0:00 We're now able [music] to look at coronary disease in a way that was never possible before. One person [music] who's April BSX may have diffuse severe coronary plaque. Another person may have whistle clean coronaries. And now we can know the answer to that. >> Dr. Campbell Rogers is an interventional cardiologist [music] who ran the cardiac catheterization lab at Bighamin Women's, taught at Harvard, and led cardiovascular R&D in the device world. The proportion of people who [music] present with a heart attack who have had symptoms before that heart attack is very small. It's under 50%. If we're going [music] to wait for symptoms to develop to then say, "Okay, great news. We're here. We're going to help." That's good, but it's not enough for the coronary plaque analysis, one can now
0:45 look at a patient's coronaries, and this was never possible before and measure the extent of disease. It's hard to overstate the revolutionary nature of this for coronary disease. >> Is that where we're moving where it's treating the actual plaque seen on imaging versus just the biomarker? >> 100%. It's like X-rays over 100 years ago. >> You trained as an interventional cardiologist. I believe you ran the the Kath lab at Brigham and and Women's and you taught at Harvard and the Harvard MIT sciences program. How did you first get pulled toward the cardiovascular system and then towards coronary artery disease?
1:30 >> Oh, that's a good question. Going way back in my going way back in time, you know, I think in medicine, people are drawn to things they have personal experience with and a family member, etc. often and they're drawn to things in which they see mentors. And where I trained at Brigham and Women's there was just an incredible tradition of cardiological care, cardiology care, academic work really in a worldleading way and those were the people I saw and learned from. So I was drawn to it. You know, you see great people doing amazing things and you kind of want to emulate them. And then at what particular point would you say in your career you became specifically interested in coronary artery disease and atherosclerosis? >> Yeah, very early on. I think there were two things in my experience that led me
2:17 to it. One was in my early training even as an intern fresh out of medical school in the late 1980s. I trained at the Bighgam where there was early adoption of really advanced coronary disease care including giving drugs to dissolve blood clots that were causing heart attacks when that was really new and then opening up arteries that were blocked with balloons and then with stances. And seeing that happen and honestly seeing the change that came to patients when one fixed their acutely blocked coronaries was incredible. I mean I I would analogize it to I expect what is the the gratification that for example surgeons see or orthopedic surgeons somebody comes in with a broken bone and they fix it and immediately someone is
3:04 really treated for what was ailing them that kind of rapid fixing and gratification really appealed to me and you know everyone's different and what they go into in medicine to me that ability to see a problem that's causing a problem for a patient perhaps threatening their life and then fix it I had a lot of appeal >> what What does running the Kath lab actually mean? If someone's un unfamiliar with Kath lab, does that involve some of the things that you just mentioned, balloons and and stances or what does that look like on a on a day-to-day? >> It does and it's really I think central to the hopefully the discussion we're going to have today. So, what's called that to avoid the shortorthhand a cardiac caization laboratory is a place it's think of it to a lay person it's it's akin to an operating room. So it's a place in which procedures are done,
3:50 invasive procedures and there really are two sorts. One is to diagnose what may be wrong. So somebody goes in not knowing and they have a diagnostic test. Think of it like an X-ray and then that it reveals what may be wrong. The second of the two is treatment. So often what's wrong can be fixed using similar invasive tools and that also takes place in a cardiac catheterization laboratory. I mean the word catheterization means using small tubes a catheter to go from some artery or vein in somebody's wrist during their leg and then snake them up using X-ray guidance so that they're in and around the heart in and around the coronary arteries. So that's that's quite invasive compared to what we're
4:37 going to spend a lot of time discussing today which is cardiac imaging. So presumably a lot of those patients you were seeing had fairly extensive or kind of late stage disease. Would that be right? >> Yeah. So that really you've gone right to the nub of the issue. And so here I described those two worlds of a catheterization laboratory. The diagnostic piece and the treatment piece. And they're both, you know, they involve invasive procedures involving the heart and these things have risk, etc. And the truth of the matter is that until quite recently, the only way to really diagnose whether someone to your question had extensive disease or only a little bit of coronary artery disease or actually maybe no coronary artery
5:22 disease, the only way to know was one of these invasive procedures. And so the observation that I made just in my practice in running the calf lab but also it's been published and it's well documented is as follows that if somebody goes in for historically if somebody goes in for an invasive procedure to see do you have coronary disease the answer to that that needs fixing that needs something invasive done. So, you're in an invasive procedure. Only about a third of patients who enter a catheterization laboratory with that question end up having the answer be yes. You've had an invasive procedure and we found something wrong and we're going to fix it invasively. Said differently, twothirds of patients who go in for an
6:09 invasive procedure end up with good news. Great news. You know, Mr. Smith, great news. We did an invasive test and we found out you don't have coronary disease. It's really good news. The problem of course is that to get that answer was an invasive test and it was expensive and at some risk etc. So that's it's not always people with advanced disease historically. It's often people who have no disease yet to document that they've had to go through this invasive procedure. >> For those who who aren't familiar with heart flow, how would you how would you describe heart flow? So heart flow really it's a technology which you know a couple of decades ago arose from scientists at Stanford University and initially the focus of heart flow was to
6:56 say this exact problem I just described that people go into an invasive procedure and then often are told we don't need to do anything invasively to fix it to try to improve that ratio so rather than a third to increase it to a much more frequently. If you're going to have an invasive test, it's because you need to have an invasive treatment. So, what heartflow initially our initial product and analysis was we take images gathered non-invasively non-invasively from a CT scan done of the coronary arteries. So, a non-invasive test. We take that information and then we apply artificial intelligence to the images and advanced computation to the images
