first and the transfer costs associated. So there's a bit of kind of, you know, there's a gap between the expenditure and the revenue. And that does require some additional working capital than maybe other companies, particularly if you're a tech company and you're just doing everything digitally. The cost of goods is your employee base to write and check and call and control the code. So it does require capital and the fundraising markets are tough right now. I mentioned earlier in the discussion that investors have kind of started to lose some of the focus on this space because of the lack of scale on the supply side. But it also was very clear that there's a huge unmet need here that we could solve and start really curing patients in bigger ways. So, you know, for us, it's, I guess, kind of twofold. First is try to demonstrate a clinical relevancy to this and actually get into patients. And that takes some time. going to have some data, say, beginning of next year, mid-next year, that shows that this is an effective platform and actually doing something important in patients. It can only show so much in mice and in, you know, kind of in vitro assays and plates. And so that using in patients and showing that the things you've shown preceding all the exciting data we generate date holds true and is actually useful in curing patients. And so once you've done that, then it becomes broadly acceptable to be able to use this across an industry. Until then, you're really, you're going to get groups that are willing to take risks, that are willing to, they have, their need is so poignant that they're willing to try something that can solve for that. And a lot of those beta sites that we're working with right now, they're kind of in that mix of, you know, the only options they have are workflows that just really suck. And they know that when they, even if they get good data in this first clinical study, as they grow, that workflow is just not going to be scalable. And I mentioned that 3,500 to 100,000, a lot of companies see that lesson. And so they're starting how do we create automated workflows or get involved with automated workflows early. But not everyone is okay taking risks on new technologies. And so once you've got that clinical data, that really opens the door to a much broader scale. So until then, it's being able to show that the price points and that the kind of the pull through on consumables that we're going to be selling is kind of the bags that are used and that the early adopters are in spaces that are relevant and useful. And then getting a funding round basically to be able to kind of bridge from from now to clinical data and being able to then have capital to start really scaling. And just like every other industry, hiring sales reps, hiring marketing, and just really trying to get this in front of as many cancer research centers and academic centers that are focused on this area, as well as a lot of pharmaceutical companies that are focused on these therapies. Dustin Weaver Yeah. Brian Feth And doing that early. Part of the challenge and I guess trick in this field is you can't just walk in to an approved product and say, use my thing instead. Because the FDA made a decision on this workflow and gave you an approval. And if you want change, you have to go and lobby the FDA to allow a change. And if it's a significant change, like a new tool, they're going to make you redo it to the trials to be able to show that the data is consistent or better. And so a lot of the groups that are focused on new technologies are not the ones that are struggling with throughput right now because it's a real pain for them to change. They're looking at startups and mid-stage companies or therapies that are coming at these big companies, but earlier in the pipeline where they can spend the time to do the process development, make sure it works well, and then move it into patients. So you're starting early. You get locked in when you move into the clinical studies. Ideally, they love it. And then you're moving out commercially with them. And, you know, they're buying a lot more systems to go to all these different sites. Dustin Weaver You're kind of offhanding just talking about all these different segments of what it takes to get in here. And I really want everybody to understand how much time we're talking about here, because like a typical clinical trial or clinical study that you are referring to is not two weeks long. Right. This is this is something that's going to take some time. What what is that standard kind of time for you guys? Brian Feth I mean, probably the shortest that you could do a clinical study in is like six months. Most people that are doing and that would be like a phase one and a really rapid one. Most people are doing, you know, two year, three year clinical studies. And it depends on the indication you're going after. can be, you know, some of the limitations is the amount of time it takes to recruit patients. So that can be a challenge. And then, you know, there's a money aspect to this. If you want to recruit more quickly, you have multiple sites that requires, you know, duplication, triplication of your costs. So there's kind of this trade-off of how expensive the study is going to be versus how many patients do you need to recruit and how challenging is it to recruit. So yeah, it's, you know, for us to get clinical data we're talking about next year, the assumption is within a kind of a 12-month frame, we'll be able to get some phase one data. And that's, I think, realistic, but as we move into later stage studies, those can take years. We actually have some experience. There's a company that was in the news lately. They were just acquired by a private equity firm called Bluebird Bio, and we had done some work with them starting in 2016, 2017. They developed an assay to validate sickling of their... So let me just step back. This company makes a gene therapy where they modify a patient's gene in a blood stem cell so that a patient that has sickle cell anemia where the red blood cell sickles is fixed. And when they do that gene modification, it doesn't get to every cell. And so they want a certain number of cells that have been properly gene modified before they put it back in the patient so they don't have to come back and do it again. And so they use our system to verify or validate that the gene had been edited in a certain amount of the total cells, and then they release it back to the patient. And so we started off kind of in this preclinical before they even treated patients. They bought a couple of our systems and were using it to develop the assay. Bought some more systems when they moved into their next stage of clinical studies. And that took a number of years. And they got their approval at the end of 2023. And then there's more systems being bought now. So it is a long timeline. And the goal for us is to not just pick a couple of winners, but really invest broadly and that some of those are going to come through and be winners. Dustin Weaver Yeah. So and then to clarify and just understanding, you know, again, for the audience, just because you're in these clinical trials for what you're doing and how you're you're working through the process, you still have a product that is able to be sold to other other entities that would still be able to use that during the process. So you're not limited. It's not like you're making zero revenue in this time. You have the ability to bring on revenue as you are building through this process. I Brian Feth think that's a really important point. It differentiates us from a lot of other startups. Dustin Weaver Yeah. Brian Feth And I think there's some benefit that we've been around for a while and that we listen to customers and kind of ultimately got to where we are because of that. But what we ended up having is now this little small scale system that allows us to do kind of plate-based and small scale work development. And then they can bridge into this bigger GMP platform. Dustin Weaver Right. Brian Feth And that little small scale has been around for a number of years. We've sold quite a few. We have over 200 that we've sold out into the market, and they're being used for a range of different applications. Growing tumors, which was our original business. Doing stem cell differentiation, which is what Bluebird was doing effectively. And then doing a lot of this immune cell screening and kind of process development that we just were talking about that bridges into this larger scale GMP platform. Dustin Weaver It's so important, too, because like just as, you know, somebody from the outside looking in, what's scary in this situation and where I think a lot of startups might fail in this in this space is that when you have all your eggs in one basket and then you get in a situation where investors are no longer necessarily excited about the space, you can almost just run dry and die on the vine, as it were, because you didn't have the forethought to think about, well, how am I going to make sure that I make it through all these years of clinical trials? and FDA approvals and re-approvals and retrials and things along that lines. Because when you're taking an initial investment, you as the CEO have to think about not only the time in which it takes to get to point C, but the contingency plan of if C now turns into all, how are you going to get and bridge that gap, right? So I think it is a differentiator. I think it's a strong differentiator that you are able to create a system, a product that is able to be sold ancillary to everything else that you're doing and almost supportively to build that and drive the revenue that is then supporting the research to make the ask of investors a heck of a lot less stringent when they're looking at your P&L or they're looking at what your forecast and futures look like. They see it in a different methodology to then just being that standard research that you're doing for the main GMP platform, right? Brian Feth Yeah. I mean, it helps that the platforms are very compatible and they kind of pull through on each other. But, you know, it is the first platform really is kind of a standalone team that focuses on upgrades, service and support installation. We even manufacture it now ourselves in San Francisco. Dustin Weaver Nice. Brian Feth You know, and that that team kind of on its own is a is kind of a little P&L. And then, you know, it's not enough to support the broader team that we built to launch this clinical platform. But, you know, it's obviously offsets and is very helpful. Yeah, Dustin Weaver well, it looks Brian Feth good. It's kind of creating the seats for all of the new Foundry customers that are going to be coming through. Most of the groups that have adopted the Foundry already started off with these smaller scale, we call them Odyssey, small Odyssey systems. Dustin Weaver And so the big flagship is the avatar, correct? Brian Feth Yeah, the brand kind of underlying the three products is avatar. It all kind of uses the same hyperbaric hypoxic culture environments. And then the three brands, it's Avatar Odyssey. It's the small-scale PD avatar AI, which is this drug screening platform designed to do kind of rapid assessment of potency. And then the third product is the avatar foundry, and that's the GMP clinical scale system. Dustin Weaver Okay. So there are the three verticals that you're currently working with. aside of the large verticals in terms of how you're scaling and growing things. You have the Odyssey that's working. It's providing real solutions to problems that are out in the market today. We have not talked about the avatar AI. And I think, you know, you and I briefly touched on this as we were prepping for this meeting. But I'd love to hear a little bit more about that, because I think that that is something that is very intriguing as well, if you can. Brian Feth Yep. Yep. So they all feed each other, which is nice. So Avatar Odyssey adopters are often thinking about, what do I need to do to get to larger scale? And so they ultimately are thinking about the foundry, but it's a lower cost buy-in. A lot of them need to test potency of their therapies. And they can do that. They can take the cells out of the Odyssey and do that in a, let's say, a tomater or a plate reader or some other tool. But it's logical that you just want to do it in line with the culture so that you don't have to take it out. You can just kind of automate the process of looking at killing. So we built the Avatar AI really as a generational upgrade to the Odyssey. It's the same chassis, same box, but inside, instead of having shelves for plates and flasks, there's a little kind of electrical impedance reader that we built in that allows us to look at changes that occur across the bottom of a well. So you basically run a current across the bottom of the culture well. And if there are things that are sticking to the bottom, like a target cell population, it disrupts that signal. And so then you can kind of tear it, normalize it, and then you add your drug. And then as the drug starts killing, those cells detach and you get a change in impedance again. And so you can measure very specifically based on electrical change, how many cells have you killed using your therapy. And we organize, we use a 96-well plate and you've got enough wells that you can kind of create the right organizational map on that plate to be able to get really accurate counts of how many cells you've killed. So that's a very useful tool in just kind of getting these nice real-time killing curves because you're getting data continuously. So you can see kind of if you start to, if this therapy starts to become, you know, suppressed or just exhausted, like it'll just, you know, it'll stop killing. So it's a nice way to be able to test potency in these immune cell therapies where otherwise you'd have to like tag the cells. And when you tag them, you put a fluorescent mark on them. It changes their behavior. The other thing that it really allows us to do is in the manufacturing world, we talked about how long the sterility testing is and like all of this quality control stuff that happens after the culture. And the FDA has been asking for better potency assays. Right now, we don't have a good way to know if the cells are actually going to kill or not. Because, you know, when patients come in, some have been on multiple years of chemotherapy. They're old. The immune cells are just not doing well. Dustin Weaver Right. Brian Feth And so even if you can kind of get close to the quantity you want, the quality may not be great. And you may need to go for longer or you may need to try again or redose. That's, you know, save enough to the cells to do that. So having a way to measure potency actually in the manufacturing area is an important thing. And it just needs to be fast and easy to use. And so that's what we've been focused on is applying the tools of machine learning to look at kind of libraries of these killing curves and being able to start making calls probabilistically. Like the way that this curve is starting to plot, the likelihood is it's going to be below the threshold or the likelihood is it'll be above the killing threshold that we want. And so that's kind of the vision and focus for that product, which is can we turn this into a rapid release assay to say these cells are potent enough and you can release them to the patient? Dustin Weaver I have a random question. Can you use the patient's immune cells in these tests to figure out exactly how much of the drug or how much of the process they are going to need to be effective? Is that something that you guys do? Brian Feth You would have to do, I think you'd have to do it empirically where you kind of look at, I mean, every patient's tumor is different, how much it's read, how much there is. So I guess the question would be based on your clinical study that you've done and the range of patients that you've evaluated, what is kind of the effective, the range of effective dose? And then you could start making decisions based on, you know, that range, whether or not this patient will need to be at the upper or lower end of the range. Dustin Weaver Yeah. Brian Feth You know, the FDA is, they're used to managing processes by adherence to a strict process. The thing does this, this, this, and this. How many times does it come out of range? And that's how the FDA wants to monitor manufacturing. The reality is when you're dealing with patient cells, there's young patients, old patients, big patients, small patients, and like they all are different. These are living therapies. And so in some ways, like the regulation kind of should look more like a surgery where the doctor is making decisions in kind of real time. But in the manufacturing process, you've got enough input-output sensors. You can start to look at how that thing is growing, maybe the metabolic activity, how is it metabolizing sugar, and start to make some decisions about the quality. That can change. Maybe you grow it for 14 days instead of 12, or maybe you grow it for 5 instead of 3, or you cut it off after 3 days when you normally wouldn't. So I feel like there's an opportunity. If you set, you have a really thoughtful way of setting your endpoint criteria for dosing, like let the process self-optimize into that. And, but we're just not, the FT is not, they're not ready for that. And the field for AI is just so new still. Dustin Weaver Yeah. Well, and that's so intriguing to me because ultimately in my thought process, in my non-medical brain here, I'm thinking of how you optimize the process to make it make sense for each individual patient. Right. And if the product, you know, is being limited by the time in which it takes to make it and the efficacy, I'm not even going to try that, efficacy. There we go. If the efficacy of the product is also an inherent issue across that timeline, if you knew on an individual patient basis, if you could track and trigger and figure out what their dosage really was, then on the grand scheme of things, you're not having to waste time on the development of that treatment for that individual. And that's where my mind was taking it. I'm sure there's probably a lot of red tape that you would need to go over and go through to get to that point. It just, you know, as technology changes, it seems like it might be a good way to help speed up the process. Brian Feth Yeah, yeah. I totally agree. I think it's coming. Dustin Weaver Yeah. Brian Feth It's just, you know, a lot of it's new. Dustin Weaver So we Brian Feth just got to give it some time to work its way through. But, you know, to credit to the FDA that we've had a CAT meeting with them. We talked to them about this concept. They all kind of understood it. We're agreeing, you know, agreed that this is something that needs to happen. So, you know, I think they're and they're working on, you know, continuous improvements using AI already. There's been a number of kind of requests for, you know, comment on some of the things they're thinking about with how to use AI effectively. So it's Dustin Weaver probably not just putting it into chat GPT and asking a Brian Feth question. Yeah, well, Dustin Weaver that's a great segue, though, into the next section of the podcast. And this is what we like to call the market edge. So what's one trend in the industry that you really see as being a front runner for what is going on with your business? Brian Feth Yeah, one of the main trends I see is something we've actually already talked about, which is, you know, for the last 12 years, these processes have been like two weeks long. Actually originally built the platform to make sure that we could accommodate the volumes that are needed for that length of a process. And really over the last, you know, call it three to five years, a couple of groups, Novartis, University of Pennsylvania, they started, you know, kind of pioneering these short workflows. And we were fortunate that we had designed the system to also be able to adapt and use smaller volumes. And so I think this kind of all of the existing systems that are just built for a billion cells and large volumes, they're going to lose value as the market shifts to shorter manufacturing. You still have to have a place for the cells to reside and it's still really important that they don't kind of become damaged during this process of modifying outside the patient. But that's a big change that's occurred, which is this trend toward short manufacturing. Another one that's really interesting, and it has been a little bit stuck because of how traditionally how manual and big these workflows are. Right now, if you want to go get CAR-T, you have to go to one of the major medical centers that's offering it. And as I mentioned, there's a lot of patients that can be cured with this stuff. So everybody, as soon as they were just, you know, people realized how effective these things could be said, how do we get it to be closer to the patient? How do we get it to be, and you know, transportations also cost time. There's a process which probably won't go away of freezing these cells before you ship them and when you return them. So, and that damages the cells to some extent. So there's some costs to having it remote. And a lot of times, you know, the patient needs to drive to wherever it is. And there's a big patient cost to that. They have to be there for a while to just make sure that the therapy was administered safely and that they're in good shape. And it can be several hours away. Looking for ways to kind of decentralize the manufacturing, because right now it's manufacturing. The administration points happen at key medical centers. The manufacturing currently is in single facilities at huge scale. The thinking is, like, if you can decentralize this into a trailer, into a less of a clean area within a hospital because it's closed, maybe it doesn't need to be a hyper-clean environment, then it can live really close to the patients. And particularly, a lot of these workflows are