Interview: What actually causes high cholesterol? | Hei Man Chan

0:01This is TED Health, a podcast from TED, and I'm your host, Dr. Shoshana Ungerleider. Heart disease might sound like something to worry about later in life, but here's a surprising fact. The buildup of bad cholesterol in arteries can begin as early as your teens and 20s. In fact, studies show signs of early heart disease in young adults with poor diet and low physical activity, even if they feel perfectly healthy. That's why understanding how cholesterol actually works in the body and what you can do about it now matters more than you might think. This TEDx lesson cuts through decades of confusion around eggs, fat, and cholesterol to explain what really impacts your heart.

0:40So if you're looking for clear, actionable ways to protect your cardiovascular health for the long haul, start here. And stick around after the lesson for an interview with Dr. Eric Topol, one of the most influential voices in medicine today. A world-renowned cardiologist, scientist, and best-selling author, Dr. Topol has spent his career pushing the boundaries of how we understand the human heart, our DNA, and the powerful potential of artificial intelligence in healthcare. From leading groundbreaking clinical trials to reimagining the future of personalized medicine, his insights may change the way you think about your own health and what's possible in medicine tomorrow.

1:19But before we dive in, a quick break to hear from our sponsors.

1:29And now, your TED-Ed lesson of the day.

1:53increase cardiovascular disease risk. But it turns out that the reality is more complicated. Let's delve into digestion to understand why. As your body processes a meal, it emulsifies fats into lipid droplets that can then get absorbed in your small intestine. These lipid droplets make their way to various tissues, including your liver. There, specialized cells unpack and reassemble them into lipid protein packages called lipoproteins that are optimized to flow through your bloodstream. Lipoproteins are composed of different ratios of protein and fat, so their density varies. VLDL, or Very Low Density Lipoprotein, is mostly packed with fatty molecules, including triglycerides and cholesterol.

2:52VLDL delivers lipids to your muscle and fat cells to be stored and used for energy. Afterwards, it's converted into LDL, or Low Density Lipoprotein. LDL is informally known as bad cholesterol because it's associated with cardiovascular diseases. This is because LDL is more prone to accumulating on blood vessel walls, which can cause inflammation. The inflammation attracts specialized white blood cells, called macrophages, to remove the LDL buildup. But absorbing too many lipids can compromise macrophages, causing more inflammation and triggering more macrophage recruitment. As the cells clump together, they form a fibrous, scab-like cap, effectively narrowing the artery.

3:49This can interfere with blood flow and weaken vessel walls, making clots and ruptures more likely. High-density lipoprotein, or HDL, on the other hand, packs more protein. It picks up excess LDL in the blood and brings it back to the liver, where it can be excreted. Because HDL is anti-inflammatory and keeps LDL in check, it's commonly called good cholesterol. Doctors order lipid panels to understand the concentrations of different kinds of lipoproteins and their building blocks in your blood. All lipoproteins are essential to your body's functioning, but it can become a problem if the different kinds get out of balance.

4:40Some people are simply predisposed to having higher cholesterol. For example, certain gene variants cause LDL receptors to behave abnormally, leaving more LDL circulating. Otherwise, different lifestyle factors can influence blood lipid levels, and food is one of them. So, what's the deal with eggs? Their yolks are indeed high in cholesterol, which prompted the American Heart Association's Three Egg Weekly Limit in 1968. But while some of the cholesterol in your blood comes directly from cholesterol found in food, your body actually makes a vast majority of it from the different fats you consume.

5:27Saturated and trans fats found in animal, palm, and coconut products can drive your liver to create more VLDL and LDL packages. Unsaturated fats, meanwhile, are abundant in foods like fish, nuts, and seeds and have anti-inflammatory properties that help counteract LDL buildup. A food's fat composition has a greater effect on blood lipid concentrations than its cholesterol levels. So, the high cholesterol found in egg yolks isn't actually of too much concern. Egg yolks contain some saturated fat, but not too much, and they also contain unsaturated fats. Generally, an egg a day is considered nutritious.

