Exploring the Bio-Convergence Boom

Michael Marks: Hi everyone. This is the Tech Surge Deep Tech podcast presented by Celesta Capital. Each episode we spotlight issues and voices at the intersection of emerging technologies, company building, and venture investment. I'm Michael Marks, founding managing partner at Celesta.

In this episode, we're going to explore one of the areas in technology I'm most excited about, what we call bioconvergence. This is a growing area of healthcare that brings together high technology and biology, where companies are driving big improvements in health outcomes by merging engineered technologies and biological science disciplines. Some example application areas include drug discovery, diagnostics, and imaging. We have a fantastic panel who are sitting at the forefront of bioconvergence to help us explore this topic. Each of these guests are working to help solve some of the biggest challenges we face in human health.

The listeners will understand that these three people are way above my pay grade. It is an honor for me to be in your company. First is Dr. Jim Rothman. Jim is one of the most distinguished biochemists and cell biologists in the world. He is the Sterling Professor of Cell Biology at Yale and previously taught at Princeton, Stanford, and Columbia. He is the former Chief Scientist for GE Healthcare. He has received numerous awards for his achievements, including the Nobel Prize in Physiology and Medicine in 2013. 

Next is Dr. Melanie Mathieu, Chief Technology Officer and Founder at drug discovery startup Prellis Biologics. Melanie is a trained immunologist and previously held prestigious research positions at UCSF, Stanford, and UC Irvine. 

And finally, Daniel Dornbusch, is the CEO at CRISPR gene editing startup Excision. He has more than 20 years in the biotech space, previously holding roles at Novartis, Genentech, and a number of startup companies. 

I'm sure you'll agree with me this is a distinguished panel. So thank you all for joining us. 

All: Thanks, thank you. Thank you. An honor to be here. 

Michael: OK, I would like to start by reading a quote from Walter Isaacson, who wrote The Code Breaker, about Berkeley's Dr. Jennifer Doudna, who did pioneering work in biochemistry and genetics and is credited with co -inventing CRISPR gene editing technology.

Here's the quote: “The first half of the 20th century featured a revolution driven by physics. The second half of the 20th century was an information technology era based on the idea that all information could be encoded by binary digits known as bits and all logical processes could be performed by circuits with on off switches. In the 1950s, this led to the development of the microchip, the computer and the internet. When these three innovations were combined, the digital revolution was born. Now we've entered a third era. A life science revolution. Children who study digital coding will be joined by those who study genetic code.” 

I love this concept of the atom, the bit, and now the gene igniting the great technology revolutions of the past century and the next century to come. So Jim, you've been deeply involved in this space for years. Do you agree with Isaacson that we're on the cusp of a life science revolution? And if so, why now? Why is this the moment we are poised to see big breakthroughs? 

Dr Jim Rothman: Well, you know, it's hard to disagree with Walter Isaacson and actually I don't disagree with him really, but I think it's just so hard to kind of call what's really going on when you're in the middle of it. And we're kind of in the middle of it. So, you know, if you try to unpack some of that with the idea of gaining some insight into what we might be going through, you know, if we were back in the 1600s, we'd be blown away by the idea of mechanical force. And there was a revolution in mechanical engineering. Then there was another revolution, the steam engine, right? Which was, we think of as the steam era, it was very important if you lived between 1800 and 1850 that your world completely transformed, including huge turbulence in society related to the entire basis of economies being replaced. Okay, the Industrial Revolution. There's no doubt that we're living through something similar now, but if we had this discussion, and I guess they didn't have podcasts back then, but if we had a podcast back in 1840, we would have been talking about steam, steam, steam, okay? What we were really talking about was energy. Only people didn't think of it as the energy revolution.

So as I look at it, there's been a revolution that we actually call physics, just the concept of physics that dates back to really Newton, if you had to pick a starting point. And the concept of energy was implicit in force equals mass times acceleration, energy equals work times distance. And then we realized later that we could benefit from chemical energy even before we thought of it as chemical energy. That would be steam. Later it was gasoline. Then we realized that various forms of energy could be interconverted. That was really Maxwell's great discovery, electrical energy. Then we had the electrical era, and then we had the nuclear era, but actually there was one core concept in the whole revolution, which is that energy can take many forms, can be utilized in many modalities, and is conserved, which is the real discovery, and can be therefore interconverted. So why do I say all that? Because it's a long -winded kind of an analysis. Because we're in the same sort of situation now, where we were chatting before we came in here about hardware. Oh man, it's all hardware, right? But then it wasn't hardware, it was software. Well, now it's getting to be hardware again, kind of. But it's in service. Why is that hardware so important? Because we're trying to scale to deal with data. We're trying to convert data into information. you know, if we had this discussion in 1960, it's the computer era, okay? Then it was the information era. I'm not really sure what era it is that we're in right now. There's probably a concept like energy but it will probably not be over soon. And if I had to guess today, it's not a computing revolution or a digital revolution, it's an intelligence revolution. And it's not data, it's not information, it's intelligence. 

