Kathiresan and Topol on Genomics of Heart Disease

; Sekar Kathiresan, MD


August 02, 2017

Focusing on Heart Attacks Among the Young

Eric J. Topol, MD: Hello. I'm Eric Topol, editor-in-chief of Medscape. I'm privileged today to speak with Sekar Kathiresan from the Broad Institute, who heads up the Center for Genomic Medicine at Massachusetts General Hospital, which is not even a year old, and who also is on the faculty at Harvard Medical School. Sek, you've done some remarkable things to advance our knowledge in cardiovascular genomics. In fact, you're my go-to guy.

I'd like to start with your background and how you got into this area. You grew up in Pittsburgh, went to Penn for undergrad, and then to Harvard?

Sekar Kathiresan, MD: I graduated from Harvard Medical School in '92 and have stayed there since. I did internal medicine (clinical cardiology) training, and I was a chief resident in medicine at Mass General. I started my research training in 2003 after all those years of medical school and clinical training. It was originally supposed to be just a 2-year stint in genetic epidemiology, but I ended up liking it so much that I spent 5 years as a postdoctoral fellow—2 years at the Framingham Heart Study and 3 years at the Broad Institute, learning human genetics. I got all of the foundation for genetics research during that experience.

I started my own lab in 2008. The whole time, we've been focused on trying to understand why some people have heart attacks at a young age, specifically looking at the genetic basis for premature myocardial infarction (MI).

Leveraging the World's Genes

Dr Topol: In addition, you've established worldwide collaborations of people doing similar things. How did you start that?

Dr Kathiresan: That's an interesting story. I started in this work in 1997 as an intern at Mass General, recruiting patients who'd had an MI prior to the age of 50 for men and 60 for women. A faculty member there, Chris O'Donnell, started that project and got me involved. Over the subsequent 6 or 7 years of my clinical training, we recruited about 500 such patients at Mass General. I realized quickly that it wasn't going to be a sufficient sample size to make the kind of observations needed to understand the biology of the disease. It's a complex disease; a few patients were not going to help solve the problem.

In the mid-2000s I worked with David Altshuler. He was my mentor, and he encouraged me to reach out to people around the world who had similar collections of patients. As a postdoctoral fellow, I emailed investigators in Malmö, Sweden, who had a similar collection. They had published their findings. I said, "Do you want to work with us?" They invited me to Malmö, and I went. We ended up partnering with six or seven other investigators to start what we called the Myocardial Infarction Genetics Consortium. That's been the foundation for all of our work on heart attack genetics.

Around the same time, I started a similar consortium for looking at cholesterol level genetics. That has now expanded to more than 50 centers around the world.

Getting at the Truth About Lipids

Dr Topol: There is a real misconception that heart attacks and coronary disease are tightly interwoven with lipids and cholesterol, but plenty of people who have virtually normal or even better-than-average lipid profiles wind up having heart attacks. Where do you see this field going in terms of better understanding the non-LDL cholesterol—or other lipid—foundation for MIs?

Dr Kathiresan: I'll share with you what we have learned about heart attack genetics over the past 10 years. Doing something unbiased, in the sense of looking across the genome and asking, "Where in the genome is there risk for heart attack in terms of cases versus controls?", we have learned that several previously known pathways show up. For example, one of the top results in any genetic analysis for heart attack is LDL cholesterol and several genes related to LDL cholesterol. In addition, we've been able to clarify some controversies in the lipids area.

It was unclear when I got into the field which of the two—HDL, the so-called "good cholesterol," or triglycerides—was more important. When I was in medical school I was taught that anything that raised the good cholesterol must be good for you. Our genetics have shown that is not the case. Basically, HDL cholesterol is a very good marker of risk but it's unlikely to be a causal factor. We published a genetics study[1] a couple of years ago that challenged the conventional wisdom and suggested that drugs that raise HDL are not going to work. We actually had a hard time publishing that study; it took a couple of years, but since then, there have been five randomized control trials of medicines that have tried to raise HDL cholesterol.

