For episode 18, we sit down with David Berry, CEO, and Board Director at Valo Health. Having founded over 25 companies, find out what crucial piece of advise he received early on in his career as well as what lead to the success of Valo’s first in human for retina trial last year. First In Human is a biotech-focused podcast that interviews industry leaders and investors to learn about their journey to in-human clinical trials. Presented by Vial, a tech-enabled CRO, hosted by Simon Burns, CEO & Co-Founder & guest host Co-Founder, Andrew Brackin. Episodes launch weekly on Tuesdays & Thursdays.
Simon Burns: David, thank you for joining us on First in Human.
David Berry: Thank you. It’s great to be here.
Simon Burns: You have an incredibly impressive background, GP at Flagship, CEO of Valo. You’re doing fantastic work. You’ve been recognized in a, in a whole bunch of different ways. Tell us a bit more about your background and what led you to starting Valo.
David Berry: I had the pleasure of doing my undergraduate at MIT. I did an MD at Harvard Med, and a PHD at MIT in biological engineering working with Bob Langer and Ram Sasisekharan. What was great about the educational background is those are just areas where the sky is the limit. The opportunity is what you seek. Bob Langer gave me a piece of advice early on in my career, which is that all problems are equally hard to solve, so why not go after big ones. And it’s been something that’s really stuck with me.
After school I joined Flagship Pioneering. I spent a decade and-a-half with the firm. It’s been fascinating because the mission and vision behind Flagship is to do unprecedented things. To imagine the future that we want and built it. I had the privilege of working with a number of great people over the years taking what were big ideas at the time felt completely foreign, and trying to figure out how we can build them into a reality where some of them, when we talk about them today, almost feel like they were obvious to a point that people have actually said, “that idea was obvious when you created it.” If you actually had the conversations at the time it wasn’t. But it’s fine.
That’s been been really exciting to me. What ended up taking me to Valo was, in 2018, and 2019, it occurred to me that there was a transitional opportunity that was emerging at a scale frankly I haven’t seen before. That was really exciting. What that is is when you think about how the tech world has disrupted a range of industries, there’s a very similar pattern in these cases, which is, you often take an industry that has siloed data, and a siloed operational structure. When you can get the right scale and quality of data, with an ability to build an integrated architecture, an ability to use that data across the architecture, you can effectively build a new model, one that is, instead of locally optimized, systems optimized. We’ve seen this over and over in the tech world because when you take that systems optimization, you can then effectively use it and become a new standard.
Pharma has been historically recalcitrant to this kind of approach. But, what we saw was that the data framework for that was starting to change around 2018-2019. That led me to say, “Wait a second, there’s an opportunity to build a big and different company here.” If we had that opportunity to build an integrated end-to-end drug discovery and development capability that’s anchored around human data, we could reduce the cost of developing drugs, reduce the time, and increase the probability of success.
If we could do it at scale, it could transform the way we think about drug discovery and development. That opportunity, where you could see something that could benefit patients en mass, where you could think about the positive benefits across an industry, it was tremendous from my perspective, so got me incredibly excited to focus on it.
Simon Burns: Do you have a count of how many companies you’ve started at this point? It must be over 25. We would love to hear your insights having seen that many biotech companies that you’ve been a part of the founding story of. What are the key lessons, the key insights, that you give to early stage biotech companies who are seeking your advice?
David Berry: Every company that gets created is its own being. It has its own way of success and its own way of failure. Learning about those processes is incredibly important for the company.
One of the things that we did at Flagship was we built, effectively, a stage gate model for a company creation. This model is is a four step model that basically works through an exploration, a proto-company, a new co, and a growth co. The intent on this is to really take out a number of those generalizable fail modes of companies.
The exploration is, you start with a big question. You do a couple things. One is you separate truth from belief: that is every field is filled with dogma. Dogma is there because it’s the language you need to succeed in those fields. It’s not necessarily right.
I can give you great examples we get taught all the time that we use three to 10% of our brains. That comes from science that was done in Germany in the 1800s where blunt instruments were put into the skulls of rats to see changes in the way they function and it worked in about three to 10% of the volume.
We’ve all heard those numbers. It’s interesting because it’s been [00:05:00] passed down. Reality is you go into a PET scan or you go into an FMRI machine, we know humans are using all of their brains. But yet, that number continues to exist. At the same time, it’s not things that seem esoteric. Kindergartners get taught that cancer is caused by genetic mutation. This goes back to 1973, when trans genes were inserted into cells were shown to be sufficient to create cancer.
You need ” necessary” and “sufficient.” The “necessary” wasn’t demonstrated. No disrespect to the science. It just took off. Of course, we all know that genetic mutation is found quite often with cancer. But there’s cancers today that have only had epigenetic changes and not genetic changes. It’s not to say we should upend the whole field. But, you need the right facts to start.