I don't think we're but at the beginning of it. And that has yet to really impact on biology. 

Michael: Well, that's actually a very interesting answer, thank you for that. A perfect segue to when talk about artificial intelligence, can you talk a little bit about Alpha Fold? And what impact that might have on life science? 

Jim: Well, alpha fold is actually a fantastic discovery. What it is, is essentially a computer product, an artificial intelligence product, that on your laptop in a matter of minutes, if you feed it the DNA sequence of a whole new organism, it will tell you what the final structures are of maybe 15 or 20,000 of the proteins encoded by the genome, just like that. It will do the work that we scientists would have done in the lab over perhaps 10 years and 10 minutes. That's amazing, okay? And it has enormous prospects for drug discovery and for basic science and everything in between. 

The thing about it though, if I can say, is that we don't know how it works. That's the key thing. It's hacking. Okay, so in my mind, this is the first example that you can hack biology. Because in biology, we are always having to figure how things work through causation. And then we can kind of figure out how to develop a drug, right? But we can't do one without the other. But these guys have figured out a way to use AI to essentially take all the data that's around and then use it productively to produce the structures of proteins without actually understanding the causative principles. Now, if you could replicate that over and over again in more complex domains, and this is an awfully complex domain, wow! That could completely change the development cycle and the paradigm of biology. Now that would be a biology revolution. 

Michael: That's awesome. Well, how about taking the bait then, Daniel? Let's see, what do you think about that? 

Daniel Dornbusch: I love the way you describe us, Jim, I couldn't agree more. I'm also curious to hear what you think about the manufacturing side of this. I often get the question, well, the Human Genome Project finished whatever it was, 2000. And why haven't we seen every disease cured, because we now know the genome?. For us, where we are, a lot of it had to do with manufacturing. How do you reproduce that? How do you make it faster? 

So, Illumina deserves the market cap that they've built so far because you can now sequence the genomes so quickly. But for what we're doing in trying to build therapeutics and gene therapies, we needed a whole series of different and more advanced hardware, software. Really, I think the bioinformatics gets passed over really quickly. 

So I'm talking about things like single cell sequencing, digital droplet PCR, which Rose did an incredible job with, even mass cytometry. How do we use new tools to interrogate these things and maybe build the databases necessary for Alpha Fold to do what it does and for some of the other software applications? But I'm going to come back to the software side.

Maybe we turn this back over to you, Michael, if you can talk about manufacturing some of these things and reproducing. 

Jim: Hey, Daniel, you're here to answer the question. 

Michael: Actually, now that we're this far off script, Melanie, you want to weigh in? 

Dr. Melanie Mathieu: Yeah, I think that's a really interesting point. I'm glad you brought up the Human Genome Project because when I was prepping for this podcast, I was thinking about how we got all this information. It launched in, what, the 90s, 1990, and it finished in 2003. And as Jim said, it's not necessarily a data revolution. We have that data. It's more about what do we do with it now and the intelligence to understand, okay, why is this meaningful? And obviously we had to sequence more than one person to figure out why it was meaningful, but that application I think is really critical.

Michael: This is all fascinating. 

Jim: Well, can I try to generalize this a little bit? 

Michael: Yes, you can. 

Jim: I've been asked the question in similar circumstances to this. You know, I just don't get it because, you guys, know so much, you biologists, your research costs a lot. Every time you get into another drug development product project, you know, you still have a high failure rate and you're spending billions on it anyway. And it sort of seems unsustainable. And like, what's wrong with you guys? Because the engineers., you know, they manage to get out their products pretty regularly. takes a three or a five year cycle. The R &D costs are therefore less. The return can be quite spectacular. So, you know, what's wrong with you? Why haven't we seen that sort of impact? Why don't we know, as you said, Melanie, so much more given all the data that we have, right? 

And there's a core difference between tech, and this comes to the point about bio -convergence, because when we talk about bio -convergence as bringing together essentially two cultures, right? We have the tech culture rooted in engineering and it's also rooted in scale and in manufacturing strategies. And then you have the life science culture, which is rooted in medicine, complexity, individuality, and in biology. And on the face of it, you know, that convergence is inherently difficult. Now, the reason it's difficult is that in biology, there's a linkage between understanding, you have to understand the biology and do and understand the basic biology at the same time that you're really developing the drug or the assay for the drug or the ability to manufacture the drug, as you say, Daniel, and that makes it inherently less reliable. So as a result of the fact that you're doing the science at the same time as you're doing the tech, when you're doing drug discovery, drug development, it's inherently a lower probability of success, inherently costlier and a bit more, maybe more than a bit more of a wandering enterprise. 