7:42 and derive information about blood flow into the arteries. So if someone is told, hey, you know what, you may need an invasive procedure. We're worried you have coronary disease that may need an invasive fix. We can say ahead of time, you can have a CT scan. You then have the heart flow. The product that that's associated with is called fractional flow reserve or FFR abbreviated. We can provide this FFR information. And you know what? If it's not terribly bad, you don't need to have an invasive procedure. You need medical therapy and etc. Symptomatic control, but you don't need invasive procedure. So it can help some of those people in that the twothirds of people that would otherwise go and have in an invasive procedure done and and not require some type of
8:30 surgery. >> Correct. And it turned out we reduced that the what's called a negative invasive angio rate meaning negative meaning you don't need something fixed. We reduced it by about 2/3. So made we showed you know multiple big clinical studies. This made a real difference for patients. And just to be clear, I would presume most of these people, if not all of these people that were going in and having an an invasive kind of procedure to look at the the plaque, the blockages in their arteries, these were people that had symptoms. >> That is correct. By and large, the overwhelming lions share. Yes. And that's actually kind of dogmatic in coronary disease. I'm sure we'll talk more about this, but historically that in the absence of symptoms, it really
9:16 didn't make sense to think about aggressive treatment. >> So the blood flow analysis, which is where heart flow kind of started. That analysis is more important for the person who already has moderate to severe plaque and you're looking to see whether that plaque is kind of affecting them in a functional capacity. It is more important for deciding is it functionally impacting them today. Yes. But I want to caution it's not more important for the long-term health of the patient where there are other metrics looking at the actual disease at the plaque which can dominate. But for deciding does somebody need an invasive procedure to put a stent in. And let's
10:02 be clear, stances are incredibly effective, as is bypass surgery, at limiting symptoms for people who have them, at improving outcomes, preventing future heart attacks, probably preventing death from heart attacks in a subset of patients. They're incredibly effective, but there are also therapies for people who don't need stances. >> Is it the plaque level, Campbell, like the absolute amount of plaque that is responsible for blood flow restriction? And is that something that's been quantified where you know okay once the average person has tends to have this amount of plaque in this artery we tend to see you know a a kind of blood flow restriction which is causing symptoms. >> The short answer is no that there is not
10:49 a direct link and one person may have a large amount of plaque which is not actually limiting blood flow. Another person may have a relatively modest amount of plaque, but you know, bad luck. It's all in one place and it's narrowing and limiting blood flow quite a lot. >> Today in in practice, and I'm not sure if this is in the guidelines or not, but when when would someone be referred for the that blood flow analysis that you guys do? So today coronary CT angography is indicated for people with symptoms suggestive of coronary disease meaning chest pain or being out of breath with exercise feeling palpitations there a whole bunch of very sometimes vague symptoms which can reflect coronary disease. So a coronary CT angagram is
11:35 indicated. In fact in the guidelines it is the preferred test above things like stress testing and nuclear testing and all those other somewhat call them old-fashioned tests. CTNogram is done and then heart flow the FFR component is indicated in the guidelines if someone has a a narrowing between 40 and 90%. meaning the artery used to be 0% narrowed and now it's narrowed down to 40 or even 90% narrowed. That's the range the guidelines indicate should have an FFR done. >> And back to your earlier point, you're saying that that's it's been validated or shown that this can be beneficial in changing the kind of treatment or
12:21 further investigations that a cardiologist may refer their patient to. >> Yes. and most specifically whether or not they may benefit from an invasive procedure. Yes. >> Talk to me about when heart flow decided to add on this additional assessment of plaque. So you've begun looking at at blood flow as you mentioned and then you've shifted and a lot of our conversation today is going to be on on plaque analysis. Was that always the kind of destination? Was it starting with blood flow and then moving into this kind of detailed quantitative analysis of of plaque? I think you know for for the company from the very beginning our mission has been to improve diagnosis and care of patients
13:07 with coronary disease and that includes where we started which is okay let's make sure the right people are triaged into or out of invasive procedures but all along I think the recognition has been that coronary plaque is present and I'll give you a couple of factoids One is that we've talked a lot today about symptoms. Does someone have chest pain? Blah blah blah. The proportion of people who present with a heart attack who have had symptoms before that heart attack is very small, it's under 50%. So if as a field and for us as a company, if we're going to improve care of patients with coronary disease, if we're going to wait for symptoms to develop to then say, "Okay, great news. We're here. We're
13:53 going to help." That's good. But it's not enough and the identification of coronary disease earlier using coronary CT and geography the same basic tool we use for FFR you can see the plaque that's how coronary CTA was developed and so I think from the very beginning there was a realization look we need to sort of start off with a you know as a startup company we were a startup company start off with a single product make sure that it's validated and make sure it's effective and in the guidelines and getting out there. But all along there was this notion coronary CT and geography provides incredibly rich information that we should be able to mine apply different sort of analytic techniques to and provide back to clinicians and to patients.
14:39 >> Let's start with the difference between a coronary CT angography standard type. Perhaps some of the listeners have even done this themselves and you typically you get a report back and that report says you know maybe it says that you have no disease atherosclerosis or it says mild or moderate or or severe and often there's a coronary artery calcium score done at the same time. So perhaps you get a number for for that. How is that scan, that investigation, and that report different to the heart flow analysis of these scans? >> Let's separate it into three different buckets rather than two because you mentioned calcium scoring which is a very important and useful tool just to be clear. So calcium scoring has been
15:27 done historically and is done today using a CT scanning machine. Same machine, same type of machine, but I would say simplified compared to what we're going to talk about next, which is a coronary CT angagram. It's simplified because it doesn't require intravenous injection of X-ray die, iodine contrast doesn't require that and it can be done very quickly and it's relatively ubiquitous. One can get it quite easily. So that's calcium scoring and that provides information. Is there calcium? Let's like mix up your bones. Is there calcium in the coronary arteries where of course it shouldn't be and if there's calcium in the coronary arteries you know that's not not how God made us that indicates there's some coronary disease.