really kind of heavily process-driven. And so if you can just automate that process and put it at the hospital, most of it can happen locally. And you can start treating patients at community hospitals and kind of regional centers. There's a number of groups that are working on that. One of our partners, LabCorp, is thinking about how to use their kind of amazing logistics You know, there's a drop box everywhere for lab core diagnostics. It is, you know, centralized facilities where these things go to, but they're, you know, you drop them in, you have results the next day. I mean, they're just really incredible at logistics and quality, making sure that the vial, you know, ultimately is correct and gets the right data to you in your inbox the following day or to the doctor's inbox. And so, you know, they could be a really critical piece to making sure that the centralization of the current is successful and having this really rapid information back on how the manufacturing is going. There's also groups that are developing these kind of really clean, effectively portable GMP manufacturing centers, the ones called GermFree that we've partnered with. And then you need a workflow that's pretty automated, that has robotics that can link things that are not necessarily automatically linked. And so we've got a partnership with Thermo Fisher to link into some of their instruments up and downstream. another partnership with a group called Cellular Origins that makes robotics. So we're kind of working on ways to take our instrument and really being able to forcibly realize the vision of decentralization. Because if we don't do it, no one else is really moving forward independently with this. You need all of those components. And it really needs to be somebody that understands the manufacturing workflow and can kind of link these things together to get a good outcome. Dustin Weaver We can concert with it all. Brian Feth That's another important trend. Dustin Weaver That is amazing. Being able to be mobile and actually bring it to the patients. because not even where you're at in these scenarios and treatment, even if you're on second line of defense, some of these patients, the ask of them traveling hours is very hard, not easy. Brian Feth Tom Whitehead told me a story. His daughter, Emily, was one of the patients that was cured out of the first trial of this. She is in college now at University of Pennsylvania, and she was five when all of this happened. She's been a real successful symbol in this field. Tom indicated that he gets calls all the time now from people that want to know how they can get involved in this. They're patients, they need, or they have a family member and they're trying to get into trials or they're trying to get treatment, to get to a treatment center. And as we mentioned, like the queue is potentially long and some of this given supply constraints. So he mentioned to somebody in, I think it was in Pennsylvania, that there's a medical center that, that administers this. It was about two hours from the person that called him and the person said, "Well, it's a long drive." The doctor says they can do another round of chemo. Like, chemo's not curative. And if anything, you know, it's kind of the opposite. You know, lose your hair. Maybe it pushes the disease back a little bit. So, you know, convenience, while maybe it shouldn't be, it is a big deal. And it can be, particularly for patients who, you know, it's a major inconvenience. And if their family members want to come with them, stay with them, support them, it could be a major inconvenience. Dustin Weaver Let's move on to the section that we like to call hitting the mark. And this is an opportunity for you, Brian, to just tell us how you guys are really nailing it. What are you guys doing over at Excel Bioscience that is top class? What is the thing that you're most proud of at this moment? And I know we've talked a lot about everything that you're doing, and it's hard to pick kind of one thing out of that. But I'd love to know what you feel like you guys are doing the best at. Brian Feth Yeah. You know, in the stage of company we are right now, every single relationship we have matters so much. And I think the team has really stepped up to make sure that our ability to kind of support, provide, you know, kind of a strong backbone to kind of every system we install, make sure that we are going through the process of running every experiment that these groups are doing and make sure that when we hand it to them that they're going to be successful and we know exactly what the likely outcomes are going to be. Just really, I think, that convergence of the engineering team that's built the thing, the process development team internally here that's going through and validating each one of those workflows, and then the customer relationship group that's kind of really going to each of those partners and understanding what their needs are and kind of trying to adapt the things that we're doing here to make sure that they translate well. And it's a small team, and just the cross-communication has been really successful. And even in small teams, sometimes you can get siloed. And I think it's a testament to, one, the quality of the team that we've built. Every player right now is just grade A, really top tier individuals, top of their field. And it shows. And I think the ability then to work seamlessly without drama and, you know, it's just a really functional team. And that's allowed us to deliver, I think, a really great beta program to date. And now we're kind of shifting this, you know, to move more toward commercial and making sure we have infrastructure set up to do marketing, to make sure that we are able to deliver kind of a scale of outreach to new partners and then translate all the data that we've been generating because we have tons of data and all of it is coming from external partners. And then, you know, we've got a lot of data from the last 10 years as well. And kind of moving all of that into kind of a coherent message has been something that the team has done really well as well. Dustin Weaver I would say that that is a good point to have. And from an outsider's perspective, looking in and being in similar shoes of building out teams developing different departments and to work in concert with one another. What I hear in that is that your leadership is really doing a lot to build that. So I would give yourself a pat on the back because that means that something's going right over there. If you feel like you're getting through a day without any drama and nobody's siloed, that's a big win in anybody's book. Doesn't matter what you're doing. But that actually leads us directly into the signature segment of Bullseye Breakdown, and that is the best bad advice. So Brian, I have a question for you. What is the best piece of bad advice you have ever received and how did not taking it benefit your company? Brian Feth So, you know, the science is kind of a star driven event. People want to publish. That's the currency. People are focused on performance of individuals and kind of the ego emerges as a result of that. And so you've got that. That's true of science. And that's actually true in consulting as well. There's a little less of an emphasis in banking and consulting on kind of good quality management and just more about, you know, GSD. I'm just trying to make progress as quickly as you can. And I'm sorry if I hurt your feelings. And so we had somebody that was underneath me. I was kind of early manager learning how to manage effectively. And this individual was kind of slow to respond to requests for information and kind of, you know, we're working on scales where everybody's there until 2 in the morning. And people disappearing for a while is not a helpful thing. And so I was asking for advice. How do I, you know, kind of course correct this individual to kind of get them to be on our timeline? And he suggested that I just grab them by the shirt, you know, just really shake them. Basically. Say, listen here, F-tard. kind of bleep a little bit there and i i just walked out of the room like i don't know what i just i don't know what kind of management advice i just got but i'm definitely no it's not correct Dustin Weaver that's not the right option how'd you handle it i Brian Feth just you know i mean i literally just ignored him um i did have a conversation with uh with the member my team member and um just explained to that I probably used comedy a little bit. Like I had just gotten some really bad advice and I'm going to try to do better. I think when you can relate the behavior to the impact on the team, then it becomes less about them and more about how it's impacting everyone else. And if they know other people are aware as well, I think that could be reinforcing. In the end, put them on PIP and then I think ultimately termination. So it didn't work. But sometimes people aren't the right fit. And that's happened a number of times at this company where people are different than what you thought. Interviews are not always a great proxy for what the real performance is. It is important, as the dogma says, to kind of try to make those changes as quickly as you can because it does impact the culture of the team. If people see that somebody's not the right fit, maybe they're sandbagging or not interacting well with others, Dustin Weaver that can just destroy. Brian Feth It's very toxic. As hard as it is to let people go, doing so quickly can be the right decision. And so I often think about building teams as kind of pruning a bonsai. You know, it's a slow process. And sometimes it can change the form very quickly. If you clip correctly, if you cut off things you shouldn't, it can damage the team. But the whole time you're trying to grow and make it better and stronger. Dustin Weaver That's beautiful. That's a great analogy. I like that a lot. All right, well, let's move into the takeaway shot. Brian, this has been a really great conversation. I feel like we've had a lot of view into what it's like to be in the biotech space and specifically cancer research. If you could leave our listeners with one piece of advice, what would it be? Brian Feth Yeah, I think, you know, this kind of idea of grit is getting some, you know, kind of circulation in the startup circles as being kind of a Dustin Weaver core requirement Brian Feth to be successful. And I think that is there's a lot of truth in that. It's not giving up and not kind of taking no for an answer. And practically, that means having plan Bs, even though it costs some money and time to do that and creating flexibility in your business plan. I also think also kind of business dogma, but listening to customers allows you to have that flexibility correctly. So the fact that we were out in the field, you know, starting at our early relationships with some of the pharmaceutical companies, they just said, gee, I'd like to buy that thing. And I'm doing these things with it. And I thought, OK, well, there's a phenotype of somebody that wants to buy it that does this thing. Maybe there's more. And, you know, as we went, we said, OK, Salt Therapy seems particularly interested in what we're doing for these things. Maybe, you know, how big is that market? And I think you just don't know that stuff unless you're out talking to people, listening and seeing the data. And so I try to be, you know, on as many of these partner calls and customer calls and data review calls. I can't because you learn so much. And I think the team has done that as well. And so that's allowed us to kind of continue to be flexible. And as the business has hit dead ends and that happens sometimes, you know, investors decide they're not interested in the flavor. thing that they were interested in, now they're interested in this. And you don't want to kind of just go to cater to whatever the investors are doing, but being able to tell your story effectively and about how what you're doing can make an impact and maybe altering the application space just a little bit can be really important to being able to make sure that the company continues to have financing to grow and to be able to find the right customers that are willing to pay for what you've built. Dustin Weaver I think Brian Feth that is Dustin Weaver a very intelligent comment. And you can't build a solution to a problem that doesn't exist. You're never going to sell it. Brian Feth That's right. Dustin Weaver So you always have to be listening to your space, your customers, your potential clients and hearing what their problems are so that you can effectively build a solution to something that the market actually needs. And if you're spending time on something that doesn't exist, then you're never going to be successful. so Brian Feth I agree with it I think it's easy to wander into spaces that are not repeatable and scalable and that's the other thing you want to find the answer to the solution to the problem that's going to continue to that's a big one for a lot of people that you can kind of repeat rinse and repeat with so yeah it takes some work and a lot of listening to do that and sometimes hard decisions Dustin Weaver well cool thank you very much for being with us today Brian this has been a really insightful conversation I am so glad that you are doing the work that you're doing for cancer. Again, as a human being of this earth, thank you for what you are doing and the work that you're putting in because I think as you continue to grow the company and the process and everything that you're working so diligently on, we are all collectively going to be in a better space in the fight against cancer. Brian Feth Well, I appreciate you giving me a platform to talk about what we're doing and it was fun. Dustin Weaver Wonderful. Let me do this outro real quick and then we can, I want to hear about your bonsai background because I don't know if you can see, I'm a water plant guy. If you can see in the background, I have like That one's cut off. I'll show you in a second. Let me run the script real quick. That's a wrap on today's episode of Bullseye Breakdown. A huge thank you to Brian Feth for sharing his insights on the future of biotech and the incredible work happening at Excel Biosciences. If you found this conversation as fascinating as I did, make sure you check them out at xcelbio.com to see how they're pushing boundaries of cell therapy innovation. If you enjoyed the episode, do me a favor, subscribe, leave a review and share it with someone who needs to hear it. You can also follow me on all the major social medias and or through LinkedIn. Give me a shout. I'd love to hear what you're working on. As always, keep aiming high, stay curious, and I'll catch you next time on Bullseye Breakdown.

Transcript Adam Welcome to the SciMed Biotech podcast. I'm Adam Stasho. My guest today is Brian Feth, CEO and co-founder of Xcell Brian Bio. Brian, welcome. Thank you. Appreciate it, Adam. Adam So, Brian, I was wondering, I think we're going to talk some about cancer care today and some of the new tools and technologies that are out there. I was wondering if you could start at a high level and give us a quick overview about what you guys are doing there at Xcell Bio. Brian Yeah. So our business is focused on creating technologies that enable this kind of new class of cancer Brian care called cell therapies or immune cell therapies to be able to be delivered to patients. And we enable, you know, kind of from the early stage discovery, research, process development, assay development, screening, as well as GMP manufacturing. So our tools kind of span that continuum of research in that field. And we do have kind of a range of different customers and partners using the products, ranging from groups like Bluebird Bio, where they use our technologies as a potency assay for their sickle cell anemia gene therapy product. We have some other systems that are doing give a delta T cell work. So a whole range of applications within the field of kind of gene and cell therapy for our technologies. Adam And what kinds of things are you doing differently from existing technologies that, you know, what are your platforms offer that are new or different capabilities that other, that people might, like Bluebird might not Brian have if they're not working with Brian you? Yeah. So maybe there's a, why don't we step back a little bit? Because I think there's something to be said about just where the field is at in general for cancer care, which I think is pretty interesting. So really there's been this evolution of kind of cancer care going from, you know, small molecule chemical based approaches to either just trying to target mechanisms that would slow down cell growth or take advantage of cell growth. Maybe something that's more targeted like a radiation where you're kind of disrupting DNA and trying to target the localization where cells are damaged. But those have a lot of side effects. And they mainly, you know, there's a lot of off-target issues. They kill any cell that's kind of fast-growing. And so you get hair loss and taste buds and other things that occur as a result of that that are damaged. The field kind of emerged over the last 20, 30 years into kind of what we think now bioprocessing, where there's a focus on antibodies and other types of peptides that can kind of more specifically target and exploit mechanisms that are different about the tumor cell than about a normal healthy cell. And while those have been successful, you know, it's kind of picking off, you know, slivers of indications oftentimes. Sometimes you get bigger swings, but it tends to be that every cancer is a little different and it can be hard to kind of really nail specific cancers with targeted therapies in a broad way. And so really over the last several decades, but really the last 10 years, there's been this kind of emergence of understanding that the immune cells actually play a critical role in keeping cancer out of your body and getting rid of cells that are damaged in some way. And it's only once the cancer cell has kind of developed this capability to inhibit or create this kind of suppressive signaling against the immune system, that then you get the immune cells that are no longer able to take care of it Brian and you get tumors that are run amok. And so the field over the last 10 years really developed these first antibody approaches to try to disrupt the cancer and the immune cell interference that the cancer cell delivers. And those are called checkpoint inhibitors. And it's a class of therapy that's been very successful and is often used alongside chemo and other types of therapies, targeted therapies, to enable those therapies to work better. And then the latest of innovation, which is where my company's focused, is looking further at the immune cell and saying, can we actually use the modern tools of genetic science and genetic modification to try to deliver more targeted signaling to these cells so that they have more specificity to go after and kill these cells that they should be not just awakened to, but actually aggressively going after. And the way that we've been doing that is by taking a patient's immune cells out of their body, typically T cell population, modifying them either with like a virus, like a Luntivirus or with CRISPR and other types of gene editing sometimes used. And then because virus is expensive, because usually we just need to start at small populations in order to have good efficacy in delivering the gene, we're then growing them up and then putting them back in the patient. And so there's kind of a long way into the end time. It's a very expensive process. It's historically been very manual. Brian But it's like curing, you know, half the patients for blood-based cancers that it's being used in. And so very exciting technology. You're often getting 80%, 90% complete response rates after use. So within a couple weeks, you're seeing really clearance of the leukemias. And then usually about half the time it will come back, and the other half the time you just don't see it again. So very dramatic outcomes compared to what we had been talking about before, which is several months of life extension, maybe a year or two with some effective therapies. So this is kind of a very exciting field. And there's been over the last few years a lot of attention as the first couple of therapies were approved in the end of 2019. And one of the challenges for the field is that these cells are living drugs, and they get unhappy if you put them in an environment where they're not comfortable. And scientists historically have been very used to dealing with cell lines, things you can put in a CO2 incubator, and just, you know, they can be harder to kill. They tend to, you know, after decades of being used to being grown in these environments, they're comfortable. Brian Human primary cells are just very different. They're very challenging and happy. And so our technologies are really focused on getting these cells equipped to be successful in the patient. And so we focus on making sure that the cells are getting adapted metabolically to the environments that they need to actually generate and use energy in. And so the metaphor I often use is kind of two different ones. One is if you take a fish out of water, you leave it out for a while, and then you throw it back in, it's going to act stunned for a while. It might never recover fully if you've deprived it enough oxygen. So there's kind of that paradigm that's happening. And then the other is if you're a cyclist and you're competing for a professional race, probably they're training in the mountains. And there's a reason for that. there's an atmospheric benefit that you're getting by having a lower oxygen level. And as a result, kind of forcing the body then to accommodate and create more red blood cells to have better distribution of oxygen. And so you're using the environment to create a competitive advantage. And that advantage gives you advantage. You know, it helps you to be equipped to be a