We didn't know that this was possible. We'd hypothesised about it, but to actually see it in a clinical trial gives people so much hope that we truly are ending heart disease.
For a long time, the school of thought was that if arterial plaque started developing in your blood vessels, that was a one-way path. We could hopefully slow it down, possibly even stop it, and stabilise that plaque by having some calcium come down onto it, so that it doesn't rupture.
But we didn't know that we could significantly reduce that existing plaque. Yes, there was some evidence of marginal benefits, but drastic reductions were more speculative [1]. And even if it was possible, we didn't necessarily know how to get there.
Well, now a new study from Denmark has shown that we can remove plaque. Not just stop new plaque from forming, but actually take away plaque that's already in our blood vessels. And it takes a specific strategy [1].
Table of Contents
- Plaque, Even When You Do Everything Right
- The Calcium Paradox
- What We Already Knew
- The FLOWPROMOTE Trial
- How Far Can We Trust It?
- What I Do in Clinic
- If a Statin Isn't an Option
- Why I'd Rather Never Need It
- References
Plaque, Even When You Do Everything Right
Why does plaque develop in our blood vessels? You'll know this already: it's due to things like high blood sugar levels, not being insulin sensitive, high inflammation and high blood pressure. All of those pour fuel on the fire. They speed up plaque formation, and that plaque can then rupture and cause a heart attack.
Now, one of the concerning things is that even if we control those risk factors, we can still see plaque developing. In the PESA study, researchers took middle-aged bank employees in Spain who didn't smoke and had normal blood pressure, blood sugar and cholesterol. They scanned the arteries in their neck, belly and legs, and checked their heart arteries for calcium. Half of them already had plaque somewhere. Even among the ones with ideal numbers, about a third did. And the higher their LDL, the more likely they were to have it: about 1 in 10 people with an LDL of 60 to 70, rising to nearly 2 in 3 at 150 to 160 [2].
Because of PESA and many other lines of evidence, including Mendelian randomisation studies, where people are born with genes that lower their LDL for life, and clinical trials where we actively lower LDL cholesterol and ApoB and see lower rates of heart disease, we can say with confidence that LDL cholesterol and ApoB, a count of the particles that carry cholesterol into the artery wall, are causal risk factors for heart disease [3, 4]. That's not to say it's the only factor. We still need to look after our blood pressure and insulin sensitivity, as well as ApoB.
The Calcium Paradox
How do we lower ApoB? We start with diet, but that can only go so far. We've entered a world where we can enhance our body beyond just diet, exercise and sleep. The first enhancement we can reach for is statin medications, which reduce the amount of cholesterol the liver produces.
But is it safe to push blood cholesterol that low? On this point, nearly every cell in our body can produce its own cholesterol. The liver and the gut make most of it, and send the extra around the body in case other tissues need a temporary top-up [5]. The amount of cholesterol in our blood is a small fraction of the body's total cholesterol, so lowering the cholesterol in your blood doesn't starve your cells of it. But it is the cholesterol in the blood that can wreak havoc with our vessels.
But surely lowering cholesterol will destroy natural hormone production and our memory?
A trial of 1,200 people on a very strong cholesterol-lowering injection found no effect on memory [6]. Another found no change in sex or adrenal hormones, even with LDL pushed extremely low [7].
But one of the interesting findings when we use statins is that, in the short term, they seem to increase the amount of calcium in the blood vessels. That's the body stabilising the plaque, laying calcium on top of it so it's less likely to rupture. So if you had a coronary artery calcium scan, your score could actually look worse, even though your chance of a heart attack had gone down [8, 9].
And that brings us back to the question. Can we actually start to remove some of that plaque, not just stabilise it?
What We Already Knew
One of the first big trials to show it came in 2006. About 350 people with heart disease took rosuvastatin 40 milligrams for two years, and their LDL fell to about 61. An ultrasound probe, threaded inside one of their heart arteries, showed the plaque had shrunk. The typical person lost about 7 percent of it [10].
Over the next decade, other trials found the same. Plaque shrank in roughly six or seven in ten people on intensive treatment, but only by a few percent. And those studies looked at one stretch of one artery, often in people who were already on a statin before they started [11, 12, 13].
Meanwhile the trials looking at heart attacks kept telling us the same thing: lower is better. Earlier this year a trial called Ez-PAVE randomised about 3,000 people with heart disease to an LDL target under 55 or under 70. To get LDL that low, a statin on its own often isn't enough, so doctors added a second enhancement: ezetimibe. Ezetimibe tells the gut not to absorb as much cholesterol. Because it works on a different pathway to a statin, the two work really well together to lower LDL cholesterol and ApoB. Over three years, 6.6 percent of the lower-target group had a major heart or circulation event, against 9.7 percent. That counted deaths from heart disease, heart attacks, strokes, stents or bypasses, and hospital stays for unstable chest pain. It was partly funded by a company that sells a statin-and-ezetimibe pill, and everyone knew which target they were aiming for [14].