6:14But it's not just what fats you eat, but what you eat them with. Fibrous fruits, vegetables, and whole grains help lower LDL levels by affecting how fats are absorbed in the digestive tract. Overall, getting plenty of unsaturated fats and fiber seems to help keep LDL in check. And other activities can also impact lipid levels. It's not yet totally clear why, but resistance training and aerobic exercise can lower LDL and increase HDL. This might be because they use more blood lipids for energy and boost the activity of HDL-producing enzymes. Meanwhile, smoking increases LDL buildup, perhaps by unleashing unstable free radical that damage lipids and vessel walls.

7:13But when lifestyle changes don't help to manage cholesterol levels, as can be the case when genetics are at play, we luckily have some pretty effective therapies. Statins are drugs that block a liver enzyme required to produce LDL. Other cholesterol medications increase the production of enzymes that break down triglycerides and promote LDL receptor activity, helping clear it from the bloodstream. When it comes to cholesterol health, it's important to keep the whole picture in mind, on and off the plate, so things don't get all, you know, scrambled.

8:06This TED-Ed lesson was created by Heyman Chan and directed by Igor Chorich. Now I'm honored to welcome Dr. Eric Topol to take us deeper into this conversation. A trailblazing cardiologist and scientist, Dr. Topol has authored four best-selling books, published over 1 ,300 peer-reviewed papers, and helped pioneer life-saving heart medications like TPA and Plavix. As the founder of the Scripps Research Translational Institute and one of the most cited researchers in medicine, He brings invaluable insights into how AI, genomics, and digital tools are revolutionizing the way that we prevent, detect, and treat illness, including heart disease.

8:46Eric, in your latest book, which I love, by the way, called Super Agers, you highlight people who are living longer and better. So one of the big takeaways I got is that aging well isn't just about avoiding disease. It's about how we live every day. And so much of my work around end of life is really geared around that conversation. So how closely are heart health and brain health linked? And what does that connection mean for how we think about longevity? Well, they're tightly intertwined, just like the brain and the heart, the brain and the immune system, the brain and the gut. There's just this tremendous amount of interdependence, and they all play a role in the heart disease story and a prevention story.

9:28As I cited in the book, at least 80 % from what we know now today of these plaque progression, rupture, cracks of the artery could be prevented. But the risk factors are more extensive than they used to be. They now include things like ultra-processed foods. We also have learned about, for example, sleep. So when we sleep, we rely on clearance of our waste products, toxic metabolites, each day that build up, right? And that is the glymphatic system in the brain. Now, we need a lot of deep sleep because that's when these waste products are eliminated. Well, if you don't get enough deep sleep and you don't have good sleep regularity, which are intertwined, you're on a schedule.

10:19You're on your circadian rhythm. Well, that is a big effect on cardiovascular age-related disease and events. So a lot of people don't know that. It's actually across the board for the three big cancer, cardiovascular, and neurodegenerative. I think a lot of people could intuitively link sleep regularity and deep sleep with prevention of Alzheimer's disease, but the data are really strong for cardiovascular as well. Yeah, you know, we've done several episodes on this show talking about sleep, the connection with essentially our entire health over our lifetime. I want to shift to technology, especially artificial intelligence.

10:59You've written for a long time and spoken so clearly about the promise of AI in health care and actually how it's playing out. But I think for most people in health care in particular, even for physicians, right, it still feels very abstract, like something happening in research labs or behind closed doors of big tech companies, but not yet in the doctor's office. When do you think these tools will start to make a tangible difference for everyday patients, say? Well, it's essential we move that forward. And you're bringing up such an essential path here because we're talking about multimodal AI, which means lots of different modes of data.

