Abstract 4370263: Cost-Effectiveness of AI-Enabled Coronary Plaque Analysis for Management of Stable Coronary Artery Disease

S Suzanne Baron (Massachusetts General Hospital, Boston, Massachusetts, United States) C Campbell Rogers (Heartflow, San Carlos, California, United States) D Daniel D'Attilio (HeartFlow Inc, Mountain View, California, United States) S Souma Sengupta (Heartflow Inc, Mountain View, California, United States) T Timothy Fairbairn (Liverpool Heart and Chest Hospital, Liverpool, United Kingdom)

Abstract

Introduction: Recent clinical studies have established that non-calcified plaque in the coronary arteries plays a greater role in risk of myocardial infarction (MI) than calcified plaque. AI-enabled coronary plaque analysis (AI-CPA) derived from coronary computed tomography angiography (CCTA) measures non-calcified plaque in addition to calcified plaque to help guide medical management. The cost-effectiveness of AI-CPA has not been established. Methods: We developed a Markov cohort model to determine the cost-effectiveness of AI-CPA (Heartflow Inc.) compared to CCTA-only based on data from the FISH&CHIPS study, which consisted of patients undergoing CCTA for suspected stable coronary artery disease (CAD). Total Plaque Volume (TPV) was measured using AI-CPA and divided into stages: 1-100, 101-250, 251-750, >750mm 3 . TPV stages were used to guide lipid-lowering therapy (Table 1). Cardiovascular (CV) outcomes included all-cause death and MI. Costs were derived from US claims data. Utilities were derived from peer-reviewed literature. Results were reported as incremental cost effectiveness ratios (ICER) over a 5 and 10-year time horizon. Sensitivity analyses were performed to assess the impact of key variables on the model output. Results: Clinical event rates were projected to be lower with AI-CPA tailored treatment at both 5 years and 10 years (Table 2). AI-CPA patients accrued an average of 0.19 (95% CI: 0.15 to 0.24) and 0.61 QALY (95% CI:0.47 to 0.76) relative to CCTA-only at 5 and 10 years, respectively. Estimated total costs were higher for AI-CPA relative to CCTA-only by $1,519 (95% CI: $1,109 to $1,896) and $1,698 (95% CI: $901 to $2503) at 5 and 10 years, respectively. The ICER at 5 and 10 years was $7,982/QALY and $2,765/QALY, respectively. 100% of the Monte Carlo simulated ICERs were <$100,000/QALY for both time horizons. costs was $17,519 (95% CI: $12,951 to $22,879) and $59,653 (95% CI: $46,358 to $75,690) higher for AI-CPA than CCTA-only at 5 and 10 years, respectively. Conclusion: An AI-CPA guided approach to medical management of stable CAD may be cost-effective compared to CCTA-only over time due to a reduction in CV events.

Article Details

Journal Circulation
Volume / Issue Vol. 152, Issue Suppl_3
Published November 04, 2025
ISSN 0009-7322
Publisher Lippincott Williams & Wilkins

Journal Info

Circulation

Lippincott Williams & Wilkins

ISSN: 0009-7322 Health Sciences

Authors (5)

S

Suzanne Baron

Massachusetts General Hospital, Boston, Massachusetts, United States

C

Campbell Rogers

Heartflow, San Carlos, California, United States

D

Daniel D'Attilio

HeartFlow Inc, Mountain View, California, United States

S

Souma Sengupta

Heartflow Inc, Mountain View, California, United States

T

Timothy Fairbairn

Liverpool Heart and Chest Hospital, Liverpool, United Kingdom