Michael: Well, thank you for that. That's also a very good segue to Daniel and Melanie. You're both working in the applications of these areas in an important way. So let's hear from you about your technology. So let's start with you, Daniel. Your company Excision is using CRISPR technology, which I mentioned was discovered through Dr. Doudna's research. Can you tell us how you're using CRISPR at Excision? 

Daniel: Absolutely. Can I just build on a little bit of what Jim just said? Because so much of what you are saying rings true to how we try to build therapeutics. So for just an example of what we're building at Excision, we're taking some of the foundational technology that Jennifer Doudna built over at UC Berkeley. One of our licenses comes from her and for the Cas12 family. So this is one of the original CRISPR that was described in that 2012 paper that ended up with the Nobel Prize. Described SOCAS9, which is a single bacterial protein, or a few bacterial proteins, SA, SP. Now there are hundreds, thousands, probably tens of thousands actually of variants that people have found. This has borne a whole cottage industry of people either looking for rare, CRISPR nucleases in places like tops of mountains and bottoms of oceans to developing novel ones even down the street from where we're sitting. 

So we use a whole variety of these in a way that is unique. We're trying to deactivate viruses, which is actually relatively similar to how CRISPR evolved in nature. CRISPR evolved in bacteria to defend itself against bacteriophage, which is viruses which attack bacteria. The innovation that the team in Kamelka Lele's lab over at Temple University came up with is how to use gene editing to deactivate viruses. But really simply, if you make a single cut in a virus, the virus mutates. We've seen how many variants of COVID -19 we've gone through in the planet in the last few years.

Other viruses are very similar. So what they figured out is how to use CRISPR and simply make more than one cut. Removing large sections of DNA, we can knock out activity. So that's what we're doing. But in order to bring it from that sort of relatively simple concept to proof points along the way, and now we're in clinical trials in the US, there's a lot of things we have to build along the way.

How do we measure it? How do we see these things are working? How do we make sure that we can ensure safety across species and then into humans? We're seeing wonderful safety, but it took time and we've had to build many things along the way. And now what we're doing at Excision and the reason we're in our next fundraising is for these next clinical trials for the efficacy stages of clinical evidence. 

Michael: Well, talking about using these new technologies for drug development. Melanie, how about if you tell us something about your work? 

Melanie: Yeah, that's right. And what Prellis is doing is that we've brought human biology into the lab, specifically the human immune response. And this is critically important because if you think about how the human immune system works, it is one of the primary producers of all therapeutics, antibodies, cytokines, even human cell responses that kill tumor cells, et cetera. And there was really no way to examine this in vitro before. We could culture human cells in lab, of course, but we could not get a fully adaptive human immune response to novel antigens, novel targets, et cetera. 

And so what we've done at Prellis is we've enabled that fully adaptive human immune response. And I kind of think of it almost like a natural AI or a native AI. We feed it something. We feed it antigens. We feed it proteins of interest and it gives us antibodies out that bind to them as a solution. And so that's really kind of what we've enabled in the lab. 

Jim: Kind of hacking the immune system if you want to think about it that way, right? 

Melanie: Exactly. We've enabled large scale hacking of the human immune system and we're collecting all the data along the way. 

Jim: It’s very impressive. As Melanie knows, I'm on the board of the company and I've been deeply impressed by what the company's doing. 

Michael: Completely fascinating. I want to turn it a little bit from to how we're using the technology and talk about what are the impacts on real world problems. I mean the role of health and medicine has so many challenges today, from the cost to access to labor shortages to quality of care. So let's talk a bit more about the problems we have the opportunity to solve with this new wave of health focused innovation that we've been talking about. So Daniel, let's start with you. The problem you're solving at Excision is very clear, cures for infectious diseases. 

So how have you determined which viruses to target? What are the results so far? How optimistic are we for where this might head in terms of real solutions? 

Daniel: You characterize this really well, Michael, that I think everyone on the planet is touched by medical need and true new innovation that needs to relieve suffering from cancer to viral diseases and others. In fact, there are several viral diseases that cause cancer. One of the diseases we're working on is hepatitis B, which is the leading cause of liver cancer in the world. So how we are going about this as using CRISPR, which is really the only nuclease we've been able to use effectively to make these multiple cuts and deactivate viruses. That was a transformative discovery for us. 

The founding scientists at Excision had already developed many of the models of some of the things that Melanie is talking about that are necessary. For us, it was cell lines, animal models, and the assays to test these for HIV, for hepatitis b and for herpes virus. We'd love to be able to apply this to a whole range of other viruses which cause all sorts of problems around the world. But starting with those three, because we always have to focus on something to begin, we've now moved those through and have some very impressive activity. So we can show cutting of these viruses in animals. We're the first company in history, and the only company in history to cure an animal of HIV. Single administration of CRISPR, removing HIV, deactivating it. And then we did a number of experiments, published this in Nature a few years ago, that showed there was no active HIV left in these animals. 