We then start asking a bunch of ” what if” questions which allows us to imagine the future. But we need to create something tangible. We use basically a methodology where we can engage the community and start figuring out where out hypotheses are right and wrong.
The second most renewable resource in the world is people’s willingness to tell you how bad your idea is. It’s great feedback because you can then start iterating your ideas really quickly. We hone it down to one, and then prototype it. It’s a proto-company. That allows us to test the company very directly and very quickly. It’s like prototyping a car or anything else that you would machine and in that case, you ask the question: What’s the key fail mode? Entrepreneurs don’t like asking this question because if you get the negative answer, you have to shut down the company.
But for us, it’s an experiment before it’s a company. That allows us to go after these key fail modes. Then, we go and start building the company as a new co. During that phase, it’s important for us to have it under our own supervision. Something fascinating happens, always within the first 18 months of a company. You get the unexpected bit of data. You never know when it’s going to be or what it’s going to be. But data’s data and you gotta follow the data. When something new and unexpected comes up, you have to embrace that moment and figure out what it’s telling you around what the platform might actually have.
After that, you start realizing where the sea legs actually are and that’s when we can convert it into a growth co, which is what you would think of as a standard company that’s growing. Now, the methodology builds off of a lot of experience of starting companies and recognizing where traditional issues are.
What we’ve tried to do is to schematize as much of it we can to take out that generalizable risk. The real lesson that I’ve taken over-and-over is: you have to follow the data. We all have best intents, the desire to make things successful, but, the probability that we predict to the exact perfect path to develop something on day one, with a very limited amount of information, is close to zero.
Once you embrace the fact that there’s learning opportunities all over the place, you make yourself into a learning person and the organization into a learning organization. And it just grows that much faster.
Simon Burns: 2023’s been an exciting year for all things AI. We’re all talking about, Chat GPT, about TP three and four. The rumors of all kinds of incredible capabilities coming around. But. Biology and chemistry’s very different. You guys are focused on mapping a lot of human biology and chemistry through the power AI. Give us a sense of some of the key challenges in building AI in your space. What, if anything, do you think is changing because it’s in the advances we’re seeing coming out of opening out of other companies?
David Berry: Chat GPT’s a great example. I love what it’s doing and where we can all see some of the transformations that are going to come out of this. It’s a really really powerful tool. But, it’s imperfect, right? If we go in and we ask it to write a 750 word essay, the sentences are great English but the content sometimes is off. That’s okay because we have the ability as people who review that to then fix it up. it’s a huge time saver. I love it. It’s great because it’s showing off the power of AI.
Drug discovery and development is similar and different. We can use the same types of data, obviously it’s different in structure. We have to integrate it across a different set of frameworks. We can start understanding how do we design molecules? How do we design drugs, and use AI to give us a learning capability that just exceeds the scale of what humans can traditionally learn because of the massive scale of data that goes into chemicals, right? There’s 10 to the 63rd possible chemical structures. There’s no way chemists can imagine all those structures and put them all in their head. No disrespect to them. It’s just that’s not how the human brain is designed.
We can start experimenting on much larger domains of chemistry. Exploring much larger numbers, of variants, and that’s great and incredibly powerful. Where there’s a huge difference, and this is where it’s important, is that if your chemical structure’s a little off, you can cause massive toxicity in a patient.
We don’t have that ability to put it out there and and, see what happens. We have to put it out there and put all the rigorous testing around it. That gives us our own feedback cycle, which is really important. Gives us a different architecture of data. That’s why we’ve seen the evolution in this field seemingly a little slower than what we’ve seen in the world of something like, Chat GPT or DALL *E or any of these other types of approaches, which are great and I’m really excited by.
It’s that rigor that we need to make sure we have on the first, time that it comes out, to make sure we’re treating patients well, we’re treating [00:10:00] them safely, and we’re ultimately making effective drugs. I expect Chat GPT to get there over time. Where we’re getting prose that looks better than anyone living can make, just like what we’ve seen with Go. What I fully expect that we’re going to see with drug discovery and development. We just have to make sure we put the right boundary conditions because, as we all know, “first do no harm.”
Simon Burns: Of course. Last year was very exciting for Valo. You guys had your first in human for a retina trial. A lot of progress clinic on heart failure. Give us a sense of what’s going on in the clinic for you and what are you excited about?
David Berry: We have two, phase two molecules. We’re very excited about them. One of them is a ROCK-12 inhibitor that we’re developing for, non proliferative diabetic retinopathy. We’re very excited about this for a couple of reasons. One is because patients who have diabetic retinopathy, and this is about a third of patients who have type two diabetes don’t have great treatment options today. They can go through surgery or they can get the delivery of drugs, and better to have drugs than not, that are administered through a direct injection into the eye.
