A few weeks back I wrote an article about the drug class known as “PCSK9 inhibitors”, and how they seem to lower the risk of a heart attack, and how there is even preliminary evidence to support the notion that they prolong life. The article wasn’t particularly popular among some readers of this blog, presumably because it was construed as supporting the LDL hypothesis (i.e. the commonly held notion that it’s a cholesterol transport molecule called “Low Density Lipoprotein”, or “LDL” for short, that causes heart disease).
For those who are unaware, the LDL hypothesis basically says that heart disease happens because LDL somewhow ends up in the arterial wall, after which it is oxidized, which starts an inflammatory reaction that gradually leads to the hardering of arteries and eventually to bad things like heart attacks and strokes.
As I’ve written about before on this blog, the LDL hypothesis is bunk. There is by now a wealth of evidence showing that LDL has little to do with heart disease, such as this systematic review from BMJ Evidence Based Medicine, which showed that there is no correlation whatsoever between the amount of LDL lowering induced by statins and other LDL lowering drugs, and the benefit seen on cardiovascular disease risk (if indeed any benefit is seen – it often isn’t).
Dr Malcolm Kendrick, the person I primarily turn to when I want to understand anything to do with heart disease, has just come out with an excellent new book, in which he goes in to significant detail about why the LDL hypothesis is wrong, and what the mass of evidence shows actually causes heart disease. The book is well worth a read for anyone with an interest in understanding heart disease (which I assume is everyone who follows this blog).
The ultra-short elevator pitch version of what he argues in the book is that heart disease is what happens when damage to the arterial wall occurs at a faster rate than repair can happen. That’s why everything from sickle cell disease to diabetes to high blood pressure to smoking to rheumatoid arthritis to cortisone treatment to the cancer drug Avastin increases the risk of cardiovascular disease – they all either increase the speed at which the arterial wall gets damaged or slow down its repair. It’s why heart disease (more correctly called “cardiovascular disease”) only affects arteries (which are high pressure systems) and not veins (which are low pressure systems), and why atherosclerosis (the hardening of the arteries that characterizes heart disease) primarily happens at locations where blood flow is extra turbulent, such as at bifurcations.
This alternative to the LDL hypothesis is known as the “thrombogenic hypothesis” of heart disease. It’s actually been around for a long time, first having been proposed by German pathologist Carl von Rokitansky in the 19th century. Von Rokitansky noted that atherosclerotic plaques bear a remarkable similarity to blood clots when analyzed in a microscope, and proposed that they were in fact blood clots in various stages of repair.
Unfortunately, at the time, von Rokitansky wasn’t able to explain how blood clots ended up inside the artery wall, and so the hypothesis floundered for a century and a half (which is a little bit ironic when you consider that no-one knows how LDL ends up inside the artery wall either, yet that hasn’t hindered the LDL hypothesis from becoming the dominant explanation for how heart disease happens). We now know the mechanism by which this happens: cells formed in the bone marrow, known as “endothelial progenitor cells”, circulate in the blood stream and form a new layer of endothelium on top of any clots that form on the artery wall after damage – thus the clot is incorporated in to the arterial wall.
In spite of the fact that probably at least 99% of cardiologists still believe in the LDL hypothesis, the thrombogenic hypothesis is actually supported far better by all the available evidence. While the LDL hypothesis cannot explain why any of the risk factors listed above increases the risk of heart disease, the thrombogenic hypothesis easily explains all of them.
One of the things Malcolm Kendrick mentions in his new book is the molecule Lipoprotein(a), a.k.a. Lp(a). It has been found to correlate far better with risk of heart disease than LDL does. Interestingly, it is in fact identical to LDL, with one small difference – it has an extra protein, called Apo(a), bound to its surface. This gives it a function that is quite different from LDL. As mentioned, LDL transports cholesterol in the blood stream. What Lp(a) does instead is to stabilize blood clots. Why does this matter?
Because whenever the arterial wall is damaged, a blood clot forms to prevent you bleeding out. The arterial wall then, as mentioned, rebuilds itself on top of the blood clot, and the clot is gradually degraded and absorbed by the body. People with high levels of Lp(a) have more stable blood clots, which means that their blood clots are absorbed more slowly. In other words, arterial wall repair happens more slowly, so people with high Lp(a) have more atherosclerosis and therefore more heart disease.
What does all of this have to do with PCSK9 inhibitors?
PCSK9 inhibitors work by increasing the expression of the LDL receptor on the surface of liver cells. The traditional LDL hypothesis explanation for how PCSK9 inhibitors reduce heart disease is that the increased expression of the LDL receptor results in increased uptake of LDL from the blood stream and therefore lower levels of LDL in the body. There is, however, a tantalizing alternate explanation for how PCSK9 inhibitors reduce heart disease, that fits perfectly with the thrombogenic hypothesis.
As mentioned, LDL and Lp(a) are virtually identical. And the LDL receptor is not able to differentiate between them, so an increased expression of LDL receptors will result not just in a decrease in LDL in the bloodstream, but also in a decrease in Lp(a). That is why the PCSK9 inhibitor evolocumab has been found to decrease Lp(a) by around 30%. It’s likely also why there is any correlation at all between LDL and heart disease – the LDL receptor is responsible for removing both LDL and Lp(a). But it’s the Lp(a) that’s contributing to heart disease, not the LDL. It’s a classic case of correlation being mistaken for causation.
So, that’s just a few thoughts on how I think PCSK9 inhibitors work, and why this doesn’t imply support for the LDL hypothesis. For a more in-depth discussion of the thrombogenic hypothesis and the evidence in favour of it, as well as a deep-dive in to why the LDL hypothesis is utter nonsense, I strongly recommend Malcolm Kendrick’s new book “The clot thickens”.