16:12 So that's a calcium score. The next of the three is then okay a coronary CT angiogram and that's a step up in terms of detail and information available. So a CT angagram requires an introvenous an IV to be placed in your arm and then X-ray die iodonated die to be given not radioactive die let's speak clear but iodonated die to be given and as the dye passes through the heart vessels the CT machine operates very quickly and takes pictures of the coronary arteries so they're outline the inner the channel is outlined with dye and that provides incredibly rich information Are there narrowings as we talked about? Is there plaque? Does one have a congenital
16:58 anomaly? Say you were born with coronary arteries of a certain unusual shape. Sometimes that poses risk etc. So that's the coronary CT and the third of the three to your question is okay well what does heart flow provide? So heart flow provides information derived from that coronary CT angagram from the second of our three along this journey. this coronary CT angiogram images they're they're sort of the e they are x-ray images they come to heart flow we do an analysis involves a lot of artificial intelligence and a lot of computation it involves an individual trained analyst for every single case going through and quality checking quality assurance etc and then information goes back to the
17:45 clinician who originally ordered the CTO who's taking care of the patient And that information includes and it goes back in about an hour between an hour and an hour and a half. So this is a very fast process. The coronary CT and is done. Information helps heart flow. We send back information to the clinician and it says are there blood flow limitations etc. Is there plaque present? And again and really importantly let's measure how much plaque there is and this is really the nub of the issue for this discussion. How much plaque is there quantitatively? Not on some sort of scale, but and what's it made of? Is there calcium? Are there signs of plaque that's higher risk? All of that we provide back for clinical use. >> Okay. So, let me throw this back to you
18:30 just so that we're we're all on the same page here. The the AI coronary plaque analysis or AI CPA, which is what heart flow is doing, a more detailed understanding of someone's plaque burden. So the the standard CT angiogram kind of gives you an indication of you know do you have mild disease or moderate or severe but this deeper analysis tells you more clearly the absolute amount of plaque you have but also the type of plaque the composition of that plaque and so my next question to you Campbell is with that information what's the benefit of having that information over just the information in a traditional CT angography >> I was trying to think of analogies in medicine that sort of would be relevant
19:18 and these tools to quantify coronary disease in a way that cannot be done without them that can't be done by somebody just reading the CT scan as a clinician. An analogy would be something like with the invention of X-rays over a hundred years ago. Prior to the invention of X-rays, if someone fell down and hurt their leg, clinician wouldn't know, is the bone broken. Physician could say, "Look, I'm going to do a physical exam. I'm going to assess it. I can't be totally sure often if the bone is broken." All of a sudden, an X-ray is done and now there's no question. It's there. It either is or it isn't. And the clinician has immediate new information. That's how I think about these tools for
20:03 assessing coronary disease using AI from CTA for so for the coronary plaque analysis. One can now look at a patient's coronaries and this was never possible before and measure the extent of disease, not guess about it, not rely on the patient's risk factors or the fact that they had a family member who had coronary disease, but it's patient specific. You either do or don't have it. And if you do, here's how much of it you have today. It's hard to overstate the revolutionary nature of this for coronary disease. It's very exciting as a time in the field. >> And does understanding the the type of plaque that that someone has, if it's low attenuation plaque or non-calcified
20:49 plaque versus calcified plaque, does that help you determine that patient's risk? Does it help with risk assessment? >> It does. And I'm going to give a a couple of I would say editorial comments as well. It does. So there is very good data that the and this is not won't surprise anybody, but the more coronary plaque one has, the worse the prognosis, the higher the likelihood of eventually developing heart attack or developing chest pain or or or needing a stent or dying from heart disease. So the more and specifically the more plaque without calcium, the more non-calcified plaque. But here's my editorial comment. This is an area now that these tools exist and they're quite new that we as providers need to have a lot of humility because
21:36 we're now able to look at coronary disease in a way that was never possible before. So we are going to learn. So for anyone to be dogmatic and say, "Okay, great. I know exactly what you know this type of plaque is associated with this type of outcome and this change over time. The science is in development and the clinical studies are ongoing using these tools and again I think humility for us as in the field humility is a good thing to have because we're going to learn a lot about coronary disease that we have not been able to know before. >> Are there some some established or accepted numbers when it comes to to plaque volume? I think listeners of this show will have heard previous guests talk about, you know, young adults in their early 20s in the Korean War who,
22:24 you know, post post autopsy showed that they had atherosclerosis at a very young age and that this is a disease that occurs across a lifetime. It's about lifetime kind of exposure and burden. And if atherosclerosis isn't your cause of death, most people are going to die with some degree of atherosclerosis. At least that's what I've had previous guests kind of share with us. And you can comment as to whether that's something you you agree about. But is there an established amount of of plaque that you would say is where things become problematic? >> The short answer is no. I agree with everything you said about the prevalence of atherosclerosis. those early Korean War victim data studies are are solid and they looked at the aorta and saw
23:10 plaques, fatty streaks in the aorta of those soldiers and it's very common and it's gotten more common for a whole variety of reasons over the last few decades. So I think the way that we look at it is there's an age component to your question. So one way to think about it is and we have for our plaque analysis we now have this information. It's part of what you know the clinician gets back and it's as follows. Okay, I have a 40-year-old male with this much plaque. We have looked at a population of nearly 300,000 people and I can tell you from that population amongst 40year-old males, this example patient has more plaque than 90% of them. That tells you something. So, let's say I
23:55 I'll make up the numbers. Let's say they have 100 cubic millimeters of plaque. They're 40 years old. in that again let's put some in the 80th percentile for 40-y old men. If you then switch and say it's no longer a 40-y old man now, excuse me, now it's a 70 year old man. They're no longer in the 80th percentile. They may now be in the 20th percentile. So you need to know the age, the sex of the patient and couch it at least in terms of that framework. >> That makes sense. So it's it's not just how much plaque do you have, but how much plaque do you have at a at a certain age and then comparing that and contrasting that to the your your peers at the same age. What can you tell us Campbell about how accurate reliable these plaque measurements are against
24:42 the gold standard? So intravascular ultrasound, that invasive kind of assessment where you're getting in and really being able to understand exactly how much plaque someone has. How does this this AI analysis of CT angography compare? >> So that's a critical question because look, one could bring forward a tool and say, look, great news, we're able to extract information about plaque from CT. The immediate question any clinician and any patient should ask is okay good. You say you can do this. Prove it. Tell me if you say I have 100 cubic millimeters of non-calcified plaque. How do you know that's true? How do I know I can trust that number? So you put your
25:28 finger on assignment. The answer is in clinical studies. As in anything in medicine, the proof of the pudding is in clinical studies comparing to a recognized gold standard. And the gold standard as you referenced is intravascular ultrasound which involves putting a catheter down the coronaries with an ultrasound transducer on it imaging plaque that way. That has been done for decades. It's a standard very invasive but a standard tool. So for example for heart flows plaque analysis we've done a clinical study totally prospective totally blinded where we where patients agreed consented to have an ultrasound done and to have a CT done. And then in a totally blinded way, we compared the two. So we had a core lab take the ultrasound, say, "Okay, here's how much plaque Mr. Smith has."
