Interviewer: Michelle L. O'Donoghue, MD, MPH; Interviewee: Calum A. MacRae, MD, PhD


January 26, 2017

This feature requires the newest version of Flash. You can download it here.

Michelle L. O'Donoghue, MD, MPH: Hi. I'm Dr Michelle O'Donoghue, reporting for Medscape. I am here in New Orleans at the American Heart Association (AHA) Scientific Sessions. Joining me today is Dr Calum MacRae, my friend and colleague who is chief of cardiology at Brigham and Women's Hospital. He is also a professor of medicine at Harvard Medical School and the fortunate recipient of the One Brave Idea grant. For those of you who have not heard of this grant award, it is a $75 million grant from AstraZeneca, AHA, and Google, who have invested in the idea of trying to cure or at least greatly advance research in heart disease. Welcome, Calum.

Calum A. MacRae, MD, PhD: Thank you, Michelle. Nice to see you.

One Brave Idea Grant

Dr O'Donoghue: Good to see you, too. Is that a fair introduction for what the grant's idea is all about? Why don't you take us through it?

Dr MacRae: I think it's a reasonable place to start. The award came out of an idea that Andy Conrad at Google Life Sciences, now Verily, and Nancy Brown [AHA] had, which was to put a grant together to try to make a stepwise change in a particular subset of heart disease. Obviously, the area that they chose was coronary disease because it still remains the largest and most important cause of morbidity and mortality in the developed world. AstraZeneca joined in and added $25 million, so the pot went from $50 million to $75 million. The goal was to not only have a game-changing scientific advance but to change the way science is funded and how it's done—and to use this as a means of exploring those questions.

From my standpoint, the exciting thing is to be able to take scientific ideas and build partnerships in ways that are not possible with conventional award mechanisms. That makes it a very exciting opportunity.

What if we were to move back much earlier into the primordial phases of the disease, which we know exist?

Dr O'Donoghue: Many people were thrilled to hear about this opportunity when it was announced last year. What was your proposal? How do you think it differed from the rest of the group?

Dr MacRae: There was an amazing group of applicants. One of the things we focused on was the gap between the scale of what is possible in genetics and genomics and the relatively limited space that people have explored in terms of bedside phenotypes. If you are going to understand a very complicated disease process, you need to characterize it in many dimensions.

For many years, a difficulty with coronary disease in particular has been that lots of the fundamental metrics of coronary disease are invasive, or they require the disease to present. There isn't a very good way of quantitatively characterizing it in everybody in a noninvasive strategy. Perhaps, and most importantly, it is very difficult to make the negative diagnosis.

Our approach was that we have learned a huge amount in the last 2 decades about coronary disease—lesion formation, inflammation, plaque rupture, thrombosis, and myocardial infarction. What if we were to move back much earlier into the primordial phases of the disease? We know these phases exist from both autopsy data from the Vietnam War and the work done by investigators such as John Deanfield in London who looked at vasomotor reactivity in the offspring of individuals with premature coronary disease.[1] We know that there is a signature in teenagers, even in a very restricted space. We thought that if we can bring cell biological tools that have been identified in a whole host of animal models, in vitro models, and cell science, and we can begin to build technologies that can measure those in the clinic and measure them very, very early, that is a way of us characterizing fundamental biological differences between those who then go on to have heart disease and those who do not.

That is the strategy, in a nutshell. There are obviously many different ways that you could do that and many different technologies that you might use to bridge divide. A nice thing about having the resources and the flexibility that this remarkable opportunity brings is to be able to test a few things, see what works and what doesn't work, and not be afraid to fail but consistently try to bring real biological insights into the clinic at an early enough stage to identify disease.

Full Spectrum of Disease and Approaches

Dr O'Donoghue: Is the focus primarily on primary prevention? Are you trying to find those people who are going on to develop heart disease, or will it be across the spectrum?

Dr MacRae: It's going to be across the full spectrum. Ultimately, the core idea is that if you understand the different forms of disease by characterizing them well enough, that makes the genetics more accessible, and it makes all of the downstream mechanistic investigations much more accessible. Think of how much we have learned from looking at the biology of only 1 or 2 subforms of coronary disease. Almost everything we know is based on models that are driven by the low-density lipoprotein (LDL) receptor loss-of-function mutations. By looking at a different phenotypic space, this allows us to begin to think about other pathways that might be important in coronary disease and to use the paradigms that have been so successful in LDL-associated biology to begin to drive other biologies home—to identify patients earlier, which will help with prevention; to make negative diagnoses, which will help with understanding the genetics and the biology; and to identify mechanisms that will lead to new drugs and new cures.

