Study Data Show AI-Powered Coronary CT Angiography Improves Risk Assessment

An analysis of registry data indicates that Cleerly’s artificial intelligence (AI)-based quantitative Coronary CT Angiography (AI-QCT) significantly improves risk discrimination compared to models combining other clinical predictors such as CAD-RADS 2.0, coronary artery calcium score (CAC), and DUKE CAD index for predicting Major Adverse Cardiovascular Events (MACE).

Clinical evidence supporting this conclusion came from the CONFIRM2 registry data, “Prognostic Value of AI-based Quantitative Coronary CT Evaluation Versus Standard Qualitative Evaluation,” as well as five additional research abstracts.

“This research shows we can now identify patients at risk for MACE with greater precision than ever before,” said principal investigator Ibrahim Danad, M.D., Ph.D., of Radboud University Medical Center. 

CONFIRM2¹ is a multicenter, international, observational cohort targeting 30,000 patients, with more than 8,000 participants currently enrolled, that aims to perform comprehensive quantification of CT angiography findings and relate them to clinical variables and cardiovascular clinical outcomes. Investigators analyzed a symptomatic subset of the cohort with available manual CAD-RADS and CAC scores. AI-QCT was applied to quantify plaque burden, morphology, composition, and luminal obstruction across the coronary tree, and predictive modeling was performed using quantitative diameter stenosis and non-calcified plaque volume.

“Plaque burden quantification by AI-QCT provides critical insights that standard CT evaluation will miss, especially in patients with mild to moderate disease who represent a significant portion of our patient population,” stated principal investigator Alexander van Rosendael, M.D., Ph.D., from Leiden University Medical Center.

Key findings include:

• AI-QCT provided incremental prognostic information over CAD-RADS, coronary artery calcium scoring, and the Duke Index 
• The addition of AI-QCT significantly improved risk stratification compared to the CAD-RADS (AUC 0.81 vs 0.79, p<0.001), the CAC score (AUC 0.79 vs 0.70, p<0.001), and the DUKE Index (0.81 vs 0.76, p<0.001)
• The greatest benefit of the addition of AI-QCT was observed in patients without severe stenosis

Five additional abstracts covering advances in stenosis quantification, diagnostic performance across patient populations, and sex-based differences in plaque assessment include:

• Optimizing Stenosis Quantification by a Novel Interpolation-Based Approach: Diagnostic Performance Data from CREDENCE and PACIFIC-1
• Diagnostic Performance of a Novel AI Algorithm for Determining Coronary Ischemia (AI-QCT ISCHEMIA) According to Patient Comorbidities
• Pooled-Analysis Comparing Accuracy of Cleerly ISCHEMIA vs. FFRCT
• Sex Based Differences in Association with Plaque Burden and Ischemia — a CREDENCE Trial Substudy
• Differential Atherosclerotic Burden and Ischemic Risk Across Major Coronary Arteries: a Comparative Analysis Using Quantitative Coronary Computed Tomography Angiography; a CREDENCE Trial Substudy

Cleerly iaims to eliminate heart attacks by creating a new standard of heart disease care. Through its U.S. Food and Drug Administration-cleared solutions driven by artificial intelligence, Cleerly supports comprehensive phenotyping of coronary artery disease, as determined from advanced noninvasive computed tomography imaging. Cleerly’s approach is grounded in science, based on millions of images from more than 40,000 patients. Led by a clinical and technical team, Cleerly enhances health literacy for all stakeholders in the coronary care pathway.

Reference
¹ Van Rosendael AR, Crabtree T, Bax JJ, Nakanishi R, Mushtaq S, Pontone G, Andreini D, Buechel RR, Gräni C, Feuchtner G, Patel TR, Choi AD, Al-Mallah M, Nabi F, Karlsberg RP, Rochitte CE, Alasnag M, Hamdan A, Cademartiri F, Marques H, Kalra D, German DM, Gupta H, Hadamitzky M, Deaño RC, Khalique O, Knaapen P, Hoffmann U, Earls J, Min JK, Danad I; CONFIRM 2 investigators. Rationale and design of the CONFIRM2 (Quantitative COroNary CT Angiography Evaluation For Evaluation of Clinical Outcomes: An InteRnational, Multicenter Registry) study. J Cardiovasc Comput Tomogr. 2024 Jan-Feb;18(1):11-17. doi: 10.1016/j.jcct.2023.10.004. Epub 2023 Nov 10. PMID: 37951725; PMCID: PMC10923095.