The future of healthcare must integrate biomedicine and technology proactively and preventatively.
In this episode of Bite the Orange, Dr. Calum MacRae, Vice Chair for Scientific Innovation at the Department of Medicine at Brigham and Women’s Hospital and Co-Founder at Atman Health, talks about bringing genomics and genetics into clinical practice and drug discovery with the help of technology. Dr. MacRae explains why zebrafish is very valuable as a model organism due to its genetic similarities to humans and capacity for large-scale disease modeling. Many have been resistant to the concept of AI in drug discovery, but Dr. MacRae believes it has become more transparent over time and will play a transformative role in this type of medical setting. Furthermore, he highlights the need for broad representation in clinical trials to ensure equitable access to care and drive innovation in healthcare.
Tune in to learn more about the potential of genomics and genetic technology in healthcare!
FULL EPISODE
BTO_Dr. Calum MacRae: Audio automatically transcribed by Sonix
BTO_Dr. Calum MacRae: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.
Emmanuel Fombu:
Welcome to Bite the Orange. Through our conversations, we create a roadmap for the future of health with the most impactful leaders in the space. This is your host, Dr. Manny Fombu. Let's make the future of healthcare a reality together.
Emmanuel Fombu:
Good morning, good afternoon, good evening, ladies and gentlemen. Welcome to another episode of Bite the Orange. And this episode is a particularly special one because there's someone that is our special guest today, someone that I looked up to you for a very long time, over several years, and he's quite a humble man. And he actually responded to me several times, and we were scheduled multiple times, and he still made time for me, and I'm honored to have him on the show. If you don't know who he is, then you better know who he is. And today, I'm honored to have on the show Dr. Calum MacRae. Welcome to the show.
Dr. Calum MacRae:
Manny, great to see you again. How have you been?
Emmanuel Fombu:
I've been great. I've been great. So tell us about yourself. For those that don't know you. What's your story?
Dr. Calum MacRae:
Yeah, so I am basically a physician-scientist who works at the interface between traditional medicine and biomedicine and tech. Working in the Harvard system, I have done a lot of early discovery work on drug development, but in the last few years have been trying to move a lot of what I do to really improve the way in which we integrate biomedicine, technology, innovation at the bedside and beyond, and have been working with large tech companies, have been working with large pharma to try and make things happen around individuals in health, wellness, and in disease.
Emmanuel Fombu:
That is quite interesting. I remember several years ago, I was in Birmingham in the UK, and I was interviewing a cardiologist out there, and he happened to be a friend of yours, which was quite random, and he knew you, and you guys grew up together with school together, which is quite interesting. So your background, you're a cardiologist?
Dr. Calum MacRae:
I'm a cardiologist, I originally trained in cardiology in Edinburgh and London in the UK. Then I did a PhD in genetics, and then I've been working doing both cardiology and genetics for the last 20 years in the Harvard system, working both with human genetics but also with modeling disease in a variety of different settings, including high throughput screening and zebrafish.
Emmanuel Fombu:
Which is quite interesting. Talking about genetics and genomics, since the Human Genome Project happened, I think the way society has actually changed quite a bit. So tell us what you envision the future of healthcare to look like. You have this great, fantastic degree, PhDs in genetics, you are the guy that we should ask this question.
