Artificial Intelligence for Identification of Patients with Increased Risk of Severe Cancer Therapy-Related Cardiac Dysfunction Following Anthracycline Therapy - 26/07/25

Doi : 10.1016/j.amjmed.2025.06.035

Isabel G. Scalia, MD ^a, Juan M. Farina, MD ^a, Milagros Pereyra Pietri, MD ^a, Patrick Sarkis, MD ^a, Niloofar Javadi, MD ^a, Nadera Naquib Bismee, MBBS ^a, Taylor Viggiano, MD ^a, Cecilia Tagle-Cornell, APRN, CNP, MS ^a, Laura Koepke, APRN, AGACNP-BC, M.S.N ^a, Courtney Kenyon, PAC, MS ^a, Barbara Novais, APRN, CNP, DNP ^a, Mohammed Tiseer Abbas, MD ^a, Balaji K. Tamarappoo, MD, PhD ^a, Steven J. Lester, MD ^a, Imon Banerjee, PhD ^b, Ramzi Ibrahim, MD ^a, Carolyn Larsen, MD ^a, Kwan S. Lee, MB BCh, MD ^a, Reza Arsanjani, MD ^a, Chadi Ayoub, MD, PhD ^a,^⁎
^a Department of Cardiovascular Diseases, Mayo Clinic, Phoenix, AZ, USA
^b Department of Radiology, Mayo Clinic, Phoenix, Arizona, USA

^⁎Corresponding author: Division of Cardiovascular Imaging, Mayo Clinic, 5777 East Mayo Boulevard, Phoenix, AZ 85054, USA.Division of Cardiovascular ImagingMayo Clinic5777 East Mayo BoulevardPhoenixAZ85054USA

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Saturday 26 July 2025
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Highlights

-	Artificial intelligence applied to 12-lead electrocardiograms can predict the risk of systolic dysfunction
-	This tool can accurately predict risk of cardiotoxicity in patients on anthracycline therapy
-	Higher-risk ECG-AI scores were associated with worse clinical outcomes

Le texte complet de cet article est disponible en PDF.

Abstract

Background

Early detection of cancer therapy-related cardiac dysfunction (CTRCD) after anthracycline exposure is critically important in minimizing morbidity and mortality. Artificial intelligence models applied to electrocardiograms (ECG-AI) may allow for early identification of CTRCD and improved outcomes.

Methods

Patients treated with anthracycline therapy between 2002 and 2022 across three tertiary centers were evaluated. Characteristics, echocardiograms pre- and post-chemotherapy, and outcomes were reviewed. ECG-AI predictive scores for left ventricular systolic dysfunction (LVSD) within one year following treatment were collected. ROC analysis was conducted for accuracy of ECG-AI score to detect severe CTRCD (left ventricular ejection fraction <40 %).

Results

Overall, 3439 patients were included, mean age 60.2 ± 14.1 years, 53.6 % male. Severe CTRCD was present in 114 patients. ROC analysis of ECG-AI scores post-initiation of therapy was superior to that of pre-anthracycline ECG-AI scores and had moderate accuracy for detection of severe CTRCD (AUC 0.761). An ECG-AI score >3.0 % was independently associated with significantly poorer survival at one year (HR 2.19, 95 %CI 1.92-2.51) and five years (HR 1.69, 95 %CI 1.54-1.87) post-anthracycline therapy.

Conclusions

ECG-AI indicating increased likelihood for LVSD post-anthracycline therapy accurately detected severe CTRCD. Clinically, this tool may allow early diagnosis and treatment of high-risk patients and may reduce unnecessary surveillance in those with lower risk.

Le texte complet de cet article est disponible en PDF.

Keywords : Cardiotoxicity, Anthracycline, Artificial intelligence, Electrocardiogram

Plan

INTRODUCTION

METHODS

Population

Echocardiographic Evaluation

ECG-AI Risk Score for Severe CTRCD Pre-Anthracycline Exposure

ECG-AI Risk Score for Severe CTRCD Post-Anthracycline Exposure

Declaration of AI-assisted technologies

Acknowledgements

	Funding:This publication was supported and funded by Mayo Clinic Arizona Cardiovascular Clinical Research Center (MCA CV CRC). We are thankful for their generous support. Contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the MCA CV CRC.
	Conflicts of Interest: The authors have no conflicts of interest. A/Prof Chadi Ayoub is supported by the Mayo Clinic Clinician Engaged in Research award.
	Authorship: We approve that all authors had access to the data and a role in writing the manuscript. IGS: Writing – review & editing, Writing – original draft, Validation, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. JMF: Writing – review & editing, Writing – original draft, Project administration, Methodology, Investigation, Formal analysis, Data curation. MPP: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data curation. PS: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data curation. NJ: Writing – review & editing, Methodology, Formal analysis, Data curation. NNB: Writing – review & editing, Methodology, Formal analysis, Data curation. TV: Writing – review & editing, Methodology, Formal analysis, Data curation. CT: Writing – review & editing, Methodology, Formal analysis, Data curation. LK: Methodology, Investigation, Formal analysis, Data curation. CK: Methodology, Investigation, Formal analysis, Data curation. BN: Methodology, Investigation, Formal analysis, Data curation. MTA: Writing – review & editing, Methodology, Investigation, Formal analysis, Data curation. BKT: Writing – review & editing, Visualization, Investigation. SJL: Writing – review & editing, Visualization, Supervision, Project administration. IB: Writing – review & editing, Methodology, Investigation. RI: Writing – review & editing, Validation, Formal analysis. CL: Writing – review & editing, Validation, Methodology, Investigation. KSL: Writing – review & editing, Validation, Methodology, Investigation. RA: Writing – review & editing, Writing – original draft, Visualization, Supervision, Methodology, Investigation, Conceptualization. CA: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Formal analysis, Conceptualization.