Risk Stratification of Cancer Therapy-Related Cardiac Dysfunction Using AI-Enhanced Electrocardiography
1 other identifier
observational
31,486
1 country
1
Brief Summary
This study investigates the use of AI-enhanced electrocardiogram (ECG) for risk stratification of cancer therapy-related cardiac dysfunction (CTRCD) before the initiation of cancer therapy. The study includes patients treated with anthracyclines, HER2 inhibitors, or immune checkpoint inhibitors (ICIs) at Severance Hospital between May 2006 and November 2022, who underwent an ECG within 90 days prior to chemotherapy. The primary goal is to evaluate whether AI-ECG can accurately predict the risk of CTRCD and compare its performance to existing risk stratification models. In addition, we aim to assess whether the variation in AI-ECG scores between pre- and post-chemotherapy assessments could serve as a predictor of CTRCD. Eligible participants are adults without prior heart failure, cardiomyopathy, or myocarditis, and with baseline left ventricular ejection fraction (LVEF) ≥40%. For trajectory analysis, only patients with an additional ECG within 90 days after chemotherapy are included. The primary outcome is the development of CTRCD within 12 months after the last treatment cycle (and no more than 24 months after the first). The secondary outcomes are severe CTRCD (LVEF \<40%) and all-cause mortality. This study aims to validate the clinical utility of AI-enhanced ECG as a simple, accessible, and cost-effective tool for predicting CTRCD across diverse cancer treatment regimens, including newer immunotherapies.
Trial Health
Trial Health Score
Automated assessment based on enrollment pace, timeline, and geographic reach
participants targeted
Target at P75+ for all trials
Started May 2006
Longer than P75 for all trials
1 active site
Health score is calculated from publicly available data and should be used for screening purposes only.
Trial Relationships
Click on a node to explore related trials.
Study Timeline
Key milestones and dates
Study Start
First participant enrolled
May 1, 2006
CompletedPrimary Completion
Last participant's last visit for primary outcome
December 31, 2022
CompletedStudy Completion
Last participant's last visit for all outcomes
December 31, 2022
CompletedFirst Submitted
Initial submission to the registry
March 30, 2026
CompletedFirst Posted
Study publicly available on registry
April 7, 2026
CompletedApril 13, 2026
April 1, 2026
16.7 years
March 30, 2026
April 7, 2026
Conditions
Keywords
Outcome Measures
Primary Outcomes (1)
Incidence of cancer therapy-related cardiac dysfunction (CTRCD)
CTRCD defined as ≥10%p drop in left ventricular ejection fraction (LVEF) from baseline to 40% to 49.9% OR \<10%p drop to 40-49.9% with a reduction in GLS by \>15% OR new LVEF reduction to \<40% from baseline LVEF, OR hospitalization for heart failure or diagnosis of cardiomyopathy defined by ICD codes,, diagnosed within 12 months after the last treatment cycle and no more than 24 months after the first cycle cardiotoxic cancer therapy. LVEF is assessed by either echocardiography or MUGA (Multi-gated acquisition nuclear imaging) scan.
From initiation of first chemotherapy up to 24 months.
Secondary Outcomes (1)
All-cause mortality
From initiation of first chemotherapy up to 24 months.
Study Arms (2)
High baseline AI-ECG risk for CTRCD
Patients classified as high risk for CTRCD based on baseline AI-ECG LVSD prediction probability before initiation of cancer therapy.
Low baseline AI-ECG risk for CTRCD
Patients classified as low risk for CTRCD based on baseline AI-ECG LVSD prediction probability before initiation of cancer therapy.
Interventions
This is a retrospective observational study using existing clinical data. No intervention or diagnostic procedure is applied to participants.
Eligibility Criteria
Patients who were prescribed anthracycline, HER2 inhibitor, or ICI agents and have ECG prescription within 90 days before their first chemotherapy treatment at Severance hospital from May 2006 to November 2022.
You may qualify if:
- Patients prescribed anthracycline (doxorubicin, daunorubicin, epirubicin, aclarubicin, idarubicin), HER2 inhibitor (trastuzumab, emtansine, tucatinib, trastuzumab deruxtecan, pertuzumab), or immune checkpoint inhibitors (ipilimumab, nivolumab, atezolizumab, pembrolizumab, durvalumab, avelumab)
- ECG performed within 90 days prior to the first chemotherapy treatment
- Age ≥ 19 years
- For trajectory analysis: patients with an additional ECG within 90 days after the first chemotherapy
You may not qualify if:
- Age \< 19 years
- History of heart failure, cardiomyopathy, or myocarditis (confirmed by ICD codes)
- Prior exposure to anthracycline, HER2 inhibitor, or immune checkpoint inhibitor therapy
- Baseline left ventricular ejection fraction (LVEF) \< 40% on echocardiography within 1 year prior to chemotherapy
Contact the study team to confirm eligibility.
