Episode 68: The Future of Precision Medicine, AI, and Biotech with Key Opinion Leader Veronique Bouchet

It's hard to understand how AI and precision medicine really impact patients' lives. Well, today Precision Life tells us how they are doing just that.

You're listening to BIO from the Bayou, empowering biotech leaders with the tools, assets, and expertise to improve healthcare outcomes.

Each episode provides insights and information to expand your network, advance your product line, and help your company thrive.

Serving the entire Gulf South, but recorded in the heart of New Orleans.

Welcome back to BIO from the Bayou. I'm James R. Zanewicz, the Chief Strategy I'm James R. Zanewicz, Chief Strategy Officer at Tulane Medicine. And today's guest is Dr. Véronique Boucher. Véronique is the Chief Medical Officer at Precision Life, a tech bio company that utilizes artificial intelligence, or AI, to accelerate drug development and personalize healthcare. Welcome to the show, Véronique.

Thank you. Thank you for having me.

I'm so excited to get to see you because we got to see each other in person at BIO on the Bayou, and now we're getting to do the podcast. And you have had such an amazing career in pharma, biotech, and digital health. So you really have kind of spanned the where we are now in advancing healthcare. And we've talked a lot about AI in medicine on this show and are really intrigued by the growth of this technology. And so I kind of have a 2-part question for you. First, I'd love to know from your perspective why it's so crucial for us to tackle complex and chronic diseases today. And then I'd love to get your take on how AI can help us do that. So maybe if we tackle the disease states and why it's important for us to address those first and what kind of drives your passion in that, and then how you're leveraging technology to do it, that'd be great.

Sure. So if we look at what's happened to date, the Human Genome Project, the precision medicine approaches it's enabled, well, that's really transformed cancer care. But these approaches haven't been as impactful in these major complex chronic diseases to date. And that's because, and when we're talking about complex diseases, what we're talking about are cardiovascular, mental health disorders, dementia, diabetes, endometriosis. They're caused by multiple factors: genetic, epidemiological, environmental factors. And the traditional precision medicine tools— whole genome sequencing, GWAS, PRS— they just don't detect sufficient signal in those diseases. And that's because they're much more heterogeneous and polygenic than most cancers. They're difficult to diagnose, and they're often poorly managed, and they affect billions of patients. They account for 80% of healthcare spending. And in the US, latest figures show that's $4.8 trillion.

Oh, wow.

Yeah, that equates to 18% of GDP. So—

And you said trillion and not billion with a T, right?

Trillion.

Yeah.

With a T. Yeah. So to address your first question, that's why there's an urgent need to address the rising costs of these diseases, because as far as we're concerned, the only way is up at the moment with an aging population.

Mm-hmm.

So it's key we solve them. So to answer the second part of your question, what can AI deliver? So to deliver precision medicine beyond oncology, we need a much better understanding of the different biological mechanisms driving these diseases within subgroups of patients. That's precision medicine, right? So first, we need access to large patient populations that have well characterized and multimodal datasets. Then we need novel analytics that can identify combinations of these factors, genomic and non-genomic features. So we can then stratify what are very complex diseases into multiple distinct forms. And that's where AI comes in. At PrecisionLife, we have a novel AI-enabled platform It's a combinatorial analytics platform, and it allows us to put patients into very distinct subgroups with different disease causes and influences. I'll give you an example. In early onset Alzheimer's, what we've identified is that it's not one disease. There are 6 major subgroups of patients, each associated with a distinct biological function. which will determine, of course, how they respond to specific drugs. And we call this mechanistic stratification. And we've done this stratification to subgroups now in well over 50 chronic complex diseases. And all these subgroups can then be studied. And we can find novel targets for drug discovery or repurposing. But we can also identify the patients most likely to respond to a given drug. based on their personal genetic makeup, phenotype, comorbidities, and co-prescriptions. But what's really cool is that we can also use AI to innovate into clinical practice. Because what we can do is predict patients at risk of developing either a disease or its complications. So, for example, in type 2 diabetes, we can identify very early the patients most likely to suffer renal failure and who could benefit from a drug called finerenone, which has been shown to reduce its progression. So you can see that in the health sector also, there's a fantastic opportunity to improve use of resources and reduce costs. But what really matters is making a real impact on patients' lives.

So you're actually able to kind of take what we've traditionally had out there and make it better and more effective and perhaps figure out when to use that, more effectively. Is that kind of what I'm hearing from that?

Yes, absolutely. I mean, that's the principal tenet of precision medicine, isn't it? It's finding the right drug for the right patient at the right time. But we can extend that to the right interventions. And we're doing a lot of very exciting work at the moment looking at, say, healthy aging, because we can start to identify active protective biology. So you could identify the measures that you need to stop diseases happening, which is really exciting.