So if a lower LDL means fewer heart attacks, does it also mean more plaque shrinking?
The FLOWPROMOTE Trial
The new Danish trial, called FLOWPROMOTE, gives us a look at exactly that. They took 104 people with newly diagnosed narrowing in their heart arteries, who weren't on long-term statin treatment, and randomly split them in two. Half got atorvastatin 40 milligrams, a standard high dose. The other half got a stronger combination: rosuvastatin 40 milligrams, the top dose, plus ezetimibe 10 milligrams [1].
Then they scanned everyone's heart arteries with a CT angiogram at the start, at nine months and at eighteen months, using the same scanner settings each time. The person measuring the plaque didn't know which treatment anyone was on, or which scan was which [1].
LDL fell from about 120 to 70 milligrams per decilitre on atorvastatin, and from about 127 to 42 on the combination. In millimoles, from about 3.1 to 1.8, and 3.3 to 1.1 [1].
And the plaque shrank in both groups. On atorvastatin, total plaque fell by about 14 percent. On the combination, it fell by 21 percent. And it happened fast: on the combination, much of that fall had already shown up by the nine-month scan [1].
But the crucial part is what the plaque is made of. The calcified part went the other way: it grew by about half, in both groups, and how low LDL went didn't seem to change that. That's most likely the stabilising we just talked about, calcium being laid down over what's left. What matters more is the soft, non-calcified plaque. That's where the fatty core sits, and in a large Scottish study, the fattiest soft plaque on a CT scan was the strongest predictor of a heart attack [15]. And the soft plaque fell even further: by 20 percent on atorvastatin, and 28 percent on the combination [1].
And the lower someone's LDL went, the more their plaque tended to shrink [1]. As the lead author, Martin Bødtker Mortensen, put it: "The lower the LDL-C, the greater the plaque regression." [16]
To put that in perspective, the 2006 ultrasound trial took two years to shrink plaque by about 7 percent, in one stretch of one artery [10]. Here, the combination took off about 20 percent, across the whole heart, in a year and a half. Even the atorvastatin group, at an LDL of 70, lost about twice what that 2006 trial did at 61. So the gap isn't only about LDL. The scans were different, and these people were new to treatment. It's a rough comparison, but it's a big step.
How Far Can We Trust It?
But can we trust this trial? What's the whole picture, and who funded it?
It's small. 104 people. And plaque wasn't the trial's main question. The main question was whether blood flow through the narrowed arteries would improve more on the stronger treatment. It didn't. Flow got a little better in both groups, with no real difference between them. So the authors themselves say the plaque results should be treated as exploratory [17].
Everyone knew which tablets they were taking, so it wasn't a true double-blind study. The person measuring the scans didn't know, though [1]. And these measurements are rough. When the same team scanned the same patients twice, an hour apart, one person's plaque reading could differ by up to about 90 cubic millimetres, out of a typical total of around 260 [18]. Across a whole group that noise mostly averages out, and the gap between the two treatments still held. But for any one person, a single reading going up or down doesn't tell you much.
And the gap between the two treatments depends on how you measure it. The 21 versus 14 percent is the total amount of plaque. Measured instead as the share of the artery wall taken up by plaque, which is how the older ultrasound trials did it, plaque fell from 31 to 29 percent on atorvastatin and from 30 to 26 percent on the combination. That difference between the two groups wasn't statistically significant [1].
So I'm confident plaque shrank in both groups. That's the really exciting part. I'm less sure exactly how much extra the combination added. But it makes me glad that I myself am on combination therapy of rosuvastatin 5 milligrams and ezetimibe 10 milligrams.
And these were people starting treatment for the first time. The authors say we can't assume the same results for people who've already been on a statin for years [1].
So who funded it? Is this just big pharma trying to sell drugs? The trial was paid for by a Danish charitable foundation, the Novo Nordisk Foundation, which had no role in the analysis [1]. And these drugs are off patent and extremely cheap. At costplusdrugs.com, rosuvastatin 5 milligrams and ezetimibe 10 milligrams each cost less than $10 a month (checked 6 November 2026: $5.45 and $6.38 for 30 tablets, before shipping) [19, 20].
One author earns royalties from the hospital that makes the plaque-measuring software, and the lead author has had speaking fees from drug companies, including Amgen, which makes a PCSK9 inhibitor [1, 17].