26:14 We took the CTA, we said, "Okay, here's how much plaque we think Mr. Smith has." And we compared the two. And they actually compare extremely well, and this is all published in peer-reviewed literature, etc., etc. And it has to be prospective because if you go about it and say, "Well, look, you know, we happen to have a few patients who had intravascular ultrasound and they may have had a CT as well. So we're going to look retrospectively and see how they compare. There are all sorts of biases either intentional or unintentional which can seep into such an analysis. So we're firm believers in prospective data looking at this correlation of the two. It's like X-rays over 100 years ago. It's no longer guesswork. One person's sore leg after they twisted their ankle may have a [music] broken bone. Another it may just be a strain. One person who's a BSX may have diffuse severe
27:01 coronary plaque. another person may have whistle clean coronaries [music] and now we can know the answer to that. So why wouldn't we use that information? Now I hasten to say we need to [music] g continue to gather the clinical evidence that these are the right choices so we can compel patients, physicians, guideline [music] writers, etc. to move in this direction. It's not just take my word for it. We have to prove it. But to your question, is this the direction it's going? Absolutely. >> A quick message from me before we get into today's episode. We are running an audience survey right now and I would really value just 2 minutes of your time. You know that there are ads on this podcast. It's how we keep the lights on and allows me to spend my time researching and then ultimately sitting down and chatting with true experts who aren't going to dish up wishy-washy
27:47 science claims. And we want to make your experience with these episodes better, not worse. And to do that, we need to understand you a bit better. Who you are, what you're here for. It helps with the decisions that we make around brands to work with and it helps us tremendously on the part that I care most about the guests we bring on and the topics that we dig into. The more that we understand what you find useful, the more we can make episodes you actually want to listen to and episodes that you benefit from. It's quick, it's easy, and it is one of the most meaningful ways that you can support the show. And again, it takes about 2 minutes. So, if you have a moment now, please head to voicingange.dia/simon and fill it out. Again, that's voicing.
28:36 SSimon and fill out the 2minut survey. Thanks for being here. It means a lot. Now, let's get back into the episode. the keto CTA paper, which I know you're familiar with, and they ran some of their scans or all of their scans through through heart flow in their most recent preprint. It was interesting to me that their primary outcome used Qangio, which I hadn't heard of, and I believe it's just it's a research tool. It's not something that's available to patients or or individuals. Do you know much about Qangio and how how does that compare to heart flow in terms of the way that it analyzes plaque? >> Well, I'll tell you what I do know and again there are others I'm sure who who with much more experience with it. To reiterate what you said, Qangio is
29:22 largely a research tool used by core labs has been used for decades for core labs for measuring plaque. It's quite labor intensive so it doesn't lend itself to sort of routine clinical use based on that. It also involves quite a bit I believe of operator interpretation. So kind of some subjectivity or risk for interobserver variance. So I think some of that may seep but I I don't know certainly used widely. There are lots of very good publications which have relied on on Qangio as a research tool. Yeah. >> Back to your earlier comment around a coronary artery calcium score. You mentioned there's kind of three three different buckets of things that we can measure. coronary artery calcium score traditional CT angography and then then the AI kind of driven analysis of the CT
30:09 angog what would you say to the person who's had a coronary artery calcium score done and it was zero and so they have been led to believe that means that they're at very low risk of cardiovascular disease and don't have atherosclerosis. Is that the case? And is there value in someone with a coronary artery calcium score of zero doing this AIdriven analysis of a CT angagram? >> I would separate the two conclusions you just posited. One is they have very low risk and the second is they don't have coronary disease because I'll answer those two differently as conclusions from a calcium score of zero. calcium score of zero does impart very low risk
30:56 and that's been proven across large studies etc. It does not say the patient does not have coronary disease. Like those are two different things. Somebody certainly can have plaque in their coronaries that is non-calcified. So a calcium score, you know, it is invisible to a calcium score. That does exist. And I think I would say to a patient like that, look, your calcium score is zero is great. That's awesome. Your prognosis, your risk of having a heart attack and dying or having needing a stent, blah blah blah over the next several years is very low. full stop. It does not tell you you don't have coronary disease. So you should see if you know this is a concern. You should understand your risk factors. You should understand your lipid levels. They should talk to a physician. And
31:41 depending if you are someone who has very high risk despite a calcium score of zero, there may be a role for imaging with with CTA. Again, I would emphasize though that today CTA is not indicated in the guidelines. It's not paid for by insurance companies, blah blah blah, unless a patient has symptoms. Doesn't mean it's not used in those patients without symptoms. But just for so everyone's aware, for patients without symptoms, it's currently not indicated, which is a whole different discussion that we can have. >> Tell me about that. Who do you think is the ideal person or persons that can benefit whether it's risk assessment or treatment or both from having a a CT
32:28 angog done and then going that next step and having the AI analysis done? >> Certainly it's patient let's start with the sort of the easy ones. It's patients who have symptoms of coronary disease who who aren't who are not acutely ill but have symptoms. That's the guidelines. It's class 1A. It's the top ranked test. But that's that's I think that's not the core of your question. For patients who don't have symptoms, I think there are certainly patients who are at very high risk. People who are in high-risisk occupations and you think of pilots and first responders and so forth for whom it may be warranted to understand do they have coronary disease. Next would be people who have coronary calcium. And let's talk about this. So, we've said calcium scores are great. If
33:14 they're zero, your prognosis is good. But if they're not zero, they indicate that you have a patient has coronary disease. And then, and it doesn't have to be more than a calcium score of over five or 10. Any coronary calcium period means you have coronary disease. And in that setting, I think there is a relatively an increasingly strong case to be made that understanding not just the calcified plaque, but also the non-calcified plaque can be super helpful. And if the calcium score, especially if you look at the guidelines, anywhere from say one to 300, there's a lot of subjectivity in