cyclist in the mountains, but it also gives you this additional advantage when you come down out of the mountains. So to your question, what do we do differently? Our technology is focused on really improving the quality and potency of the cell therapy product and then automating that workflow. And between those two things, that opens a lot of doors to trying to really advance and scale these therapies, Brian which has been the real challenge. It's been a supply-side challenge in this industry so far. All right, so that makes a lot of sense. You know, looking at ways T-cell therapy Adam is quite effective. and also quite expensive, so then finding ways to make it more successful, more of the time successful, meaning the cells are still alive when you give them back to the patient, and also cheaper and faster, seem like good goals. One of the questions I have maybe about the fish metaphor is, the answer might be both of these, but you were talking a little bit about the stunning the fish when it goes back in the water. To what degree are these primary human cells experiencing difficulty surviving when they're out of the human? And to what degree is the problem? They're fine when they're out of the human, but when you put them back, they're having the problem. Brian Yeah, this is where the metaphor doesn't quite hold up. Because what we're actually doing is giving them a lot of oxygen and a lot of sugar when they're outside of the patient and they're in this culture. Because they grow fast when you do that. And the kind of the name of the game is, you know, really how many cells can I get in X period of time? And ideally, the shorter the period, the more cells, the better. But one of the reasons that we were kind of dosing at these high volumes in the game is to try to get as many cells as you can is because they don't all act very effectively. Brian And so you're just trying to kind of use numbers as a way to get at least some benefit out of these therapies. And they have been reasonably successful. Thankfully, the blood is pretty high oxygen comparatively. We're breathing 21% oxygen. The blood's kind of in that 10% to 15% range. So having these cells kind of at the 21 or sometimes groups even hyperoxygenate the cultures, it doesn't seem to affect them that much because the metabolic shift isn't very great as they move back into an environment with less oxygen. but it becomes a real problem for treating solid tumors, which are very hypoxic. And the other element, if you can create more potent cells, you just need fewer of them. And so one of the areas that we've been focused on is this kind of idea of how many cells can I get over time. We said, well, what if you work at it from the other side, which is how many cells do I actually need if they're X potent? And so can we make these cultures much shorter? Can we take what has historically been a 10 to 14 day culture time and turn it into a day or three days by just metabolically priming them? Forget about expanding them. Let's try to get as good a starting population as we can. Let's get that gene editing efficiency as high as we can, something our technology helps with. And then let's metabolically prime them so that they're in fighting shape whenever they go into the patient. And that way you just need fewer of them. So that's kind of the thought process around, you know, trying to solve some of the needs in the space around how do we scale this to hundreds of thousands of patients as these therapies are moving earlier line and, you know, continuing to get these successful. And it's really shortened the manufacturing time. Let's get it as short as we can so we can stack more of those patients on the same instruments over time. Let's improve the potency so that when they're in there, we get as high complete response rates as we can. and let's do it in a way where we can create more flexibility or plasticity in the memory population so that you're just not shocking them and driving them all to affect yourselves, but kind of leaving a population that can have some self-renewing properties so you get more Adam of an entrenched response over time and more of a persistent response. That's going to address part of some of the patients who are having issues with relapsing after, you know, you're saying they're getting good results, but about half of them are seeing the cancer come back. And maybe if we have memory cells that can survive for longer and replicate longer, then they'll have a better outcome in those cases. That's right. Yep. The goal here is to not have to Brian redose patients. It's ideally Adam a one-time transplant Brian effectively, a one-time infusion. And then you'd hope to, you know, the patients that are living for a decade plus now, You know, they have residual memory cell populations that are continuing to survey and make sure that they remain disease-free. And, you know, the problem is really exasperated the solid tumors. So you can kind of use that as an example of what's happening in the blood-based solid tumors. It's so difficult to get the cells to go to these really nasty hypoxic tumor sites. They just would rather be anywhere else. It's important that basically to counter this challenge even to get them to localize correctly, we're just flooding as many cells as we can. And so TIL therapy, which Steven Rosenberg has been working on at the NIH for several decades now, they got their first approval with a company called Iovance this past year. Dosing volumes for that type of therapy are in the 50 to 100 billion range, which is like two orders of magnitude more than we're using for blood-based cancers. And as a result, you can imagine you need leaders and leaders of media to culture these cells. They use feeder layers, which can be expensive and take up a lot of space as well. So it's an expensive process, and it's challenging to do it in a small instrument. So that's part of our goal is to be able to shorten that time. That's a month plus of culture right now. If you can shorten that, get it in small volume, you also make that much more broadly capable of addressing needs. So it's the supply-demand. Supply is really the constraining factor in our field right now. And the opportunity to just bring this to solid tumors and to all of the blood-based cancers just not being treated right now that could be, there's just this huge opportunity to impact patients. And, you know, I like the idea of Adam supply and demand being the issue. And one approach to that is let's just produce more culture incubators and more physical facilities so we can produce more of them, which has some benefit. But if you're able to make the process much more efficient, then that's sort of to everybody's benefit. You can probably do well as a company, but also we can treat a lot more cancer patients and the cost to each patient is lower. There's a chicken and egg with a regulated industry like ours. Brian A good example is Legend, which is a partner with J&J. They have a drug called Carvicti, which treats multiple myeloma. It just went from a later line therapy to a second line therapy this past year. And their addressable patient population shifted from thousands of patients to over 100,000 patients. And they were producing 3,500, basically patient doses a year and need to figure out how to scale up. Brian And that's a real challenge because they're locked in to some extent what they got approved with a number of years back, which is not a very scalable workflow. And so they're trying to apply robotics targetedly where they can. And they did announce, to your point at a recent conference, that they're building a bunch more factories and just basically trying to replicate the existing manual workflows. But you're not getting economies of scale. Adam It's Brian a very expensive way to try to meet that. But that's how they're going to get to blockbuster drug status is by just, you know, the heavy slog. And they are interested in new technologies. It's just you have to go back and redo a lot of the clinical data to be able to integrate. And it really alters the workflow. So in our company, while we're very interested in supporting those larger companies, and we've met with the FDA and have talked about this in CAT meetings ourselves, we're also focused on earlier stage companies that are still in that process development stage and are making decisions now. Do I just get patient data in a manual workflow, or do I invest in something that's more scalable that can pace with me through clinical? Adam Yeah, that makes sense. So perhaps we can move on for a moment and talk about the trending question in everything these days is AI. Is there a place for AI? Is that a thing that you guys are thinking about? Are there ways that that could help you be more efficient and get these culture times cheaper or faster? Yeah. I Brian mean, as you guys know, software is the backbone for almost everything. And there's always a lot of data that comes off of process. So, you know, given that we're focused on drug screening applications, potency screening and cytotoxicity applications, as well as manufacturing, there's definitely a role for AI as you're generating data. You know, particularly if you have control of the inputs, you can start using that and kind of a nice optimization. So one of the things, actually two of the things we talked about with the FDA was kind of taking the way that we measure potency using, you know, basically live cell-based potency assays. And we use that to measure the efficacy of a CAR-T using either kind of EDT ratio assays or serial killing assays, which are two kind of common assays in our field. We automate that process and we can get data out. And we're running that alongside the manufacturing workbook so we can start evaluating the quality of the product. How potent is this product kind of as we go through this process and then as a release? And then there's other things that we're collecting just in the manufacturing process, like pH, dissolved oxygen. We're getting metabolic information like lactate. You're getting valuable cell kind of information. And then you can add in other stuff like any kind of phenotypic information that you're getting from flow. And all of that data can help you start to make smarter decisions about what needs to happen in the culture to optimize it. And so for us, you want to optimize the quality of the cell and maximize the quantity within a constrained, ideally within a constrained time period. And that's, you know, in traditional bioprocessing, pretty reasonably easy to do. In the world of cell therapy, the patient is the main variable. You got young patients, old patients, patients that have been on chemo for a couple of years, others that are fairly naive to Brian a lot of therapies. And there's certainly criteria that gates whether a patient is eligible for therapy, but it is a huge mix. And some patients take forever to grow, and other patients are pretty rapid. So the kind of the vision for our company where I think there's a real opportunity from a data play and where we've already started investing is this IO equation. You're controlling for unique inputs that allow you to improve the quality and alter the quantity. And then you're using that information in tandem with the real-time readings you're getting off the culture to then just self-optimize that culture. And AI can allow you not only to make those smarter gating decisions, but in some ways you should be able to just let it run and self-optimize to a specific endpoint you're trying to get to. And if that means the patient is ready in three days versus 10 days or it's ready in one day, depending on the workflow, then you should be able to do that. And it may be that you can select a really release patient based on when they're ready in terms of a metabolic profile on a potency endpoint. And so we've been thinking about in the potency assay ways to do that in a cell-free way that can make it a little bit more scalable and easier for clinicians to use. and in the manufacturing, really making it a hands-off manufacturing approach, where there's no real requirement to have a human involved, other than just making sure that the kind of release criteria endpoints are met. So I think it's a big deal for healthcare, which is, you know, AI is challenging, because you don't often know what's driving outcomes. You don't, you know, it can be a confluence of things that allow you, that the AI kind of picks to drive towards specific endpoints. And that's challenging for regulators that are used to looking at this is the process, this is your adherence to the process. Do we trust? Because you often don't know as much about the end product. So I think there's going to be a change in our field as AI becomes a little bit more accepted. The FDA has some time to kind of think about how to regulate it effectively. I think we'll start seeing a lot of technologies removing a lot of the human involvement Brian in it and producing the data in a way that is going to be useful for clinicians and useful patients to have better outcomes. Yeah, that makes sense. So I'm Adam going to sort of try to make up an example of what I think you, one of the things you might have just described, and you can tell me if I'm missing any pieces there. But you might take some T cells from a patient, use CRISPR to modify them. Now you have to start growing them. You stick them in an incubator. And potentially the incubator has some knobs and levers you can control pH and oxygen and some of the other variables that are going to help decide how effectively these cells grow. A piece that I'm interested in that I didn't really realize is that there was so much variety and how well a patient, you know, cells might grow from patient A to patient B, I was sort of thinking, like, there might be an optimal set of conditions, and that's that, but you're saying perhaps not. So maybe within the incubator, you've got some AI technology that's trying to look at, you know, it adjusts the knobs automatically and looks at how the growth goes over some period, and then adjust them again and adjust them again until it finds the fastest growth, not only for, in general, but for that particular set of cells. and also keeps it within certain criteria, right? We need to make sure that it's not, I don't know, if it growing too fast would cause a problem or whatever the criteria are. How accurate or how fair is that of a description? Yeah, I mean, I think that's Brian pretty accurate. I mean, you started with the gene editing, which is an important parameter. So if you get a low gene editing efficiency out of a patient for some reason, you know, that may, and you've got to target, you know, end amount of total number of gene-edited Brian cells that you want to administer, you know, that could, one, determine, you know, what you need to do in terms of length of time, but it could also be that, you know, that you're influencing certain, you know, you can use the culture environment to influence some population expansion, and so if you need a different shift, like a different mix of cells, more near-used cells, For instance, drop the oxygen down and allow that if you need more cells out of the gate and then, you know, maybe you have a higher oxygen, more pressure, which is one of the things we use in the culture. So there's a lot of you have a lot of like levers that you can use to optimize the culture. And even like duration of frequency of rocking, you know, I think a lot of the field believes that cells are best expanded through kind of a more static culture. But there is a role for mixing them up, allowing them to refresh. So there's just tons of parameters that you can alter to speed up, slow down, and change the population mix, shift metabolism that I think right now we use as part of our process development with partners and customers. And then they just lock it. And you do your best to try to create a broad enough kind of bell curve that you're getting most of the patients to be successful in that. But you're going to lose some outliers potentially. You're going to have some people that just are maybe growing, overgrowing the culture, and you have more than you need in terms of sales. You have some that are just going to be lost in the culture, and that's true right now in existing workflows. So the idea here is to be able to have a very flexible system for process development. And then currently we lock in for manufacturing, but my dream is to have a system that continues to be able to optimize on a per patient basis just so that no patient is kind of left behind in this. If you don't have to go back and ask them to, a lot of patients just don't have the time. Some patients are dying before they receive the therapy. These are often very late stage patients currently. And so if you can shorten those cultures and if you can give them the best shot of having a dosable cell base before the first time out, you want to do that. And when Adam we're talking about AI and FDA approval, one of the things, and you probably know more about this than I do, but one of the things that comes to mind is the more the models can be built in a way that provides you lots of detail about what happened and why, I would think the more likely FDA would be to go along with it. if you throw it to an AI model, it adjusts levers and comes out and it's like it's done, and you have no idea how long it was at this pH and how long it was at that pressure and that oxygen level. It's harder to swallow that. Adam You can provide all the information about what the levers and knobs were doing, what was the pH at every minute of the day, and then also potentially why those decisions were made. If the model can tell you, I'm increasing the pressure at this time because I'm sensing that thing, then you're getting closer to the point where it's the same as if you had a human standing there adjusting everything, where they theoretically could explain exactly why they adjusted everything at that point. Or the technician couldn't explain what they did and didn't keep track of what they did, in which case that wouldn't be so great either. Yeah, having more explainable answers seems like it would be beneficial. Yeah, no, I think that's a great, that is a great insight. And that was Brian actually the feedback that we received, which is, you know, there's still very much an interest in understanding what's driving specific changes to be made. Brian And then, you know, ultimately being able to kind of validate that that fits within kind of the range of things you might expect. it doesn't have to be rules-based, which is kind of what's currently happening, but it does need to be traceable and kind of transparently understood what happened. And so, you know, I think there is a big role to play, particularly for software groups that are focused on data logging, focused on kind of cloud-based connectivity to manufacturing, to be able to start having that traceability and making sure that any kind of decisions, in theory, you could run through a filter before the AI makes a decision to change something. It can also have a human, at least as a review or a validation. So yeah, there's a lot of shades in this where they just don't want the FDA's guidance is going to be. You know, it can be full, partial, you know, kind of lightly guided. So there can be a lot of opportunities. But in any case, it's going to be, you know, I think an important base of technology to be able to ultimately get to better patient outcomes. I mean, it just feels obvious to say that, but when you incorporate some flexibility in decision making when you've got a lot of input variability, Adam that's just naturally going to lead to better outcomes for patients. That certainly makes sense. Well, perhaps we can touch on one more topic before we close out today, and that is point of care. I was wondering if you can tell me about what you're seeing, sort of where things are headed with point of care and actually things from, you know, the patient's point of view once we get out of the out of the lab. Brian Yeah, Adam yeah, this is one of the more interesting aspects of cell therapy and Brian kind of how the field of cancer care has evolved. You know, a lot of the therapeutics traditionally have just been central factories trying to take advantage of these big economies of scale to make the lowest marginal cost product they can so that they can efficiently distribute this to patients. And while the price is not wholly dependent on the cost, certainly the more expensive the therapy is to make, it's naturally going to have just a more expensive price point. So there's some opportunities here. With cell therapy, the field is still fairly new. Most of the manufacturing, because it's been so manual to date and because the technologies really weren't, they were kind of adapted from other fields, they are still very centralized in the therapy makers that are producing these. And then there's a handful of academic sites that are kind of trained and able to deliver these therapies through an infusion and then monitoring for a couple of adverse events that might be at risk of cytokine storm. On the whole, these are very low toxicity, very safe therapies. And so the opportunity here is to try to reduce the need to have a long transport chain involved. Some of that, it may still end up having cryopreservation as part of it, but some of it may enable more fresh samples to be used. And then more broadly, there's elements within the workflow that if you can just automate them and link them together, maybe through some just basic robotics of which those are available now, then that whole workflow in theory doesn't need to kind of be central. And one of the interesting things is, you know, there's like the base of population is fairly distributed. You can have, you know, one of our advisors is Tom Whitehead, who's Emily Whitehead's dad. This is one of the earliest patients that received CAR-T and is alive today and doing well. And his comment was he gets calls from patients all the time, and