Dr. Calum MacRae:
Manny, thank you for saying that. I think there's lots of guys you can ask, but I'll tell you what I see, having worked at this space for quite a long time. So the technology piece, sequencing, and genomics, that's been essentially mature for almost 15 years, the technology is very well developed, everything about it is scalable, and yet we failed to get it systematically into clinical use. So the main reason for that is the same thing that I think has blocked implementation and innovation in medicine for a long time, and that's the existing workflow, that everything is very contingent on the way in which patients meet doctors and doctors meet patients. And so, in essence, for the last 200 years, physicians have sat in their offices and waited for people to come to them, or they've, in the setting of the emergency room, waited for something to happen acutely and then hospitalized the patients, and I think what we've started to see is that that leads to a very lopsided flow of information. And so, by combining the existing things that we know are important in biomedicine with technology, we can begin to distribute the information and the workflow that we need closer and closer to the patient, and importantly, do it upfront, do it before something happens, do it proactively, do it preventatively. And so what we've been working on over the last couple of years is thinking about how do we build technologies that allow that to happen. We were fortunate enough to work on a very large project in the American Heart Association, that was also funded by AstraZeneca and Quest Diagnostics, and that allowed us to build some of the platforms that you need in order to identify who will benefit from genetics, make sure that the information flows not just to their physician, but also to the patient and their family, and then to allow the family members to connect the dots and come back to the center with all of the information that you need in order to use the genetics without having to spend 10, 12, 15 years training everybody in the entire ecosystem to use the information. You can give people the data in the right format, you can pull data in the right format, if you just use modern technology, and you begin to give people the infrastructure support that they need to actually make things happen in the real world. And so that's what we spent basically the last 5 to 7 years doing, since you and I last spoke, is trying to build out tools and technologies that allow that to happen, first of all, in genetics and genomics, but then ultimately more broadly in clinical care. Again, also been fortunate enough to, in the interim, start a care delivery company with two of my colleagues, Rahul Patel and Rahul Dayal, called Atman Health that actually automates that last mile of care, not just the genomic and genetic testing, but completely separately from that is able to take existing guideline-directed therapies, and again put the right information in the hands of the right people so that everybody can really get access to the same level of care, and I think that's one of the things that we're beginning to see. You mentioned COVID. I think one of the things that really happened during COVID is people started to realize there are these huge variations in the ways in which people access care and the type of care that they get in the speed at which information and innovation flows through the system. And so part of what we've been doing is trying to build the sort of highways that allow that to happen in the way that it happens in every other industry in our lives. That's the most amazing thing is, there's no other area of our lives that is stuck in the 19th century except medicine.
Emmanuel Fombu:
Which is quite interesting. I know you mentioned the zebrafish earlier, I've been at the World Medical Innovation Conference that Harvard hosts every year, and zebrafish was something that was very big and something I know you do a lot of work on. So what is zebrafish?
Dr. Calum MacRae:
So zebrafish is, first of all, I suppose there are three reasons that we use it. One, it's small. Two, it's transparent. And the third, it has a diploid genome very similar to humans. So a lot of other transparent vertebrate animals have polyploid genomes or very complicated and distinctive ways of managing their reproduction. And so the nice thing about the fish is it's very similar to humans in terms of its body plan and its genetics that lets you model diseases at very large throughput. We're able to build, for example, a couple of hundred different disease models and then try and map the genetics of disease onto the therapeutics. And we've successfully completed, and in the middle of commercializing several, suppressors of disease. So lead compounds that we can then move closer and closer to the clinic identified in the fish. I think, as you probably remember, ten years ago, the fish was very much in vogue, then as people, like everything people, didn't quite know how to use it, didn't know where to put it in the entire drug development pipeline. But in the last couple of years, we've started to see a real resurgence, and now people recognize that it's a very useful model to test large numbers of hypotheses, large numbers of compounds early on before you focus down on later stages of drug development. And so we've seen a huge interest, particularly in the setting of AI-based drug discovery. So one of the things that's quite interesting with AI-based drug discovery is you get a lot of candidate compounds for candidate diseases, so you need a system that actually allows you to model those at the right scale. Otherwise, you hit a roadblock as soon as you start to move into in-vivo modeling. And so that's where we're seeing a major resurgence in the zebrafish as a tool in drug discovery and drug development.
Emmanuel Fombu:
Which is quite interesting because you mentioned AI in drug discovery, I have spoken to a lot of people in the R&D side of things that are resistant to the idea and concept. So where do you stand on that?