Dr O'Donoghue: You have many different people from different universities and different research groups who are all participating in this endeavor. Is that right?

Dr MacRae: That is absolutely right. That was another exciting thing: the chance to build partnerships in different ways. We have brought together a diverse group of folks from Massachusetts Institute of Technology who are experienced in fabrication, device and biomedical engineering, and moving concepts into prototyping and beyond with clinical enterprises at both Stanford and in Boston. We are thinking about population studies through the Framingham Heart Study, the Million Veteran Program, and other programs for questions that we can only ask using a particular group of investigators.

We have some very creative and dynamic investigators. We have a fundamental genomics biologist from Toronto, Fritz Roth, and we even have a venture capitalist. One of the very appealing things about this was the ability to draw in individuals who were not in the original consortium. We have already identified Mark Huffman at Northwestern University for his incredible work using social networks in ischemic heart disease. We are going to collaborate with Mark as we begin to think about how we can do the things that I said we were going to do.

Dr O'Donoghue: There is so much additive value when you get multiple people collaborating. It's such a unique opportunity to form these bridges, put our minds together, and think about the best way to tackle heart disease. If it becomes too isolated, you just end up looking at the problem in a single way.

We have the ability to crowd source. If we reach a roadblock, we have the ability to say, "Who can help us overcome this roadblock?

Dr MacRae: I could not agree more. In fact, one of the most appealing things for me is that we have the ability to crowd source. If we reach a roadblock, we have the ability to say, "Who can help us overcome this roadblock?" If it's an industrial partner, a patient group, or another research group, we can go to them and say, "This is what we can bring to the table to help us get beyond that roadblock." I'm excited that it changes the dynamic of how you think about doing research. I hope that others will agree.

Dr O'Donoghue: Because so many people are involved, it helps to address one of the criticisms raised when the idea of the grant was first pitched, which was that it could be a teenager in the Midwest who wins this award. When somebody at Brigham and Women's Hospital and Harvard Medical School got the grant, some people said, "Oh, there is nothing unpredictable or out of the ordinary about that." But your point is that it's not just about Brigham and Women's Hospital. There are a lot of minds at work and a lot of novel concepts that are being introduced.

Dr MacRae: I will point out that as far as coronary heart disease is concerned, I am almost a teenager from Wisconsin. You're absolutely right that one of the very appealing things about this is that we're also changing the paradigm of how we work together. As I said in the opening sessions [at AHA], many of the people that we will work with were in the room, and we want to collaborate with people who were perhaps critical upfront. They have experience and skills that we are going to need in order to solve some of these major problems.

Dr O'Donoghue: Finally, a lot of people are talking about precision medicine, big data, and leveraging big data. What does that mean in tangible terms? I feel like it's one of those buzz terms that people are vague about.

Dr MacRae: People do tend to use those terms in a generic fashion. The key thing about data is understanding what the data mean. I was originally trained in biophysics and signal noise. I am very conscious of the fact that part of the problem with big data is the mismatch that I spoke about at the start. We have these amazingly comprehensive data sets in genomics on how cells are connected to each other, the ways in which the world around us is beginning to feed into an Internet of things. Yet, in the human biological space, we are still restricted to measuring more precisely things that, in many incidences, were first discovered in the 17th and 18th centuries. A part of solving the big data problems—for medicine, at least—is to bring in new data types that collect information in a directed but more agnostic fashion and try to understand how different data types contribute to the information content you need to solve the problem.

That is perhaps not so apparent when you use the term in its broadest sense. You need to really understand. You need to featurize the data. What are the data types that are really contributing the most information content to the problem that you want to solve? Having all of the data is fantastic and allows you to solve lots of problems. You still need to think systematically about how to get the most information out of the data that you have for the problem that you are trying to address.

Dr O'Donoghue: Calum, when we are at the AHA meetings next year, will that be too soon to start seeing some fruitful results from this grant?

Dr MacRae: We don't know what is going to happen or when. I suppose that is the real excitement of all research. I do hope that we are going to move quickly. We are going to communicate directly to the entire research community exactly how we are progressing. People will see that we are changing the way that we address problems, and hopefully we will see some early results within the first few months. I look forward to communicating those as they arise.

Dr O'Donoghue: Terrific. Congratulations, Calum. This is a huge undertaking and a huge honor.

Dr MacRae: Thank you very much, Michelle. I am honored and humbled to be doing it.

Dr O'Donoghue: Joining me today was Calum MacRae at the AHA Scientific Sessions. I am Dr Michelle O'Donoghue reporting for Medscape.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.