11:40So it could be electronic health record, could be sensors that you're wearing. It could be your environmental conditions, air quality, air pollution, that sort of thing. It could mean your images, scans that you've had and your labs. And if you've had any genetic genomic data and on and on. Now, no human being can process meaningfully all that data to say, are you at risk? and is your risk being mitigated by the things that you're doing to prevent this or at least deferred for many, many years? So this is what has changed the whole landscape. If we didn't have AI, we couldn't talk about prevention.

12:22I think this is the next big frontier of AI in healthcare. Now, I'm not sure that the medical world is ready for this yet, but once we get multimodal AI models validated for this precision forecasting at the individual level, and we're seeing some of that already happening, this would become the norm that every person, even young, would say, these are your vulnerabilities, and this is how we're going to prevent that from happening. Once we do that, prevention is how many thousand times fold better than reactive treatments, which at best we're trying to treat a chronic disease after it's already manifest, this has got to be the way forward.

13:07The problem is we've been talking about prevention from the beginning of time in healthcare. But really, we didn't have those layers of data like organ clocks, body-wide clocks, these pace of aging, these protein markers, this multi-cancer early detection, all these technologies, which in and of themselves, we couldn't say that much. But if you were to pick one use of AI that could really help people, this is the one. We'll see if the system, which certainly has not been set up through incentives to focus on prevention, will be interested in doing that. I think it'll be done at other countries.

13:47Sure. Our system is not well positioned for this. But in countries where prevention would be the big payoff and that there's no insurance companies, there's no billing, that's where we'll see this take hold and sooner rather than later. So how do we balance the excitement about AI in medicine with the real need for, of course, clinical validation, safety and equity? And is there anything that worries you about the way AI is currently being developed or deployed now in health care? Well, I don't think there's that much deployment yet. And the reason for that is there's not that much compelling data.

14:26Like, for example, the most impressive data we have is using AI for mammography. to pick up 20, 25 % more cancers, important cancers, that would have been missed by radiologists. And to look at a normal mammogram and still say that this person's at higher risk, this woman in the next three or four years, which there's a commercial capability for doing that now. Now, in this country, the only way to get this is through one of the radiology scan networks, rad net. You have to pay extra. Which is crazy. The point is machine eyes can see things that radiologists and other humans will never see. There's an 80 ,000 plus women, Sweden, randomized trial that prove this is the way to go.

15:17And have we incorporated that in the United States? No. So if we haven't incorporated that, even though there's over 1 ,000 or 1 ,100 FDA cleared AI algorithms, most of these have never been published. Doctors can't look at the data, health systems can't look at the data. And even when they can look at the data, it's not like rigorous prospective type stuff that we would find really compelling. So we need to up the compelling evidence, up the transparency, then we'll get more implementation. But ours now in this country is much lower than it could be. Well, gosh, I so hope that's all coming in hopefully the near future, right?

16:00So for people listening who want to find physicians who are informed about genomics, right, who are comfortable using digital tools or AI, you know, the things that are available in their care, what do you recommend? How can patients sort of advocate for themselves in this rapidly changing landscape where I know for sure medical education is way, way far behind. Well, what you're bringing up is one of my most profound frustrations throughout my career. It's why I wrote the book, Super 8 Years, because people should know about this. But unfortunately, the medical community is short on physicians who are keeping up with all this.

16:46There's so much going on here, the multi-cancer early detection blood tests that are getting better and better all the time. Instead, what's filled this space are these longevity companies, anti-aging supplements, these influencers who are hawking the supplements, who call themselves longevity experts, but really it seems like their primary interest is selling. and this is what's taking up this space because people who can afford it are looking for an edge of how they can have a longer healthier life and what they're resorting to are these people who claim to be the cognoscenti even though they don't have evidence they may be pitching peptides to inject yourself with who have no data or data like take rapamycin which is it works really well in mice, but there's no convincing human data.