So, Véronique, PrecisionLife is focused on finding better treatments through the AI-driven innovation engine that you have. Can you explain how your approach differs from, and perhaps even improves on, traditional methods in drug development?

What we're doing is quite unique in that we can address the major weaknesses the current AI models and the GWAS-based tools have, which especially prevalent in complex chronic diseases. So instead of machine learning, what we use is a unique combinatorial analytics approach. And what this does is allow us to analyze huge multidimensional patient datasets, but in a very computationally efficient way. So what we can then do is hypothesis-free identification of combinations of features. That means you don't introduce bias that is introduced by machine learning. And these combinations of features could be just combinations of SNPs that together are associated with a specific disease phenotype in subgroups of patients. And I mean, this could be disease diagnosis, severity, specific symptoms, Or even a quantitative trait such as level of drug response, which we've also done. And the key difference with our platform is you can look at the nonlinear effects of interactions for anybody who's into genetics. So you can look at the interactions between multiple genes and then other epidemiological environmental factors as they interact across very complex genetic and metabolic networks. So, what does this mean? We can find lots more drivers of the disease biology, better stratification of patients according to the mechanisms underlying their disease. And that's a key advantage. Machine learning, PRS risk models built on GWAS, as I was hinting at before, they cannot reconstruct these very, very critical nonlinear interactions. So, they don't have predictive power in multifactorial diseases. Well, especially across populations with different origins. And they work from population averages. So, they're not useful for stratification of individual patients. Another advantage, I mean, if you can imagine the combinatorial complexity in these datasets, I mean, we routinely deal with disease studies that require testing of 10 to the 40th potential combinations. But we can reduce this in a manner that keeps all the biological relevant signal, and runs very complex disease studies in just a few days. So that even with all the extensive statistical validation, interpretation, evaluation of results, et cetera, we are able to deliver groundbreaking discoveries at exceptional speed. But it, it isn't just the speed. In all our case studies, what we've managed to reveal are Many more replicated genetic associations than GWAS, and therefore much deeper disease insights. And as I hinted at, both causal and protective effects. But we can do this from much smaller patient datasets than all the alternative AI drug discovery approaches. So that's what makes it really exciting.

So you're able to do more with less and perhaps impact lives in an even greater way, which is fantastic and gives people a lot of hope who may not have had as much for their disease state otherwise. And so I think what you guys are doing at Precision Life— Yes, absolutely. Is incredible. And I want to turn away from the company for a brief second now and talk a little bit more about you and your role as a chief medical officer. We hear about those roles all the time, but I think a lot of folks outside of the space or even adjacent to the space may not understand what a CMO really brings to a company. So how do you help translate that science, that research into a commercial reality that affects patient lives?

Well, we are a tech bio company, so the role is quite different from the traditional CMO role in a biotech. That's taking drugs into clinical development, because what we do is collaborate on R&D with biopharma partners. My role is to deliver our biopharma portfolio strategy and support the diagnostics and healthcare product strategy development. I'm a member of the senior executive team, and I also sit on the board. So, of course, I'm also involved in defining the long-term vision and the strategy of the company. As a tech buyer, we have very limited resources, and we do need to focus our efforts on commercial opportunities that will create revenue. So, we do a lot of analysis to identify the key unmet needs for precision medicine in these diseases, and then develop business cases to enter new TAs or markets. Then what we do is design and interpret what we call disease studies. That's how we run the analysis of the data so that we are able to deliver novel or repurposing targets, patient stratification biomarkers, clinical decision support tools. So we're constantly looking for data, and we work with the commercial team also on data access, but also the research collaboration agreements. To date, I think we've got over 30 data agreements and 11 commercial biopharma agreements. My team's carried out over 50 disease studies, and we store all the insights from these into a proprietary disease bank. Then we can look at the influence and penetrance of hundreds of mechanisms and novel targets in all the patient subgroups across multiple diseases. And that allows us also to find new indications for existing drugs. So, you can see how we then are able to translate all the inverted comma geeky stuff and data into actionable drugs and products that can make a difference in drug discovery and also patient care. And of course, I'm also responsible for publications, grant applications, developing relationships with key opinion leaders, because we need to access data and we need help interpreting our results, of course. Also got responsibility for collaborating with patient-facing associations. And of course, attending, presenting at meetings and conferences.

Yeah, so you're busy.

Yep.

I love that you kind of ended on collaboration. It's the perfect tee-up to where I wanted to take us next, because collaboration is so essential for advancing precision medicine, whether that be collaborating between different disciplines like AI and biotech, or collaborating to pull partners together in different ways. And I know you have to do that a lot in the role that you're in. So what advice would you give to people who are looking to collaborate in the space of AI and precision medicine? And what are some of the key factors you see for success of those collaborative partnerships?

