Machine Learning Identifies Clinical and Genetic Factors Associated With Anthracycline Cardiotoxicity in Pediatric Cancer Survivors

Marie A. Chaix; Neha Parmar; Caroline Kinnear; Myriam Lafreniere-Roula; Oyediran Akinrinade; Roderick Yao; Anastasia Miron; Emily Lam; Guoliang Meng; Anne Christie; Ashok Kumar Manickaraj; Stacey Marjerrison; Rejane Dillenburg; Mylène Bassal; Jane Lougheed; Shayna Zelcer; Herschel Rosenberg; David Hodgson; Leonard Sender; Paul Kantor; Cedric Manlhiot; James Ellis; Luc Mertens; Paul C. Nathan; Seema Mital

doi:10.1016/j.jaccao.2020.11.004

Machine Learning Identifies Clinical and Genetic Factors Associated With Anthracycline Cardiotoxicity in Pediatric Cancer Survivors

Marie A. Chaix, Neha Parmar, Caroline Kinnear, Myriam Lafreniere-Roula, Oyediran Akinrinade, Roderick Yao, Anastasia Miron, Emily Lam, Guoliang Meng, Anne Christie, Ashok Kumar Manickaraj, Stacey Marjerrison, Rejane Dillenburg, Mylène Bassal, Jane Lougheed, Shayna Zelcer, Herschel Rosenberg, David Hodgson, Leonard Sender, Paul KantorCedric Manlhiot, James Ellis, Luc Mertens, Paul C. Nathan, Seema Mital

School of Medicine

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

Background: Despite known clinical risk factors, predicting anthracycline cardiotoxicity remains challenging. Objectives: This study sought to develop a clinical and genetic risk prediction model for anthracycline cardiotoxicity in childhood cancer survivors. Methods: We performed exome sequencing in 289 childhood cancer survivors at least 3 years from anthracycline exposure. In a nested case-control design, 183 case patients with reduced left ventricular ejection fraction despite low-dose doxorubicin (≤250 mg/m²), and 106 control patients with preserved left ventricular ejection fraction despite doxorubicin >250 mg/m² were selected as extreme phenotypes. Rare/low-frequency variants were collapsed to identify genes differentially enriched for variants between case patients and control patients. The expression levels of 5 top-ranked genes were evaluated in human induced pluripotent stem cell–derived cardiomyocytes, and variant enrichment was confirmed in a replication cohort. Using random forest, a risk prediction model that included genetic and clinical predictors was developed. Results: Thirty-one genes were differentially enriched for variants between case patients and control patients (p < 0.001). Only 42.6% case patients harbored a variant in these genes compared to 89.6% control patients (odds ratio: 0.09; 95% confidence interval: 0.04 to 0.17; p = 3.98 × 10^–15). A risk prediction model for cardiotoxicity that included clinical and genetic factors had a higher prediction accuracy and lower misclassification rate compared to the clinical-only model. In vitro inhibition of gene-associated pathways (PI3KR2, ZNF827) provided protection from cardiotoxicity in cardiomyocytes. Conclusions: Our study identified variants in cardiac injury pathway genes that protect against cardiotoxicity and informed the development of a prediction model for delayed anthracycline cardiotoxicity, and it also provided new targets in autophagy genes for the development of cardio-protective drugs.

Original language	English (US)
Pages (from-to)	690-706
Number of pages	17
Journal	JACC: CardioOncology
Volume	2
Issue number	5
DOIs	https://doi.org/10.1016/j.jaccao.2020.11.004
State	Published - Dec 2020

Keywords

anthracycline
cancer survivorship
cardiomyopathy
echocardiography
genomics
machine learning
risk prediction

ASJC Scopus subject areas

Cardiology and Cardiovascular Medicine
Oncology

Access to Document

10.1016/j.jaccao.2020.11.004

Cite this

Chaix, M. A., Parmar, N., Kinnear, C., Lafreniere-Roula, M., Akinrinade, O., Yao, R., Miron, A., Lam, E., Meng, G., Christie, A., Manickaraj, A. K., Marjerrison, S., Dillenburg, R., Bassal, M., Lougheed, J., Zelcer, S., Rosenberg, H., Hodgson, D., Sender, L., ... Mital, S. (2020). Machine Learning Identifies Clinical and Genetic Factors Associated With Anthracycline Cardiotoxicity in Pediatric Cancer Survivors. JACC: CardioOncology, 2(5), 690-706. https://doi.org/10.1016/j.jaccao.2020.11.004

Chaix, MA, Parmar, N, Kinnear, C, Lafreniere-Roula, M, Akinrinade, O, Yao, R, Miron, A, Lam, E, Meng, G, Christie, A, Manickaraj, AK, Marjerrison, S, Dillenburg, R, Bassal, M, Lougheed, J, Zelcer, S, Rosenberg, H, Hodgson, D, Sender, L, Kantor, P, Manlhiot, C, Ellis, J, Mertens, L, Nathan, PC & Mital, S 2020, 'Machine Learning Identifies Clinical and Genetic Factors Associated With Anthracycline Cardiotoxicity in Pediatric Cancer Survivors', JACC: CardioOncology, vol. 2, no. 5, pp. 690-706. https://doi.org/10.1016/j.jaccao.2020.11.004

