Exome Sequencing Has a High Diagnostic Rate in Sporadic Congenital Hypopituitarism and Reveals Novel Candidate Genes

Abstract Context The pituitary gland is key for childhood growth, puberty, and metabolism. Pituitary dysfunction is associated with a spectrum of phenotypes, from mild to severe. Congenital hypopituitarism (CH) is the most commonly reported pediatric endocrine dysfunction, with an incidence of 1:4000, yet low rates of genetic diagnosis have been reported. Objective We aimed to unveil the genetic etiology of CH in a large cohort of patients from Argentina. Methods We performed whole exome sequencing of 137 unrelated cases of CH, the largest cohort examined with this method to date. Results Of the 137 cases, 19.1% and 16% carried pathogenic or likely pathogenic variants in known and new genes, respectively, while 28.2% carried variants of uncertain significance. This high yield was achieved through the integration of broad gene panels (genes described in animal models and/or other disorders), an unbiased candidate gene screen with a new bioinformatics pipeline (including genes with high loss-of-function intolerance), and analysis of copy number variants. Three novel findings emerged. First, the most prevalent affected gene encodes the cell adhesion factor ROBO1. Affected children had a spectrum of phenotypes, consistent with a role beyond pituitary stalk interruption syndrome. Second, we found that CHD7 mutations also produce a phenotypic spectrum, not always associated with full CHARGE syndrome. Third, we add new evidence of pathogenicity in the genes PIBF1 and TBC1D32, and report 13 novel candidate genes associated with CH (eg, PTPN6, ARID5B). Conclusion Overall, these results provide an unprecedented insight into the diverse genetic etiology of hypopituitarism.