Hypertrophic Cardiomyopathy Genotype Prediction Models in a Pediatric Population

Randa Newman, John Lynn Jefferies, Clifford Chin, Hua He, Amy Shikany, Erin M. Miller, Ashley Parrott

The Toronto Hypertrophic Cardiomyopathy (HCM) Genotype Score and Mayo HCM Genotype Predictor are risk assessment models developed to estimate a patient's likelihood of testing positive for a pathogenic variant causative of HCM. These models were developed from adult populations with HCM based on factors that have been associated with a positive genotype and have not been validated in external populations. The purpose of this study was to evaluate the overall predictive abilities of these models in a clinical pediatric HCM setting. A retrospective medical record review of 77 pediatric patients with gene panel testing for HCM between September 2005 and June 2015 was performed. Clinical and echocardiographic variables used in the developed models were collected and used to calculate scores for each patient. To evaluate model performance, the ability to discriminate between a carrier and non-carrier was assessed by area under the ROC curve (AUC) and overall calibration was evaluated by the Hosmer-Lemeshow goodness-of-fit statistic. Discrimination assessed by AUC was 0.72 (P < 0.001) for the Toronto model and 0.67 (P = 0.004) for the Mayo model. The Toronto model and the Mayo model showed P values of 0.36 and 0.82, respectively, for model calibration. Our findings suggest that these models are useful in predicting a positive genetic test result in a pediatric HCM setting. They may be used to aid healthcare providers in communicating risk and enhance patient decision-making regarding pursuit of genetic testing.