@article{fa3c75cf317c4833951fca3be6f0df96,

title = "Machine Learning Identifies Clinical and Genetic Factors Associated With Anthracycline Cardiotoxicity in Pediatric Cancer Survivors",

abstract = "Background: Despite known clinical risk factors, predicting anthracycline cardiotoxicity remains challenging. Objectives: This study sought to develop a clinical and genetic risk prediction model for anthracycline cardiotoxicity in childhood cancer survivors. Methods: We performed exome sequencing in 289 childhood cancer survivors at least 3 years from anthracycline exposure. In a nested case-control design, 183 case patients with reduced left ventricular ejection fraction despite low-dose doxorubicin (≤250 mg/m2), and 106 control patients with preserved left ventricular ejection fraction despite doxorubicin >250 mg/m2 were selected as extreme phenotypes. Rare/low-frequency variants were collapsed to identify genes differentially enriched for variants between case patients and control patients. The expression levels of 5 top-ranked genes were evaluated in human induced pluripotent stem cell–derived cardiomyocytes, and variant enrichment was confirmed in a replication cohort. Using random forest, a risk prediction model that included genetic and clinical predictors was developed. Results: Thirty-one genes were differentially enriched for variants between case patients and control patients (p < 0.001). Only 42.6% case patients harbored a variant in these genes compared to 89.6% control patients (odds ratio: 0.09; 95% confidence interval: 0.04 to 0.17; p = 3.98 × 10–15). A risk prediction model for cardiotoxicity that included clinical and genetic factors had a higher prediction accuracy and lower misclassification rate compared to the clinical-only model. In vitro inhibition of gene-associated pathways (PI3KR2, ZNF827) provided protection from cardiotoxicity in cardiomyocytes. Conclusions: Our study identified variants in cardiac injury pathway genes that protect against cardiotoxicity and informed the development of a prediction model for delayed anthracycline cardiotoxicity, and it also provided new targets in autophagy genes for the development of cardio-protective drugs.",

keywords = "anthracycline, cancer survivorship, cardiomyopathy, echocardiography, genomics, machine learning, risk prediction",

author = "Chaix, {Marie A.} and Neha Parmar and Caroline Kinnear and Myriam Lafreniere-Roula and Oyediran Akinrinade and Roderick Yao and Anastasia Miron and Emily Lam and Guoliang Meng and Anne Christie and Manickaraj, {Ashok Kumar} and Stacey Marjerrison and Rejane Dillenburg and Myl{\`e}ne Bassal and Jane Lougheed and Shayna Zelcer and Herschel Rosenberg and David Hodgson and Leonard Sender and Paul Kantor and Cedric Manlhiot and James Ellis and Luc Mertens and Nathan, {Paul C.} and Seema Mital",

note = "Funding Information: The study was supported by the Canadian Cancer Society, Toronto, Canada; Canadian Institutes of Health Research, Toronto, Canada; the Pediatric Oncology Group of Ontario, Toronto, Canada; the Ontario Institute for Cancer Research, Toronto, Canada; Children{\textquoteright}s Cancer and Blood disorders, Toronto, Canada; Ted Rogers Centre for Heart Research, Toronto, Canada; and the Labatt Family Heart Center at the Hospital for Sick Children, Toronto, Canada. Dr. Sender is Senior Vice President, Medical Affairs for Pediatric, Adolescent and Young Adult Oncology at NantKwest. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Publisher Copyright: {\textcopyright} 2020 The Authors",

year = "2020",

month = dec,

doi = "10.1016/j.jaccao.2020.11.004",

language = "English (US)",

volume = "2",

pages = "690--706",

journal = "JACC: CardioOncology",

issn = "2666-0873",

publisher = "Elsevier Inc.",

number = "5",

}

TY - JOUR

T1 - Machine Learning Identifies Clinical and Genetic Factors Associated With Anthracycline Cardiotoxicity in Pediatric Cancer Survivors

AU - Chaix, Marie A.

AU - Parmar, Neha

AU - Kinnear, Caroline

AU - Lafreniere-Roula, Myriam

AU - Akinrinade, Oyediran

AU - Yao, Roderick

AU - Miron, Anastasia

AU - Lam, Emily

AU - Meng, Guoliang

AU - Christie, Anne

AU - Manickaraj, Ashok Kumar

AU - Marjerrison, Stacey

AU - Dillenburg, Rejane

AU - Bassal, Mylène

AU - Lougheed, Jane

AU - Zelcer, Shayna

AU - Rosenberg, Herschel

AU - Hodgson, David

AU - Sender, Leonard

AU - Kantor, Paul

AU - Manlhiot, Cedric

AU - Ellis, James

AU - Mertens, Luc

AU - Nathan, Paul C.

