Diagnostic Exome Sequencing in Persons With Severe Intellectual Disability

Abstract

Severe intellectual disability (IQ <50) affects about 0.5% of the population in Western countries and carries a high health burden. In developed countries, most severe forms of intellectual disability are thought to have a genetic cause, but the cause is still unknown in 55% to 60% of patients. A diagnosis and understanding of a genetic cause may offer information on the prognosis, preclude further unnecessary invasive testing, and lead to appropriate access to medical and supportive care. De novo point mutations in more than 1000 different genes may cause intellectual disability. This study was undertaken to assess the role of de novo as well as X-linked and autosomal recessive inherited mutations in 100 patients with unexplained intellectual disability, using a family-based exome-sequencing approach. These individuals had prior extensive clinical and genetic evaluation, but no definitive etiologic diagnosis was possible. They had exhausted current diagnostic strategies, a common situation for patients with severe intellectual disability. The 53 female and 47 male patients had unexplained severe intellectual disability. Forty-one patients were 10 to 20 years old, 37 were younger than 10 year, and 22 were older than 20 years. Patients had undergone genomic profiling and targeted gene tests, with metabolic screening if needed, but no etiologic diagnosis was possible. Genomic DNA was isolated from blood, and exomes were enriched. Possible de novo mutations were selected by excluding variants inherited from either parent, and candidate recessive and X-linked mutations were selected by segregation analysis. Candidate de novo mutations were validated by conventional Sanger sequencing methods in DNA samples. High-throughput resequencing was used to confirm new candidate genes in 765 individuals with intellectual disability. Mutations were classified based on 7 factors: function of the affected gene, effect of the mutation on the codon, in silico prediction of the functional effect at the amino acid level, evolutionary conservation, brain expression patterns, analysis of gene ontology terms, and use of animal models. An average of 24,324 genetic variants were detected per patient. The automated prioritization scheme identified 690 candidate de novo mutations (average number per patient, 7; range, 2–20). Sanger sequencing confirmed the presence of 79 de novo mutations affecting 77 genes in 53 patients (range per patient, 1–4). Conclusive genetic diagnoses were obtained for 10 patients with de novo mutations in known intellectual disability genes and for 3 male patients with severely disruptive, maternally inherited mutations in known X-linked intellectual disability genes. No diagnostically relevant, inherited autosomal recessive mutations were identified. A diagnostic yield of 13% was obtained from mutations in known intellectual disability genes. Twenty-four novel candidate genes affected by de novo mutations were identified. DYNC1H1, GATAD2B, and CTNNB1 are novel intellectual disability genes, which raised the diagnostic yield of exome sequencing to 16%. Exome sequencing can be used as a diagnostic procedure for patients with severe intellectual disability of unknown cause. The diagnostic yield of 16% may increase as methods improve to identify other genes associated with intellectual disability.