33:59 there of how much therapy should someone be on, medical therapy. Once you get over 300 or 400 of calcium score, the guidelines are pretty aggressive and most people would say you should be treated aggressively. But in that kind of call it the middle range, there's a lot of ambiguity. The final point I would make is and this is really for clinicians. It's that in treating coronary disease with medicines, talking with a patient, there's a conversation about look, you know, whatever your LDL is high and you need to take a medicine to lower it or you have risk, you need to take medicine. And for many patients, it's super abstract. It's kind of like high blood pressure. You can't feel it, you can't see it. And so the adherence of patients with medicines is notoriously poor. So we see value in
34:48 potentially using the CT androgram and the plaque measurements to go to a patient as [laughter] a physic for a clinician go to a patient and say look it's not just your LDL measure here's your coronary disease here it is I can show it to you now here's why you need to do the following to take medicines modify lifestyle and so forth >> is this where cardiology is moving do you think and is there evidence to support this idea that you treat the the plaque and not necessarily just the number. So if we say for example, we take a 30-year-old who goes in and sees his doctor and his Apo is super elevated, tries to [snorts] make some lifestyle changes, it's still elevated. His doctor speaks to him about, you know, lowering that with with a drug and
35:36 he's left wondering, well, do I have disease? And we know that some people for whatever reason they have elevated APOB and are not getting the same disease as the next person who has the same APOB. So is is that where we're moving where it's treating the the actual plaque seen on imaging versus just the the biomarker? >> 100%. And it's back to the analogy earlier again. This is where we need humility. It's like X-rays over 100 years ago. You know, it's no longer guesswork. One person's sore leg after they twisted their ankle may have a broken bone. Another may just be a strain. One person who's apx may have diffuse severe coronary plaque. Another person may have whistle clean coronaries. And now we can know the answer to that. So why wouldn't we use
36:22 that information? Now I hasten to say we need to get continue to gather the clinical evidence that these are the right choices so we can you know compel patients, physicians, guideline writers, etc. to move in this direction. It's not just take my word for it. We have to prove it. But to your question, is this the direction it's going? Absolutely. >> And what are the the downsides, barriers or or risks of scanning more people other than you know there's a cost and accessibility component? Are there any safety issues with with performing more CT angogs like radiation or other things that like the intravenous iodine contrast? >> There are radiation is the main one I'll talk about in a second. The contrast, you know, for people who are not allergic to contrast medium, which some people are, is really quite benign. The volumes of contrast used are really
37:09 small. So that really is important, but it's relatively minor. The radiation is not minor. And although these CT angagrams are quite parsimmonious when it comes to how much radiation they use, any radiation is greater than zero. So it is a factor to consider. There's quite a literature on the tiny effect of these doses of radiation on long-term risks, for example, of cancer, but there is a literature on it and people need to be aware of that. But this very very modest doses of radiation. >> We're going to walk through my own scan results and and the heart flow analysis of these. And I for background for listeners who haven't caught my previous episodes, I've had a scan back in 2024 and then I had a follow-up scan in 2026 which was about 16 months later. Before
37:57 we get there there, Campbell, just a little bit more information that I'm interested in on exactly what happens when when the lab sends you the the scan. Can you just walk us through in a little bit more detail how how does Heartflow perform that analysis and how how important is the machine that's used for the scan the types of images that come through the overall kind of protocol at the center where you have your CT angiogram. So what the logistics of what happened our CT androgram is done the images come to us and the first thing we do is we do an assessment of quote image quality and that means are the images sharp I call them you know
38:44 loosely speaking are they sharp enough for us to then do our detailed AI analysis and the answer to that is not always yes it's yes it is the images are good enough about 97 a little more than 97% of the time but it's not 100% And the reason that other two or 3% we don't provide an analysis is the images are just too fuzzy. There may have been, you know, an inadequate medications given at the time or other factors that may factor into that. Once it passes that bar though, it then goes into our automated process with the people periodically doing QC checks etc. as I described. Now to your question about the technique at the site, it does make a difference. the you know each site has its own unique scanner and there are a
39:29 whole variety of different types of CT scanners. They have their own unique procedure if you will for exactly how that camera how the scanner is used, what medications are given etc. So we at heart flow spend a lot of time working with CT imaging sites with our customers to get them to really understand here's what you need to do to make sure the images are optimized. So that's point number one. it does make a difference. The second is if one is comparing and we're going to talk about this in your case Simon. If one is comparing two scans over time in the same patient, has the disease gotten worse, better, stayed the same? You know, back to my fracture analogy, is the bone healing or is it not healing? The more one can have the same scanner, the same technique then
40:15 and now makes a difference. For sure. >> Yeah. And I'll I'll point out to the listener because of that I had both my scans at the Lungquist Institute in LA. So the same center and and I actually requested to make sure it was the exact same machine and and protocol to to try and reduce some of that variability there. [snorts] So let's let's get into to my results. Campbell, perhaps we start with the baseline and then we can walk through the the changes that were observed on my follow-up scan, which again was about 16 months later. >> Some may say, "Oh, we don't see that [music] or that doesn't sound like coronary disease." Let's have the humility I've been asking for. We've never [music] had tools to look at this. So, how do we know that this hasn't