Dr Topol: It's been a big bust.

Dr Kathiresan: It turns out that we probably were on the wrong side of the seesaw. When HDL is down, triglycerides are up. People thought that HDL was what was important. The genetics now strongly point to triglycerides-rich lipoproteins.

We have LDL and we have triglyceride-rich lipoproteins. The other key factor in the lipids space is something called lipoprotein(a). The genetics are compelling that these three things are very important for heart attack. The surprising thing has been that of the 55 gene regions we've identified for heart attack, only about 40% point to things that we already knew about. Another 60% don't relate to any of the known risk factors, like blood pressure or cholesterol, suggesting that there are new mechanisms for atherosclerosis. As a community, we need to figure those out.

The Statin Hypothesis

Dr Topol: For example, the common variant of 9p21, a 60 kb noncoding region, has nothing known to do with cholesterol, and we are still working on what it really means, right?

Dr Kathiresan: Yes. At Scripps, you played a big role in trying to sort that out. It's been 10 years and it's been very challenging. None of this is going to be easy. Cholesterol was hypothesized to play a role in heart attack more than 100 years ago, and some people are still debating the role of LDL cholesterol. This isn't going to be straightforward, but it does suggest that there are lots of other mechanisms.

Dr Topol: That's obviously very important because Brown and Goldstein, the famous Nobel Laureates who were instrumental in the development of statins at the turn of the century, published an editorial in Science, "Heart Attacks: Gone With the Century?"[2] That was the notion that statins would be widely used and that we would stamp out heart attacks. That hasn't exactly happened, although there has been a reduction in large ST-elevation infarcts.

Dr Kathiresan: There are a couple of issues. Their hypothesis is sound; it says that if you start treatment early enough, and if the LDL is low over an extended period of time (30-40 years), you won't develop atherosclerosis. They based that hypothesis on model organism work but also on human genetics. People who carry mutations that naturally lower their LDL to very low levels lifelong rarely develop atherosclerosis. Societies like rural China, where LDL is very low, have very little atherosclerosis. It is a very good hypothesis and we still have to test it. We don't know.

Dr Topol: If you could do it at birth...

Dr Kathiresan: If we could do it safely...

Dr Topol: And safely—right.

Dr Kathiresan: Even if you do that, there are still several other elements or pathways. We are seeing now, in the United States at least, a transition from risk that was driven over the past century by blood pressure, smoking, and LDL, to this century, when risk is basically being driven by abdominal adiposity, insulin, and triglycerides—the cardiometabolic axis. That's what we're seeing with the obesity epidemic. LDL levels are coming down and heart attack rates have come down as a result, but we have the countervailing force of cardiometabolic disease. That's where triglyceride-rich lipoproteins come in—insulin and so forth. This is on an incredible rise in the United States and also worldwide.

Why DNA Isn't Destiny

Dr Topol: One of the most seminal studies in the three decades during which I studied cardiology and coronary heart disease was one that you and your colleagues published last November in the New England Journal of Medicine.[3] In that study, you had the genetic risk scores, so you knew the various polygenic markers and could separate people into low, moderate or intermediate, and high risk, and you showed the titration of high risk—which has never been done before, genomically—with better lifestyle.

A Cell editorial[4] published very soon after your paper said that diet and exercise will save us all.

I want to get your thoughts about this. These days, if people knew that they were at high risk without any connection to family history, blood pressure, or LDL, they could benefit from this knowledge and this could be a way to promote, for them in particular, a healthy lifestyle.

Dr Kathiresan: Thank you for your kind words about the paper. The work started with a very simple observation. In my preventive cardiology clinic at Mass General, we have patients who come in and say, "My father died of a heart attack at age 50. I am doomed." They feel that DNA is destiny for this disease. We wanted to address that if you are at high genetic risk, can you overcome or counterbalance that risk with a favorable, healthy lifestyle? We've known for many years that a favorable lifestyle is associated with a reduced risk for coronary heart disease. In the context of genetic risk, how do they interact?