These chronic disorders are just far too complex for any organization to solve on their own. And so you're going to have to have many different types of collaboration to be successful in this space. So you need to think quite broadly about the collaborations you need. So if we Start with data access. You need to spend a lot of time identifying the best possible multimodal data. What we found for us to be successful for genomics, ideally this should be whole genome sequencing data. Doesn't need to be 30x coverage. Low pass is great. Genotyping is also feasible. Ideally, you'd want longitudinal treatment response, comorbidity data from EHRs. And you want to access populations with diverse genetic ancestries. That's going to become more and more important.

And it's harder to find, right? Because those places are traditionally under-resourced. Often that data is not as accessible or not in the right format for you to use is what I'm hearing.

Exactly. You need to understand that replication is always something that pharma partners and academics want to see that you've done. So you need more than one dataset. For each disease, because you need to validate the results. The other thing to realise that well-characterised datasets, as you just quite rightly mentioned, they're not always publicly available. They're not always in dbGaP or UK Biobank. So you have to collaborate with international academics and clinicians. So you can access not just the data, but also their deep disease knowledge, and also be able to benefit from their influence in the medical and scientific community.

Yeah.

So while we know about the disease drivers we find, and we know quite a lot about them, we can't be experts in 50 diseases. That's just not possible. So if you're a company working in this space, you're going to really need to collaborate with PIs who are specialists in their field, who understand what you do, because not everybody will do, and who are willing to collaborate and can apply the insights we bring And ideally to co-author publications, because that's really important. One thing we found, and I'd really advise people to look at this when they structure data agreements, is building some quality control checks before you pay a lot of money, because not all data is what it appears to be when you start looking at it critically.

Yeah.

The other thing is charging for data. I mean, companies and academics should be really reasonable and realistic about the price. It's just simply not that valuable until it's been analyzed. and you generate new insights. So, you know, think about risk-sharing models with startups who really can't afford to pay what Big Pharma might pay. And even talking to Big Pharma, they're not willing to pay some of the astronomical prices either. Then the other thing is we are dealing with diseases where existing in vitro and animal models, they just don't work. So you need very early on to identify the best academics or startups who are developing novel assays and models for target validation, because you'll have to do some collaboration agreements with them. And that could be specialized CROs, but a lot of these models are actually sitting in top universities.

Like Tübingen University, of course, right?

Exactly.

Have to get my plug in.

Exactly. Well, that's why it's so great to have people like you introduce us to the right people, because we can access— well, first of all, find out, and also access key collaborators that we would not necessarily know about. Look for some of the AI companies that have got complementary capabilities. You're not going to do it all. So novel chemistry, AI-enabled antibody generative companies, we work with AppScience, and we do some quite creative deals as well. Look at the way that you could do co-development, because neither company will have a lot of money. I think it's really key to publish your results and try and publish early and present them at major scientific and investment conferences. It increases the trust that people have, especially if you've published in high-impact peer-leveled papers. What else? Basic, your website. It's gonna be good.

Have a decent one, right? Yeah, it's like, it's so funny.

Explains early on who you are. I don't want to have to click through 5 things. And more importantly, what problem are you solving and what's your competitive advantage? It's— there's so many people who claim to be AI these days. Make sure it's very clear very early on. And then do your homework before reaching out to partners. Look at their websites, go to conferences where the R&D and the BD folk are attending so you can grab them as they come off the platform before they disappear. Try and get their business cards, but mostly find out what is it that your potential partner is looking for and how can you help them meet their goals, and then you customize your outreach accordingly. It's very useful if you've got a board member or your own personal contacts, if you've been working in pharma for a while, to get warm introductions. Major piece of advice, do not go to the CEO directly. CEOs and R&D people hate that, and it'll get absolutely nowhere.

Yeah.

They're too busy, and they're not the right people. It's probably going to be in our R&D, either the head of R&D or the head of that particular research area. And when you get to that first meeting, just listen to what they're trying to solve and see how you can help. 2 more things I'd say. PPI, that's patient and public involvement. That's massively helpful. And it's now an essential requirement for research funding. Really collaborate with patient organizations, especially the ones that have funded research. They are Key to get on side, not just for their data, but they'll help you disseminate your results. And they also are fantastic at supporting grant applications. They add a lot of kudos.

And for people who don't know how to do that, places like the Milken Institute can be incredibly helpful with them getting access to those because they maintain all those relationships and are happy to help pull those patient populations and those patient advocacy groups into biotech research. So I always love saying that and giving them a plug because they do such great work.