Dr. Calum MacRae:
So I think it's always wise to be cautious, but it's also important, I think, to understand what it is that you can accomplish with new tools and technologies. And to be honest, AI has been around probably for 25-plus years. We've certainly used variations of AI in some of our image analyses in the fish for over a decade. I think it's important with AI, it's not on its own a solution. It's a technology that allows you to optimize, and it allows you, if you close the loop, to constantly improve. And I think there are lots of great examples in the rest of our lives where AI has been really transformative. So, for example, even the last time you and I spoke seven years ago, it would have been difficult for weather forecasters to predict more than 3 to 4 days ahead accurately, and now there are 10, 15, 20-day forecasts that really are quite precise, and that's because of the ability over the last decade for AI-based models to continuously optimize predictions. And I think you're going to see some of that exact same approach in islands within medicine in the short term, but in the long run, as we begin to improve the data we collect, as we get more computable data in medicine, and as we begin to think about how we can make AI more transparent and safe, I think you're going to see it implemented much more widely and much more rigorously in different settings in medicine, including in drug discovery. At the end of the day, one of the problems with AI is making sure that you have the information content to solve the problem that you have, and AI is a very good tool for trying to identify just how much information is in a particular data set and can it actually inform a particular outcome. And then, if there isn't enough information content, you can begin to add information and optimize this for the solution. And I think that process is really what I mean by AI. I think there's a lot of hype around one-and-done type solutions with AI, and I don't see that even as a strategy that AI has been really designed for. It's much more of a continuous optimization process, and in order to do that, you need real-world data to get the algorithms that you're using to drive the outcomes. And that's how we've looked at AI, and that's how we've built our platforms to incorporate AI. And I think you're seeing that pretty widely in many other areas of human endeavor, and it'll come to medicine pretty quickly. Physicians tend to want things that serve the physician, and so a lot of the initial deployments of AI have been how do you reduce the workload on the radiologist or how do you improve the precision or accuracy of the radiologist. And those are definitely very useful tools, but at the end of the day, you're still left feeding information to a human who makes human decisions. And so one of the things we're going to have to overcome is how do we get humans out of the loop, but yet retain the safety that you need in order to work in medicine, there's no minimal viable product in medicine. You need a complete solution pretty much on day one.
Emmanuel Fombu:
With that being said, actually, right before you came on, I had another interview with a doctor out of Cameroon, where my family is from, and I was asking him if they had any kind of EHR system in Cameroon, he said no. And there's a big challenge in clinical research where minorities, especially African-Americans or Blacks in general, are underrepresented in clinical studies. My grandmother actually died from heart failure herself, which got me interested in studying cardiology myself. And so, with that being said, are you doing any kind of work with developing countries in general or?
Dr. Calum MacRae:
100%, one of the mission statements of Atman was to build a platform that would allow essentially everybody in the world to have access to the same level of care. And I think, as you pointed out, one of the biggest problems at the moment is that there are large pockets of the world where there's just no information whatsoever, there's virtually no care delivered. And so what we've begun to do is to think about ways that we can democratize access by using community-based individuals who can then look after their entire community with the information that we're able to provide through our platform. But importantly also, by doing that, you can even direct the information that you need to collect in order to address the problem. And so I see that sort of last mile out also being the first mile back in. And I believe that one of my hopes is that, particularly countries where there has been less development of a medical ecosystem and infrastructure will be able to bypass a lot of the traditional on-premises technology and go straight to the patient. Ultimately, that's one of the things that we really want to do. And I think, Manny, you, and I have spoken on and off about the difficulties with heart failure trials. How do we get access to the people without going through multiple steps, without increasing the cost of trials? And so one of the things that we've been working on is how do we start to build clinical research operations that go directly to the patient and their local provider, rather than having to go through a sort of network of recruitment centers? How do you find people where they are and help them engage with the research that they need, wherever they are in the world? And a lot of that involves, as I said, with the genomics, but it occurs in every other setting, giving the information, liberating the information so that the patient and their local provider have access to the tools that they need to operate at the very highest level. And I think if we can let information flow like that in a way that it has done, nobody's listening to music on different devices, nobody's using different communication tools, no matter where they are in the world. How do we start to make those technologies pervasive in medicine in ways that are both carefully supervised, but also that that accelerate the propagation of simple metrics that we know work? It's amazing that even in the US, in a fully insured population, only about 60% of people ever get a second prescription for a statin after their index event. You'd like to think that none of that type of variation would exist in the most highly developed countries in the world, but it's pervasive. And so we have to find solutions that work for everybody, and that's what we're trying to use technology to do in a very precise fashion.