41:01 happened before? We don't know. And here we have really good imaging done [music] at a wonderful place in Los Angeles that does worldclass imaging. you have all of this AI overlay and okay here's exactly what [music] happened. So let's use that to understand both for you as a patient what's important and then for the field how can [music] these tools perhaps be used more broadly. >> All righty friends, quick one before we dive back into this episode. Summer sleep can be brutal and I want to give you a better tool than just blasting the air conditioning all night long. Here's the thing that many people miss. Your core temperature has to drop for you to fall into deep sleep. And what actually matters there is the surface you're lying on, not just the air in the room. That's the gap eight fills. It's a smart mattress cover that sits on your
41:46 existing mattress and cools or heats each side independently anywhere from 12 to 43° C. If you've got AC, the pod works right alongside it. You're cooling the bed directly instead of the whole house, so you can ease off the thermostat a little and take some load off that summer power bill. And if you don't have AC, this is the one. It turns the bed from a hot, restless, sticky, kick the sheets off situation into somewhere comfortable you actually sleep through the night. My side runs cool. My partner's runs a little warmer and nobody has to negotiate. Here's the really clever part. Eight adjusts the temperature of your side of the bed according to your body temperature, the room temperature, and the stage of sleep you're in. Cooling the sleep surface
42:33 before you get into bed helps you drift off faster and supports a deep slowwave sleep. And keeping the temperature in the right zone overnight means you're not getting the thermal disruption that negatively affects REM sleep later on, the stage of sleep where memory consolidation takes place. Use my code the proof at 8.com/theproof for up to $500 off. The 4th of July sale ends July 12th. You get 30 days to try it at home and return it if you don't love it. Again, code the proof atleep.com/theproof. We all know that hydration matters. That's something we're taught from an early age, especially if you're someone who exercises and sweats regularly. What's often overlooked, however, is that hydration is more than just
43:20 drinking water. It's electrolyte balance, particularly sodium, that drives water absorption and ensures it's properly distributed across your cells and the fluid around them. For me, that means a serve of element 60 to 90 minutes before I play padell or do any sort of vigorous exercise, either a can or half a sache. And if I'm out there for over an hour and sweating, I'll have more during the breaks. I'm religious about it. I'll even get whoever I'm playing with on the electrolyte program. anything for a little extra competitive edge over our opponents. Each serving delivers a meaningful dose of electrolytes without the artificial ingredients, food dye, and other unnecessary ingredients that you find in many sports drinks. Element was built around the research showing that for active people, optimal sodium intake
44:06 often sits well above standard government recommendations. They've just released 12 oz sparkling, the same electrolyte ratio as their 16oz can in a slim format. Each can delivers 500 mg of sodium, 100 mg of potassium, and 30 mg of magnesium. There are four flavors: black cherry salt, lemonade salt, orange salt, and pineapple salt. Receive a free element sample pack with any order at drinkelement.com/simon. That's d r i n klm nt.com/simon. So maybe walk us through the topline takeaways from the 2024 the baseline scan and for those who are watching on YouTube you'll be able to to see this on screen as well. >> Sure. So Simon what you're looking at
44:53 here is your heart your coronary arteries and this is the heart flows so-called user interface. So your clinician would have access to this and be able to go in and look at it just as we're going to do now. So what you see here, it's a heart flow analysis. Here is your date of birth and here's your age as of when the scan was done, December 3rd, 2024. So to start with, what you see on the left are the coronary arteries. These kind of snaky looking things. This big tube in the middle is the aorta, which brings blood out of the heart to the rest of the body. And the coronary arteries branch off of it, as you see. And what I'm going to do is I move it. You can see there are multiple arteries that bring blood to the heart. Your right coronary artery is shaped like a C over here. Your left anterior descending is here.
45:38 And your left circumflex is this one that comes down around the back side of the heart. Now, what you're looking at, you see these areas of yellow. And that's where you have coronary plaque just as we've been talking about. So, we can actually zoom in on that and look at that and say, okay, here's where some plaque has developed over time over your 38 years. And then if we said okay amongst to our earlier discussion amongst 38-y old men where does this put you and we have a wealth of data in 38-y old men and so this is actually if you look down here at the bottom the total plaque that you have is 116 millimeters cubed that puts you in the 89th percentile meaning you have more plaque than 89%
46:25 of 38-year-old men whose scans have come to us. So that's point number one. Point number two is we've talked about flow and fractional flow reserve. That's this. I'm going to click this little lever up here. And this shows, hey, great news. There is no limitation of blood flow. If there were, it would appear yellow or red, but there's nothing. It's all blue and blue green, which is all normal. So your blood flow is normal to your heart despite having plaque. We're going to switch that back off. And then the last thing we can do is we can say, well, what is this plaque made of? So now what I'm going to do is I'm going to go to this middle panel and I'm going to grab the with my mouse this little slider. These three stripes in the middle are your left main and left anterior descending arteries strung out
47:11 from the beginning at the top to the more terminal ends at the bottom. And we're going to sort of scroll down like this. I'm moving it down and I'm going to draw your attention. I'm going to pause. Pause here. draw your attention to this little bow tie over here. This shows exactly where, as I scroll, where in your artery we're looking. And finally, over here to the right shows this is where the plaque is. Here's the artery. This kind of gray shape here, which is bringing blood to the heart. And here's the plaque. It looks sort of like a cap on top of the artery. It's not narrowing the lumen, which is great, not limiting blood flow, but there's plaque there. And just to then quantify this, if we're going to go back to the very beginning, you in 2024 had plaque
47:59 in your left main artery. That's right here. You can see a little bit of calcium, mostly non-calcified, which is yellow. Calcium is blue. And then the plaque goes away. And then down here in your left anterior descending, you have another area of plaque. Right? So that's as of 2024. That was what was seen. >> Campbell, just a quick question on the left anterior descending where you were showing that cap of of plaque and you mentioned that blood flow is not affected. Is that typical in in early stage disease where the plaque is kind of growing outwards out of the vessel rather than inwards and and causing obstruction? >> Yes, it's not it's not universal, but