We found that if you are at high genetic risk, based on 50 different DNA markers, you could cut that risk in half by having a favorable lifestyle that included not smoking, regular fruit and vegetable intake, maintaining an ideal weight, and so forth. It was a very sobering message in some sense and a good public health message—that if you are at high genetic risk based on, let's say, family history, you should not take this DNA-as-destiny approach. Rather, you do have control over your health, specifically by trying to practice these healthful behaviors.

Mimicking Mother Nature

Dr Topol: It transcends the Framingham Risk Score era because now you have a way to gauge risk and it can be titrated, so it was a big step forward. I also want to get into the idea that you can protect your heart disease risk naturally—that is Mother Nature. Previously you've talked about APOC3 and a startling finding about these homozygotes that you identified in Pakistan. Would you tell us that story?

Dr Kathiresan: You wrote many years ago about protective mutations. When we think about genetics, we think automatically about risk, but actually there is a big value of genetics in finding people who are naturally protected because of a mutation, and the main value is that you could hopefully develop a medicine that might mimic that mutation. If you can do that, then you can transfer the benefit that nature gave just to that one rare person to the entire population. That's the concept.

There's a very good example in the cardiovascular space with the gene PCSK9, where this held true. We set out a couple of years ago to ask whether there are other examples. The first that we found was the gene apolipoprotein C3. This is a gene that has been known about for 30 or more years. It's a gene that puts a brake on your body's ability to handle dietary fat. When we eat a McDonald's burger, right after the meal, the triglyceride level goes up two- to threefold. The body has to clear that fat and the APOC3 protein actually dampens your ability, or puts a brake on your body's ability, to clear it.

We found that about 1 in 150 people in the United States have a favorable mutation that gets rid of one of the two gene copies of APOC3. These individuals have lost a "bad guy" in their blood, and therefore they have lower lifelong triglyceride levels and about a 40% lower risk for heart attacks. That immediately suggested that if you could develop a medicine that got rid of APOC3, you might be able to reduce risk for heart attack.

One of the other key features of this paradigm is finding individuals who lack both copies of that gene. Sometimes you would call them "human knock-outs." Why do you want to know that? If there's a person walking around who naturally lacks both copies of that gene, and they are healthy, then that immediately says that you could pretty safely treat somebody with an inhibitor of that protein and not have a lot of adverse effects. It's not a complete predictor, but it's pretty close.

We set out to find these individuals. We looked at more than 100,000 people in the United Sates of European ancestry and did not find a single person who lacked both copies of APOC3. It turns out that there are people in whom both copies are gone, but that property tends to happen more when the parents of a child are closely related to each other—for example, first-cousin marriage. In some parts of the world, it is actually fairly common. It's not taboo as it is in the United States. Pakistan is a country with the highest proportion of marriages that involve parents who are closely related. We went to an investigator in Pakistan, a collaborator who had recruited a large study of heart attacks there, and we did sequencing of APOC3 in more than 20,000 people. We found four individuals who completely lacked the gene.

Dr Topol: It was striking that these people, first with low triglycerides, also had no triglyceride elevation when they ate a fatty meal.

Dr Kathiresan: It's fascinating. This was a small fishing village. My collaborator, Danish Saleheen, had a mobile truck to do studies. They went out to the fishing village and recruited family members in whom gene copies were present and those with both copies gone. They gave both groups of individuals a fat challenge and then took blood samples every hour for 6 hours. In all of the people who had APOC3, the triglyceride levels went up (like they would in you and me), but in the people who didn't have the gene, the triglyceride levels did not budge at all after the fatty meal. This gives us some insight as to why people are protected from heart attack.

How You Lower Triglycerides Matters

Dr Topol: It's interesting, because it flies in the face of so many studies where they lowered triglyceride levels and findings were very disappointing—there was little clinical effect. But this is a different target, of course.

