Lessons from the biology of CETP inhibition

A very warm welcome. My name is John Kastelein and I'm the chief science officer of New Amsterdam Pharma. We are developing an inhibitor of a protein called cholesterol ester transfer protein that I will abbreviate from now on with CETP. CETP inhibition has known a long and rocky road and I would like to explain to you why this new CETP inhibitor is going to make up for the mistakes made in the past. These are my disclosures, and as I said, I'm the chief science officer of New Amsterdam Pharma and a consultant to many companies in the cardio metabolic space.

Inhibiting CETP raises HDL cholesterol. And these inhibitors were initially developed based on the premise that increasing HDL would prevent coronary artery disease. We now know in 2022 that that was wrong. In fact, raising HDL cholesterol does not prevent heart attacks or strokes. But these investigations were so large that everyone in those days believed that raising HDL would actually help prevent coronary disease.

It was not appreciated at the time, that if you have a mutation in the CETP gene, a loss-of-function mutation, that you not only have increased HDL, but you also have lower LDL, lower non-HDL and lower apoB. And we now understand that actually the raising HDL part doesn't do anything for coronary artery disease but in fact, it is the part of CETP inhibition that lowers LDL cholesterol and lowers apoB, that actually makes it work.

Now, what do you need in order to move a drug into phase one, two and three? You need genetic validation. And what is very interesting is that we have genetic validation in the left-hand panel since 2008, where Thomson in the JAMA described that if you're born with a loss of function allele in CETP, that means that you have lower CETP activity, that you have less coronary disease.

And in fact, it was Brian Ference in 2017, who showed that the more potent your CETP inhibition is, the more you are protected against coronary disease. So, between 2008 and 2017, 11 years have gone by and during those 11 years, multiple studies have reconfirmed that if you're born with low CETP, you have much less risk for a heart attack and a stroke.

And as you can appreciate from this slide, the CETP inhibitors, in yellow, lie on the regression line between non-HDL and heart disease. Meaning that their LDL lowering is as much worth, as LDL lowering conferred by statins, ezetimibe or PCSK9 monoclonals.

And now it is very apparent that the reduction in atherogenic lipoproteins drive the protection by CETP inhibition, not the changes in HDL cholesterol. Now, it immediately raises the question, does it matter how LDL is lowered?

And what you see here on this screen is by Brian Ference work, with a long list of different genes. And as you can see, the odds ratio reduction is basically identical. So the answer is no, it does not matter how LDL is lowered. If it is lowered you're always paid back by a reduction in MACE. And that's a very important anathema that we are going to test in multiple large-scale trials.

The biology of CETP is very interesting. When you inhibit the transfer of cholesterol from HDL to LDL, the consequence of that is that there is less cholesterol in the liver. Therefore, the LDL receptor is up-regulated and LDL and apoB containing lipoproteins, are removed. But the consequence is also that there's more pre-beta HDL produced and that pre-beta HDL is able to efflux cholesterol out of beta cells, actually in the pancreas, contributing to less type 2 diabetes and possibly also in the brain. Now, this is something that is a little bit outside of the scope, but we think that there's good science to link CETP inhibition and Alzheimer's disease.

Now, what have we learned from previous CETP inhibitor outcome trials? The first thing we've learned is that torcetrapib was the wrong choice. In fact, as you can see here, torcetrapib increased blood pressure and not a little bit. Actually, in the large trial, there was an increase in systolic blood pressure of more than 15 millimeters of mercury, in almost 10% of the atorvastatin group, increase in aldosterone, decrease in potassium. And it proved that all of these negative things actually had nothing to do with CETP, but they were the consequence of this molecule, which is now called an off-target side effect.

Then the next drug came, dalcetrapib. Dalcetrapib was a very safe drug but it only raised HDL. It didn't lower LDL. And so what you can see here is the HDL increase in the Dal-Outcomes trial, about 35 to 40%. And here, you can see the LDL changes. And in fact, there were no LDL changes. We now know that if you do not confer a change in LDL, that this is what you can expect, no effect on MACE. So, that was the second problem. The second CETP inhibitor was very safe, had no side effects but didn't lower LDL cholesterol and therefore had no effect on MACE. So, it's very obvious, and the Dal-Outcomes trial contributed to the insight, that increasing HDL cholesterol does not lower CHD risk, which in itself is, of course, very important.

Then the third trial was with a drug called evacetrapib. Evacetrapib had an LDL lowering that was estimated by the sponsor to be around 31 to 32%. Unfortunately, that LDL cholesterol lowering was assessed with the direct LDL assay that totally overestimates it. We now think, actually on the basis of post-hoc analysis, that evacetrapib only lowered LDL by about 20 percent.

And this trial actually, as you can see here, this is the assay where there was a 31% reduction of LDL cholesterol overestimation. And if you overestimate the LDL lowering of your drug, you're going to underestimate what you need in terms of numbers of patients and duration of trial to prove it. And this trial was stopped prematurely for lack of efficacy but the trial was basically stopped after two years. Now, if you only lower LDL by about 20 percent, you are as good as ezetimibe.

And we have an outcome trial with ezetimibe called IMPROVE-IT. And there the curves only diverge basically after three years.

So, I think we can safely conclude that the first drug had off-target effects, the second drug didn't lower LDL, and the third drug actually had the trial too small and too short to appreciate the effect.