48:44 it's quite common. The term of art, if you will, the clinical term is remodeling. So the artery remodels and gets bigger so that it kind of knows plaque is growing. It wants to keep it from narrowing the lumen. So the whole artery gets bigger in a way that can accommodate the plaque yet not compromise blood flow to the heart. >> And just to be clear when we think about the risk of that plaque, one aspect of that is is it affecting blood flow? And in this case you're saying it's not. But is there also some risk in whether that plaque kind of ruptures and ends up traveling down through the artery and causes a blockage elsewhere? >> Yes, exactly. And it's a great point. It's a great illustration of what coronary disease risk is. So, let's talk
49:31 about that. The primary mechanism of heart attacks is that a plaque such as this, and you use the term rupture, would rupture. All of a sudden, in the blink of an eye, for reasons that remain not fully understood, the plaque will suddenly kind of pop as if it were a blister. And it will cause a blood clot because all of a sudden all of this cholesterol and all this abnormal disease tissue is exposed to the blood, which is flowing right here. and a blood clot forms and then then there's a problem with blood flow because now the arteries totally blocked or nearly totally blocked and the left anterior descending brings blood down is the biggest of the three coronary arteries
50:16 in in you and in most people and it brings blood to a big chunk of the pumping chamber the left ventricle of the heart >> and that I think will be something that we might come back to that when I discuss a little bit later my decision around how I'm going to treat this and it'll be something I unpack further in an episode with Dr. Dan Sofer who's helping me with the the treatment side of things. But let's move through to 2026 because this so far sounds a little bit morbid and I'm sure people want some good news. >> All right, so Simon, now we've switched and now we're looking at your second scan which was done about 16 months after the first one. So now we're up here in April 2026. You've now had a birthday. So you're 39 years old. And a
51:02 couple of things changed which is really interesting and uh remark I say remarkable. As a reminder, here is the coronary arteries. They look the same as they did in terms of their basic shape and structure. We still see here's this plaque in the left anterior descending this yellow area. But a couple of things have changed. Number one, if we look down here at the total volume of plaque, it now has fallen to 60. your percentile, both you had a birthday, so you're older, and the amount of plaque has gone down. So now you're down in the 80th percentile. You were in the 89th percentile, as a reminder, a year and a quarter ago. And then what we're going to do again is look down by scrolling down this. The first thing I want to highlight, and I'll show you sideby-side
51:48 images in a second, is that in your left main artery, there had been a plaque before, and we don't see it anymore. Oh, it's gone. was small but it's gone in your left anterior descending will come down and this area where there had been sort of that big thick plaque we talked about it rupturing and so forth appears less it's you know the yellow area is less and this quantification of plaque in the left anterior descending is less than it was before especially this non-calcified yellow plaque. So again, we'll look at them side by side in a second in an exact comparison, but it's really interesting to see this difference. And I want to hearken back on one of the things I said earlier,
52:34 which is these sort of learnings. You know, some may say, "Oh, we don't see that or that doesn't sound like coronary disease." Let's have the humility I've been asking for. We've never had tools to look at this. So, how do we know that this hasn't happened before? We don't know. And here we have really good imaging done at a wonderful place in Los Angeles that does worldclass imaging. You have all of this AI overlay and okay here's exactly what happened. So let's use that to understand both for you as a patient what's important and then for the field how can these tools perhaps be used more broadly. >> Just a quick question on the the percentile. So you mentioned there that on the 2026 scan with the reduction in in plaque which was about a 40%
53:20 reduction or maybe a little bit more during over that 16-month period. So annualized would be a little less than that. You said it put me at about the 80th percentile. Do you know Miami heart that that cohort? I looked at that and at least in their youngest age kind of bracket they had 40 to 44 year olds. They didn't have a bracket under that. that level of total plaque volume put me at about the 50th percentile. Is that just differences in in kind of the the population characteristics in those in the two different groups, the all of the people that you've scanned versus their population? >> It's possible. I think again I'm not as familiar with I'm familiar with the publication for Miami heart that used I think a different tool that didn't use the heart flow plaque analysis which you
54:06 know as I mentioned has been validated is the only one that's been validated prospectively so I really can't speak to the you know are there differences in population is it the tool that's being used that is a little you know is overalling disease in that case etc so your percentile drops I don't know >> and then just in terms of of validation We were speaking before about validating against intravascular ultrasound. And so presumably, and I haven't dug into these studies, but presumably that's validating in subjects that have quite a lot of plaque. And I've had several people kind of comment on my level of plaque and saying, "Look, in the scheme of things, it's it's very small amount. It's a it's a mild amount of plaque." And one of the questions I had for you
54:52 was when you're using AI to analyze plaque in someone who has a small amount of plaque, how accurate is that measurement of plaque? And is there is there likely any kind of noise as you get down to small amounts of of absolute plaque levels? >> To speak specifically to our data against intravascular ultrasound, which you asked about, the way that we did that prospective only prospective study was as follows. We looked at individual plaques. So the patients tended to be patients who were undergoing invasive procedures. You're absolutely right. So overall they perhaps had more disease or were a population of higher disease than would include you. But the way we looked at it was by lession. So IVIS is a very
55:37 specific tool. I'm going to image from here to here and quantify the plaque. And we did the same. So the comparisons we made and were published were at the lesion level meaning that even though the patient may have had more lesions the perleion agreement is what we're talking about here. It's a very sightsp specific disease and that's what we've quantified and validated. So in that paper there are lesions ranging from very small plaque volume such as you have to bigger lesions with more plaque such as people with higher disease burden might have. And what do we understand about measuring change in in plaque over time with with heart flow? So using me as an example again here, I I measured 16 months apart from baseline