Perhaps just as importantly was demonstrating the safety side of this. This is always one of the things, anytime you're gonna edit DNA in someone, you have to demonstrate safety. In fact, this is the same with any drug that we call the therapeutic window, the efficacy versus safety. We're targeting viral DNA, which gives us a very large window of DNA sequences to target that don't resemble anything in the human. So we've never seen an off -target effect in the animals, which are now people that we've dosed. So we've cleared our phase one, two trial in HIV in the United States. We've dosed six patients. We just published that a few weeks ago at the large gene therapy conference of the year. 

So we've shown safety. We're showing some early signs of efficacy, which was the point of the first clinical trial, which needs to be safe, showing biodistribution, and now we can move on to broader clinical trials. 

Michael: Well, we're all pulling for you for sure. Melanie, so you talked about having a human immune response in the lab. Can you tell us how that works, you've to overcome to make that happen? 

Melanie: Absolutely. And then I want to go back to what Daniel's talking about, because that's just fascinating. 

So, how do we make this work? We bring together a few different technologies. So a lot of physics, a lot of biology, a lot of human immunology. And really, I think it comes down to the biophysics of how the cells are interacting with each other and enabling that to look a lot like a native human tissue. Immune cells are primed to respond. They respond throughout our body and almost every tissue. But the lymph node is really as one of my favorite professors used to say, kind of the bar where the cells gather to exchange information. It's where they gossip about what's out there. Right. And so recreating that kind of interface between the different cell types enabled us to recreate that bar. We set up a bar for the immune cells to chit chat with each other and then we gave them the antigen and then we get the native immune responses out of that. 

Michael: What are the commercial implications of this? How do we make this commercially available? 

Melanie: Right. So why this is commercially important and why it's so revolutionary in the field of drug development is that a standard antibody development process, discovery, etc. takes about a year and a half to two years to really get a good antibody that's going to have high therapeutic potential. Because we're not doing it in animals, because we're human from the get -go and we kind of sidestep a lot of the issues of the animals. And because we're using human cells to start to recreate these lymph node organoids, we get a massive pool of diversity. We condense this timeframe down to about two months and deliver better antibody libraries than what you will get out of animals. And so this enables kind of a lower risk approach to selecting targets, doing testing, developing antibody libraries, because you're not investing such high personnel overhead and cost overhead into the development of this. 

Michael: Well, thanks for that. actually like to, you know, it's really cool to talk about the technology. We're all technologists. We can do this all day long. But Jim, if I might go to you to talk a little bit about what downstream looks like and what I mean by that, you've had a pretty unique background of being both on the research side and the commercial side in science and healthcare. How do we ensure that these breakthrough technologies become available to everyone and not just to the rich? 

Jim: Well, you know, that's the problem of our time in healthcare. The reason that we haven't solved for it is that our system in this country is terribly fragmented and the component parts are at odds with each other and have incentives that are competing. And I think we all understand that. But as tough a problem as that is, it's only part of it. I think if you had to look at any one part that's sort of technology -based as compared to societal, political, it would be cost of goods manufacturing. 

Michael: That's what I would think 

Jim: And especially as that relates to the complex biological products. So, Daniel, one day when you're giving viruses a double haircut, Snipping away at them in the clinic, you know, I mean the price is going to be considerable and some of that is of course related to patents and and what what a company can get and deserves to get based on the amount of effort and the research that's gone into it and all the failures I think we all understand all that. The manufacturing costs are a significant part of it. So if we're talking about expensive biologics, yeah, I mean it's really true that manufacturing costs a significant part of them, and especially for the first generation of gene therapies. I mean, I've seen numbers to suggest that actually every sale comes at a negative gross margin. Not obviously in every case, but even in the very first cases. 

So part of it is going to come from scale, simply as the markets scale up, the usual, the prices are going to come down to some degree, but that will be offset by the cost of goods. So different answers for different categories of product. For the biologicals that are based on viruses and virus -like production, I think we will see economies of scale, but not as much as we'd like to see, that they're going to be inherently expensive. 

But I think the answer is going to have to come from distributed local manufacturing rather than large centralized manufacturing. For monoclonal antibodies and the like, we've seen an almost complete failure of biosimilars, which have been attempts to lower the cost because of the regulatory issues that relate to the complexity of product. 

We're going to need to have a technology that normalizes the product itself. believe, for example, carbohydrate chains and the variation that are probably the single biggest cause of batch losses in manufacture. I don't think that's beyond the realm of technology to achieve. Overall, the largest answer is going to come from finding solutions that are disposable, localizable, and in effect, will come from the conversions that you started out with with micro. 

