Serum Proteins Predict Treatment-Related Cardiomyopathy Among Survivors of Childhood Cancer
Description
This dataset accompanies our publication in JACC: CardioOncology (https://doi.org/10.1016/j.jaccao.2024.10.004) titled “Serum Proteins Predict Treatment-Related Cardiomyopathy Among Survivors of Childhood Cancer”
Background: Anthracyclines are effective chemotherapy agents but cause cardiomyopathy as a major late effect in adult survivors of childhood cancer. Early detection using biomarkers is critical for targeted interventions. We performed untargeted mass spectrometry-based proteomics profiling to identify proteins predictive of cardiomyopathy.
Study Overview:
-
Population: 75 asymptomatic survivors with subclinical cardiomyopathy and 75 matched controls (no cardiomyopathy) from the St. Jude Lifetime Cohort Study (SJLIFE) used for discovery; 23 survivors with severe or symptomatic cardiomyopathy and 23 matched controls used for independent validation.
-
Data: Untargeted mass spectrometry profiling of 867 serum proteins.
-
Analysis: Differential expression analysis followed by conditional logistic regression with a least absolute shrinkage and selection operator (LASSO) penalty identified 27 proteins that best discriminated cardiomyopathy cases.
-
Outcome: The 27-protein model effectively discriminated survivors with symptomatic or severe cardiomyopathy requiring heart failure medications.
Contents:
This Zenodo deposit includes:
-
Prediction Code
-
Scripts used for data cleaning, LASSO selection, model fitting, and prediction generation.
-
-
Protein Data (867 proteins)
-
Normalized expression matrix of 867 proteins for all patients in the discovery and validation cohorts.
-
-
Patient Metadata
-
De-identified metadata including cardiomyopathy status and grade for each patient
-
Citation:
Poudel, S, Shrestha, H, Pan, Y. et al. Serum Proteins Predict Treatment-Related Cardiomyopathy Among Survivors of Childhood Cancer. J Am Coll Cardiol CardioOnc. 2025 Jan, 7 (1) 56–67.
https://doi.org/10.1016/j.jaccao.2024.10.004