EVAPORATE: Icosapent Ethyl May Slow Atherosclerosis Progression

(UPDATED) New, 18-month data from the EVAPORATE study suggest that part of the mechanism behind the beneficial effects of icosapent ethyl (Vascepa; Amarin) may be that the drug slows the progression of atherosclerosis.

The findings, seen in patients with high triglycerides similar to those included in the REDUCE-IT trial, confirmed what was seen in an interim analysis at 9 months with this agent, a pharmaceutical-grade omega-3 fatty acid formulation. Specifically, compared to those who received placebo, patients who received the study drug saw significantly less progression in fibro-fatty, fibrous, low-attenuation, total noncalcified, and total plaque volume on multidetector CT.

“Now we can say with confidence that at least one mechanism of benefit of icosapent ethyl is a reduction in the atherosclerotic burden,” said Matthew Budoff, MD (Harbor-UCLA Medical Center, Torrance, CA), who presented the results at the virtual European Society of Cardiology 2020 Congress. “Obviously, exactly how it slows atherosclerosis will require yet further research,” he told TCTMD.

The results were also simultaneously published in the European Heart Journal.

Since its release on Saturday, some critics have raised concerns about the differences in baseline characteristics between the study arms and whether assessments were blinded during follow-up.

Icosapent Ethyl Seems to Block Plaque

For EVAPORATE, researchers used similar entry criteria to REDUCE-it to enroll 80 patients with elevated triglyceride levels (135 to 499 mg/dL) who were on stable statin therapy, had LDL-cholesterol levels greater than 40 mg/dL but no higher than 115 mg/dL, and had at least one coronary stenosis of 20% or more on CT. Patients were randomized to receive icosapent ethyl 4 g/day or placebo for 18 months and were assessed by serial multidetector CT at baseline, 9 months, and 18 months.

Among the 68 patients who completed 18-month follow-up, mean age was 57.4 years, 54.4% were men, and mean baseline triglyceride levels were 259.1 mg/dL. The authors state that there were no significant differences in baseline plaque volumes or characteristics between the study groups, although this is the aspect of the trial now being debated on Twitter. The most common plaque type was fibrous in both the study (74.7%) and placebo groups (57.9%). Low-attenuation plaque (the primary endpoint) was least common, making up only 5.1% and 6.5% of baseline total plaque, respectively.

At 18 months, significant drops in plaque progression were seen for all plaque types in the study arm compared to placebo except for dense calcium. Results were adjusted for age, gender, diabetes, hypertension, and triglyceride levels at baseline.

Percent Change in Plaque Volume at 18 Months

This kind of CT angiography-based analysis “opens the door for looking at mechanistic studies more,” Budoff said. “We're doing a study now looking at semaglutide [Ozempic; Novo Nordisk] and its effect on atherosclerosis progression to try to see if the benefits of GLP-1 receptor agonists are atherosclerotic or not. I know other investigators are looking at other therapies, including statins, so I think that this will become a new mode to evaluate plaque progression noninvasively, transitioning from the old days of intravascular ultrasound, which I think now is largely outdated due to its invasive nature and high cost.”

The next step will be to “look at a different cohort in a larger study to see if we can both replicate and refine some of the results that we found in EVAPORATE,” Budoff said. “But I think that further studies would look more at the cellular level of atherosclerosis. There have been some benefits with oxidation and some benefits with inflammation, and I believe both of them may be contributory here.”

Criticism Over Lack of Normalization

Following the presentation of the EVAPORATE data, Twitter discussion over the study’s methodology turned acrimonious, questioning how baseline differences in plaque volumes were “normalized,” as well as whether physicians reviewing CT data had been aware of treatment allocation.  

TCTMD reached out to Hector M. Garcia-Garcia, MD, PhD (MedStar Washington Hospital Center, Washington, DC), who serves as director of his institution’s core lab, to get some clarity around the concerns. Garcia-Garcia said this lack of normalization—a statistical process standard for all intravascular imaging studies of this kind—makes the results of EVAPORATE “uninterpretable.”

“I don't think there is a methodological issue,” he clarified. “I'm not even suggesting that there is an issue in the execution of the trial. I think it's simply a statistical approach which is very standard for intravascular imaging studies that was not applied on these data, and that if you do that your conclusions may not be as they are presenting them. What I'm afraid is that maybe the conclusions are not going to hold if you do the normalization or the adjustment that we always do for IVUS.”

In response, Budoff argued that due to the intrinsic differences between coronary CTA and IVUS, with the former measuring the entire coronary tree and the latter only imaging limited segments, the EVAPORATE data were properly adjusted. “The shortcomings of intravascular ultrasound (only having limited segments of the coronary tree to evaluate) requires normalization,” he said in an email. “We use coronary CTA, and the methods we apply are quite standard. We have published literally dozens of studies of plaque progression with CCTA over the last 10 years and the methods we employ with CCTA do not involve normalization.”

The researchers did perform an unpublished secondary analysis using normalization, he added, and the results did not change. “Luckily, Dr. Garcia-Garcia’s concerns should be completely assuaged, as the results held up with normalization.”