17:39It could be very dangerous to suppress your immune system. No one even knows the dose. No less can we can't even measure the immune system in people on any routine basis. So there are so many dangerous things out there like get a total body MRI. And so those things that without evidence are getting traction. That's why I think what's going to happen in this country is people are going to be the grassroots type of go to the doctors i want this i want that you know like for example cardiology do have i had a lipoprotein a because we're going to have drugs that suppress that it's a very important risk factor we didn't ever have anything for it but we will soon or how low should i go with my ldl or you know so many other things like inflammation markers in my arteries ct scans that do show inflammation of the artery wall so there's so many things we could do that we're not doing where the evidence exists and this is where i go cuckoo because you know i see all these ads for this anti-aging supplement crap and peptides and you name it and at the same time we've got good stuff out there so what we're having is just the opposite of your question, is people embracing things that lack evidence, but where the evidence does exist, and it's pretty impressive, we don't have uptake.

19:03It's a wild and crazy world out there. And for people who don't have a background in medicine or in scientific literacy, it's really hard to tease out what's real from what's not. And I I found your book so helpful. For me, I'm obviously a physician. I'm somebody who carries a BRCA2 mutation. So I've only recently learned about that and had to navigate a lot on my own, understanding risk, weighing prevention strategies, making really hard decisions that aren't always clearly supported by typical care. Your work really embodies the promise of personalized or precision medicine. But for many of us, that still sort of feels like a buzzword.

19:44In practical terms, what would truly personalized care look like in someone's real-life medical journey and what needs to happen for something like genomic information, like a BRCA mutation status, to be used more routinely and meaningfully in preventive care? Well, I'll give you an example that I went through in the book for BRCA2, which is not that uncommon a mutation. In Iceland, which has 300 ,000 people, most of them have their genomics. And it was shown through death certificates that people, when they knew their BRCA2 status, it would translate to seven years more of healthspan. Now, it isn't just women with breast cancer.

20:28BRCA2 increases risk for other cancers, like, for example, prostate cancer in men. And men are often silent carriers of BRCA2 to their offspring. Exactly what happened to me with pancreatic cancer in my father. and we had no idea that we, no breast or ovarian in the whole family. This is classic. And what's amazing to me is that the public doesn't know about it. The little bit people know is that BRCA2 breast cancer, that stands for breast cancer. They don't know that, just like what happened with you, it could be an increased risk for pancreatic cancer. It is just as frequently transmitted through men as women.

21:06You know, this is amazing. We have this research domain, which is like Earth 1. You know, all these great papers and top-tier journals that establish all these facts. And then Earth 2 is the uninformed public, including the uninformed medical community, largely, right? You know, you can get a genome sequence, a whole genome sequence. It's down to about$200 now. It's cheaper than a lot of blood tests that we commonly get. No less scans. Of course, you know, there are people that say, I don't want to know. I don't want to know. But, you know, the more ways we have to prevent diseases like Alzheimer's and breast cancer and whatnot, the more people will want to know.

21:48You know, this is sad because genomics should be part of our regular assessment of people to help establish what they're at high risk for. And the data are abundant and so impressive. And it's just it takes years. When I wrote Creative Destruction of Medicine many years ago, I said, you know, we should be using telemedicine. And it sat still for almost a decade until we had a pandemic where you couldn't go to the doctor. And so now we had telemedicine. This is the problem. Do we have to have a pandemic for people to change their ways? It's crazy. Yeah. No, it really is. I think there's a lot of drivers at play that maintain the status quo in medicine.

22:36But I think the more that, you know, people can be talking about the fact that access to these validated tests are cheap, it's an easy test to do, it's actionable information, typically, if your insurance will cover it, right, if you qualify. So you've been such a strong advocate for research and public investment in innovation. Why is it so important, zooming out here, especially now, to double down on science and discovery? And what's at stake as a country, as a world, if we don't do that right now? Well, you touched on it, which is, you know, this personalized, individualized medicine. But we're finally, you know, at the brink of being able to make it in standard practice.