Michael: Well, one of things for, Daniel, mean, obviously, if you can cure a disease as opposed to it being a managed care situation, that obviously will reduce costs. But why don't you address that? 

Jim: That's a really good point. 

Daniel: I think you're absolutely right. Today, certainly, the costs of these complex, high tech therapies are exorbitant and coming down. But it also depends on what are the incentives to bring these down. I mean, we can talk about mainframe computers, classic Clay Christensen disruptive, right? This is what are the incentives and where, frankly, where's the money in bringing down the cost. Now, for some of the original, and this is true for maybe any new therapy and new technology, we start with something very small, whether it's mainframe computers and giant universities. Start with only a few, they're extremely expensive. Now, how do we bring these to larger markets? So one of the things, larger markets, one of things we're trying to do is get into markets with millions of people rather than just thousands or tens of thousands in case of hemophilia, which are the original indications, which CRISPR is approved for now. How do we get it to millions? This brings incentives to manufacturers to bring the cost down from the very expensive to, truly, how do we make this a dollar? We have to get there someday? It does take time. 

Jim: It's long way to go there from a million. Now there are several. But I see this is more like the automobile. How do we break? There's so many technologies to bring automobile costs down. not everyone, but many people can afford them in different ways. There's lots of technologies now that are finally. was promises of gene therapy saying, we'll bring down the cost. Some of these are only happening now

adherent cells to suspension to cell lines that are 10 times more productive. I could say that what we're doing is we're focusing on this in a way that we naturally would coming from highly developed countries and rich countries, which is, you know, we're talking about the very high end drugs, OK? Actually, globally, there's another level of problem, which is workaday drugs that are, you know, pills like statins, or blood pressure control, relatively simple things that we take for granted, most of us, that are off patent, et cetera, are still far too expensive than the Global South, for example. So there's another level of manufacturing improvement, which I think, even for small molecules, we make them in these large factories on an enormous scale using tons and tons, literally thousands of tons of organic solvents, creating waste problems and so on. When in fact, the technology exists today to make the very same molecules in a much more compacted, condensed and flexible way using enzymes. 

Melanie: So almost like a point of care manufacturing system. 

Jim: Yeah, almost like that. And it becomes more like a biological manufacturer on a limited scale in a kind of disposable bag. And that technology hasn't been adequately invested in today. It requires more R &D. But actually, I do think that there's a huge opportunity to improve public health. Not just in the global south, but also in our own countries. 

Michael: Well, I'm going to come back here and talk a little bit about regulation, which has a big impact here. But before I move into that, Melanie, so you have in your company, in your lab, your technology, the ability to really to create drugs at a much, much lower cost. You have a point of view about how we get from there into people's hands, whether it's in rich countries or poor countries, at the lower cost levels that really should result from what you're doing.

Melanie: Yeah, so I think there's a long history of therapeutic development that is extremely expensive initially, not widely distributed and not available. And then as time goes on, costs come down, things go off patent, things are generally more widely available. Jim brought up a fabulous point that, you know, we don't really have point of care production. And I would love to see a future where a pharmaceutical distribution center could do their own point of care on demand with a prescription coming in. That would be game changing for global health. Absolutely. What we are doing that I think is going to have a major impact is that we are reducing the time for discovery and reducing the risk of discovery because we're reducing the overall cost and overall personnel investment.

And I think that there is room in the world of bioconvergence with all the data that we have for more intelligent development. So we reduce clinical trial failure. That's a major area of waste in therapeutic development. So all of these both enable costs to come down, time to be compressed in a way that's revolutionary compared to prior methods of drug development and higher risk ventures to be taken on. 

Let me give an example. If you look at clinical trials in clinical trials .gov, the majority of cancer clinical trials are going after the same target. Different bivalent configurations and maybe different clinical protocols behind these clinical trials, but they're all still going after the same molecule in general. And I think that soaks up a lot of research investment and time into one area because it's relatively low risk. We know these work. We know that they may offer a better solution for certain cancers. However, it does suck the oxygen out of the room for other therapeutics that could be developed and may be useful in different populations. And so I think by reducing the risk of going after some of these other targets, reducing the risk of going after different diseases, effectively tackling other areas that may not have been so attractive in the pharmaceutical business. 

Michael: Well, let me throw something out there. And it's not something we've previously talked about really, but as we get to lower cost development of drugs, that means that we will be able to develop drugs for much smaller populations. Do you guys have thoughts about the regulatory framework, which really is going to have to change if we have that? We can't wind up in a situation where we have a drug for, you know, a hundred thousand people. Curious what you guys think about what changes need to happen and can they in the regulatory framework? 

Daniel: Actually, our interactions with the FDA have been outstanding on this.

Michael: Let's hear about it. Let's hear about that. 