Yeah, very good point. They're a great organization for that. And then finally, in any collaboration, vital to find like-minded people who can work with you and who understand your approach. You need to find the right person basically in the organization who's gonna become your product champion and help get the deal done. And the deal to me, it isn't over until the fat lady sings, as they say in opera speak. Make sure also once you have got a deal signed that you really set up the alliance management side of things so that you deliver on your deal and you deliver these key milestones in the right time. So that's also very, very important.

I love that. And I feel like you've given us a masterclass in a lot of the pieces that allow a tech buyout company or quite honestly any company to become successful. But I'm curious, How have you pulled all this together at Precision Life and had any insights that perhaps have led to a new treatment or an improved patient outcome? Do you have any examples that you can give us and give us kind of another plug for what Precision Life has been able to accomplish?

Yeah, I can give you a great example in ME/CFS. ME is myalgic encephalomyelitis. It's also known as CFS or chronic fatigue syndrome. It's a very serious long-term illness which affects multiple body systems, as I'm sure you know, including the nervous and immune systems. And it's incredibly complex. It's got many different causes and risk factors. Now, what's been really exciting in over 30 years of global research, until we came along, not a single replicable genetic finding from any mega GWAS study had been reported. We carried out a study, in a UK Biobank cohort. And we identified 14 novel genetic associations with this disease. And we've used these insights into some of the genetic drivers to stratify patients. And what's really exciting, that we've also been able to link the underlying biological mechanism driving the disease in specific subgroups to the way they present in the clinic. And what we call this is mechanistic stratification. This mechanistic stratification has a predictive value that's equivalent to monogenic risk factors such as BRCA1 in breast cancer. So, it's pretty reliable. What we're doing now is we've reduced this to practice as a cheap and fast saliva genotyping test. We call it a mechanostic test. And this can capture both the disease risk and the patient's underlying mechanistic cause of disease. And therefore, we can help identify likely effective therapies. And we have a partnership with the Metrodora Institute in Salt Lake City to do clinical validation of our mechanostics in a 1,000-patient clinical trial. Now, within just 6 months, we've started to return results to patients in the clinic. And patients' care is being changed as a result of the disease insights we're providing. But even more importantly, the patients are finally getting recognition and a much better understanding of their symptoms after years of being stigmatized, including being told, oh, what you've got is psychosomatic. Do you know, you are lazy, you are trying to avoid work. You can see that what we are doing has got the ME/CFS community very excited, and they're very, very supportive of the work that we are doing. We're also looking at identifying active protective biology, and then potential drug repurposing, as well as novel targets that we can put through a trial to very quickly see if we can improve lives of these patients. And next along the list with these mechanostics will be long COVID and endometriosis. So, we're actually impacting patients' quality of life and fast-tracking potential treatments. where there was no knowledge before. So that to me is really exciting.

And this is where I think the application of AI in healthcare, tech bio, is just so important and so transformative for the industry and why I'm grateful that we have companies like PrecisionLife out there. And so with that, as we talked about, a lot of what brings things forward are those collaborations, that network, and that ability to connect with folks. As your good friend and mine, Chris Yochum, said to me when he first found out that you were coming over for Bio on the Bayou, he's like, she is European biotech royalty. How did you get Dr. Boucher? And so, so now that everyone knows that you're European biotech royalty, I want to find out, you know, we're heading into 2025. Are there any upcoming events over the next year you could highlight where people could find you to connect, to learn more about Precision Life, to meet you, to collaborate, to see how you can get things done together and help move healthcare forward for patients? What will you be attending in 2025 where people might find you?

We'll be in San Francisco for the JPMorgan conference in January, and I'll be in the bullpen.

Oh, I cannot wait to see you there. That's where I'll be. Excited. The bullpen there, some of our favorite people. So good that we'll be there together.

We've got the AI in Drug Discovery conference in March 2025, if you happen to be in London. There's also the World Congress on Endometriosis in Sydney in May. I mean, we've got a raft of events coming up, and they're all Going to be listed on our webpage, precisionlife.com/events. And of course, you can follow Precision Life on LinkedIn.

That's fantastic. And it's a great place for people to check back often to find out where you'll be and where they can connect with you. And I want to take a moment and just thank you for taking so much time out and talking with me. I'm going to be sure to keep an eye on you and on Precision Life in the future. I love seeing what you're doing. Cannot wait to see how many more lives you change. Is there anything else you want to say to leave us with today?

Thank you very much for having me. It's been a pleasure, and I look forward to meeting up with you very soon.

Definitely. And as always, we're going to have Dr. Véronique Boucher's contact info and informational links for Precision Life and lots of other things we talked about posted in our show notes. I want to encourage everyone to check them out and learn more about you and Precision Life. And thank you once again, and have a great day.

Bye.

Thanks for joining us for Bio from the Bayou, and we hope you'll join us again.

To never miss the latest updates in biotech, hit the subscribe button now.

We'll catch you on our next episode of Bio from the Bayou.