Risk factors for 1-year allograft loss in pediatric heart transplant patients using machine learning: An analysis of the pediatric heart transplant society database.

BACKGROUND: Pediatric heart transplant patients are at greatest risk of allograft loss in the first year. We assessed whether machine learning could improve 1-year risk assessment using the Pediatric Heart Transplant Society database. METHODS: Patients transplanted from 2010 to 2019 were included. The primary outcome was 1-year graft loss free survival. We developed a prediction model using cross-validation, by comparing Cox regression, gradient boosting, and random forests. The modeling strategy with the best discrimination and calibration was applied to fit a final prediction model. We used Shapley additive explanation (SHAP) values to perform variable selection and to estimate effect sizes and importance of individual variables when interpreting the final prediction model. RESULTS: Cumulative incidence of graft loss or mortality was 7.6%. Random forests had favorable discrimination and calibration compared to Cox proportional hazards with a C-statistic (95% confidence interval [CI]) of 0.74 (0.72, 0.76) versus 0.71 (0.69, 0.73), and closer alignment between predicted and observed risk. SHAP values computed using the final prediction model indicated that the diagnosis of congenital heart disease (CHD) increased 1 year predicted risk of graft loss by 1.7 (i.e., from 7.6% to 9.3%), need for mechanical circulatory support increased predicted risk by 2, and single ventricle CHD increased predicted risk by 1.9. These three predictors, respectively, were also estimated to be the most important among the 15 predictors in the final model. CONCLUSIONS: Risk prediction models used to facilitate patient selection for pediatric heart transplant can be improved without loss of interpretability using machine learning.

Treatment Strategies for Cardiomyopathy in Children: A Scientific Statement From the American Heart Association.

This scientific statement from the American Heart Association focuses on treatment strategies and modalities for cardiomyopathy (heart muscle disease) in children and serves as a companion scientific statement for the recent statement on the classification and diagnosis of cardiomyopathy in children. We propose that the foundation of treatment of pediatric cardiomyopathies is based on these principles applied as personalized therapy for children with cardiomyopathy: (1) identification of the specific cardiac pathophysiology; (2) determination of the root cause of the cardiomyopathy so that, if applicable, cause-specific treatment can occur (precision medicine); and (3) application of therapies based on the associated clinical milieu of the patient. These clinical milieus include patients at risk for developing cardiomyopathy (cardiomyopathy phenotype negative), asymptomatic patients with cardiomyopathy (phenotype positive), patients with symptomatic cardiomyopathy, and patients with end-stage cardiomyopathy. This scientific statement focuses primarily on the most frequent phenotypes, dilated and hypertrophic, that occur in children. Other less frequent cardiomyopathies, including left ventricular noncompaction, restrictive cardiomyopathy, and arrhythmogenic cardiomyopathy, are discussed in less detail. Suggestions are based on previous clinical and investigational experience, extrapolating therapies for cardiomyopathies in adults to children and noting the problems and challenges that have arisen in this experience. These likely underscore the increasingly apparent differences in pathogenesis and even pathophysiology in childhood cardiomyopathies compared with adult disease. These differences will likely affect the utility of some adult therapy strategies. Therefore, special emphasis has been placed on cause-specific therapies in children for prevention and attenuation of their cardiomyopathy in addition to symptomatic treatments. Current investigational strategies and treatments not in wide clinical practice, including future direction for investigational management strategies, trial designs, and collaborative networks, are also discussed because they have the potential to further refine and improve the health and outcomes of children with cardiomyopathy in the future.

Current State of Pediatric Cardio-Oncology: A Review.

The landscape of pediatric oncology has dramatically changed over the course of the past several decades with five-year survival rates surpassing 80%. Anthracycline therapy has been the cornerstone of many chemotherapy regimens for pediatric patients since its introduction in the 1960s, and recent improved survival has been in large part due to advancements in chemotherapy, refinement of supportive care treatments, and development of novel therapeutics such as small molecule inhibitors, chimeric antigen receptor T-cell therapy, and immune checkpoint inhibitors. Unfortunately, many cancer-targeted therapies can lead to acute and chronic cardiovascular pathologies. The range of cardiotoxicity can vary but includes symptomatic or asymptotic heart failure, arrhythmias, coronary artery disease, valvar disease, pericardial disease, hypertension, and peripheral vascular disease. There is lack of data guiding primary prevention and treatment strategies in the pediatric population, which leads to substantial practice variability. Several important future research directions have been identified, including as they relate to cardiac disease, prevention strategies, management of cardiovascular risk factors, risk prediction, early detection, and the role of genetic susceptibility in development of cardiotoxicity. Continued collaborative research will be key in advancing the field. The ideal model for pediatric cardio-oncology is a proactive partnership between pediatric cardiologists and oncologists in order to better understand, treat, and ideally prevent cardiac disease in pediatric oncology patients.

Administration of Dexrazoxane Improves Cardiac Indices in Children and Young Adults With Acute Myeloid Leukemia (AML) While Maintaining Survival Outcomes.

Anthracycline-induced cardiotoxicity remains a significant contributor to late morbidity/mortality in children and young adults with acute myeloid leukemia (AML). The cardioprotectant dexrazoxane can be used as prophylaxis to diminish risk for cardiomyopathy but whether it affects risk of relapse in pediatric AML is unclear. Our institution adopted the use of dexrazoxane before anthracyclines administration for all oncology patients in 2011. We compared patients with AML (ages, 0 to 21 y) who received or did not receive dexrazoxane during the years 2008 to 2013. In total, 44 patients with AML (ages, 4.5 mo to 21.7 y) were included. We identified no statistical difference in 2-year event rate (62% vs. 50%, P=0.41) or 2-year overall survival (69% vs. 69%, P=0.53) between patients receiving (n=28) or not receiving (n=16) dexrazoxane. Ejection fraction (P=0.0262) and shortening fraction (P=0.0381) trended significantly higher in patients that received dexrazoxane compared with those that did not receive dexrazoxane. Utilization of the cardioprotectant dexrazoxane before anthracycline chemotherapy in pediatric patients with AML demonstrated no significant difference in either event rate or overall survival relative to institutional controls and seems to improve cardiac function indices. Further studies in this patient population are needed to confirm these findings.

Early therapy-related myeloid sarcoma and deletion of 9q22.32 to q31.1.

Survival following childhood neuroblastoma is improving with low rates of secondary myeloid neoplasms. We describe a 13-month-old male with intermediate risk neuroblastoma who developed an isolated scalp therapy-related myeloid sarcoma (t-MS). Developmental delays and two distinct malignancies prompted constitutional evaluation. Chromosomal microarray identified a 7.3 Mb deletion of 9q22.32 to 9q31.1. He remains in remission 11 months following hematopoietic cell transplant. Unusual presentations of rare diseases necessitate a multidisciplinary approach and adaptation of standardized protocols to accommodate increased risks imposed by genetic variants.