Blood Tests to Predict Heart Attacks

Researchers in the Scripps Translational Science Institute in California have developed a blood test that may be capable of predicting an imminent heart attack.  Publishing in Science Translational Medicine, Eric Topol and colleagues built on early work that shows populations of unusual endothelial cells in blood samples taken from heart attack patients.  These cells seem to come from the linings of arteries.  So to find out why this was so interesting, Chris Smith spoke to Dr. Topol.....

Eric -   Well, we have a very big unmet need and that is today, we can't diagnose heart attacks that are incubating.  It's very easy to diagnose a heart attack where there's already been damage to the heart muscle.  So there's enzymes in the blood and there's a cardiogram that shows that, but the limiting factor is that we can't tell when the artery has cracked before the blood clot has formed.  It's the blood clot that causes the stoppage of blood flow to the heart muscle, so we want to know when that crack is occurring which is a precursor to a heart attack.

Chris -   So how did you approach that problem? 

Eric -   We know that when someone has a heart attack, we have a narrowed area of a blood vessel which has an atheromatous deposit in it and for some reason, this atheromatous deposit ruptures, or cracks open, and this forms a blood clot which then blocks the vessel.

Chris -   So with that in mind, how did you approach say, well is there a way to try and pre-empt when this might be about to happen and who's at risk?

Eric -   Yes, so back in 1999, there was the first paper ever that showed that these cells that were presumably coming from the artery lining could be found in the bloodstream before a heart attack.  And so, that was a very provocative paper that sat dormant for well over 12 years. We didn't really have until more recently the ability to unequivocally identify these cells as truly coming from the artery, and also to zoom in on them and to do such things as sequencing and elaborate studies to understand what these cells are all about.

Chris -   So what did you actually do?  You have a group of patients who have a heart attack, you have another group of patients who are equivalent to the first group that haven't had a heart attack and you compare the cells in the blood of both?

Eric -   Right, so normally, there's very few of these cells, if any, are in a healthy person and the cells are very elliptical, with one nucleus, they're very consistent. Whereas in the heart attack individuals from their early minutes of their blood had nothing to do with the heart attack per se, these are cells that clearly have been present for some days prior to that.  But what was so unique about these cells is they were giant, they were very distorted, and in clusters.  And this is the first time that that's been demonstrated.

Chris -   So let's piece this back together then.  You get these blood samples from people who have had a heart attack and it looks like these cells were probably in circulation prior to the time they had the heart attack.  There could therefore be a warning sign that something is about to happen.  How do you think they got from the diseased artery into the test tube that you tested?

Eric -   Well it's pretty straightforward.  You know, there's a crack that's emerging in this inflamed segment of an artery from the surface of the heart and as the crack is growing, before the blood clot forms, these cells are just getting shed from a spot right into the blood and of course, once they're in the blood, just getting a blood tube sample is a window into that process.

Chris -   It's reassuring that you found only a small number of these cells in healthy people.  But of course, I wonder whether you included in that control group, people who might have other risk factors but aren't having a heart attack because - is there going to be a gray area where you'll have people who have artery disease but they're not at imminent risk of a heart attack but your test looking for these cells might say they are?

Eric -   Well we haven't seen any gray area yet.  You know, we've extended this in many different patients and we even took the healthy people and we kept having them come back which had never been done before to see how stable the finding is of their absence, or relative absence, of these cells.  So, we tried to drill down on that quite a bit, Chris.

Chris -   The big question must be, well how far in advance can we predict a forthcoming event and therefore intervene meaningfully in these people?

Eric -   Well you know, this was a segue to a much more simple, quick and hopefully quite inexpensive test.  The work that we did was too laborious, trying to isolate all these cells, but now that we've done genomics of these cells, and have a much easier signal to work with, we will take that to the emergency room setting to validate a practical test.

Chris -   Wouldn't it be more meaningful to take a large group of individuals who are at risk and then just bring them back into the clinic on a weekly basis and get blood from them, and then marry up those blood samples with the ones who do then over time going to have a heart attack?

Eric -   Yeah, that wouldn't work too well because you'd have to do that in thousands of people every week to get the few that are going to have a heart attack.  It's not so easy to find people who are about to have a heart attack using that route.  But if you go to the emergency room setting where people - for example in the US, there's 3 million people coming in with chest pain or tightness, or pressure, thinking that you might have a heart attack each year - and out of those, so many have no damage to their heart muscle but in fact are the exact kind of people that we're trying to identify because they are having this precursor event.  So that's a much better way than just taking people who are perfectly stable with no symptoms, having to come back to a clinic every week.  But, you're bringing up another point.  This is a one off test.  In order to have the blood under surveillance all the time, in high risk individuals like you were outlining, that will require a sensor embedded in the blood which we're working on as well that would communicate to one's Smartphone to give a ring tone that a heart attack is incubated.

Chris -   I'm also getting at the point that we're coming up with all of these coronary risk profiles and we're saying to people, "You have a 10 % risk or you have a 15 % risk" and we're using this to inform what drugs we put people on.  If we've got an even more acute measure, these are the individuals who are not just at risk but these are really elaborating a heart attack that's in evolution then they could actually be stratified for even more intense treatment and a heart attack could be prevented in those individuals.

Eric -   You've nailed it.  The point is, is that once you know the heart attack underlying process is ongoing, it's a heart attack waiting to happen, then the main thing is to prevent the blood clot.  If we prevent the blood clot, then the chance of preventing the event is exceptionally high and naturally it becomes a new goal.

Ben -   Eric Topol from the Scripps Translational Institute in La Jolla, California.