Daniel: So they've given some draft guidance. And those are some very broad words in terms of the FDA. But they've actually published some draft guidance on the ability to change submissions and drug applications based on different sequences for different people. So one of the challenges and one of the expenses of pharmaceutical manufacturing traditionally over the last 100 years is once you set a drug formulation and manufacturing schema, you have to stick to it through the entire process. And if you want to change, frankly, anything, a single reagent, a single technological advance, you've actually got to go back and restart. And it takes a lot more time and money. So being able to change something on the fly, as you're describing, would add so much to contribute to lowering costs, speeding new therapies, and to develop individual therapies. I think the FDA sees this. And Peter Marx's team over there at the moment understands this really well.

Now, the devil's always in the details on how do we do this. So Excision's experience with the FDA has been outstanding. This is going back conversations over many years now in that we explained what we wanted to do. They clearly understood it and then encouraged us to have a more comprehensive, almost more aggressive position. For example, the traditional way to do a safety study. Once you have a new drug, you test it in a person, you do what's called a phase one study, which is healthy individuals. For our indications, we were trying to cut out HIV. Well, the FDA clearly understood. Doesn't make any sense to try to cut out HIV in someone who's HIV negative. That doesn't make any sense, and you're just exposing someone to potentially unnecessary risk. So they said, we should look at people only who are infected with HIV so that we can determine whether this is safe in these individuals. And let's not give them a dose that may not be able to show any efficacy, can we cut any HIV, but let's give them a potentially therapeutic dose.

We completely agree. This is exactly the right way to do it. And it actually probably advanced our technology a couple of years. That's fantastic. That's a surprise to me. It's outstanding. We have other companies that are in front of the FDA that have not had that experience. 

Michael: So I'm glad to hear you have. Melanie, any thoughts on this? 

Melanie: Yeah. The modern FDA is a little more amenable to personalized therapeutics. And I think that's the right way to go because there's a lot of advances that will come from understanding the basic genetics behind different responses to different therapeutics. I think of COVID as an example. We even have different responses to different viruses. And there was some recent, very elegant work done at UCSF showing that there is a genetic basis for some people having completely asymptomatic COVID infections. They really don't get sick. And some people, you know, having a devastating course of disease.

The more that we can focus on that, understand it, and maybe do a little bit of semi -personalized immunotherapy and have an FDA that understands that, the better for everyone. 

Michael: That's great. Let's all hope this is how things turn out to be. So the last topic I would like to cover is look forward a bit. So let me start with you, Jim. You have a lot of experience in academic institutions and there's lots of new research going on. Are there things that you think that we need to do better connect, if you will, the academic institutions, the work that's being done in those institutions with the companies, with the commercial applications? And is there a change in mindset that you think needs to happen, or has that already happened? 

Jim: I think that it's already happening. There's sort of two questions there. Yeah, it's certainly happening that academic medical centers and so on would recognize they should have and they desire to have a prominent role in development of applications of the science that they discover and foster. So much so that it's actually, in a way, becoming counterproductive. The university is where the discoveries typically have been made that broadly illuminate whole areas of biology and medicine and therefore enable pharmaceutical developments ultimately. For example, Daniel, the CRISPR technology that Doudna, Charpentier and others discovered came from university -based studies of how some very odd maritime organisms managed to survive and other ones didn't, what turned out to be a virus -like infection, right?

 But what we're seeing in the modern medical environment and medical schools is a kind of almost desire to take the place of a pharmaceutical company and it's a bit misplaced. And what I mean by that is that, is there recognition? Yeah, there's an enormous amount of actually pressure within medical school environments now to do translational research that has an application. And there's nothing wrong with that. In fact, it's really good. There was too little of that in the past. Now it's reached the point where it's frankly, crushing fundamental basic research. And yet there's a point of confusion about rules. And I see this very frequently, which is that there's somehow the illusion that drugs can be developed within universities. It's just not really possible. It's not culturally possible, it's not financially possible, it's just simply not possible. There have been a few one -off examples, but the result is that resources are actually being taken away from true innovation to focus on drug discovery without a license, if I can put it that way. Imean, pharmaceutical companies are really very well equipped, not just financially, but culturally, in every other possible way, within a professional fashion, to go after drugs. 

And as Melanie said, it would be nice if they could do it in a broader way, in a more inclusive way. But because it's such a high -risk venture, of course pharmaceutical companies cluster around low -risk areas. There's no avoiding that. That's the market. The real answer in the end that will elevate success rates, whether they're in academic medical centers and actually enable academic health centers to begin to really meaningfully discover drugs and change the curve on success rates in clinical trials, the single biggest lever, even more than manufacturing, probably even more than the market considerations, is just gonna be discovering targets. When you get a failure of a clinical trial, it's not because the drug company was incompetent, they made a bad chemical. The chemical did exactly what it was supposed to do. It addressed the target, and yet the patient didn't respond or respond well enough. Why? That's what we mean by saying it wasn't a good target.