they're making a decision about whether they drive two hours to go to the hospital at UPenn or if they, you know, stay at their house and go to their local clinic to get another round of chemo. And, um, always, you know, there's, there's a real lifestyle, um, lifestyle's the right word. There's, there's a real, um, both economic and, um, you know, kind of, uh, social impact, impact of having to go somewhere to receive treatment that, um, is at play. It's a real one in patients. And I think, um, you know, from the outside, we say these therapies have to potentially cure, but the reality is like, can I take off my job, you know, take out work to do this stuff. Do I need family there to support me? Like, do we need to pay for a hospital? How far is like, can I make that drive multiple, can they make that drive multiple times a day? And the reality of that is it's, it's Brian not always easy. And patients, um, Brian more and more we've been seeing and I've been hearing through Tom often should pick the path of least resistance. And it's a therapy that's not going to help them necessarily or cure them may delay onset. Um, but it doesn't get them, you know, what they need. And as these therapies move to first line, there's going to be a real problem. And so the opportunity here is to bring the therapies closer to the patient. And so that's where we've been kind of focused is how do we be able to create an automated workflow with our technologies integrated that are living either in the hospital, point of care, or just outside the hospital in some sort of either hyperlocal or regionalized decentralized manufacturing. and what are the key bottlenecks in being able to deliver that. One of our investors is LabCorp. They've got this kind of world-class capability of sample logistics. They're able to get a sample from the smallest town Brian in Texas to a central location in Illinois or Michigan and be able to process that by the next morning to have data back in your email or in your clinician's email to provide them the outcome of that data. And so there's an opportunity to leverage them as investors and them as partners for us to be able to have kind of this really rapid feedback loop on what is the quality of the manufacturing process, how potent are these cells, being able to take a lot of that characterization and just channel that to a central location with them as the analytical backend, as well as do release testing. And then from our side, we're focused on kind of automating these workflows and having them able to drop in. We've been working with some partners that are specialized in clean rooms, some of them mobile. And so we think there's an opportunity. The fields have been talking about it, trying to get these therapies closer to the patients. There hasn't been like this obvious way to do it. There's been maybe one main announcement with Galapagos and the Blood Centers of America where they're working together to try to bring this to Blood Centers of America clinics for infusion and processing. We think we have the opportunity as kind of instrument developers to be able to try to own the integration of that workflow and partner with groups that can bring it, you know, basically to the clinician and to the patient. And then this kind of black box on how do we deal with some of these requirements for rapid sterility testing and other things that are challenging to do. And you're kind of otherwise asking the hospital or having specialized equipment locally to do that. You know, this kind of outsourcing model that we're thinking about with LabCorp as important partners, I think we'll make this feasible. So we're excited to see that change from going about talking about it in the industry to actually starting to pilot Adam this approach and demo. So that's on the horizon for us. Yeah, OK, so that makes sense. You know, we're starting to have these new technologies that have better outcomes that are cheaper, but still they are far away from the patients. So it sounds like you might be saying one or both of getting better transportation logistics where you can get samples, get blood T cells from the patient to a place that can process them and back to the patient as fast and quickly, you know, easily and cheaply as possible. Same Adam time. Right. And part of that is get it, you know, is can we get it better logistics by sharing with other people who already have good logistics? And the other one is, can we get our machines smaller, cheaper, in more places? You know, I know that you're not saying this at this point, but sort of like you can imagine the pathway being getting it to the point where every pharmacy, you know, on the corner, you can take over, they can take a quick blood sample, give them some T cells, and they culture them and give you back the, you know, the results the same day. So I know we're not there yet, but sort of as we go along the spectrum, that's where things are headed and we can certainly see the benefits of how that, you know, health patients. Yeah. Well, you know, Walgreens announced an initiative. So it's, Brian you made the comment corner story that they are interested in being a front end for this. They announced that they were setting up a central facility somewhere else in the country. And so it's still going to be a long turn on the Brian process. So I think, you know, really it needs to be ideally local. So it's rapid, not just the infusion and then freezing and sending. but if you can have everything close by, you know, the truth is in a small town, you know, you don't need a huge capacity to be able to process patients that are in need of these therapies. And so if you have a bunch of little decentralized hubs that are kind of all cloud-based controlled, have the same, you know, infrastructure, same as all, you know, all automated, you're not having any kind of variability in terms of the, you know, handling or anything like that, Like you should be able to lock in kind of this reproducibility across the, you know, really across the country, but have them, you know, kind of deal with local capacity in a smaller scale way. So you don't need these big ballrooms and large facilities and tons of people there. So that's kind of the vision. We actually just announced today a collaboration with a hospital in Adam Australia, Brian the Royal Perth Hospital, with our local affiliate in Australia, Xcell Biosciences Australia, Xcell Bio Australia. And that is working directly with a hospital system there to do a rapid manufacturing on these two workflows. These are the billion plus, the 100 billion cells I was mentioning that Steve Rosenberg's lab had been working on for the better part of a couple of decades now. So, well, in Australia, there's, you know, there's a lot of melanoma patients and there's a real big unmet need and how to how to fight, Brian particularly the aggressive metastatic version of that. And TILs have shown to be effective in that. So these hospital systems are running these very manual processes, but that's just not, you know, they're not skilled at that. That's not, you know, kind of, this hospital system is very good at that, but they're looking for automation to be able to scale it and not have to have as much kind of manual involvement in the process. So, and then this is a replicatable model. They actually have a clean room and are doing it at the hospital, and then they're going to be outfitting Adam other hospitals to do this type of capability Brian across Australia. Adam Yeah, sounding great, sounding exciting, and it sounds like you're making good progress very quickly. Brian, you mentioned TIL therapy a minute ago, and I was wondering if you could just take a minute to tell us a little bit more about what that is. Brian Yeah, TIL Adam therapy, Brian I mentioned, is kind Adam of pioneered by Steven Roseburg, among others. Brian But he did a lot of early clinical studies showing that basically if you take the tumor, solve the tumor out of a patient, and you digest it or just break it up, usually mechanically just with a knife, and then you extract out the T cells from that. If you activate those T cells, some cytokines, then they often will become active and aggressive against that tumor again when you put it back in the patient. And because they've been living in that tumor, they know how to identify it, and they can be very successful in actually eliminating the cancer. So a company got approval for the first TIL therapy commercially this past year called Iovance, and it is targeting skin melanoma and has had really nice success. Not nearly as much success as what we've seen in terms of the response rates for blood-based cancers, for CAR-T therapies, but I think really exciting as a great starting point for these therapies to show that they can work, that they Brian can knock back the tumors for most patients, and that in some cases they're curing patients. There's an individual named Chris White who's one of the patients out of their studies that's alive and doing well. And it's been really great to be able to hear his stories about the challenges of going through this process, the chemo, a lot of the things that while it's extended life made life terrible in a lot of ways. And just the joy of finding this therapy that has just eliminated the disease completely. And it's unusual. Everybody knows somebody that's been touched by cancer Brian to see the disappearance of the disease when you can't cut it out. That's what people think about curing right now. It's like, can you cut it off? And so this is, he said that he still has these calcified places you can see, I guess, probably with an MRI or a CT scan. where the tumors used to be, where the necrotic core of the tumor was. Those still exist in his body, but the cancer is gone, and the thing that was killing him is gone. So it's pretty amazing. And I think there's likely a Nobel Prize coming for this technology. And I think skin melanoma is just kind of the tip of the iceberg. There is a whole range of solid tumors that there's opportunities for this to be successful in. But it won't be successful if it requires 100 billion cells and a month and a half of culture and, you know, many, many leaders of media. So as the technologies get better, I hope ours is at the center of it, as technologies get better for expanding these and culturing these cells to be effective. And as the therapies get more potent so you don't need as many, you know, there's an opportunity to wrap this into a really nice, tight, automated workflow. And then, you know, there's supply side, you know, the demand side is millions of patients. If these therapies continue to be successful, it's a tall tumor. And you're just going to need technologies that allow that to scale as much as you can. So all the groundwork we've been talking about with decentralized distribution of these therapies and decentralized manufacturing, the kind of shortening of the timeline of culture, the automation. I think all that's going to be really important to have that tool set ready to be able to employ and be able to deliver Adam TILs at the scale that it's going to be delivered at. That makes sense. Can I just ask one clarifying question? You mentioned, in this case, taking T cells from a patient and a solid tumor cells and activating them with cytokines, and then they go back in and start fighting more aggressively. In this case, are they just activating cytokines, or are they also genetically modifying them before putting them back in the patient? Well, there's 1.0 Brian and 2.0. So the TILS 1.0, which is what was approved this past year, they don't actually do any genetic modification, which is one of the reasons the hospital in Perth can just work on that process themselves. They're not violating IP per se. There's certainly clinical thresholds. You want to show that your process is going to not create problems for patients. And so in Australia, they have hospital exemptions. And in Europe, they have hospital exemptions. So these patients can be enrolled. The hospital is getting paid to treat the patients. And they're kind of generating this clinical base to show that the process is effective. In the U.S., they don't really do that at this point. So you've got to do the clinical studies. You've got to pay for it. And then once you get approved, you can start charging patients. so it's happening more naturally these are the places where the hospitals can start receiving money immediately for the work they're doing and in some places it's also grant funded and the governments are investing in this but for and so for 1.0 there's some variance to the process but they're kind of all driven off of the NIH process that was developed by Stevens the 2.0 there's some interesting scenes happening where they are looking for ways to to put a payload on them like you would with a car and do genetic modification to deliver that. And that's kind of the, that's where a lot of the companies are Brian focused right now that are working on tills. I think, you know, our technology and the opportunity that we're looking at is that, you know, this 1.0 workflow is really effective. It just sucks. You know, it's just a really big hassle, very hard to keep as reproducible. And it's just that it's a very large scale. So from a process standpoint, we can make that process very consistent, short, kind of rapid workflow with small numbers of cells. It starts to look different. And you may be able to increase outcomes just through that. And then obviously our technology, we focus on making the cells more potent, which is one of the reasons they're using so many of these cells. So if you can make the cells more potent, maybe you can get those complete response rates up just in this 1.0 generation. And then the technology will obviously work well for 2.0s as a sponsor, working with sponsors with our technology is the format. Brian, thanks so much for taking the time to come and talk with me today. This has been really interesting, and I appreciate you telling us all about it. I appreciate the opportunity and the platform. It's an exciting area that cancer therapies have managed to move into over the last decade. When we started the company and as I went through my own academic training, the excitement was around small numbers of targeted therapies that seemed to have impact on portions of indications and several month life extensions. And I just think we've really moved into a new era where there's, I think, a lot of hope. It's not quite a silver bullet, but the immune system is a very sophisticated organ, I guess. And that menagerie all have specific roles, communicate in different ways to each other. And we're just really learning about that in depth and starting to be able to exploit the things we're learning to provide these really amazing outcomes for patients. So I think we're just right at the beginning of immunotherapy and kind of cell therapies in general. And things like CRISPR and the gene editing tool sets we've been making strides with over the last decade have enabled a lot of that. Large genomic capabilities for sequencing have enabled a lot. So the tools that we've kind of started to pull together over the last decade, two decades, are really starting to bear fruit Adam for patients. Yeah, that's fantastic. It's a great place to be and so many things to look forward to and work towards. Well, thanks again, Brian. My guest today was Brian Feth, CEO and co-founder of Xcell Bio. Thanks again, Brian. Brian Thanks, Adam. Appreciate it.

Transcript Chris Delightful to introduce you all. Some new faces. For you, Brian, if this was a relationship, this is the 12th anniversary, which would be silk, some sort of silk gift. Brian I didn't Chris get anything for you either, so we're all square. So my name is Chris Spivey. I'm the Director of Industrial Relations and Strategic Alliances for the Pharma Sciences One Group at MJH. Matthew, I'll ask you to introduce yourself. Yeah, I'm Matthew Matthew Lauber. I am a senior director at Waters Corporation. Waters is known for chromatography the world over. Our chromatography serves a lot of the chemistry, manufacturing, control work in the world for pharmaceuticals, and I oversee a portfolio that develops and then supports the chemical and biological tools, including the things inside of the columns of HPLC and the like, for the analysis of biomolecules. and expanding portfolio, giving you a recent promotion as well. Tom Tom? Yeah, hi guys. I'm Tom. I'm the Chief Commercial Officer of Ori Biotech. So we're a technology company in the cell and gene therapy space looking to enable patient access to cell and gene therapies. I've been in the cell and gene therapy sector about 15 years now, so it's great to be here. Looking forward to the discussion. Yeah, it's Brian good for the first thing we'll talk about. Brian? I'm Brian Fath. I'm the CEO and co-founder of Xcell Biosciences. And we are a technology business also in the cell and gene therapy space. We focus on therapeutic potency and cell therapy products and have designed a suite of technologies to enable improved potency and measurement of that potency. Chris Yeah, with interesting sensors and analytics, which comes into this conversation as well. So not necessarily for comment, but I would like for the record to point out that Bill Cassidy quickly slow walked his conversation about the HHS secretary, RFK Jr., and the revision or re-examination of the childhood vaccine thing. So he was a holdout for that Senate confirmation. And just for the record saying, I'm disappointed that he was so quick to do his Neville Chamberlain appeasement speech. So a lot of news headlines and a lot of political stuff, so feel free to intersperse them. The first thing really though, Sel and Gene, that I want to talk about is Pfizer has discontinued its hemophilia treatment, BECVAS, and it was the last in its portfolio. In conjunction, one of the pioneering companies, a local store, Bluebird Bio, was sold for the princely sum of $30 million to the Carlyle Group, which has more money than God. So the prompt really that I like to sort of look at from a couple of different angles is the struggles potentially that the gene therapy commercialization path has taken, but also its relationship with big pharma. So Matthew, I don't know if you have a feeling about that. Yeah, Matthew many feelings. Pfizer certainly made their choices, but they're very rational choices once you begin to dissect it all. I was thinking over the weekend about how I was at the Pfizer Symposium talk at ASGCT last year, and that was just a month after the commercialization. A couple interesting things. First, the science, the clinical science was amazing. They featured well-known Dr. Roland Herzog. So it was clinical this and efficacy of that, pre-existing immunity, very very logical approaches and built on lots of data hugely impressive and they were already even using a special serotype of of a b for the um backbiz so that's the starting point that wasn't that's not even a year ago yet but really seems for rational reasons Pfizer's made their choices we could probably share a panelist discussions on on why um big pharma is a it's going to play a big and cell and gene therapy going to market, but large pharma will also very dutifully usually set up multiple programs. So in the end, it comes down to medicine and choices and the receptiveness of the patients, in which case there's a monoclonal antibody here to be talking about. That's a good point. But Chris haemophilia is not a small rare disease either in some senses. So Tom? Yeah, I agree. I mean, I've actually had the pleasure of working with Bluebird for the past Tom almost 12 years now so they were a client of mine in my previous business and you know i think they certainly weren't short of very talented people very purpose-driven people very energized um but like you say i think starting with the science last year i had the pleasure of meeting a guy by the name of jim jimmy ola hair who was one of the first recipients of the vertex product for sickle cell cast jav during clinical development so the rival product to the bluebird product. Last year, end of last year, he climbed Kilimanjaro, the first ever sickle cell patient to climb a mountain. Anything above, I think, 8,000 feet in elevation is basically a death sentence. Kilimanjaro is 20,000 feet. So I think it's hard to argue that the scientific efficacy, these products are remarkable. But the issue is commercial viability. We haven't yet focused on how for selling gene therapy, we can get commercially viable products. And most of that is rooted in they're too expensive to manufacture, they're too hard to manufacture, they're too low quality, the costs are way too high. And my view is that until as an industry, we start to focus on this problem, we start looking at just therapeutic efficacy, these products, we look at actually manufacturability and commercialization, then we're always going to be stuck repeating this same issue we've got with Bluebird, where the products, they work, they get to market, they have three commercial products, remarkable, and they're sold for $30 million, like you say, it's a travesty. Yeah. I mean, it doesn't look like it's getting better anytime soon. Um, it's going to Brian continue to be tough out there for most of the reasons that Tom just said. There's some other elements as well. And I think, um, you know, one of the things that needs to be mentioned here is that they had some factor nine expression over time that seemed to indicate that the efficacy was waning. Um, and that's, you know, also a problem that, um, you know, their concerns about broadly for gene therapy. How successful were you in inserting the genes or modifying the genes that you needed to modify? And you're paying a high price for what might be diminished efficacy and requirements to redo the therapy over time. And particularly if you've got alternative therapies that can at least be effective in reducing symptoms, a lot of patients will go for that. I think one of the interesting