23:19if we embrace it and then we allow it and we invest further in it. The biggest thing is preventing diseases. That's where it's at. And we finally, you know, 25 years after the human genome sequence, we can do this. It wasn't just the human genome sequence. It's the proteomics, the proteins. It's the AI. It's the cells. All these big advances, understanding particularly the immune system. So if we really want to prevent diseases and keep people healthier much longer without these age related diseases, we would invest heavily in it. Now, the problem is that we have big investors, billionaires who are investing in these companies to reverse aging.

24:05They want a potion or a strategy that will just be like, you know, Benjamin Button and make them younger, you know. and that can be done in mice but whether it can be done in people is still uncertain but you know when these folks were young they wanted to be rich and now that they're rich they want to be young this is a problem is that they have put billions of dollars into these companies and a lot of them are doing elegant science work but the problem is they're years away they may never work They may be too risky. Some of these induce cancer in the animal models. And so what if we took that investment and put it into this?

24:45That's what we should be doing. If we could get the billionaires to come together like they're doing for reversing aging and instead say, hey, what about some funds for preventing age-related diseases? And the problem, of course, is we've seen gutting of NIH and public health funding at the same time of this unique moment in medical research. So we need new funding sources. And there's never been a better time in respect of momentum or a worse time with respect to the administration's decision to cut research support. Building on that, can you give us an example of a scientific breakthrough that only happened because of long-term investment and vision, something that wouldn't exist without decades of research behind it?

25:34Yeah, I think the biggest one I've seen in recent years is the P-Tau 217. So neuroscience is fascinating to unravel the brain and all that. But when you can get a marker, a protein marker, that is as good as one from the cerebrospinal fluid, so it's just a blood test, it's as good as a PET scan for a beta amyloid or a PET scan for tau protein, these misfolded proteins in the brain. These are PET scans of the brain, which are very expensive and a lot of radiation. You can get this from a simple, low-cost blood test. It's the culmination of decades of work. And of course, there are other neuromarkers like this from the blood, not just P-Tau-217.

26:22There's P-Tau-181, GFAP, NFL. But the point is, we have a marker. So if you're 50 years old, you would know 20 years in advance if you're at risk for developing Alzheimer's. So you would know at a young age, 40 or 50, when you're still cognitively at the top of your game, that you're at risk. And you have all that time. If you were just starting medical school today, what would you study? And what skills do you think this next generation of clinicians will need to thrive in this new era? AI, I would know how to use in the nuances of multimodal AI and collaborate with computer scientists who don't understand medicine.

27:06And I don't fully understand these models. So, you know, it could be a great synergy if I was a medical student. And what I would really do is say, here it is 2025. This is the beginning of a new year in medicine. The much talked about with no action, you know, no walk in the talk, prevention, prevention of age-related diseases. And I would spend my career on that. What I've been working on for all these years with genetics, genomics, sensors, and now AI, Yeah, it would be, no, the whole package and getting really well versed in each of those areas. Because understanding the human condition and human layers of data, which include our environmental data, things like out in nature, social engagement.

27:57You know, it's not just hardcore stuff. It's also a lot of things that we used to think were soft and now have become a strong body of evidence. that's what I would be working on because I think this is a turning point. Unfortunately, it should happen much faster. It should happen now. But hopefully, we're laying the groundwork for that. Dr. Eric Topol, I am so grateful to you for this conversation, for your incredible work. I would recommend your books to anybody who's interested in how to live a better life starting today. So go check them out. You're very kind, Shoshana. It's such a delight for me to talk with you.

28:39Hopefully we'll stir it up out there. Yes, thank you.

28:49And that's it for today's episode. Thank you so much for listening. TED Health is a podcast from TED, and I'd love to hear your thoughts about this episode. Send me a message on Instagram at ShoshanaMD. This episode was produced by me, Shoshana Ungerleiter, and Jess Shane. Edited by Alejandra Salazar and fact-checked by Vanessa Garcia-Woodworth. Special thanks to Maria Lages, Farrah Degrange, Daniela Balarezo, Constanza Gallardo, Tansika Sangmarniwang, and Roxanne Heilash.