So what we have to do is come up with ways to get better targets. Well, with all of the knowledge that we have in the causative biology that we've been developing, you know, over the last 20 plus years since the genome, we haven't succeeded in doing that. Actually, we failed. Okay, so what we do is we turn to the newest game in town with hope and admiration. It's called AI. And we look breathlessly at it. And then we first believe that it will be the answer. I don't know actually of any better hope, today. And if it succeeds even 10 % as well, as we hope it will, it'll help us choose better targets. It'll make the economics much more realistic, even possibly for schools of medicine. Certainly will allow pharmaceutical companies to take on a broader range of targets. So that's, I think that's what we really have to push for. 

Michael: Well, as we're getting to the end and predicting the future, and we're talking about artificial intelligence and every podcast in the world has to talk about artificial intelligence.

Jim: But I can testify that all four of us are here in person. And that we are all real sentient human beings. And none of us are avatars or artificial products. 

Michael: Well, as you guys all know, there's this whole discussion, this philosophical discussion about if artificial intelligence is going to destroy the world and all this stuff. Daniel, there's also a discussion about the ethical implications of gene editing. I mean, are we going to get to a place where, you know, people pick their physical characteristics and pick what sports their kids are good at and all that kind of stuff. And that's in a similar range of discussion as we have in artificial intelligence. What is the implications of these technologies in the future? Do you have any thoughts about that? 

Daniel: Yeah, certainly. There's no shortage of sci -fi movies that has to do with gene editing and how this can go wrong. And same with AI. think, Jim, you're exactly right in that new discoveries, in different places, universities or small labs or even potentially someone's garage, as you can, there's lots of articles about CRISPR being developed in people's garages. You can do some of this work, but this is not, as we like to say in the industry, pharmaceutical grade. The implications of this are all of the different controls you need to have to actually have a therapeutic that doesn't hurt anybody. I this is really the purpose of the FDA, going back to your regulatory question. How do we not hurt people? Make sure it's safe. That's job one, or at least as safe as possible. You can look at chemotherapies and say, that's not safe. But where is the benefit versus risk? It's really about that. so bringing it back to your question of what's the benefit versus risk, clearly trying to use these gene editing technologies to cure disease and relieve suffering, I think we can all say, yes, that's going to be a useful one. Could someone make me smarter with gene editing? I love blue eyes. Maybe in the future in some way. 

But this is the challenge of science. If you ask a scientist, could this be true? Or a physicist, these are my favorite things. Could this be true? The answer is always, well, it could be true, but in what type of universe? And it may have to go into a new string theory to prove it. The point I'm trying to make is every real lab I've seen is trying to relieve suffering and cure diseases. Do we need guardrails? Absolutely. That's why there have been, at least in most of the developed countries, and at least it's nice to see China put people in jail who are trying to do something else.

So even countries that we tend to say, they don't follow the Western rules. No, I think most of the planet is all on the same page. 

Michael: So I can't, not gonna be able to take a drug to be six feet tall, darn. 

Daniel: Not today!

Michael: All right, in wrapping this up, I would like each of you to make one prediction of a medical breakthrough, drug breakthrough, healthcare breakthrough that will become the most valuable in the world, just, take a shot at it, Melanie, go first. 

Melanie: Wow, that's a tall order. So I'm going to go back to what I was talking about earlier, where I talked about semi -personalized medicine, where we start to categorize people better. We understand how therapeutics will work in them better. We'll have better results. You clinical trials are run and they look for some sort of statistical significance above placebo. And often it's not a great p -value.

But we know it kind of works and understanding why it works for some people, how it's working for some people, being able to better select for those patient populations will be hugely valuable in the future. 

Michael: Daniel? 

Daniel: On a global scale, unfortunately, some of the lowest tech can be the most meaningful. Like clean water for most of the world can relieve so many ailments and so many diseases. So in terms of dollar per impact per person on the planet, we're probably looking at everything from mosquito netting to clean water to sanitation and antibiotics. 

Melanie: I'd like to change my answer. 

Daniel: That's only because I have a degree in public health. 

Melanie: He's correct. 

Daniel: What I'd like to see is what Melanie's building, and this is the bending the curve piece. One of the biggest challenges of pharmaceutical development is why is it so expensive? Why does it take so much time? And getting more expensive, taking more time over.

This is going the wrong direction in terms of bringing new therapies to people. Why? Well, the technology is getting more expensive. And the FDA says, well, we now know how to measure these things. So you have to go do this to make sure it's safe. And so the next thing you know, your development timelines have advanced in multiple years in tens or hundreds of millions of dollars more. How many billions of dollars now? The $3 billion to bring a product to market? OK, so the new technologies that I'd like to see are the ones that can bend those curves. How do we bring it down orders of magnitude to do that? So part of that, I agree with Melanie, is make it individual. The other is what technologies can apply, AI, new or what you're building is the database is necessary to make it faster, better, cheaper to bring new therapies. 

