Prepared by: Dr. Selim Süleyman Sert

Name of the study: Prognostic performance of CCTA derived physiology and plaque characterisation in the CAREER study

Published in Congress: EuroPCR 2025

Link: https://media.pcronline.com/diapos/EuroPCR2025/19491-20250522_0900

Introduction
Coronary computed tomography angiography (CCTA) is widely used as a non-invasive diagnostic tool for coronary artery disease (CAD). Recently, artificial intelligence (AI)-enhanced physiologic metrics such as CT-derived fractional flow reserve (CT-µFR) and plaque characterization from CCTA have emerged as promising predictors of outcomes. However, the optimal use of these parameters to safely defer invasive coronary angiography (ICA) remains unclear.

Objective
The CAREER study aimed to evaluate the prognostic value of CT-µFR and AI-based plaque composition analysis in predicting the need for coronary revascularization, and to determine whether invasive angiography can be safely deferred based on these non-invasive metrics.

Methods

• Population: A total of 307 patients were enrolled in the CAREER trial.
• CT-µFR Analysis: 267 underwent core-lab analysis; 254 patients (723 vessels) were included in the final analysis.
• Follow-up: Patients were followed for up to 4 years. The primary endpoint was a composite of revascularization within 3 months post-CCTA and adverse outcomes (cardiac death, non-procedural myocardial infarction (MI), or any revascularization) during follow-up.
• Imaging & AI Analysis: CT-µFR and plaque features (lipidic volume, calcific burden, plaque burden, etc.) were automatically assessed using AI.
• Statistics: ROC curves and multivariable Cox regression models were used for prognostic analysis.

Results

• Primary endpoint incidence was significantly higher in patients with CT-µFR ≤ 0.80 (87.2%) than those with > 0.80 (21.7%) (HR: 8.51, p < 0.001).
• Revascularization within 90 days was more frequent in the CT-µFR ≤ 0.80 group (86.4%) than in the > 0.80 group (17.1%) (HR: 9.82, p < 0.001).
• A combined CT-µFR > 0.80 and total CT-µFR > 2.65 identified patients with a prognosis similar to those who underwent physiology-guided intervention. This strategy could safely defer ICA in ~40% of patients.
• In deferred vessels, the following parameters predicted adverse outcomes:
• CT-µFR ≤ 0.85 › HR: 6.10 (p = 0.003)
• ΔCT-µFR ≥ 0.05 › HR: 17.59 (p = 0.006)
• Plaque burden ≥ 64.3% › HR: 18.13 (p = 0.006)
• Lipidic volume ≥ 6.45 mm³ › HR: 7.13 (p = 0.004)
• Calcific volume ≥ 38.88 mm³ › HR: 5.82 (p = 0.004)

Conclusions
CCTA-derived CT-µFR and plaque characterization parameters provide strong predictive value for determining the need for coronary revascularization. Patients with CT-µFR > 0.80 and total CT-µFR > 2.65 can be safely managed without invasive angiography.

Commentary
The CAREER study demonstrates that CCTA, enhanced with AI-based physiologic and compositional assessment, allows personalized, evidence-driven decision-making in coronary artery disease. The rapid (2-minute) automated analysis is clinically practical and could reduce unnecessary invasive procedures, offering a cost-effective and patient-friendly alternative in appropriate cases.