things, we also have worked with Bluebird for quite some time, And, you know, they have a much more rigorous assay than the Vertex assay. They both use a protein-based assay to assay for whether or not the protein exists and to what abundance. But Bluebird uses a cell-based assay as well that allows them to have a much more high-fidelity, low-variance assay to determine if they successfully modified the genes. And I think having potency assays is one of the real challenges in this field and one of the real needs in this field in order to kind of demonstrate that you effectively edited the genes and ultimately that you're not going to have to redose. Now, there is a model now of kind of how some of these therapies could be treated if the FDA were to kind of realign itself around these therapies differently, which is really a success-based outcome. It's, you know, surgeries are often paid for over time. They're paid for based on success. I mean, maybe there's a model here that's some sort of hybrid where you could regulate it more like a surgical procedure, given that these are really custom individualized therapies and, you know, allow the hospital to have some kind of benefit in that outcome as well, because you do really need intensive monitoring over time and you need everyone to be aligned around the kind of inevitable success of these with a single dose. So I think that there's going to have to be some changes at the regulatory level as well. Chris That's a very interesting comment. And I would say an astute comment. I remember Peter Marks talking about potentially regulating CRISPR as a medical device for similar reasons, but he wanted to keep it within his own CBER group and not Matthew move Chris it out to the device group. But no one mentioned genome integration or off-target effects. So I think on the science side, we're all relatively comfortable with where we're at. It's more the business side. There's still a Brian list that's being priced in. Chris Oh, yeah. So for the time, Alan, I'm sure we're going to return to this conversation. The next thing that I want to talk about, which was very exciting for me, was comparing Roche, Illumina and a company called Icon. So you've got for Roche, I think that came from Stratos Genomics. It was, to me, echoes of their own PCR, where you're just amplifying or expanding a signal to get a better signal to noise ratio. And I'll contrast that to with its first of its kind spatial transcriptomics technology. And then ICON really, to me, stood out as an interesting new type of physics. You're using light. It's a single molecule tracking system. And it's oblique line scanning imaging, which to me is interesting if you can now visualize protein movement for various different reasons. So, Matthew, I'll start with you. It's a lot to unpack. Just pick and choose. Matthew Let's start with Roche. Interesting in the genome sequencing, what goes around comes around is probably a flippant way to be talking about this, but good for Roche to be dedicated to watch and nurture technology trends after being displaced once upon a time. So dial back 20 years, Illumina was the one doing it to them with their plans for the 454 life sciences approaches to pyro sequencing. So as some have said, technology disruption for an incumbent can kind of feel like a slow wrecking train. If you're on the tracks, you can maybe see it coming, but it might build momentum. But I'm blown away because I'm a chemist and biochemist by training. these X NTPs that they use in the expansion step of this, just fascinating. Any video online will show you this. I recommend people go check out the sequencing by expansion Tom videos from Broge. It's really cool stuff. Thank you. Tom? Yeah, I think what struck me about the news that, I mean, we're seeing across multiple sectors in biotech is really companies that are deploying advanced technologies to get deeper understanding of the fundamental science, whether that's during manufacturing or in this case during early development, and then how they're able to harness that data to either speed up drug discovery, reduce the cost, increase cycle time. I think we're seeing right now with a lot of investors out there that most groups in biotech, whether you're a therapeutic developer, you're a technology developer, whatever you're doing, having a very strong point of view on the digital aspect. So how are you, what are you measuring? How do you collect the data? And then are there ways in which you could apply large language models or other mechanisms to then start interrogating that? And this news, I think, is very pertinent, particularly for me in cell and gene therapy, where we're talking about the most complex of possible drugs, which are human cells. The deeper we can get of an understanding of that and use that data to better inform of targets or efficacy or all these other things, I think that's the next wave. And like I said, I think most companies right now need to have a point of view and Icon showed that with their fundraising, particularly that they got a huge value ascribed to the work that they had done in this space. Chris Yeah, 350 million for a Series D, it was caught my attention as well. The CMOS chip for Roche is one of the selling points. It did seem to be speeding up from days to hours the process, but I'm not sure if it gave more biological Brian insight, but maybe I misunderstood that. So, Brian? Yeah, excuse me. Yeah, it did seem like there's going to be at least a higher fidelity readout and a little bit better epigenetic profiling capability. As to how that enables science, I mean, I think time will tell, but it does seem like there's going to be at least increasing in speed and dropping a cost that's going to be a consequence of this. And I think, you know, there's been arguments about whether or not going cheaper is going to enable the field in more new ways. I think it will. I think that there's always opportunities to develop new applications if you can do it for a dollar instead of a hundred, instead of ten. I think there's an opportunity to really do some new things from a broad scale sequencing. And Illumina has had their market share chipped away gradually over time, which has been interesting to see. They were such a strong, dominant force for so long. Having another large incumbent in this space beyond kind of PacBio and some of the other groups that have come out, I think is a real threat to their business and is going to ultimately just continue the kind of price pressure downward, which is good for consumers and all the application use. On the Illumina side, especially transcriptomics, there's a number of different players out there. There's some incumbents like 10X, CS Genomics, I believe is another smaller player that's in the space there's there's a couple different applications it's clearly a hot field and is going to be really enabling for for some science to be able to as Tom said be able to get deeper and understand exactly what's happening in what areas of the tissue it's just going to enable it's a discovery tool at this point but really going to enable some new particularly in oncology some new approaches to therapeutics. Chris I saw a good 10x genomics presentation at the Precision Medicine World Conference that was updated what I didn't currently understand was possible, which was impressive. Speaking of the science, let's dive a lot deeper into an investigational antisense oligonucleotide, which is a mouthful. So this is Stoke Therapeutics and Biogen's new relationship for Dravet syndrome. So Matthew? Yeah, this one I know pretty well. It's Matthew in my backyard here in Massachusetts, not too far down the road, so therapeutics. But what's super interesting about their ASO work is that they are using ASOs to upregulate transcripts. It's a very, very unique use and really is expanding the boundaries of what are you using ASOs and SIRNA or synthetic olgotherapeutics for. It's not just about knocking down expression. So this one's quite neat that the CSO, the chief scientist behind us, she's been at work to figure out the pre-elements and their interactions on the mRNA to ASOs. So super neat. It's about a 20-mer. It's extensively modified with methyl and methoxyethyl groups, just like many others, to be optimized for binding and nucleus resistance. So a lot going on there. There definitely is a wave of oligotherapeutics coming after a little bit of gestation period over the decades. Chris Yeah, Biden has been sitting back to some degree given some of their recent sort of issues with some of their Tom neurogenerative drug type therapies. Tom? Yeah, I mean, equally, I mean, ASOs are relatively new to me. I'm sure they're not new to the field. But I think looking at some of the data, particularly in neurodegenerative disorders and dementia, I think is super exciting. I mean, there's an awful of vastly unmet need on that side and really no drugs or treatment. So anything that can address that, I think, is really interesting. The other thing that piqued my interest about this, and it kind of relates to the Bluebird discussion as well, of at what point should Big Pharma be coming in to bootstrap and support some of these companies in development? I think broadly we've seen a lot of positivity with pharma coming in at this stage to A, provide the capital, so to give them a runway, and then B, to support with some of the commercialization efforts for these products. Because I think for a lot of these companies that are doing it themselves, I mean, we like the phrase they're flying the plane and building the plane at the same time, right? They're building all these commercial systems and processes at the same time as trying to execute a commercial strategy. And I think that's what piqued my interest here with Biogen coming in, like you say, dipping a toe back in. It's really great to see that because I think they're going to add a lot of value. And if we can get some more of these therapies out, commercialized and viable, I think that's it's good for all of us. Chris Thank you. Ron? Brian Yeah, big up front, which is nice to see. The total deal size looked pretty good. I don't know what the total market size for the therapy is. Hard to say for me as a ratio, but it looked like a good deal. And, you know, I think the interesting thing about these drugs is, you know, they're more complex, they're expensive, they require, you know, kind of this spinal injection administration. And there's some long-term risk that's uncertain, but they are actually, you know, adjusting the disease physiology or the disease pathology, whereas some of the kind of existing drugs out there are really just treating symptoms, reducing seizures. So I think it looks like an exciting Chris drug and I think a nice addition Brian to see kind of an advanced therapy go into a portfolio at a good price. Chris Good. I'm glad to hear that. All right. So this next one I struggled to get people to go on record commenting about, and I'm not really sure why, but it's Thermo's sort of acquisition of solventum. That's 3M's filtration and purification business. So $4.1 billion. I know plenty of people, maybe too many people, active in this space. No one really wanted to go on the panel to talk about this. So, Matthew, what do you think? And I'll preface it by saying that it's not a trivial choice. Let's look at Ecolab and Cytiva and some of the other suppliers. Once you make a choice, it's part of your regulatory filing. So it's no footsteps backward from the cave, the old Latin phrase. Once you've made that choice, it's hard to return from it. So, Matthew. Matthew Well, I think it's Chris a Matthew lot about filling portfolios and making sure that you're horizontally exposed to the right growth areas. And Thermo, we know, does this very, very frequently. So it matches to some of their CDMO ideas with some of the basic technologies, membranes, purification media, extremely valuable pieces. And it's not dissimilar to things that the industry has seen Danaher choosing to do, like we talked about. Even on the basics of making sure inside of these massive columns that are generally used for upstream or downstream polishing, that you need the right stuff inside, so you look for the value-added additions that one could be finding. So membranes or monoliths, maybe we could even go back in time to look at Sartorius building out their capabilities here for monoliths. monoliths from via separations, but it kind of all sits. It fits the same mold. Paul, Tom Paul? Yeah, so I've been following Thermo Fisher for a long time. They have a pretty deep footprint in the cell and gene therapy space. I think I saw a lot of this is a redistribution of the capital, right? They had deployed a lot of capital pre pandemic and post pandemic into their CDMO businesses. I think they had doubled the number of people in a very short period of time. Those were extreme high growth areas coming out of the pandemic, which as we've been discussing have started leveling off. And so we've seen a lot of reductions across their viral vector businesses, across their cell therapy CDMO business and some of their other groups to, I assume, redirect the flow of capital into some of these areas that they expect are going to be more long term high growth. So I think a lot of talking about before, it's definitely not a good indication for the viability of some of these other modalities where you see big companies doubling down on existing approaches. So yeah, I thought it was very interesting. And like you say, with others like Cytiva and many of these others, it's worth watching out for because this could be coming in other groups as well, I Brian suspect. Chris Yeah, it's kind of like architects being busy being a good barometer for the future of the economy in some ways. So Brian? Yeah, I mean, having Brian some M&A in this space is great. And for both Tom and my company, having bioprocessing activity is a good thing. Maybe the bad news is it was 4X revenue off of the billions, so it wasn't necessarily a stellar sale price, but kind of digging in a little bit to try to understand what might have been the things that Thermo was looking for because I think they're going to take some pieces of it and probably jettison the things they don't want. And I think the things they're looking at, first of all, monoclonal antibody demand is really skyrocketing. and there's kind of this micro filtration and ultra filtration technology that helps with that and that's that's a decent multiple i think in that business there's single use filtration membranes cartridges those are also good multiple businesses that's useful for biologics and cell therapy and that market's growing at like 15 so there's good good growth in that market and then viral and mRNA purification technologies also get multiples and you know the the biopharmaceutical industry needs some specialized filtration technologies for this or purification technologies for this so um and our vectors continue to to be a driving you know a growth driver for the business so um i think you know at least those three areas seem like they're higher than 4x multiple in terms of what they're getting um and so it may Chris be Brian that that there's some some special interest in some of the areas within this business for Therma. Chris Very interesting. So the last bio in San Diego, antibodies was the big storyline that ties it together. So speaking of tying things together, we're going to revisit again, GLP ones each time. I don't particularly want to, but it hits the news. So thanks to potential tariffs and I guess other reasons, Lilly is planning on spending $27 billion, which coupled to a few years ago, a similar amount to expand its factory footprint in the US. So, Matthew. Matthew Yeah, it's been exciting to see what Lilly has done investing even in the Midwest, which is where I'm from, what you're talking about. To see them making extra choices this year for supporting the pipeline for their GLP ones. It's equally exciting. I think the biggest news will come when and how they establish some of their product lines for the orally dosed GLP ones. So looking forward to that. I've also been related digging into some of the pipeline that has to come next. It looks like quite a few are working on amylin, which is another dual agonist and some sequence that can be incorporated into some next generation medicine here. So Chris perhaps all Matthew of this will help for even more of these multi agonists in the future. Chris Yeah, yeah, no, it's a pathway Tom that's going to keep on giving, I think. Tom? Yeah, well, I see this as a little bit of carrots and sticks, right? We had previous carrots, we had tax incentives for companies to move manufacturing to the US and Lilly and others were taking advantage of that. And now, unfortunately, we have some sticks which are the possibility of international tariffs on companies so there's probably a good bottom line reason to move manufacturing to the us today i think it's also interesting looking at this being a weight loss drug and reduction of price i think you know the biden administration did put rules in place against medicaid and medicare so these weight loss drugs do have to be under the guise of medicaid medicare so companies looking to bring down the price to get them in the insurance bracket, I think is probably a very viable strategy giving us Zenpix the second biggest drug in the world currently, which is interesting. And I think we're seeing other companies as well. I know Apple and many other companies have started already saying they're going to move manufacturing to the US, which is probably reaction, like you say, to tariffs and sticks. But I'm sure we can get into a political debate in the last three minutes of this discussion, whether that's a good thing in the long run know. Chris Well, I'll egg Brian on in the sense that part of this was predicated upon the requirement of the tax breaks being extended. So there was a proviso in that, Brian. I meant to mention that, so well good for you for pointing the stick and the carrot out, Tom. Brian I mean, Lilly is the most highly valued pharmaceutical company in the world by quite a lot and I think they're really focused on trying to figure out how to maintain that over the next years and continue to grow it. And the main drivers of that business right now are the weight loss medications that they've got with the narrow and ZepBound. So I think protecting those assets and we've heard a lot recently about these compound pharmacies that have been producing these therapies because the demand is just so ridiculous. I know that they expected a lot when they launched these therapies a number of years ago. But, you know, it's become lifestyle drugs in many ways beyond just kind of therapeutic treatment for obesity. So, you know, I think they're trying to keep up with demand. They've seen erosion both of the prices that they can offer because of these alternative sources and just kind of an inability to keep up with the demand requirements has kind of shaken some of the investors' confidence in their ability to deliver. So I think trying to kind of reinforce just through scale to be able to continue to provide supply over the next year, which is really important for them, which is why they want this route. So it seems like a fairly direct line. You know, they had a big problem. They're still fighting compounders, even though they're no longer a shortage. And so being able to have sufficient capacity is one step, and then trying to protect their pricing is another. Chris Yeah, I think it was taken off shortage recently, and that meant that you're not allowed to compound. if I understood that correctly, which is... There's another group Brian still Chris making these. Brian I think that some of the controversy around Hemser, some of the online marketing had to do with this, that these groups are fighting that there's not sufficient supply and want to continue to exist to deliver these. Chris Excellent points. I'll throw it over to everybody for last thoughts or anything that we want to cover. But before I do, what I had meant to mention when I talked about the revision of the childhood vaccine schedule, Moderna's shares continue to plummet. And that's part of that whole thing where vaccine companies are probably looking at some sort of ontological threat from all this. So any parting Matthew thoughts, comments, Matthew? I hope those mRNA developers carry on quickly through the clinic on their neoantigen therapies. I think this will be very useful to patients and effective. Hopefully they can get to it. Tom Thank you. Tom? Yeah, I think for me as a longtime advocate of the cell and gene therapy industry, my parting words are clearly these therapies work. They work very well, but we need to refocus on commercial viability because stories like Bluebird is not going to take many more of these before investors and pharma start to lose patience in the industry. And so I think we need to redouble efforts on commercial viability and really start looking at how we can turn cell therapy from a cool science project into a viable business. Matthew Thank you, Tom Ron. Yeah, policy stone throwing at science isn't helpful. And what Brian we really need is alignment of the politicians that understand the science, the pharmaceutical industry, and the scientific community to stand up and advocate for medicines that are effective and have been for decades. And the fact that the vaccine industry is under threat and people are questioning the science, I think we can't let people get away with that. It really needs to be a unified voice around the real value of these therapies and the lives that they're saving. Chris yeah actually one of rfk's comments about all this was one um measles outbreaks are common which is not true and then two oh you might want to like uh do the therapy rather than the prophylactic super appreciate your time great conversation very interesting points of view so thank you guys appreciate you being here Brian thank you thanks Chris chris Bye.

