Evaluation of the Diagnostic Yield of Exome-Based Panels for Congenital Heart Defects in Different Clinical Settings
Maxim Verlee, Marrit M. Hitzert, Rosa Laura van Loon, Laura Muiño Mosquera, Katya De Groote, Julie De Backer, Laurent Demulier, Ilse Meerschaut, Jan D.H. Jongbloed, Yvonne J. Vos, Johannes M. Douwes, Rosalie L. Neijzen, Dennis Dooijes, Johannes M.P.J. Breur, Els De Vogelaere, Griet Vermeulen, Sofie Symoens, Paul Coucke, Wilhelmina S. Kerstjens-Frederikse, Bert CallewaertBACKGROUND:
Virtual panel analysis (VPA) of exome data is a common approach for the molecular diagnosis of congenital heart disease (CHD). However, differences in gene panel composition and patient inclusion criteria limit the evaluation of its diagnostic utility. This study aims to assess the diagnostic yield of VPA in a cohort of patients with CHD across 3 academic centers.
METHODS:
We collected clinical data including phenotypic features and family history, from 853 probands with CHD who underwent VPA analysis at the Center for Medical Genetics Ghent (525 probands; 471 genes), the University Medical Center Groningen (195 probands; 345 genes), and the University Medical Center Utrecht (133 probands; 55 genes). We evaluated the diagnostic yield by comparing the 3 centers with respect to panel composition and clinical presentation.
RESULTS:
The Center for Medical Genetics Ghent reported a higher diagnostic yield (9.9%) compared with the University Medical Center Groningen (7.2%) and the University Medical Center Utrecht (5.3%). In all centers, the diagnostic yield was higher in patients presenting with a syndromic constellation and did not differ significantly between the sporadic and familial cases. In 1.7% of the 536 nonsyndromic probands, a molecular cause was identified that typically is associated with syndromic CHD. Twelve genes showed likely pathogenic or pathogenic variants in multiple patients and contributed to 56.2% of the identified causes.
CONCLUSIONS:
We report an overall diagnostic yield of VPA for CHD of 8.6%, to which only a few genes contribute significantly, highlighting the complex origin of CHD. Since panel size, gene panel content, and local practices largely affect the diagnostic yield, we propose a (minimum) core gene panel for suspected isolated CHD, as well as a coordinated testing strategy for CHD to improve diagnosis and counseling and to catalyze collaborative efforts.