Dr Kathiresan: That's the issue. There were lots of studies over the years (particularly with fibrates and fish oils, for example). In randomized controlled trials, those two medicines lowered triglycerides but they were unable to show that they lowered risk for heart attack. The challenge is that we don't really know what the molecular targets are for those two drugs, and triglyceride metabolism is complex. You can imagine ways—and there are actually ways—that you can lower the triglyceride level, but counteract that with other bad things where the net effect might be no effect on disease risk. The way you lower the triglycerides will matter—maybe a little less so than for LDL. It looks like almost any way you lower LDL (although there are some exceptions there too) makes a difference in terms of heart disease risk. For triglycerides, it matters how you lower them.

We are seeing that there are several genes (APOC3 and a couple of others) in the pathway where there is naturally occurring genetic variation, pointing to these genes as being the way to lower triglycerides if you want to lower risk for heart attack.

Dr Topol: That's phenomenal. What we are seeing here is starting to really crack the big three: Lp(a), APOC3 (and other triglycerides), and LDL. We're going to see the lipid story become amplified. There is still going to be this other...

Dr Kathiresan: Residual risk.

Dr Topol: That's going to be an interesting enigma.

Genetics and Informed Cardiac Care

Dr Topol: Where are you going next? How are you going to keep building this? This foundation of knowledge has been extraordinary. You have been working on it for a decade. What can you do to expand this now?

Dr Kathiresan: The lab has worked on three elements during the past 10 years: discovery of new genes, understanding how they work, and then translating those findings to improve cardiac care. I see genomics and informed cardiac care going in two ways. One is identifying a subset of individuals who are at much higher risk, based on the genome. We are pretty good at that right now and I think there will be broad uptake over the next 10 years.

We'll then be able to find a subset of individuals early in life, based on their DNA sequence, who are at three-, four-, or 10-fold higher risk for heart attack. Then the question becomes, what do you do for those patients? We've already shown the value of lifestyle and probably a statin, but then the key question is, what else is there? Can we develop a medicine in the nonlipid space that can have dramatic benefit? That's what I see in the next 10 years.

On Twitter First, Then The New York Times

Dr Topol: That would be exciting. We will ultimately get there as we learn more.

Now, you are big on Twitter.

Dr Kathiresan: No bigger than you.

Dr Topol: I enjoy following you. You are great to follow because you are one of my favorite educators. We can learn a lot from Twitter. What do you like about it? Sometimes, of course, you are tweeting about the Steelers, but when you are not tweeting about the Steelers or politics, what do you enjoy about Twitter?

Dr Kathiresan: I love what you just said. Every day I learn something new on Twitter. It's a little bit of a double-edged sword. We all know about social media; it's quite addictive. I could sometimes spend an inordinate amount of time on it. That aside, I learn a lot and it's mostly about science. It's things that I would not have seen. On your feed, you transfer an incredible amount of information daily, and there are lots of other opinions. Often now it is the place for immediate news, whether it's science news or other news.

A good example: A couple of weeks ago, the topline results from the randomized controlled trial of the PCSK9 antibody were announced. I knew they were going to be announced because it was a 4 PM release by Amgen at the close of the market, so I'm waiting.

Dr Topol: The first look is going to be on Twitter.

Dr Kathiresan: Exactly. A day later it will show up in The New York Times.

Dr Topol: The pulse of our field, as you say; the amount of information that you can get through Twitter in science and biomedicine—our world—is quite extraordinary, and it's just as surprising that a lot more physicians and researchers don't use it.

Dr Kathiresan: Two of the healthiest areas are genomics and cardiovascular medicine. There's a tremendous amount of cardiology on Twitter, and of course, genomics is way ahead of a lot of other fields.

Dr Topol: It seems that way. It's some of my favorite stuff.

This has been really fun. I just cannot say enough about how much you have accomplished in such a short time to advance the field. [Heart disease is] still right there as the number-one cause of death and disability, and we still have a long way to go, although cancer is catching up and may soon overtake it in the United States.

Thanks so much for joining us. And thanks to all of you for joining us for this conversation. It got a little deep into the pathophysiology and genomics of coronary disease, but it's certainly an area that we are going to continue to build on.

Follow Dr Kathiresan on Twitter @skathire and Dr Topol @EricTopol


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.