Key result
Multi-modal ML outperforms clinical-only models in predicting late-onset cardiomyopathy, achieving an AUC of 0.93.
Why the study?
Current risk stratification using clinical factors alone shows suboptimal discriminative capacity for anthracycline- and radiation therapy-related cardiomyopathy in pediatric cancer survivors.
Does a multi-modal machine learning model integrating clinical, genomic, and imaging data improve the prediction of late-onset cardiomyopathy in adult survivors of childhood cancer compared to clinical factors alone?
Cohort (n=1,217)
Does a multi-modal machine learning model integrating clinical, genomic, and imaging data improve the prediction of late-onset cardiomyopathy in adult survivors of childhood cancer compared to clinical factors alone?
Integrating clinical, genomic, and imaging data using a multi-modal machine learning approach significantly improves the prediction of late-onset cardiomyopathy in childhood cancer survivors compared to clinical risk factors alone.
May refine cardiomyopathy risk stratification in survivors; leaves open prospective validation before clinical adoption.
10018 Background: Pediatric cancer survivors face substantial risk of anthracycline- and radiation therapy (RT)-related cardiomyopathy (CM). Current risk stratification using clinical factors alone demonstrates suboptimal discriminative capacity. We hypothesized that integrating multi-modal data (clinical, genomic, and cardiac imaging/electrocardiography) would enhance predictive accuracy for late-onset CM beyond single-modality models. Methods: We analyzed data from 1,217 adult survivors of childhood cancer enrolled in the St. Jude Lifetime Cohort (SJLIFE), accessed via St. Jude Cloud. The primary endpoint was CM, defined as left ventricular ejection fraction (LVEF) <50% or ≥10% absolute decline from baseline. Four progressively complex feature sets were evaluated: (1) Clinical-Only (anthracycline/RT dose, age at diagnosis, sex); (2) Clinical+Genomic (adding 31 protective variants from PeCan); (3) Clinical+ECG/Imaging (adding 86 signal-processed ECG features and global longitudinal strain [GLS]); and (4) Multi-Modal (comprehensive integration). Logistic regression, random forest (RF), and extreme gradient boosting (XGBoost) algorithms were compared using 5-fold stratified cross-validation. Model interpretability was assessed via SHapley Additive exPlanations (SHAP) analysis. Results: Clinical-Only models demonstrated limited discrimination (AUC 0.70; 95% CI, 0.65-0.75). Incremental addition of genomic data (AUC 0.85) or ECG/GLS parameters (AUC 0.89) substantially improved performance. The Multi-Modal XGBoost model achieved superior predictive accuracy (AUC 0.93; 95% CI, 0.90-0.95; sensitivity 0.88; specificity 0.89), marginally outperforming RF (AUC 0.91). SHAP analysis identified GLS as the dominant predictor, followed by cumulative anthracycline dose and protective genomic variants, demonstrating synergistic contribution of multi-modal features. Conclusions: Multi-modal machine learning integration using open-source St. Jude Cloud data enables highly accurate prediction of late-onset CM in pediatric cancer survivors (AUC 0.93). The demonstrated superiority over single-modality approaches supports a paradigm shift toward comprehensive risk stratification. This XGBoost-based model could facilitate precision survivorship care, enabling risk-adapted surveillance intensification and early cardioprotective intervention for high-risk patients. Comparative performance of machine learning models for 10-year cardiomyopathy risk prediction. Model Type Algorithm AUC (95% CI) Sensitivity Specificity Clinical-Only Log. Reg. 0.70 (0.65-0.75) 0.62 0.66 Clinical + Genomic Random Forest 0.85 (0.81-0.89) 0.79 0.82 Clinical + ECG/GLS XGBoost 0.89 (0.86-0.91) 0.78 0.81 Multi-Modal (Full) XGBoost 0.93 (0.90-0.95) 0.88 0.89
No takes yet. Share an insight, caveat, or question.
Masab Mansoor (2026) conducted a cohort in Late-onset cardiomyopathy in pediatric cancer survivors (n=1,217). Multi-Modal machine learning model vs. Clinical-Only model was evaluated on Cardiomyopathy, defined as left ventricular ejection fraction (LVEF) <50% or ≥10% absolute decline from baseline (AUC, 95% CI 0.90-0.95). A multi-modal machine learning model integrating clinical, genomic, and imaging data predicted late-onset cardiomyopathy with superior accuracy compared to a clinical-only model (AUC 0.93 vs 0.70).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: