A stronger statin plus ezetimibe shrank coronary plaque more than atorvastatin alone in a small trial
Table of Contents
1.0 Summary
In FLOWPROMOTE, a randomized trial of 104 people with newly diagnosed stable coronary artery disease who had never taken a statin, rosuvastatin 40 mg plus ezetimibe 10 mg reduced total coronary plaque volume by 21.1% over 18 months, against 13.6% with atorvastatin 40 mg alone. LDL cholesterol fell to 42.5 mg/dL with the combination and 69.5 mg/dL with atorvastatin, and lower LDL went with more plaque shrinkage. The results, from secondary analyses planned in advance and published in JAMA Cardiology, measure plaque on CT scans and do not show fewer heart attacks or deaths. A separate after-the-fact (post hoc) analysis of the Australian CAUGHT-CAD trial adds context on what happens to calcium inside plaque during statin treatment. The findings are encouraging for the idea that lower LDL can partly reverse plaque, but they are too small and too indirect to change the usual stepwise approach to treatment.
2.0 Key Points
- FLOWPROMOTE randomized 104 statin-naive patients (mean age 64, 79% male) to atorvastatin 40 mg or rosuvastatin 40 mg plus ezetimibe 10 mg, 52 in each group, for 18 months.
- LDL cholesterol fell from 127.4 to 42.5 mg/dL (65%) with the combination and from 119.7 to 69.5 mg/dL (38%) with atorvastatin alone.
- Total plaque volume fell by 69.8 mm3 with the combination and 39.0 mm3 with atorvastatin (difference 30.9 mm3, 95% CI 4.5 to 57.2; P = .02); noncalcified plaque fell by 82.3 mm3 and 53.5 mm3 (difference 28.9 mm3, 95% CI 4.6 to 53.2; P = .02).
- Calcified plaque volume rose by a similar amount in both groups, and lower achieved LDL cholesterol was linearly associated with greater regression.
- The trial’s results reported here are imaging measures in a small group, and the abstract gives no data on heart attacks, strokes, deaths or side effects.
3.0 Evidence base
Coronary plaque is the buildup of cholesterol and other material in the walls of the heart arteries. Earlier studies, mostly using ultrasound from inside the artery, found modest shrinkage with lipid-lowering treatment in stable coronary disease. FLOWPROMOTE looked at the entire coronary tree in people starting treatment for the first time [1].
The multicenter trial enrolled statin-naive patients with newly diagnosed stable coronary artery disease. Each had a narrowing of more than 50% on coronary CT angiography, a dye-enhanced heart scan, and a CT-derived fractional flow reserve of 0.80 or less, an estimate suggesting that the narrowing limits blood flow. The 104 patients in this analysis had a mean age of 64, and 82 (79%) were male. They were randomized to atorvastatin 40 mg, 52 patients, or rosuvastatin 40 mg plus ezetimibe 10 mg, 52 patients, for 18 months. Scans were repeated at 9 and 18 months [1].
LDL cholesterol fell from 119.7 to 69.5 mg/dL (38%) on atorvastatin and from 127.4 to 42.5 mg/dL (65%) on the combination (P<.001) [1].
At 18 months, plaque shrank in both groups and shrank more with the combination. Total plaque volume fell by 69.8 mm3 (21.1%) against 39.0 mm3 (13.6%), a between-group difference of 30.9 mm3 (95% CI 4.5 to 57.2; P = .02). Noncalcified plaque, the softer portion that has not hardened with calcium, fell by 82.3 mm3 (28.2%) against 53.5 mm3 (20.0%), a difference of 28.9 mm3 (95% CI 4.6 to 53.2; P = .02) [1].
Calcified plaque volume increased about equally in both groups [1]. That explains why noncalcified plaque fell by more than total plaque in each arm.
Total plaque regressed in 47 of 52 patients on the combination (90%) and 38 of 51 on atorvastatin (75%; P = .05). Noncalcified plaque regressed in 51 of 52 (98%) and 39 of 51 (77%; P = .001). Across the trial, lower achieved LDL cholesterol was linearly associated with more regression of total and noncalcified plaque [1].
CAUGHT-CAD offers a second look at the calcium question. That trial randomized 449 statin-naive Australians with a family history of premature coronary disease (mean age 58, 43% women) to a calcium score-informed prevention strategy including a statin, or to usual care. An after-the-fact analysis, reported by the American College of Cardiology from JACC: Cardiovascular Imaging, examined 338 participants with measurable plaque at baseline and three years. After allocation and crossover, 161 took statins and 177 did not, with 840 lesions in all [2].
Beta values below are adjusted estimates of the size of each change. Statin treatment was not associated with a change in overall lesion calcium volume (P = .22). It was associated with regression of noncalcified plaque (beta -9.83; P<.001) and with more intermediate-density calcium (beta 0.89; P = .02) and high-density calcium (beta 0.79; P = .001). Among statin users, an LDL reduction of at least 30% went with more high-density (beta 1.11; P = .03) and very high-density calcium (beta 1.68; P = .04) [2]. The authors describe this shift as plaque passivation, a move from lipid-rich plaque toward a more stable, calcified form. Their four-pattern classification found classic stabilization more often in statin-treated lesions (61.4% against 38.6%) and potential destabilization more often in untreated lesions (64.7% against 35.3%; both P<.01). They note that these patterns have not been validated against clinical outcomes and should not guide individual treatment [2].
4.0 Where it fits in practice
No change to current practice follows from these results. Statins remain the starting point, with ezetimibe added when cholesterol is not at target, as set out in Treatment Pathways. FLOWPROMOTE asks whether starting with the stronger combination would do better, and in 104 patients it did on an imaging measure. That supports the principle that lower LDL cholesterol goes with more plaque regression, but it does not show that starting with two drugs prevents events.
The design also limits what can be said about ezetimibe. The groups differed in both the statin and the add-on, so the extra effect cannot be credited to ezetimibe alone.
These data do not make repeat CT scans to track plaque a routine part of care, and calcium-density analysis is not standardized or available for everyday use [2]. In both studies, calcified plaque held steady or grew while softer plaque regressed [1,2]. That suggests more calcium on a scan during statin treatment need not mean treatment is failing, though this reading is still a hypothesis.
5.0 Open questions
FLOWPROMOTE’s plaque results are secondary end points, planned in advance, from a small trial, and the percentage changes describe scans, not patients’ health. The abstract does not report the trial’s primary end point, side effects, how well patients tolerated treatment, or funding. Participants were statin-naive with narrowings that limited flow, so the results may not apply to people already on a statin or to those with nonobstructive plaque. The abstract also gives the atorvastatin regression rate as both 76% and 75%; the counts (38 of 51) match 75%, which is the figure used here [1].
CAUGHT-CAD’s analysis was done after the fact, and treatment crossover means the comparison of statin users with nonusers was not a randomized one. Its outcomes are also imaging measures, this time over three years. The details here come from an ACC summary of the paper, not the full report [2].
Neither study shows that more plaque regression leads to fewer heart attacks or deaths. That needs outcome trials.
This summary is provided for clinical decision support only and does not replace individualized clinical judgment. Findings are summarized from the cited primary sources; consult those sources before changing practice.
Version History
| Version | Date | Description |
|---|---|---|
| 1.0.0 | 2026-11-01 | Initial release |
References
- Mortensen MB et al. Intensive vs More Intensive Lipid-Lowering Therapy and Plaque Regression in Statin-Naive Patients: Prespecified Analyses From the FLOWPROMOTE Randomized Clinical Trial. JAMA cardiology. 2026-10-30. doi:10.1001/jamacardio.2026.4788. Available at: https://doi.org/10.1001/jamacardio.2026.4788
- CAUGHT-CAD Analysis: Statin Therapy and Plaque Passivation. https://www.acc.org/Latest-in-Cardiology/Journal-Scans/2026/10/27/15/01/CAUGHT-CAD-Analysis.