Michael: All right, Jim, you got a wild forecast for us? 

Jim: Yeah, I'll try to come up with something out of the box. That would be great. I actually think that it's going to be hacking the brain. So what do I mean? I'm good to hacking. 

Michael: All right. We're listening. 

Jim: What I mean by hacking the brain is that, you know, there's actually two nervous systems that we have. There's the brain and there's the rest of it. Okay. And we tend to focus on the brain, but actually we have this entire nervous system that every organ, every part of every organ is wired with sensors that report up through the vagus nerve. I mean, the vagus nerve, like in medical school, you'll learn that the vagus nerve controls the heart rate. It controls the contraction of the intestines. 

But actually the truth is that only 20 % of the fibers in the vagus nerve are actually going outward from the brain, namely controlling. 80 % are actually reporting signals and information. Now, if you could hack those wires, if you could hack the brain to convince it, for example, that you had way too much sugar, okay, you would actually cause a reflexive change in the body probably that would lower your blood sugar. There's examples of this. There are sensors in the liver that actually report out how much sugar you have in the blood coming out of the liver. And there are other sensor, nerve cells of course, that report up and sense how much sugar you have going into the liver. And by doing a comparison, there is a continuous report to the brain of whether you need to make more sugar, mobilize glycogen, break it down, or maybe you have too much and you have to put it away. in glycogen and fat. It all comes to diabetes and metabolism. So if you could actually stimulate the appropriate nerve with an appropriate medical device and actually convince the brain that you had too much sugar on board, you might not need to take insulin or another drug to control your diabetes. 

And there are actually publications to this effect that when that is done in experimental animals, you simultaneously lower cholesterol, you'd simultaneously normalize sugar, et cetera, et cetera, et cetera, the entirety of metabolic syndrome. So there's a huge opportunity here. And why does it sound so alien when you hear about it? Because we don't know about it. Why don't we know about it? Because nobody pays for research in it. Because everybody wants to understand the brain, but everybody's forgotten actually that there's a whole other nervous system. It might only have 1 % of the nerve cells, but it might actually control 90 % of what's wrong with us. 

Michael: Well, I will say every day I go to work at Celeste Capital, we think about what an incredible world we live in, that we have all these people working on all kinds of different technologies. Today we talked about healthcare and so on. Jim, I need you to correct a number that I have used, but I have told this many times, that when you and I first met, you made the comment that if you would have asked us biologists 10 years ago, how much we understood about how biology works, we would have said something like 50%. And you said if we asked biologists today, you'd say probably less than 1%. Do I have these numbers approximately correct? 

Jim: OK, so I'm trying to think about that one. Right. So my answer a bunch of years ago would have been very granular, very specific answer. It would have been, ok, we've got the sequence of the human genome. There's like whatever it is, 25 ,000 genes or something like that. And we know specifically how, I don't know, 12 ,000 or 10 ,000 of them, we've assigned functions to them. We maybe actually even know a lot about what their protein products do. So I would have probably done a quick mental calculation. I probably came up with something like 50%. It might have been a little high because I'm an optimist. Maybe whatever it was, something like that.

Now today, as a result of all the other things that we've learned, biologists have come to the conclusion that, you know, it's not all in the DNA. The DNA is of course a great part of it, but there's a lot of individual variation and there's a lot of variation so -called epigenetics that goes above and beyond the DNA and which is nonetheless inherited. It has to do with chemical modifications of the classic DNA sequence, and it has to do with proteins that are modified that are in those modifications being associated with the DNA containing chromosome and inherited. 

And so I'd have to probably come up with a much humbler answer, okay, because by now we just keep coming up with confounding results with drugs and other biological insights that coming back to Walter Isaacson, which is kind of where we started out. We're hoping that we're at the beginning of a revolution. because it seems like what we understand at that level is less. We are understanding a lot of very specific things and that's going up, but the ability to translate it has not gone up. And so we need a revolution. We're hoping it's AI, but it may not be AI. But we do need a revolution and I'm an optimist. So I think the future of biology and medicine is really, I don't want to say endless, but I'm an endless optimist that the next 50 years are going to just tremendous. 

Michael: Well, completely fascinating. That's a great way to end this. I think we're probably all optimists here. I want to thank again Dr. Jim Rothman, Dr. Melanie Matu and Daniel Dornbusch. It's a delight to have you do this with us. Thank you very much. Thank you.

Thanks for tuning into the Tech Surge Podcast from Celesta Capital. If you enjoyed this episode, feel free to share it, subscribe, or leave a review on your favorite podcast platform. We'll be back every two weeks with more insights and discussions of all things deep tech.