Clinical Research · us
AI quantifies coronary plaque; randomized trial shows greater diagnostic certainty and increased lipid-lowering treatment
The PARAMOUNT trial suggests that incorporating coronary plaque information into chest pain assessment may change physicians’ treatment decisions. However, the benefits so far mainly involve diagnostic certainty and lipid management; reductions in heart attacks or deaths cannot yet be inferred.
Identifying the cause of chest pain after testing and starting preventive treatment early are important tests of imaging technology in clinical practice. On November 7, Cleerly announced preliminary results from the randomized PARAMOUNT trial: care guided by artificial intelligence quantification of coronary plaque provided greater diagnostic clarity and led to more patients receiving lipid-lowering treatment than usual care guided by functional stress testing. However, these results have not demonstrated that patients consequently experience fewer heart attacks or deaths.
The approach, called AI quantitative coronary computed tomography analysis (AI-QCT), quantifies plaque using computed tomography images and incorporates a plaque-based assessment of myocardial ischemia to help physicians determine subsequent management. The trial compared two complete care strategies, so even favorable results cannot be attributed entirely to the AI algorithm alone.
PARAMOUNT is a multicenter, open-label trial that randomized 530 people at 21 trial sites. Participants were symptomatic outpatients with suspected coronary artery disease, no prior history of coronary artery disease, and low-density lipoprotein cholesterol (LDL-C) of at least 100 mg/dL. This preliminary analysis included 523 participants whose data had been locked; 57% were women, and the median age was 57 years. The applicability of the results should initially be considered within this patient population.
According to Cleerly’s official website and the same company press release distributed through Business Wire and syndicated by Observer-Reporter, the trial met its prespecified hierarchical composite primary endpoint at 180 days, with a win ratio of 2.00, a 95% confidence interval of 1.37 to 2.56, and a p-value of less than 0.0001. This measure compares multiple outcomes in a predetermined order and cannot be interpreted as a halving of cardiovascular risk. The company stated that diagnostic certainty about whether chest pain was caused by coronary artery disease was the largest contributor to the difference.
Changes in treatment decisions were more readily apparent: lipid-lowering medication use in the AI-QCT-guided group rose from 31% at baseline to 75% after testing, while use in the usual-care group rose from 32% to 47%. The company also reported better LDL-C reduction in the AI-QCT group, but the press release did not provide the magnitude of the reduction. These data support the possibility that plaque information may prompt physicians to intensify preventive treatment, but they cannot yet establish how much long-term health outcomes improve.
Two further limitations should be distinguished when interpreting this trial. First, under the open-label design, physicians and patients knew which strategy was being used, which could influence diagnostic confidence and prescribing behavior. Second, the syndicated content originated from a company press release and does not constitute independent validation by another research team. Full methods, details of the individual endpoints, and safety data still await a formal paper.
The results were presented at the 2026 American Heart Association Scientific Sessions but have not yet undergone peer review. The company expects full publication in early 2027. For now, the more cautious conclusion is that plaque-guided care can change post-test assessments and medication use. Whether these changes can further reduce heart attacks, extend lives, or lead to additional testing or treatment burdens still requires more complete, long-term evidence.