Emmanuel Fombu:
Which is quite interesting. Last week I was actually in Saudi Arabia, and I had the opportunity to visit the Saudi universities down there, and they gave me a good tour, and they had this whole project called the Saudi Genome Project, and it was fascinating, fantastic labs, and the goal was to identify patients with rare disease. So there are about 50,000 samples that they have collected so far in that lab, and it's quite impressive. With that being said, if you look at a place like Africa where I had a conversation, and people say, people can't afford care, right? So everything is set up as a non-profit kind of thing, sustainable in the long run. I'm hoping that leads like yourself could join me and support this kind of infrastructure where we can actually scale this. Because what you describe, right there was a concept of decentralized clinical trials where it doesn't matter where the patient is globally.
Dr. Calum MacRae:
Precisely, I totally agree with you, Manny. I think that's what we need. It's actually one of the things that's really incredible. We deployed a blood pressure management program using the Advent Health platform in a community in Detroit where there was real stress on the system, there were housing insecure, and we were able to show that the existing guidelines just don't work. The trials that have been done in Caucasian individuals to regulate hypertension do not work in the vast majority of African Americans, but the system was able to optimize quickly to find those drugs that did work, and we were able to get people at goal within eight weeks once we had optimized the system. And I think that type of thing should never happen in the future if there's broad representation of patients globally in clinical trials, and I think I know all the large pharma companies that we work with, a lot of the biotechs that we're talking to, they all feel the same way. They want to really make sure that nobody is left behind when we start to innovate in healthcare going forward.
Emmanuel Fombu:
I agree with you on that point, right? I remember working in pharma, and I would see the maps of where studies were done, and there's a big sub-Saharan Africa, there's like zero sites, zero. I don't blame the study operators for not fighting anyone because there's no EHR kind of system or anything there, so I see a massive opportunity. And so, for anyone who's listening and you have an EHR system, there is a market there, they could expand it, actually collect data for that patient population. So if you were on the steering committee of a study, would you oppose the idea of actually having sites in Africa?
Dr. Calum MacRae:
Oh, not at all. I think, in fact, I think one of the most useful things is really to ask ourselves, why do we need sites other than patients themselves? How do we give people the tools to identify themselves as having particular disorders and then bring themselves into a clinical trial environment? Giving people access is really hugely important, and that's for both care, but also think importantly to innovation and to research. I don't think we should obviously force anybody to participate in research, but I think if we're not offering people that, we're not really looking after them, we're not really taking care of them.
Emmanuel Fombu:
And it got to the whole point of bias. If you don't have data understanding population under the AI algorithms ... completely skewed.
Dr. Calum MacRae:
100%, I think one of the most important things you learn in data science is a lot of the information content is at the extremes. And if we're not incented to make sure that the top socio-demographics are helping get data from the lower socio-demographics or just the different socio-demographics, then we're in trouble. We need as much information as we can to drive the outcomes that we all want.
Emmanuel Fombu:
From your background, you've done amazing things. You worked with Google and others like AstraZeneca, Novartis, you ... in your field, and I respect you a lot for that. You don't have to say it. I'm quite humble on your side, but that's not the point. I admire how aggressive you are in pursuing your vision and what you believe in. And I'd like for people to listen to this conversation today and realize that doctors are also innovative, but it's not just on one side, because today we hear about the word digital and health, no one talks about like genomics and everything else, but everyone is focused on CROs and EPROs and all these different pieces, but it's a bigger market of digital health, don't you agree?