Part of what we’re excited about is this drug is orally bioavailable. It gets into the eye after being administered orally. That’s exciting because we think it can offer the patients a better route of administration, lead to potentially higher compliance, and maybe that leads to a better benefit for them.
But the other thing that we did is we were able to use our computational tools to try to predict which patients are most likely to progress from one stage to the next stage of diabetic retinopathy. That helps us to identify the patients that we think are most likely to benefit from the drug.
This allows us to start matching up drug to patient benefit in ways that are better for the patient, for the health system, and we’re very excited about that. We’re actively enrolling a phase two study right now very excited to see how that’s going to go. Obviously we’re very hopeful about that.
The other one that we’re going after is a unique drug that is a functional S1P1-biased agonist. There’re S1P1 drugs out there, but, they functionally antagonize the underlying biology. That leads to a very powerful anti-inflammatory benefit that’s been used for multiple sclerosis, ulcerative colitis. Those are proved drugs that are great.
What this drug does by being selective is it actually benefits cardiovascular specific biology. We’re very excited about that. We’re developing it along what I like to call the heart failure axis; that is, we can go after indications from acute kidney injury to, heart attacks, to heart failure with preserved injection fractions, that’s very exciting to us.
What we’ve been trying to do is understand who are the patients who are most likely to benefit? And how can we administer the drug to those patients? That’s been very exciting to us because each of these is allowing us to explore and develop and validate, hopefully, computational platform in our own way.
Simon Burns: We saw you present at DPM a couple weeks ago now. We’re or, maybe a year into the the XPI bear market, M&A still not coming back. What’s your sense of all things of the current state of biotech?
David Berry: The fundamentals of biotech as a technology remain incredibly strong. I mean, As I look out at some of the drugs that are being developed, I don’t think I’ve seen a more exciting time than the present. Whether you look at things like KRAS inhibitors to what’s being developed in the CV world, it’s an exciting, time for new drugs that are being developed. We don’t have to look too much further than the COVID vaccines to say this industry is doing unprecedented things at an unprecedented rate, and that’s so exciting.
Obviously, the financial markets turned. They didn’t just turn in biotech, but they turned pretty bad in biotech .Last year was a rough year if you were a public company. Anytime people put out news they were more likely to be met with a decrease in their stock price regardless of what the news was. But that’s what you get in bear markets. What was really encouraging to me around JP Morgan was people were starting to focus in on: what are the companies? What are the drugs? What are the technologies that are going to lead to that near, mid, and long-term transformation. People really grafting that kind of future opportunity.
That’s what biotech has been about. What that opportunity to deliver transformations for patients ultimately can be. Watching the field come back to that is tremendously exciting. In any bear market there’s always doom and gloom conversations. I also heard a lot of hope and optimism and that was, ‘s incredibly exciting from my perspective.
I’m hoping that 2023 becomes another one of those years that we talk about where we’re seeing a science first year, we’re seeing a patient’s first year, and that opportunity for the industry to deliver unprecedented benefits is something we’re going to get the opportunity to see a lot more of.
Simon Burns: Lastly, the clinical trials you’re building for scale, thinking a lot about how to scale the discovery phases. Obviously, we think a lot about scale and the applications to technology in the clinical phases. Um, Where if, anywhere do you find that there’s, a kind of exciting room for technologies, room to drive efficiencies in clinical trials?
David Berry: There’s a tremendous opportunity to drive efficiency in clinical trials. I mean, I tend to break down clinical trials into two or three components, but, I’ll focus on two. One is how you design your study the other is how you execute your study. The design there’s a tremendous [00:15:00] opportunity to think about how we can match drugs to patients at the right time. That’s a big part of what we’ve been focusing on at Valo.
If you can get the right intent to treat population, you can make better trials. You can also get drugs that are designed to do better for patients. That’s really exciting. Of course, we’ve also see some improvements in the operational side. It’s not what we’re focused on, but it’s something that I’m very excited about because it’s not just about matching the drug to the patient but you gotta find that patient.
The more that you can streamline the ability to identify that patient and enroll them in a study, ,of course, the faster the study goes. It’s also beneficial for that patient because they’re looking for that treatment. So, the more that ecosystem, if I can be passive about it, is helping to match those drugs to patients, and do it faster, do it in a more seamless way, it’s better for everyone. I’m very excited about many of the companies and technologies that have been emerging to help make that from a idea into a reality.
Simon Burns: That’s great. With that, David, thank you so much for joining us. Appreciate the time.
David Berry: Well, Thank you. It’s been great to join.