What about side effects? About one in five people had their treatment changed during the trial, mostly for muscle aches or raised liver enzymes. Only two stopped their statin altogether [1].
So here's my take. We've known for twenty years that plaque can shrink a little. This trial suggests how far it can go, across all of the heart's main arteries, when LDL gets very low.
And the lesson I take from this trial is about how low LDL goes. For people who already have heart disease, the heart-attack trials back getting it under 55, and in this trial the group that got to around 42 did best [14, 1]. So how do you get there?
What I Do in Clinic
Here's what I do in the clinic. I start people on rosuvastatin 5 milligrams. If that's not enough, then we add on ezetimibe 10 milligrams. And if people are statin intolerant, which about 1 to 2 percent of people are, that's when we switch to PCSK9 inhibitors [21].
That's a much lower statin dose than the trial used, so it's fair to ask why. Rosuvastatin 5 milligrams on its own lowers LDL by about 45 percent. Going all the way up to 40 milligrams only gets you to about 63 percent [23]. And in a trial of 3,780 people with heart disease, a moderate dose of rosuvastatin plus ezetimibe prevented heart attacks and strokes as well as a higher dose on its own, and fewer people had to stop or cut back because of side effects: 4.8 percent against 8.2. That trial was funded by a company that sells the combination pill [22].
If a Statin Isn't an Option
But hang on, what are PCSK9 inhibitors? It's a class of medication that helps your liver pull more LDL cholesterol out of the blood. They're very, very well tolerated, and they don't cause the muscle side effects that statins cause in 1 to 2 percent of people [24, 25, 21].
Why I'd Rather Never Need It
So what about calcium scans? Why not wait until a scan shows plaque has developed before starting treatment? To me, that misses the boat. We want to truly practise preventive care, which means intervening before problems start. Where I do use a scan in clinic is when someone isn't sure they need treatment. Seeing calcium in their own arteries can be what convinces them. And that's in line with the latest American guideline [8].
But the study overall gives hope for people that do have plaque in their blood vessels, that through the correct treatments, we can actually reduce that plaque and make it regress.
Plaque in your arteries isn't a one-way street. And if you don't have it yet, start early and aim low. Do that, and I really think we can end heart disease.
Other studies
A few more findings from the research behind this article, which didn't make it into the video:
- The FLOWPROMOTE authors write that "it remains unclear whether the differences observed in plaque regression translate into a clinically meaningful reduction in cardiovascular events". Over the 18 months there were three non-emergency stent procedures and no heart attacks or deaths, far too few events to say anything about heart attacks either way [1].
- "Statin-naive" isn't quite absolute: about a quarter of the participants (26%) had started a statin in the three months before their first scan, with a median of 25 days. People on long-term statins were excluded [1].
- The share of people whose plaque shrank was similar to earlier trials such as SATURN (71%) and GLAGOV (64%); the authors put FLOWPROMOTE's slightly higher rates down to its patients being new to treatment [1, 11, 13].
References
1. https://jamanetwork.com/journals/jamacardiology/fullarticle/10.1001/jamacardio.2026.4788
2. https://www.jacc.org/doi/10.1016/j.jacc.2017.10.024
3. https://academic.oup.com/eurheartj/article/38/32/2459/3745109
4. https://jamanetwork.com/journals/jama/fullarticle/2556078
5. https://www.mdpi.com/2072-6643/10/6/760
6. https://www.nejm.org/doi/full/10.1056/NEJMoa1701131
7. https://doi.org/10.1161/CIRCRESAHA.115.307071
8. https://www.jacc.org/doi/10.1016/j.jacc.2025.11.016
9. https://www.jacc.org/doi/10.1016/j.jacc.2015.01.036
10. https://pubmed.ncbi.nlm.nih.gov/16533939/
11. https://doi.org/10.1056/NEJMoa1110874
12. https://doi.org/10.1016/j.jacc.2015.05.065
13. https://doi.org/10.1001/jama.2016.16951
14. https://www.nejm.org/doi/full/10.1056/NEJMoa2600283
15. https://pmc.ncbi.nlm.nih.gov/articles/PMC7195857/
16. https://x.com/MaBMortensen
17. https://www.jacc.org/doi/10.1016/j.jacc.2025.12.067
18. https://doi.org/10.1016/j.jcct.2024.09.009
19. https://costplusdrugs.com/medications/rosuvastatin-5mg-tablet/
20. https://costplusdrugs.com/medications/ezetimibe-10mg-tablet/
21. https://doi.org/10.1016/j.jacl.2022.09.001
22. https://doi.org/10.1016/S0140-6736(22)00916-3
23. https://www.accessdata.fda.gov/drugsatfda_docs/label/2012/021366s025s027lbl.pdf