Dr. Calum MacRae:
Absolutely, it's all connected. And I think that's, from my standpoint, one of the exciting things about digital health is it starts to bring together pieces of biomedicine that have been completely separate for many years. Think about it, essentially, pharma is a separate industry from healthcare. They're two completely separate industries with a chasm between them. And if we can tokenize information, make it secure, if we can begin to build the right audit components, can we bring everybody together around shared goals and shared themes? 100%. And I think digital medicine, digital health is going to going to help us do exactly that.
Emmanuel Fombu:
I know we could have this conversation for the next ten hours. I'll be in Boston in person, and I'll have a camera there, so we actually have this part two of this conversation. But before we wrap up, I want to get a commitment from you. I would love to invite you to Cameroon because I told the doctor, Ekiti Martin, before we left off the air that I will force him to bite the orange and then advocate for digital health in sub-Saharan Africa to go down and actually have conversations like this. Like, you have the expertise ... just to share the knowledge with that infrastructure. I think someone of your caliber, well respected, and I think we could start actually having them lay down the right infrastructure in place to do things that can actually bring us into those data sets and make sure that the future of humanity is great.
Dr. Calum MacRae:
100%, fully committed to doing that and excited to think about ways that we could work together to do exactly that.
Emmanuel Fombu:
Perfect. Everyone, you heard this. Dr. MacRae, we'll be in touch quite a bit and will bring a proposal to you. I'll get your insights on this. We're going to solve that problem with sub-Saharan Africa. Hopefully, we should connect with Google.
Dr. Calum MacRae:
Love to do it, great to see you.
Emmanuel Fombu:
Honored, thank you, you, too. Thank you very much.
Dr. Calum MacRae:
Thank you for the time, bye.
Emmanuel Fombu:
All right, all right.
Emmanuel Fombu:
Thank you for listening to Bite the Orange. If you want to change healthcare with us, please contact us at info@EmmanuelFombu.com or you can visit us at EmmanuelFombu.com or BiteTheOrange.com. If you liked this episode and want more information about us, you can also visit us at EmmanuelFombu.com.
Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.
Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.
Sonix has many features that you'd love including share transcripts, world-class support, automatic transcription software, powerful integrations and APIs, and easily transcribe your Zoom meetings. Try Sonix for free today.
About Calum MacRae:
Calum MacRae is Vice Chair for Scientific Innovation at the Department of Medicine at Brigham and Women’s Hospital and Associate Professor of Medicine at Harvard Medical School. He is the leader of One Brave Idea, a group of leading scientists from multiple disciplines working together to understand the earliest stages of coronary heart disease (CHD). His clinical interests include the management of inherited heart disease and cardiac involvement in systemic diseases.
MacRae is a cardiologist, geneticist, and developmental biologist who received his M.D. and Ph.D. from Edinburgh and London before coming to Boston in 1991. He completed postdoctoral fellowships in human genetics with Drs. Christine and Jon Seidman, and in developmental biology with Dr. Mark Fishman, and received additional clinical training in internal medicine and cardiology before joining the Division of Cardiology at Massachusetts General Hospital in 2001. MacRae is a leading investigator at the Brigham and Women’s Hospital Genomics Center, a principal faculty member at the Cardiovascular Research Center and the Harvard Stem Cell Institute, and an associated member at the Broad Institute.
Things You’ll Learn:
By utilizing technology and providing infrastructure support, physicians and patients can benefit from genetic and genomic integration in healthcare.
Zebrafish are small, transparent, and very oddly genetically similar to humans, making them useful model organisms for studying diseases.
AI can improve precision and accuracy in healthcare and reduce the workload of healthcare professionals, allowing them to focus on care delivery and research.
Atman Health is developing platforms that directly connect patients and local providers, increasing access to healthcare.
Digital health can bring different aspects of medicine together and unite stakeholders around shared goals, like giving people access to research opportunities in Africa.
Resources:
Connect with and follow Dr. Calum MacRae on LinkedIn.
Follow Atman Health on LinkedIn.
Explore the Atman Health website.
Discover more about One Brave Idea.