Transcript Hello and welcome to this IBTV spotlight on the rapid delivery of cellular therapies. 00:13-00:18 I'm joined here today by Sean, Edwin, and Shannon to discuss this. 00:18-00:22 And to be thinking about really how we can get cellular therapy to patients who 00:22-00:25 just don't have the luxury of time. 00:25-00:32 So to get started, discussions here at Advanced Therapies Week in Dallas this week 00:32-00:37 have suggested that as many of 20% of patients eligible for cell therapies 00:37-00:43 are actually dying while on the wait list due to manufacturing bottlenecks. 00:43-00:45 What is contributing to these delays? 00:45-00:47 >> Yeah, absolutely. 00:47-00:53 So it's a shame that many patients don't have access to the therapies that 00:53-00:56 they need to save these kind of diseases that they have. 00:56-01:01 The number one, I think, we see on our side of the things is not having enough 01:01-01:05 capacity to manufacture these drugs, both not locally where the patients are 01:05-01:08 located or even at a centralized plant kind of concept. 01:08-01:12 The other part I think is the administrative delays on getting approved for 01:12-01:16 these kind of therapies, getting the funding to fly to where these therapies 01:16-01:20 are actually manufactured, kind of administrative side of things that 01:20-01:21 are a huge problem. 01:21-01:22 >> Yeah, absolutely. 01:22-01:28 And I think actually, in a way, the 20% understates the magnitude of the challenge. 01:28-01:32 20% is the people that actually got on to the wait list in the first place. 01:32-01:36 If you actually look at the gap between the patients we could be serving and 01:36-01:39 the patients we are serving, it's probably about a factor of seven, 01:39-01:43 where seven times as many patients could be served, it's being served a day. 01:43-01:48 And in a way, slightly perversely, because of the incredible progress in biology, 01:48-01:52 that gap's getting wider and wider every year. 01:52-01:57 So we've got this, I think it's really important, this is not a marginal problem 01:57-01:58 that we can deal with over time. 01:58-02:01 This is an urgent problem that needs sorting now and I absolutely agree. 02:01-02:04 It's about manufacturing capacity and no more, no less, 02:04-02:06 we've solved the biological challenge. 02:06-02:10 We need to solve the challenges of getting therapies to patients. 02:10-02:14 Exactly the same challenge we saw with things like COVID vaccines just a few 02:14-02:20 years ago, that now we've got so good at biology, what's really important is 02:20-02:23 that we can solve manufacturing and we need to very quickly address those 02:23-02:26 bottlenecks and get those therapies out there. 02:26-02:28 We are going to help patients that can benefit from these therapies. 02:28-02:35 Bridging upon that, a lot of places that the patients are rural don't always 02:35-02:39 have access to large cities to get to and some of the major cities and 02:39-02:40 the major cancer centers. 02:40-02:47 So kind of building on that, manufacturing in one area or even regionally 02:47-02:51 is still not going to fit the void that I had said that there's so many 02:51-02:56 patients that just Tom Whitehead was down the hall this morning saying 02:56-03:00 the same thing he gets calls and they just can't make the three, four hour 03:00-03:04 trip to get into that major center where we've really got to solve some 03:04-03:05 of the scalability issues. 03:05-03:08 Excellent, thank you. 03:08-03:13 And Edwin actually to your point, the eligible patient population for 03:13-03:17 cell therapies continues to grow and is possibly beyond what our frame of 03:17-03:20 reference is and beyond the statistics we actually have to hand. 03:20-03:24 But how do manufacturing delays and the inability to rapidly scale 03:24-03:29 directly impact patient outcomes, especially for those with life 03:29-03:30 threatening conditions? 03:30-03:37 It is a new modality we're working with and actually, yeah, Shannon makes 03:37-03:41 a good point is that we didn't used to have to think about where 03:41-03:43 therapies were made and things like that in the past. 03:43-03:47 It was it was all a very tried and tested route and of course these are 03:47-03:51 different, they're individualized therapies and so that brings both 03:51-03:55 challenges in the manufacturing paradigm, but also in terms of the impact 03:55-03:59 it has, it's enormous because it is a dose per patient. 03:59-04:06 So the inability to manufacture that dose firstly can be a death sentence 04:06-04:08 for these patients, the patients. 04:08-04:12 But secondly, it's a time criticality to it and that's something else is 04:12-04:15 it's not just as simple as is that therapy there or not? 04:15-04:20 Is that therapy available in the window that the patient can receive it? 04:20-04:24 And this is an incredibly complex delivery. 04:24-04:29 It's in the case of autologist therapies, we need to schedule the 04:29-04:32 aphoresis, we then need to search for a manufacturing slot, we then need to 04:32-04:37 get that back to the patient, we need to prep the patient in that interim. 04:37-04:41 This is a homework law more complicated than delivering chemotherapy. 04:41-04:44 And so fundamentally the impact of these things unfortunately is 04:44-04:49 patients are dying, the biology is there to treat them, but we're not yet 04:49-04:53 got the systems in place to be able to deliver the therapies and save them. 04:53-04:57 And adding upon that, the manufacturing processes for all these 04:57-04:59 therapies are all different, right? 04:59-05:04 So the challenge is to even if you were to put these there, the 05:04-05:07 manufacturing process is alone and we're not talking about the logistics 05:07-05:12 for all materials or, you know, they're QA, QC, making sure that's 05:12-05:16 standardized amongst the therapy, but then you have multiple therapies. 05:16-05:19 So, you know, there may be one for a LL, another one for sickle cell, all these 05:19-05:20 things. 05:20-05:25 So not that the field is trying to standardize those because we want 05:25-05:28 to maintain the movement that we have in getting these therapies out 05:28-05:30 and more trials, et cetera. 05:30-05:35 But it's someday we will have to bring that world together and standardize 05:35-05:40 some of these manufacturing processes, even amongst autologous, you know, 05:40-05:44 allogeneic, NG therapies, there's got to be a standardization in order to scale. 05:44-05:49 And just to add to that, actually, is I think the standardization of the 05:49-05:53 process parts, trying to do that in a way that's, I tend to think that 05:53-05:56 standardization of the engineering that goes around is different to the 05:56-05:58 standardization of the biology. 05:58-05:59 I hate the idea of standardizing the biology. 05:59-06:03 That, to me, is the route to limit the innovation that can happen in the 06:03-06:03 biology. 06:03-06:06 And I think if there's something the last 10 years have taught us is, 06:06-06:08 from a biological point of view, we've become more and more capable. 06:08-06:11 But I think absolutely, we then need to standardize the things that deliver 06:11-06:12 the biology. 06:12-06:14 And I think we need to separate those two things out. 06:14-06:17 Otherwise, we create this limitation on what the therapist can do. 06:17-06:17 Yeah. 06:17-06:22 And then, I mean, that engineering set of things that equipment that goes 06:22-06:26 into it, train that personnel, the technicians that are in the lab space, 06:26-06:31 to flexibly and quickly adapt to the different treatments and therapies 06:31-06:34 effectively, efficiently, and with high quality is important. 06:34-06:40 Training is a huge gap in the field as this conference is well aware of. 06:40-06:46 And so as other groups, whether it's a lab core request and some of these 06:46-06:49 locked signs, just taking material, that's a challenge on its own. 06:49-06:54 So to think how you're going to do that at a scale for manufacturing and all 06:54-06:58 these other pieces, we need to learn from the things that they've done 06:58-07:02 that they've been successful on and bring those into our world. 07:02-07:02 Yeah. 07:02-07:05 I think it's something that's often missed when you think about a labor. 07:05-07:07 We talk about there being too much labor. 07:07-07:09 And we can cost that per hour. 07:09-07:11 And we can put that into a spreadsheet. 07:11-07:15 But it's easy to miss out that the fact that to get to the point you've got 07:15-07:19 someone that can actually manufacture these therapies is months of training. 07:19-07:23 And there's the back end of that, unfortunately, that again is often 07:23-07:27 neglected. It's not trained once and that person's then manufacturing 07:27-07:28 therapies for the next 20 years. 07:28-07:32 Unfortunately, we know in this space that average tenure and GMP is short 07:32-07:35 by some numbers, maybe 18 months. 07:35-07:41 Now, if you've got six months training, 12 to 18 months of useful actual 07:41-07:42 productivity, it's unsustainable. 07:42-07:43 It's unscalable like that. 07:43-07:49 And that's unfortunately the bits that then fall outside, I think, of the bits 07:49-07:52 that we often kind of see in computer spreadsheets are often neglected. 07:52-07:56 That training piece and the security of tenure piece, all of that really has 07:56-07:59 to be factored in when you think about the manufacturing piece therapies. 07:59-08:02 But if we think about the speed of manufacturing, so if we take kind of 08:02-08:05 labor outside of that, as you say, months and months and months of training 08:05-08:09 required in order to get to that level, where else within the manufacturing 08:09-08:14 kind of workflows or let's think of kind of the systems as a whole, 08:14-08:18 where is the kind of opportunity for potentially innovation or for just 08:18-08:20 accelerating that speed to get these therapies to patients? 08:20-08:25 Like is there a space that we can say actually there's somewhere ripe 08:25-08:28 for innovation to make this quicker? 08:28-08:32 Yeah, I would say just on the engineering side too, is the field has 08:32-08:38 relied heavily on equipment, engineering processes that were very much 08:38-08:41 biopharma based for MABS and CHO cultures. 08:41-08:45 We've really in the last five years really kind of moved away to that where 08:45-08:48 technology innovators to use the innovation point to that 08:49-08:51 are making better technologies. 08:51-08:53 And so we've seen just in the last couple of years, 08:53-08:57 specifically with the hemological cancers, they've been able to bring down 08:57-08:59 the manufacturing process to two or three days. 08:59-09:02 And we hope that rolls out into solid tumors and some other things. 09:02-09:07 But really, as I said, the biology is the key to this. 09:07-09:12 So we don't want to take away from, just to reduce the amount of time, 09:12-09:16 we don't want to take away the efficacy of the product at the end of the day. 09:16-09:20 So what you'll see more in the next couple of years are technologies that are 09:20-09:25 bringing in AI, machine learning, bringing in making more potent cells quicker. 09:25-09:30 Whether that's through media, bioreactors, whatever that is. 09:30-09:35 And then that goes into, as we're making these new facilities, 09:35-09:40 training for these new facilities that we as a field make sure that the whole 09:40-09:44 organization's keeping, our whole world is keeping up with those advancements 09:44-09:47 so that we can, one of the failures that we made 10 years ago really was, 09:47-09:52 Ted's point, we think it would be an easy tech transfer of this process from 09:52-09:54 even one side of the pond to the other. 09:54-09:56 And it's all paperwork and raw materials. 09:56-10:01 I mean, and it ended up being much more than anyone thought in that ads time. 10:01-10:06 And all the levels of inter-seeing and making some of these products, 10:06-10:11 specifically in two different regulatory environments, we're getting better at it. 10:11-10:18 And I think we can bring all those coming into one world with one mission. 10:18-10:20 And the mission needs to be, how do we treat more patients? 10:20-10:22 Back to your first question. 10:22-10:23 That needs to be the driving goal. 10:23-10:26 If we're doing something else, then we're really missing it. 10:26-10:32 Yeah, I think it's actually, I see a real positive that so far we've managed to get 10:32-10:35 as far as we can, repurposing equipment that was never designed for this. 10:35-10:39 So I think it's great that we actually now have tools that people are developing for the ground, 10:39-10:42 from the ground up for what we're trying to do now. 10:42-10:46 And that is going to mean we can make big strides forwards in a whole load of things, 10:46-10:48 including things like turnaround time. 10:48-10:54 And the same applies actually to the processes we're developing 10:54-10:57 and how we're applying this is we're at the beginning, 10:57-11:00 we've got still a lot of improvements to make and absolutely, 11:00-11:05 it's great that suddenly we're at a moment where we've got these new tools, AI, 11:06-11:09 better analytics tools and better ways to capture data to see these models. 11:09-11:13 That means that we can start to make some really big strides forwards. 11:13-11:17 And the final one actually, I think in terms of reducing that is 11:17-11:24 with the moment, because we're so labour heavy, we have got a huge amount of variability. 11:24-11:28 We talk about variability an awful lot of that gets attributed to the factors of patient-specific 11:28-11:35 product. And yes, a chunk is, but any labour-intensive process always ends up variable. 11:35-11:41 So coming from an automation background is kind of the golden rules of automation are, 11:41-11:44 you do it for one of two reasons. You do it for capacity, 11:44-11:46 which is actually the one we talk about most, but you also do it for quality. 11:46-11:50 And it's that second one that I think from the point of view of actually 11:50-11:54 increasing how quickly we can get to patients, increasing the quality of the product we can 11:54-11:58 produce, it's not necessarily changing the fundamental biology, but the moment we've 11:58-12:03 got a spread in there, we can bring that spread down and critically bring it down to the leading edge, 12:03-12:07 where at the moment we talk about maybe it's an eight or a nine-day process, of course, 12:07-12:10 in development, people have said, well, it could have been maybe five or six, 12:10-12:14 but they don't go to market with that product because they need to take into account the fact 12:14-12:17 that there is variation in there. Or if the fact we can bring that down, we can now start taking 12:17-12:22 that time out and giving it back and getting it back to the patients quicker, which again, 12:22-12:28 we know is critical to improving outcomes. And then by taking those strides technologically, 12:28-12:32 we can improve like in process quality checks, final QCHX release material faster, 12:33-12:39 you mentioned AI and being able to, I guess, better project the results of what a clinical trial 12:39-12:44 and how that process is working and developing currently. And all of those things can get the 12:44-12:52 drugs to the patient lots faster. I think one of the things we've not raised yet is surrounding 12:52-12:58 logistics. And we also haven't spoken really much to regulatory, but I think if we consider the kind 12:58-13:05 of logistical bottlenecks that are really preventing that clinical delivery, where are we with the 13:05-13:12 kind of logistical web of getting not just products out of the patient, but back into them? I don't 13:12-13:20 know. It starts off. Yeah. We know it's a giant pain to be moving these materials from the A4A 13:20-13:24 Center, the Patient Care Center, to maybe a centralized plant to have autologous therapy 13:24-13:29 created. The cold chain logistics alone are extremely costly. Continued monitoring of that 13:29-13:35 sample during trains, they're both two in from, you have the worst case scenario, your centralized 13:35-13:40 plant goes down and has errors, tracking that material, pulling the material again, if lost. 13:40-13:48 All poses severe timeline risks to getting a therapy introduced to the patient in a timely 13:48-13:51 fashion. If they missed their window, even worse before we're having to restart. 13:53-14:00 I think it's something that makes me slightly nervous. And my focus very much in what we do is 14:00-14:06 typically within the clean room and improving the manufacturing. But I see here a little nervous 14:06-14:12 that we're making immense strides there really quickly. And I worry about a situation where 14:12-14:16 in the same way, I think we got caught a bit on the hot with the biology. Suddenly the 14:16-14:20 biology is approved. We've got patients we can treat and we didn't have the manufacturing 14:20-14:24 place to do it. I worry about as an industry that the next step is okay, something we've solved 14:24-14:29 that bottleneck, we've got fully automated plants able to produce tens of thousands of batches a 14:29-14:35 year. Okay, now we're suddenly pushing a lot more pressure onto supply chain. We've got a lot of 14:35-14:40 raw materials moving, we've got patient materials moving, and we've got consumables moving. It's 14:40-14:44 pretty startling when you look at what it's going to take for some of the processing. And 14:44-14:49 some of that you can resolve with having warehousing on site. But very quickly that could become 14:50-14:55 a challenge itself. And so I think we need to make sure that what we don't do is wait until 14:55-15:01 we've solved the challenges of manufacturing felt the pain more broadly before we start looking at 15:01-15:05 acting on those solutions. I think it's really important over the next year, 18 months, we look 15:05-15:10 to make big strides and I know people are looking at it. But we've got to make sure we've got credible 15:10-15:15 solutions in place. It's not the most complicated thing. What I don't think we need new technologies 15:15-15:23 there. It's something that I think we can look to other industries. I take up the progress things 15:23-15:27 like home delivery from people like Amazon and Grocers have done over the year. We've shown we 15:27-15:35 can move kind of really complicated lot pulling around huge numbers of locations really, really 15:35-15:39 efficiently. But we've had to do that. I mean, they've put an enormous amount of effort into 15:39-15:44 thinking about how they distribute their warehousing, where they put facilities, what kind of vehicles 15:44-15:48 they've got moving between them. I'm not sure we've quite done the analysis yet in this space. 15:48-15:51 I mean, it's really important if we're not going to suddenly see us push the challenge 15:51-15:56 outside of the manufacturing space. Yeah, and the raw materials for a typical autologous or 15:56-16:04 allogeneic drug product to some of those are liquid based. So, you know, just there's so much 16:04-16:09 of a challenge with that. So we, you know, what you see the pharmacy industries move towards is 16:09-16:13 how can they, you know, formulate some of those on-site on demand. And I think again, 16:13-16:19 that's probably something that the cell and G therapy field should be focused on moving forward. 16:19-16:23 So we're not the worst thing that can happen is we don't have that raw material and we have a 16:23-16:28 patient that's waiting on their product and they've already been treated. And before, you know, 16:28-16:34 they've already gone through a depletion therapy and they're waiting on their product and we don't 16:34-16:37 have something simple in the manufacturing process, something we could have made on-demand if we would 16:37-16:45 have been ready. The challenges that we've discussed today, and we've referenced centralized models 16:45-16:50 and decentralized as well, we've just been talking a little bit about in situ. Where are we with the 16:50-16:56 conversation of decentralized manufacturing models? And where do we expect, where do we see that 16:56-17:04 conversation going from 2020 to 2025, but also beyond? Yeah, I think in the last couple of years 17:04-17:10 we've seen a lot more interest coming from both federal funded committees and agencies to more 17:10-17:14 privately funded by the hospitals themselves that are interested in bringing that kind of therapy 17:14-17:20 production to their site. Not only is it a huge way for them to bring additional patients to their 17:20-17:26 areas, but to treat their local patient population. We, if I'm a technology standpoint, I think we're 17:26-17:30 in a really good position to start answering those calls, both from a facility standpoint, 17:30-17:35 be able to place completely standalone clean rooms in their parking lots or integrated into their 17:35-17:43 facility to the equipment training and transfer from larger centralized clean rooms, as well as 17:43-17:47 you know kind of building up that equipment to answer and communicate the calls to a more 17:47-17:59 you know replaceable or scalable flexible platform. For me, I think in a way, over time I've started 17:59-18:03 to object a little bit to thinking of either centralized or decentralized. I think that almost 18:03-18:10 pitches one model against another. I think if I was to throw out another phrase perhaps we need to 18:10-18:16 think about optimized manufacture. And I think what we're going to learn is that that's probably 18:16-18:19 going to be driven by the biology again. I think putting the biology right at the center, I think 18:19-18:26 there's hugely compelling reasons to look at putting things close to patient, whether that's 18:26-18:29 in a hospital, whether that's within a city, I think that's where we've got to look at 18:29-18:35 each individual biology and work out which is best and particularly obviously the biological 18:35-18:41 drivers, things like fresh product and if that's going to be the step that's going to drive 18:41-18:44 a potency improvement, if that's the thing that's going to drive us 18:44-18:48 and be able to create a faster turnaround time because we've got better viability without 18:48-18:53 loss on both the freeze and the thaw. So I think let's look at the details and then think about 18:53-18:59 what is the optimal solution for the therapy. And I think there'll be a diversity actually. 18:59-19:02 And I think that's one of the things that is always a bit of a challenge. I think mostly in 19:02-19:05 the past we've always ended up with quite a uniform solution and one thing has kind of 19:05-19:09 stood up head and shoulders above and okay we've gone down that route. I think that'd be a real 19:09-19:15 shame in this industry. I think the opportunity we've got is to allow us to really be biology led 19:15-19:20 for the best biology to win and I think it will really go to harness the power of self-theract. 19:20-19:24 I think that's what the diversity will do and I think then optimizing those models and I love 19:24-19:30 the idea of sort of flexible delivery and flexible manufacturing methods where we can look to think 19:30-19:35 beyond centralized which I think is a limitation a lot of what we do and really choose the solution 19:35-19:44 that is right for the biology. I think we need a successful example of where we can take a therapy 19:44-19:49 and put it in a regional hospital that feeds community hospitals. There's a constant request 19:49-19:55 from the community hospitals and you know and there's a larger you know there if you want to 19:55-20:00 call them regional whatever they are back to you know using these different phrases. Community 20:00-20:05 hospitals are right for testing this theory and whether it's TILS whether it's sickle cell, 20:05-20:11 CAR-T whatever it is I think we need a couple of really good examples where they've been successful 20:11-20:17 where we've we put in all the things we've talked about facility you know training standardization 20:17-20:21 and processes blah blah blah and show that it can be done improve it to ourselves but also 20:21-20:25 prove it to the regulators because once we do that a couple of times I think then that you know the 20:25-20:32 sky is the limit. Absolutely well that's all we actually have time for within the limits of this 20:32-20:39 discussion so thank you everybody for joining us. Thank you.