AU - Mital, Seema

N1 - Funding Information: The study was supported by the Canadian Cancer Society, Toronto, Canada; Canadian Institutes of Health Research, Toronto, Canada; the Pediatric Oncology Group of Ontario, Toronto, Canada; the Ontario Institute for Cancer Research, Toronto, Canada; Children’s Cancer and Blood disorders, Toronto, Canada; Ted Rogers Centre for Heart Research, Toronto, Canada; and the Labatt Family Heart Center at the Hospital for Sick Children, Toronto, Canada. Dr. Sender is Senior Vice President, Medical Affairs for Pediatric, Adolescent and Young Adult Oncology at NantKwest. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Publisher Copyright: © 2020 The Authors

PY - 2020/12

Y1 - 2020/12

N2 - Background: Despite known clinical risk factors, predicting anthracycline cardiotoxicity remains challenging. Objectives: This study sought to develop a clinical and genetic risk prediction model for anthracycline cardiotoxicity in childhood cancer survivors. Methods: We performed exome sequencing in 289 childhood cancer survivors at least 3 years from anthracycline exposure. In a nested case-control design, 183 case patients with reduced left ventricular ejection fraction despite low-dose doxorubicin (≤250 mg/m2), and 106 control patients with preserved left ventricular ejection fraction despite doxorubicin >250 mg/m2 were selected as extreme phenotypes. Rare/low-frequency variants were collapsed to identify genes differentially enriched for variants between case patients and control patients. The expression levels of 5 top-ranked genes were evaluated in human induced pluripotent stem cell–derived cardiomyocytes, and variant enrichment was confirmed in a replication cohort. Using random forest, a risk prediction model that included genetic and clinical predictors was developed. Results: Thirty-one genes were differentially enriched for variants between case patients and control patients (p < 0.001). Only 42.6% case patients harbored a variant in these genes compared to 89.6% control patients (odds ratio: 0.09; 95% confidence interval: 0.04 to 0.17; p = 3.98 × 10–15). A risk prediction model for cardiotoxicity that included clinical and genetic factors had a higher prediction accuracy and lower misclassification rate compared to the clinical-only model. In vitro inhibition of gene-associated pathways (PI3KR2, ZNF827) provided protection from cardiotoxicity in cardiomyocytes. Conclusions: Our study identified variants in cardiac injury pathway genes that protect against cardiotoxicity and informed the development of a prediction model for delayed anthracycline cardiotoxicity, and it also provided new targets in autophagy genes for the development of cardio-protective drugs.

AB - Background: Despite known clinical risk factors, predicting anthracycline cardiotoxicity remains challenging. Objectives: This study sought to develop a clinical and genetic risk prediction model for anthracycline cardiotoxicity in childhood cancer survivors. Methods: We performed exome sequencing in 289 childhood cancer survivors at least 3 years from anthracycline exposure. In a nested case-control design, 183 case patients with reduced left ventricular ejection fraction despite low-dose doxorubicin (≤250 mg/m2), and 106 control patients with preserved left ventricular ejection fraction despite doxorubicin >250 mg/m2 were selected as extreme phenotypes. Rare/low-frequency variants were collapsed to identify genes differentially enriched for variants between case patients and control patients. The expression levels of 5 top-ranked genes were evaluated in human induced pluripotent stem cell–derived cardiomyocytes, and variant enrichment was confirmed in a replication cohort. Using random forest, a risk prediction model that included genetic and clinical predictors was developed. Results: Thirty-one genes were differentially enriched for variants between case patients and control patients (p < 0.001). Only 42.6% case patients harbored a variant in these genes compared to 89.6% control patients (odds ratio: 0.09; 95% confidence interval: 0.04 to 0.17; p = 3.98 × 10–15). A risk prediction model for cardiotoxicity that included clinical and genetic factors had a higher prediction accuracy and lower misclassification rate compared to the clinical-only model. In vitro inhibition of gene-associated pathways (PI3KR2, ZNF827) provided protection from cardiotoxicity in cardiomyocytes. Conclusions: Our study identified variants in cardiac injury pathway genes that protect against cardiotoxicity and informed the development of a prediction model for delayed anthracycline cardiotoxicity, and it also provided new targets in autophagy genes for the development of cardio-protective drugs.

KW - anthracycline

KW - cancer survivorship

KW - cardiomyopathy

KW - echocardiography

KW - genomics

KW - machine learning

KW - risk prediction

UR - http://www.scopus.com/inward/record.url?scp=85097372256&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85097372256&partnerID=8YFLogxK

U2 - 10.1016/j.jaccao.2020.11.004

DO - 10.1016/j.jaccao.2020.11.004

M3 - Article

C2 - 34396283

AN - SCOPUS:85097372256

SN - 2666-0873

VL - 2

SP - 690

EP - 706

JO - JACC: CardioOncology

JF - JACC: CardioOncology

IS - 5

ER -

Machine Learning Identifies Clinical and Genetic Factors Associated With Anthracycline Cardiotoxicity in Pediatric Cancer Survivors

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this