56:25 to to followup. Then I was I was reading the the that ACC scientific statement and they had some wording in there that suggested, you know, optimal time interval was perhaps, you know, 5 to 7 years. And then I started thinking with someone that has a small amount of plaque like myself, did I wait long enough to get an accurate reliable kind of pulse check on on what the difference in in plaque is truly >> I think that again I keep coming back to the same theme. I apologize. This is about having humility. We don't know. And in that scientific statement, you're right. There's a range. I think it goes from two years on up as a recommendation. shouldn't be done sooner than that because it makes sense sort of at a in a in a broad way of course but
57:11 there isn't any real you know large volume data on this on what what changes how quickly it changes so to say oh you know it's too soon we don't believe this we don't have an experience upon which to make such judgments I'll tell you for our product for our analysis we have a lower limit if it's anything below one millimeter of calcified plaque or you know 2 millimeters of total plaque. We don't call that as plaque. So we say okay there there may be small changes that you know we can't be totally sure of in the one or two millimeter range. That's not what we're talking about here in your case. These are changes you know of tens of millimeters of plaque. And we have a lot of confidence that given everything you said imaging site same
57:57 scanner same technique great imaging. I think no reason to disbelieve this. And I'll tell you a bad reason to disbelieve this is, oh, we just don't expect this because we haven't been able to look at it. The connection of these tools for coronary disease management [music] and good clinicians guiding patients in their preventive course is essential and [music] there's a way marker on the way to showing reduced death and heart attack rates and that way marker is lowering LDL. [music] We have already shown and published in a study called the side that use of this plaque tool led to meaningful reductions in LDL. You know, think of Apo B etc. lower lipid levels associated incontrovertibly in dozens of trials [music] for decades with lower rates of
58:44 heart attack and cardiovascular death. >> Switching gears for a second here, folks, to talk about Shopify. Shopify is the e-commerce platform that we use at 38 Terra to sell our science-backed gut health products into the US, the UK, Australia, New Zealand, and soon to be Canada. If you've ever thought about turning an idea into something people can actually buy and experience, this is one of the main hurdles that stops people building the site, taking payments, routing orders to your warehouse. Shopify makes these things an absolute breeze. It's honestly never been easier. the design and launch tools, templates plus AI, get a professional looking storefront up fast. No coding required. Then there's checkout where a lot of stores quietly lose people. Some recent tweaks we made
59:31 to our Shopify checkout have meaningfully lifted the number of customers who subscribe on their very first purchase and returning customers can check out in a single click. And if you get stuck, their built-in AI assistant, Sidekick, helps you build, troubleshoot, and keep moving. like having someone in the room who set up a thousand of these things before. All you need is the idea. Shopify handles the rest. Start your free trial at shopify.com/proof. That's shopify.com/proof. Sign up today at shopify.com/proof. The more I've learned about the supplement industry, the more selective I've become about the supplements that I put into my body. This category has a
1:00:17 real trust problem. Light regulation, easy manufacturing, labels that don't have to disclose everything that's inside the formula and so on. So, when I find a brand that is doing the work properly, I tend to stick with them. That's why I partnered with Momentous. They built what they call the Momentous Standard, which is a commitment to testing, to transparency, and label accuracy that frankly very few other companies in this category bother with. Every product is independently certified by NSF for sport or informed sport, which means it's tested for contaminants for heavy metals and banned substances and verified to actually contain what the label claims. Two of the supplements I take every single day are made by Momentous, their plant-based protein,
1:01:04 and the Creapure creatine monohydrate. The protein blends pee and rice for a very strong amino acid profile supporting muscle adaptations from resistance training and tendon ligament and muscle recovery from playing pedal. And on creatine, the evidence has been quietly stacking up over the years, not just for muscle strength and power output, but for healthy aging, particularly cognitive performance, and when combined with resistance training, bone health. Right now, Momentus is offering listeners up to 35% off your first order with the code the proof at live momentous.com. That's live momentous.com and use the code the proof. I think I know the answer to this, but something that that
1:01:49 I was thinking about when reflecting on the keto CTA study, the first study they published that it was retracted and then the preprint along with my own results. And let me just give you and you know this but the listeners some more background here. I also had my my scans analyzed by clearly and by Qangio too. Matt Matt Budof very kindly ran my scans through Qangio and that's usually or it's not something that's available to to the public. So, as we mentioned before, it's a research tool, but it allowed me to kind of compare and contrast between these different tools, which I thought was interesting given a lot of the conversation around keto CTA. And first thing I'll say is that all of
1:02:34 these analyses, they actually did agree with each other at a high level in that they all showed I have quite a small amount of of plaque. But there was some disagreement in the absolute amount of plaque, albeit all of them saying it was low, and then also the direction of of of change over that 16-month period. And it would be easy for me to just use heart flow and and kind of hope that that is the most accurate one because that showed the best result and that would be my my as a patient that would be my bias. I'd look to that one and say look that showed 40% regression. Great. So my question to you is the way that I
1:03:20 was thinking about this and also the key keto results was that both in my case and in the the case of the subjects in that study were looking at people who have low amounts of plaque scans done over essentially a year. So I was thinking based on that scientific position statement that maybe there is some noise here and maybe there needs to be some more validation between serial scanning between baseline and follow up in subjects with a low amount of pluck from these different analyses. Do you have a different view on that and why there is some disagreement in the results that I have and how I should think about which one is likely to be most accurate or reliable? >> Yeah. So I think look as you have gotten
1:04:07 a sense in this conversation we at heart take tremendous pride and put a huge amount of faith in our clinical data published great journals prospectively validated blah blah blah. We've talked about that. the other vendors in this space have not published purely prospective IVIS comparisons. So that's kind of just point number one. So we have so we stand on the evidence that
…