Transcript Speaker 1 Welcome everybody to the 10th, that's a double digits episode of Behind the Headlines. This is basically an all-star lineup, returning champions on today's panel, which I'm happy to see, and particularly because of the number of headlines. So my name is Chris Spivey. I'm the Director of Industrial Relations and Strategic Partnerships for the Pharma Health Solutions Group at MJH Life Sciences. uh i'll go clockwise so lax i'll ask everybody to Speaker 2 introduce themselves likes to go first thank you locksprin and kill partner deloitte lead our life sciences operations practice jonathan jonathan grinsstein north american editor at inside precision medicine brian brian feth i'm the ceo and co-founder at Xcell Biosciences Speaker 3 and i'm beth willers uh the principal at white matter communications i go by wmc Speaker 1 Right, so I wasn't kidding about All-Star Champion returning. You guys are awesome. So lots to get through, however, because of JPM in particular. But I want to start off with Sana Biotechnology and Allogeneic. And maybe I'll cheat and ask you, Brian, to go first, since I suspect you have something to say on this. Speaker 4 Yeah, this is a company that's been working hard on different things over time. They finally had a big win here. I do think it's a little for that type of specific, this type of therapy. There's still some unknowns 'cause it was done in the allogeneic donor derived form. And so their underlying stems derived pancreatic outlet cell is still yet to be validated in the clinic. It's a preclinical at this point. So they've got some hurdles in front of them from a kind of differentiation efficiency, maturity and stability over time, scalability and manufacturing, but really good news for them from a first shot out four weeks in. Basically, there was immune acceptance, didn't seem to create any kind of side effects, was producing protein that was indicative of insulin production. So it looks like it's working. Speaker 1 Jonathan, immunosuppressant wasn't news for this. Is that a turning point, pivotal? Speaker 2 Yeah. I mean, the, the immunosuppression is huge. I mean, it's, it's not a trivial thing at all. I mean, you know, the more people I speak to, you know, I recently had a conversation with beam therapeutics and, you know, they're pushing their entire transplantation program to be without immune suppression. It's just, it's just a much harder thing that, that on the patient. And so it's huge to be able to, to proceed without it. Yeah. Speaker 1 yeah that should shorten timelines as well if i'm not mistaken the required immunosuppressant is a lengthy process i Speaker 5 i think the the adjoining therapies um because many of these um allogeneic therapies kind of have um uh adjoining therapies that need to go along with it for managing immunosuppression and there's a whole host of side effects that that you can mitigate entirely if you're able to derive benefit directly from the platform itself that's one part the other part that this platform technology, I mean, Allergenic has been on the books for so long. And one of the significant challenges has always been immunogenicity and the fact that like this kind of paves the way for the true platform technology with Allergenic is really testament for where this is going, right? So I think it's important. Speaker 1 So I'm sure we'll return to that in the coming weeks and months. So one thing that really struck me, struck home hard for me actually was the UK Biobanks going with Olint Proteomics as their preferred platform. Syax, did you have a thought about that? Speaker 5 Yeah, I think so. There's a couple of things that in the broader proteomics world that I think this plays a very important role. And one is, I remember reading about, you know, a couple of the big tech companies making investments and leveraging the proteomics databases to identify using machine learning and AI to identify targets, which I think is, you know, especially when you see the diversity of the platforms that are available between O-Link and between a number of others, it just gives you like different facets of the data to mine from, which I think is in general, but is general great, notwithstanding the competitive nature of like innovation progressing through competition. I think that's always a great thing. I do think that it's interesting you know in the in the article they actually um or in the news piece they actually talked about it wasn't an exclusive deal it was it was a prioritized um platform so they still kept the door open for um you know multiple Speaker 4 platforms being used for generating the data set so it's going it's interesting to see how that will play out Speaker 1 interesting question yeah Speaker 4 i mean i just in our own work you often see you can you can profile transcriptome and you don't often see that I wouldn't say not often but occasionally you will see that not well translated into the protein and so even though it looks like there's an up regulation of a specific target it may not actually be regulated so I think these are very complementary what it does show is that with you know proteins have arrived at a you know kind of mature enough for population level studies which is also a big deal. I don't think it replaces genomics. I think genomics kind of are low cost, I can proceed it. But this is the answer. You know, this is ultimately what people are after when they're trying to understand expression. Everything else is somewhat of a proxy, unless you're obviously doing other things like mutational profiling or something like that. Speaker 3 Yeah, I think it's nice to see us coming to this point, like to his point, you know, proteins are are finally maturing and we're seeing that we have more information to work from. And it would be nice if we could make really educated decisions and move forward from them. But the problem is that there's still a lot to figure out how do you how do you house and move through that data and AI has its own concerns in itself and how much energy and water and things like that that uses. So I think that it would be really nice to see this mature in a way that we can continue growing and sharing and collaborating too, because the information is only going to be valuable if we have access to it as a Speaker 2 whole. Speaker 1 Jonathan? Speaker 2 Yeah, I mean, I just wanted to comment on, I think it's a good move by the UK Biobank to not limit this to thermo. I mean, it is an impressive platform. It is limited to 5,400 proteins. I mean, I make that sound like it's nothing, but it is a quarter or so of the known proteome. It also doesn't include any post-translational modifications, which in a lot of cases is all that really matters in terms of knowing whether these proteins are active or not, which is really, really important. So I think it's a step in the right direction. I think there's a lot more to be brought on and definitely we're seeing a huge move into multiomics in the precision medicine world. So yeah, exciting to see it being implemented on this scale, but I think it's just a very first step. Speaker 1 I'm glad to see proteomics have a moment, so to speak. Right. So busy schedule, GLP-1 agonists, lacks more reviews, more science. What are you saying? Speaker 5 Yeah, I think there's a published article in Nature Medicine that came out last week, or maybe this early this week. It was a comprehensive study, more of a meta study, but with the Veterans Affairs patient database of folks that were on GLP-1s. And it was exhaustive because they kind of measured risk-benefit ratios for, I want to say, 150 or so indications, which is probably the most exhaustive that has been published so far. It's been picked up pretty effectively in the market. And the higher-level message is that there's probably three higher-level messages. One I picked up on is there's obviously more benefit, and I know some of the other panelists will have thoughts on this, around where GLP-1s and the mechanism by which GLP-1s are impacting, you know, neuromuscular, skeletal, and metabolic pathways in the body, and therefore the benefit ratio is quite high. But I mean, they're also not devoid of risks and side effects. So the paper also talks about some of the significant side effects around, particularly on the end track. But what I thought was most interesting was there was a sort of a sentence I picked up on like how obesity and the and the dietary challenges that people have what this kind of proves is that the the mechanisms in the body is more kind of tied together with the neurological pathway in terms of addiction and obesity and and all of these areas and so that's a that's a pretty interesting finding I think you know they were talking about Alzheimer's so all of these are kind of intertwined and it really seems to be a very effective drug. Speaker 1 Beautiful. Thank you. Brian? Speaker 4 Yeah, the concept of a metabolic syndrome and kind of the tie together of cardiovascular type 2 diabetes, obesity, they've kind of known that there's been links, and it's taken a while for us to get to a point where these drugs are being used now for obesity versus type 2 diabetes. So, yeah, it's a powerful class. It's been around for quite a long time, and I think effectively at Pillform, there's been you know kind of once weekly to once every other week to once monthly and and ultimately trying to get this to be to be easy to use we see it starting to get the attention of the government um in terms of price regulation as well so um i think it's the class that's here to stay Speaker 3 yeah very much yeah it's a very intriguing um it's a very intriguing area right because it has so many benefits but there's also potential drawbacks we don't know what long-term side effects look like but i think that kind of speaking to what you know lax and brian said is that we can kind of see this almost moving into a preventative like a maintenance type of situation people who are pre-diabetic i personally know somebody who's in that situation who is basically trying to get things under control before they go over the edge um but there are still major side effects to these particularly I think it was laxing trying to get it into the pill form these injections are not enjoyable and sometimes they can make people really sick and so we need to figure out how do we get past that you know because the drug can't help you if you can't take the drug and that was kind of the situation so dosing and things like that I think we still have quite a bit to learn in the area but quite a bit of potential. Speaker 5 Jonathan? Speaker 2 Yeah I mean lots of interesting things with this article I mean And I mean, one thing that also comes to mind, but hasn't totally been talked about, I mean, this study was done with like the Veteran Affairs, which led to bring up, but, you know, there is now this compounded form of this that's being sold by like hims and hers and all this stuff. And it's just getting into the hands of a lot of people. So I don't know. We'll see what happens with that. You know, a lot of the positives that we're seeing did have to do with the brain, you know, it was neuropsychiatric, neurodegenerative related things and whatnot. on. The issues they did see though were with gastrointestinal issues, hypotension, joint pain, kidney stones, cancer, kidney and pancreatic issues. So I don't know, there's quite a bit to be looked at here. And then finally, I mean, at the point, also the point on the brain and the nervous system. I mean, I think the gut is the most innervated organ we have. There's some really interesting research going on there. The gut-brain axis is, you know, very, very hot. And I think there's some companies that are looking at trying to alter brain, you know, disease modifying outcomes in the brain by targeting the gut. So really interesting people to watch. Speaker 1 Well, I am glad though that we're getting to a more connected, holistic view of how the whole system seems to work. So if nothing else, that's a positive. Right. So big news, bookend meeting of the year, JP Morgan Healthcare Conference, can't ignore it. So 2024, 106 deals over 100 million in biotechnology i'll just mention a couple that were announced at jp morgan so intracellular therapies that was uh jnj 14.6 billion um for bipolar cap lighter and then there was eli 2.5 for breast cancer scorpion therapeutics and the last one gsk about a billion up front for a gastrointestinal tumor uh idrx so lax what did you hear see feel coming from uh jp morgan Speaker 5 I mean, in general, I almost feel like it's one of those ring the bell moments in J.B. Morgan that, you know, people are driving up, working up to make those announcements. Sometimes, you know, these deals kind of take long to kind of mature. But I do think that, I think that, I mean, biotech in general and pharma broadly is back to making sort of tuck-in specific acquisitions that make most sense from a therapeutic area standpoint. And this year's JPM is kind of a reflection of that, that, you know, that's kind of the going pathway for most biopharmers. I think the other thing that was unique this time around is, and at least some of the, I don't know, I wasn't there personally, but what I've heard is like a lot of the conversation around how do you work with the incoming administration to kind of, you know, keep the innovation and the R&D engine live while driving some of the affordability and accessibility questions that a new administration will definitely have a see on. So I think there's that balance that I think that was very intriguing as well. Speaker 1 Well, I agree. And I also meant to throw in NVIDIA with the partnership announcements they made, which I still haven't Speaker 4 quite understood so far. Yeah, that's actually been an interesting area that we've been focusing on as well for kind of AI's engagement with with biology. But just generally in the meeting, I think I had the same, you know, kind of sense of enthusiasm at the beginning of the week when a lot of those deals were announced, it kind of came on the back of Biogen's bargain basement offer for Sage. So it was kind of a nice push. But when you look at the deals that were done, they're largely later stage, you know, kind of plugging a hole as Lac said, trying to prevent, you know, kind of damage from a patent cliff or loss of, you know, IP elsewhere. We haven't seen kind of broad platform based or, you know, anything that's, um, at least notably early stage. So, um, I think we're moving in that direction. It feels like, you know, it feels like there's a rebound coming and, um, there's probably just a little bit of reservation as we, as we watch the new administration form. Um, although I had, I'd read somewhere that, um, the market was kind of broadly happy with, um, the Marty, uh, Macri selection for, for the, the FDA. um i guess we'll see in the coming days um i think that was last friday so pretty recent Speaker 1 don't look away we will find out actually it was a 400 billion loss in uh income last year that they need to replace in some way shape or form and beth i saw you commenting on that actually in the cycle of investment uh recently yeah Speaker 3 i've got big opinions on this i think that we uh and we kind of talked about this last time when we had a panel discussion but i think that we need to really start acknowledging the long-term nature of this industry and you know I do think that it's nice to see money going into later stage and you know items whatever that be whether it's a drug or whether it's platform systems whatever but if we don't have the money going in early stage then we don't come to the innovative place so we've been in this like push-pull where nobody wants to spend money but we know that people have money holding on to the money gets very expensive too so I think that I had really thought we would see the break in 24 and I think you really did start seeing that around June. But we need to go into a boom. We really can't hold much more than these two to three year spurts. And so I'm just ready to see what's going to happen. And I would really love to see some advancement in the tools and technologies and systems, more of the collaboration. If we are getting all of this information from data, okay, how are we making sure that everybody wins you know we all want to make money but we need to keep innovating to do that so Speaker 1 yeah actually good point and industry academia especially in the genomics field has been burgeoning somewhat this last year jonathan yeah Speaker 2 i mean um you know that i think the entire field is moving towards what's called precision medicine and then you know everything is rooted into molecular diagnostic and so uh you know it's we're seeing the implementation of genomics in the us at least across various health systems um we all i mean we were just talking about the uk also that they made a fantastic kind of move in that direction um so you know i think it'll be commonplace eventually everyone will have their genome sequenced um you know when and when that happens in life and how that's done it's not totally clear at least for the people who are older i think you know maybe a place they would need to start is with all newborns but you know that's a that's a discussion for another day um i i and um but yeah definitely going to be everywhere i um illumina definitely try to well i wouldn't say they made a grab but the announcement with nvidia was pretty huge i did want to quickly just say something about that and i i got invited to the embargoed call with with NVIDIA where they told us all the announcements going, heading into it. And so when it came out on Monday, there was a couple other announcements and I was like, oh my God, this is going to go, like, it's going to be a boom moment. And it just did kind of fizzle out pretty quickly. So it was like this false start of excitement, um, to kind of like what Beth was saying. Um, so still some good stuff for sure happened, but, uh, you know, it's like, Ooh, not, not totally there. Speaker 1 Interesting. So there's precision medicine, the new personalized medicine for profit, so to speak. Speaker 5 I wonder if like, you know, if I were to put like multiple sort of trends together, whether the direction will go towards more situational precision medicine, Jonathan, like where I'm thinking about is like, you know, if like our body, human body kind of has an evolving set of proteins that go up and down based on kind of the situation. And as this AI and AI chips become more and more accessible, it's almost like a continuous trending of what's happening with your proteomic balance in the body. And that's like a forward indicator for what kind of therapeutic conditions or metabolic conditions that you might need to address ahead of time and becomes more and more preventative. then it becomes like, you know, therapeutic. So I can see the world going that way. And I think it's something that, and with all the ease of computation and AI and ML, I think that's naturally the trend where this is going to head out. Yeah, I had a conversation with a couple companies who Speaker 2 are looking to actually do that. I mean, you know, there's the question of how do you actually implement it, getting the person's samples. I don't know, do you have a chip on someone? But, you know, sequencing some RNA, whether it's a small panel and seeing how it changes week to week and having some sort of alert doesn't seem too far in the future. Speaker 1 There's a nice diabetic drug that was supposed to activate only insufficient amounts of other trigger enzymes and things, which never really was commercialized. So I think it might be complicated, although I will also say that organoid companies, tissue dynamics comes to mind, has very good data sets now on how that fluctuates during the time. So I think you're onto something like that. So it's probably complicated, unfortunately, though. Yeah, absolutely. So Beth, you were nodding when Jonathan was talking about newborns that seemed to resonate with you. Speaker 3 Well, I mean, it's such a large discussion, right? Because then are we basically putting ourselves on services like, you know, cars? Are we doing diagnostic panels? How are we doing this? Right. But also, I think, you know, there's always going to be a place for small molecule drugs, know very particular things like you have phil you know comfortable things like we know these things work these things are good for like small ailments but when it really comes to moving towards preventative care how do we get that and i think you're seeing a lot of these models and this is kind of i think where jonathan was going is that you're seeing a lot of these models that are really geared towards maybe 45 and over that are like these concierge health models where you can pay to get all of these different scans and blood tests and this and that and basically we'll show going on in there what you need to fix which is very different than the traditional medical system that's kind of like wait until you have a problem um and we see that our medical system has had a lot of struggles recently so it's just it's a fascinating subject that's why i was like i'm like oh wow because it does we could spend a few hours probably a few days on that one Speaker 1 yeah the microbiome opens up that door as well there's uh aaron crowley talks about that yes Speaker 3 fascinating and like you said like the gut to brain just how how connected our bodies are and how much I don't think we really acknowledge that if we're treating one small symptom there but if we talk about these persistent therapies if we talked about advanced