Sponsors & Collaborators
- Yonsei Universitylead
- VUNO Inc.collaborator
- Korea Health Industry Development Institutecollaborator
Study Sites (1)
Severance Hospital, Yonsei University Health System
Seoul, 03722, South Korea
Related Publications (8)
Stein-Merlob AF, Rothberg MV, Ribas A, Yang EH. Cardiotoxicities of novel cancer immunotherapies. Heart. 2021 Nov;107(21):1694-1703. doi: 10.1136/heartjnl-2020-318083. Epub 2021 Mar 15.
PMID: 33722826BACKGROUNDLee SH, Cho I, You SC, Cha MJ, Chang JS, Kim WD, Go KY, Kim DY, Seo J, Shim CY, Hong GR, Kang SM, Ha JW, Rha SY, Kim HS. Cancer Therapy-Related Cardiac Dysfunction in Patients Treated with a Combination of an Immune Checkpoint Inhibitor and Doxorubicin. Cancers (Basel). 2022 May 7;14(9):2320. doi: 10.3390/cancers14092320.
PMID: 35565449BACKGROUNDScalia IG, Farina JM, Pietri MP, Sarkis P, Javadi N, Bismee NN, Viggiano T, Tagle-Cornell C, Koepke L, Kenyon C, Novais B, Tiseer Abbas M, Tamarappoo BK, Lester SJ, Banerjee I, Ibrahim R, Larsen C, Lee KS, Arsanjani R, Ayoub C. Artificial Intelligence for Identification of Patients with Increased Risk of Severe Cancer Therapy-Related Cardiac Dysfunction Following Anthracycline Therapy. Am J Med. 2025 Nov;138(11):1561-1568.e1. doi: 10.1016/j.amjmed.2025.06.035. Epub 2025 Jun 25.
PMID: 40578470BACKGROUNDChoi HM, Kim J, Park J, Park JB, Kim HK, Choi HJ, Yoon YE, Cho GY, Cho Y, Hwang IC. AI derived ECG global longitudinal strain compared to echocardiographic measurements. Sci Rep. 2024 Nov 2;14(1):26458. doi: 10.1038/s41598-024-78268-8.
PMID: 39488646BACKGROUNDMihos CG, Liu JE, Anderson KM, Pernetz MA, O'Driscoll JM, Aurigemma GP, Ujueta F, Wessly P; American Heart Association Council on Peripheral Vascular Disease; Council on Cardiovascular and Stroke Nursing; and Council on Clinical Cardiology. Speckle-Tracking Strain Echocardiography for the Assessment of Left Ventricular Structure and Function: A Scientific Statement From the American Heart Association. Circulation. 2025 Sep 9;152(10):e96-e109. doi: 10.1161/CIR.0000000000001354. Epub 2025 Aug 6.
PMID: 40765507BACKGROUNDGomes C,Geels J,Debray TPA,Malekzadeh A,Asselbergs FW,Linschoten M
BACKGROUNDOikonomou EK, Kokkinidis DG, Kampaktsis PN, Amir EA, Marwick TH, Gupta D, Thavendiranathan P. Assessment of Prognostic Value of Left Ventricular Global Longitudinal Strain for Early Prediction of Chemotherapy-Induced Cardiotoxicity: A Systematic Review and Meta-analysis. JAMA Cardiol. 2019 Oct 1;4(10):1007-1018. doi: 10.1001/jamacardio.2019.2952.
PMID: 31433450BACKGROUNDHaj-Yehia E, Michel L, Mincu RI, Rassaf T, Totzeck M. Prevention of cancer-therapy related cardiac dysfunction. Curr Heart Fail Rep. 2025 Feb 19;22(1):9. doi: 10.1007/s11897-025-00697-x.
PMID: 39969700BACKGROUND
MeSH Terms
Conditions
Condition Hierarchy (Ancestors)
Study Design
- Study Type
- observational
- Observational Model
- COHORT
- Time Perspective
- RETROSPECTIVE
- Sponsor Type
- OTHER
- Responsible Party
- PRINCIPAL INVESTIGATOR
- PI Title
- Professor
Study Record Dates
First Submitted
March 30, 2026
First Posted
April 7, 2026
Study Start
May 1, 2006
Primary Completion
December 31, 2022
Study Completion
December 31, 2022
Last Updated
April 13, 2026
Record last verified: 2026-04
Data Sharing
- IPD Sharing
- Will not share
Individual participant data will not be shared due to institutional policies and privacy regulations related to the use of retrospective clinical data.