medicines whatever we want to call them this is literally what we're fighting is yes we can fix that thing but what about these five other things that happen we don't want those things to happen so it is all connected um I we're just on the precipice of it. I mean, if you look at the data, if you look at everything, we're really starting to make really cool advances. But it takes also a lot of responsibility. Speaker 2 I like to joke, we're in the like the caveman painting phase of precision medicine. Like I have some idea like this person's stick figure, you know, that's like, that's where that's as far as we really are the huge way to go. Speaker 4 Even just imaging, like nobody looks in your body until you have a problem, you know, a symptom. And it's expensive right now. But, you know, it's clear that in the future, there just be body scans that are done. You learn a lot of information, you get false positives, but the better markers and diagnostics that you can run, the more you can kind of resolve those. So yeah, I would agree. Feels like we're still, you can look forward and see that we're still early days in kind of precision medicine and, you know, screening applications. Speaker 1 Actually, funny you should mention painting. Dan Pear talking about mRNA in the body was painting Speaker 2 the intestine Speaker 1 with where things are active, which struck my attention sort of thing. So moving into the unknown, the drug Medicare second set has just been announced. And I'll broaden this to that and also all new U.S. government actions coming up. I'll start with you. Speaker 5 Yeah, I mean, I think what's interesting for me is the list included diabetes drugs right off the bat. Like that was interesting. I mean, obviously we were all expecting that proliferation of Lycalve. They were going around that they would be in the list. And then I think the other thing is I think there are a couple of cancer medications in that list. And so, um, what I mean, I think it's also, I'm also curious with all the executive orders that the new administration is issuing, how this will pan out. I know this was like a tremendous set from the previous administration. So I think it remains to be seen on what the enforcement cycle around this would look like. Speaker 1 Thank you. Uh, Brian? Speaker 4 Uh, uh, the Trump administration's a little bit of an enigma on what they're going to do here. they had issued an executive order that lowered kind of rescinding a prior the prior administration's kind of lowering prescription drug costs under Medicare and Medicaid but and it was kind of designed to to test models for reducing drug pricing so it wasn't directly related to the Inflation Reduction Act but so it seems like they're moving kind of against Speaker 4 it but it also is a popular policy. And the current new administration seems like it's really focused on populist policies. I will say that Peter Kolchinsky out of the RE Capital is pretty vocal. He's been putting out a lot of papers that kind of pushing back on drug pricing being the problem. And I think really making the case clearly and well that there's a lot of social benefit and a long-term benefit through generics that's kind of fed back in that's often not incorporated into the calculation of what the real cost to society versus benefit to society is. And there's been, you know, he's been talking about insurance reform and PBM reform, and that's also come out of the Trump administration. So it wouldn't surprise me at all to see kind of a hard shift from, you know, focused on specific drugs and kind of more of a shift on kind of reform that may be supportive of the pharma industry and the business that would, you know, similarly help to reduce costs of the overall healthcare system and ultimately maybe get to lower prices by way of that. But we will see. I think there's still a lot of, you know, reducing prices for drugs that people take, particularly in Medicare, is a big benefit to political goodwill. Speaker 1 And speaking of that, you know, pricing costs, you've been vocal about uh, reduced cost of some treatments in India, for example, you know, or, I wonder how much more of that we'll see and how much that might be embraced. Speaker 4 Yeah. In fact, uh, a group out of India, I think it was, um, uh, Siddhartha Mukherjee. Um, he's his company. I can't remember the name and MUNI maybe, um, they licensed a drug out of Europe, uh, and brought it to India and have been, um, is a CAR T, um, standard product. they've been making it for substantially like a third or less maybe tenth of the cost i think um i think it's a tenth of the cost yeah yeah so um i mean it it can and should be done um i think uh there are you know patents are designed for a reason i think a really good benefit it creates market exclusivity allows there to be you know kind of strong incentive for investors to invest in these companies on a long-term basis societal good comes out after that period of time where you you get the reduction, you get competition. You don't want to break that because that really is an innovation engine. And I think people look abroad at some of the patented drugs and see that there are reduced prices there because those governments are negotiating. And yes, it really is on the back of the taxpayer and the innovation here. We're paying higher prices, effectively subsidizing. But there are a lot of other ways we can reduce the cost of healthcare here and get some of those drug prices down by reducing some of the regulatory red tape and realigning how the system kind of buys drugs so you don't have these centralized buyers in the middle. And I think there are opportunities for groups outside of the U.S. to develop drugs and this is a global economy now and I think having drugs that are being launched for reduced prices in India is a great thing and maybe those drugs can come here as well. So I mentioned also that it's happening in Brazil. Brazil is looking to adopt kind of the, they have a mandate I think to adopt the the best in class they obviously want to do that in a way that they can you know limit the expense so they're working with groups like caring cross to to bring in cell therapies at much lower prices there Speaker 3 well Speaker 1 um it's Speaker 3 uh i'm flashing back to a conversation that i was having in college about uh the bricks you know i think you're seeing a lot of markets that are finding ways to do this cheaper faster better and we need to ask ourselves how and why is that happening i also think there be like a really big prioritization that and this you know leads into a very uncomfortable discussion for a lot of people but some things just don't make money some things are there because they're better for society some things are there because they help people they keep people well and then other things do make money and so we really also have to evaluate the relationship between pharma and money insurance and money and Speaker 6 you just the overall Speaker 3 market and what we value a healthy society has it's this is another one you could spend you know hours days weeks on on the discussion but we need to make sure that if we're seeing that a drug is so aggressively lower priced in another market why so is that it because we also had rx360 and we had all of the issues with heparin and what oh wait you know we don't want a lower cost for a bad quality we don't want people dying right we want a lower cost with that same level of quality we want to make sure that this is sustainable going forward that people can get help and we also want to make sure that we don't turn like medical tourism into a standard that's another concern for me too Speaker 1 yeah anecdotally i'm hearing more about it too uh jonathan Speaker 2 yeah um i'm not really a price uh guy to have that discussion i mean i just wanted to quickly comments on like the the how unexpected things may be i mean you know some web pages went down immediately from the government site uh right after the inauguration i mean like the women's reproductive health site is completely dark now and so you know um big question mark for a lot of these things Speaker 6 yeah okay Speaker 1 so we've got just a couple of minutes left um anybody you have a thought that they want to cover something they want to make sure we don't forget to mention You're nodding, Brian. Speaker 4 Yeah, I want to come back to the theme of AI's involvement in biology, which I think is an interesting emerging trend. We knew it was going to happen. We know that investors and companies are looking for ways to get at large data sets that we generate in biology. And it's an interesting one. OpenAI partnered with Retro Therapeutics, which is developing a better process for stem cell reprogramming using derivations of the Yamanaka factors. And basically, OpenAI was able to generate differential factors from the standards that are being used that are much better, apparently. And so it's a really interesting application for AI, just creating better entry points, better tool sets to do better biology, using kind of AI-driven analytics around the structure of proteins and deducing how those factors are important in the biology and kind of coming up with better ideas. So it was an interesting deal in that Sam Altman, I think, funded Retro initially. And so there's kind of a little bit of an incestuous thing going on there. But, you know, from his side of things, saw an obvious opportunity and is leveraging AI to kind of deliver on it. On the point of proteins and, Speaker 2 you know, NVIDIA, one of their collaborations was with the ARK Institute, which is this newer kind of funding model institute out of San Francisco, very interesting, headed by Patrick Shue and Sylvana Connerman. They just had a fascinating article on protein evolution modeling. So basically, how can you make any protein you want? And NVIDIA is going to kind of flip the computational build out to take that further. Their collaboration and videos with the Mayo Clinic is all about advancing precision diagnostics. And with all the power that they're supplying, I mean, one thing that AI is fantastic at is analyzing images, right? You know, famous Silicon Valley scene of hot dog or not hot dog. I mean, we're now in the cancer or not cancer, what kind of cancer phase. And so with all the compute that NVIDIA can provide and all the data that Mayo Clinic has, they're being able to they're working towards establishing foundation models which basically allows the AI to like make decisions on all sorts of really interesting diagnoses when it comes to in this case cancer so I think it's going to be a next field that's going to really explode I think this is going to be a major push and you know who better to deal with them with the Mayo Clinic so excited to see how that all kind of unfolds and I think it'll unfold pretty quickly one thing on And I know just to quickly say is that in JPN, one thing I did notice that was kind of like a weird kind of thing, not so much in the biotech or pharma thing, but was like from people who have academic ties, the notion of what's going to happen there, like what's going to happen to the NIH, what's going to happen to all sorts of funding models. I think that was something that was definitely in the backdrop, wasn't really spoken about, but it's awesome. Speaker 1 I'm going to thank you all, smiling people, for enjoying the conversation. I get to see everybody. Speaker 6 Thank you Chris. Enjoy the shoot. Thanks Chris. Okay,

Transcript Tell us about your background. Brian: As a teenager, the loss of my grandparents to different cancers influenced my scientific journey. In college, cancer research became a focal point. My career started in strategy-focused management consulting, advising life science companies. Notably, I led the team that advised the merger of Invitrogen with Applied Biosystems. My interest in entrepreneurship grew from family history and helping lead a project funded by Coca-Cola and the Gates Foundation in Kenya and Uganda. Returning to the US, I pursued an MBA at UC Berkeley, where I met Dr. James Lim, my scientific co-founder, at an entrepreneurship event. James: I earned my Ph.D. in Biophysics from the Scripps Research Institute, focusing on single-cell analysis and cancer cell metastasis. Afterward, I did a postdoc at Harvard Medical School, where I identified and expanded rare circulating tumors. With varying degrees of success, I was able to demonstrate that we can keep these rare tumor cells alive and dividing ex vivo. My interest deepened in understanding how different microenvironments impact cancer cells. Later, I joined Lawrence Berkeley National Lab, and the collaboration with Brian led to the founding of Xcellbio. How did you go about founding Xcellbio? James and I founded Xcellbio within the Berkeley ecosystem. We also secured partnerships with a local cancer center and UCSF's genitourinary cancer research group provided initial data sets. These collaborations, along with biotech and pharma partnerships, led to a $5M seed investment in 2016. Our business model evolved based on partner feedback, focusing on selling workflows to pharmaceutical and academic customers. What got you interested in cell therapies? When we started Xcellbio, our focus was on growing tumors from patients. Our early work highlighted to us the importance of immune effector cells at keeping tumor cells at bay. We saw firsthand that you can engage and arm immune cells to target tumor cells with a high degree of specificity and effectiveness. It was incredible to observe the effectiveness of T cells killing tumor cells. This solidified our belief in the potential of cell therapies. What was the need you saw in cell therapies? Early collaborations with CAR-T cells revealed their efficacy in screening conditions mimicking the solid tumor microenvironment. We observed that CAR-Ts would get exhausted under low oxygen and hyperbaric screening conditions, compromising their effectiveness. This led us to question how we could enhance cell therapy potency in these immunosuppressive environments, ultimately resulting in the creation of the Avatar™ family of cell culturing and analytical systems. What was the first application Xcellbio went after? Initially, we aimed to enrich disseminated tumor clusters (DTCs) from cancer patient blood. We shifted focus to insights into TME-associated tumor cell and therapeutic cell performance, which was an area where several partners were offering to pay us to provide insights. The company gained traction with this enrichment idea; can you tell us a bit more about that aspect? Early collaboration with Thomas Krahn from Bayer Healthcare was pivotal. Running a blinded pilot study, we demonstrated impressive results, leading to paid partnerships and the first sale of our technology platform. Recognition of the importance of hypoxia-regulated pathways and pressure sensing receptors further boosted our efforts. Studies with Merck exploring lymphocytes in solid tumor specimens steered us towards repositioning our technologies for immune cell therapy. You recently signed a partnership with Labcorp. What caught their eye? Labcorp sought to expand their service offerings in gene and cell therapy (CGT) development. They were interested in technologies modeling the tumor microenvironment. Our collaboration involves applications providing more relevant tumor models and reprogramming immune effector cells for enhanced efficacy. The partnership expanded to include our GMP cell therapy manufacturing system, bridging preclinical research to scaled-up autologous patient data. Why are these tools to modulate conditions necessary? What might it do from a potency perspective? Conventional in vitro settings often show excellent cell therapy performance. However, the challenge arises in the suppressive solid tumor microenvironment. What’s apparent from our early investigations into cell therapies, is that only a small subset of the immune/effector populations exhibit targeted tumor killing. So, the question becomes how can you select or enrich for the active cell populations? Our systems provide a more relevant predictive readout, accounting for factors like low oxygen tension, high interstitial pressure, and low nutrient levels. The cells that survive or flourish in these conditions are “super killers”, having been conditioned in this environment. And the process of cells’ reactions to these conditions have been studied for some time and recently received Nobel Prize recognition, right? That’s right. The 2019 Nobel Prize in Physiology or Medicine was awarded to teams for discovering how cells sense and react to low oxygen levels. Then in 2021, the same prize was awarded to Ardem Patapoutian and David Julius for discovering receptors for sensing temperature and touch. So, these are no small discoveries by any means and explains how these external stimuli can result in changing cell phenotypes. Are there discernible phenotypic differences between those cells that kill cancer and others? What makes them more efficient in targeting and ultimately killing cancer? Our working theories suggest that conventional activation and exhaustion markers are poor predictors of effector cell potency. We combine cell-based killing assays with targeted immunophenotyping to identify persistent and potent effector cell subpopulations. These cells, exhibiting anti-tumorigenic properties under challenging conditions, often display unique metabolic profiles. Is this the critical step to enable cell therapies to target solid tumors? What else might be needed to unlock that space? Maintaining optimal differentiation status during in vitro CAR T cell manufacturing is crucial. However, the suppressive solid tumor microenvironment presents additional challenges. Metabolically conditioning cells through transient nutrient depletion and lowered oxygen tension produces fitter cell products. This conditioning enhances cytokine secretion, cytotoxicity, and metabolic rewiring, potentially translating to better efficacy at the tumor site. We think these functional outcomes in preclinical models will translate to better trafficking to the tumor site and better efficacy at the tumor site than current approaches offer, hopefully achieving higher complete response rates in patients. So how would a biotech go about applying your technology to their programs? Our technology spans the biotech value chain. The Avatar™ and Avatar AI™ instruments serve preclinical research, screening conditions, and assaying antitumor function. The Avatar Foundry™ enables fully closed transduction (optional) and expansion under GMP conditions for improved gene editing efficiency and greater potency for use in clinical trials and cell therapy commercialization. Do you think a one-size-fits-all approach will work for cell therapies? Unlikely. Cell therapies vary in source material, cell types, dosing regimens, and desired phenotypes. Bespoke manufacturing workflows tailored to specific therapies are necessary. Is it feasible to think about bespoke manufacturing workflows? How might this look? Yes, bespoke workflows are crucial in cell therapy manufacturing. Tailoring unit operations to each therapy's specific requirements involves optimizing steps for consistency, throughput, and quality. This approach accommodates diverse therapies with unique manufacturing processes. You’ve been doing some work along these lines as well. Can you speak to these efforts with Cellular Origins? Our focus on cell quality includes efforts to automate and reduce human interactions in our workflow. The partnership with Cellular Origins aims to create solutions for automation around our proprietary workflow, connecting upstream cell selection and downstream operations. So, a platform could be automated with, in theory, any unit operations desired to create the best process for one’s cell type or indication? Yes, we favor a modular approach where unit operations are optimized for each therapeutic strategy. Modules can connect in a sterile manner, enabling technicians or robots to perform the workflow with minimal manual manipulation. How important generally is an offering like this to the cell therapy field? What does automation and breaking the process down into unit operations allow for? This automated, modular approach is essential for modern cell therapy manufacturing. Automation increases efficiency, reduces costs, and ensures consistency. Breaking processes into unit operations minimizes batch-to-batch variability, making results comparable across batches and sites. It also allows for use of best-in-class technologies for delivering new or more powerful capabilities to the field, i.e., the Avatar Foundry’s emphasis on delivering improved capabilities around[CS1] transduction, expansion, and potency improvements. What other parts of the manufacturing workflow do you think can be improved? Are there specific unit operations you think are critical to further optimize? Genetic modification of therapeutic immune cells is critical and can be improved. While lentiviral transduction is common, it comes with complexities. Nonviral approaches, such as electroporation, show promise but need refinement to enhance efficiency and reduce cytotoxicity. So, what’s next for Xcellbio? We recently launched our beta program for our GMP cell therapy manufacturing instrument (Avatar Foundry), with Elevate Bio announced as one of a number of top-tier beta sites utilizing our platform to optimize the metabolic fitness of their (clients’) cell therapy candidates for solid tumor indications. That’s terrific. Brian and James, thanks so much for chatting with us. We wish you a productive Advanced Therapies Week!