Recurrent copy number variations as risk factors for neurodevelopmental disorders: critical overview and analysis of clinical implications Fátima Torres, 1,2 Mafalda Barbosa, 3,4 Patrícia Maciel 5,6 1 CGC Genetics, Porto, Portugal 2 Instituto de Ciências Biomédicas Abel Salazar, Universidade do Porto, Porto, Portugal 3 Department of Genetics and Genomic Sciences, The Mindich Child Health & Development Institute, The Seaver Autism Center for Research and Treatment, Icahn School of Medicine at Mount Sinai, New York, USA 4 Instituto Gulbenkian de Ciência, Oeiras, Portugal 5 Life and Health Sciences Research Institute (ICVS), School of Health Sciences, University of Minho, Braga, Portugal 6 ICVS/3B’s—PT Government Associate Laboratory, Braga/ Guimarães, Portugal Correspondence to Professor Patrícia Maciel, Life and Health Sciences Research Institute (ICVS), School of Health Sciences, University of Minho, 4710-057 Braga, Portugal; pmaciel@ecsaude. uminho.pt Received 11 August 2015 Revised 27 September 2015 Accepted 28 September 2015 To cite: Torres F, Barbosa M, Maciel P. J Med Genet Published Online First: [ please include Day Month Year] doi:10.1136/ jmedgenet-2015-103366 ABSTRACT Neurodevelopmental disorders (NDs) encompass a spectrum of neuropsychiatric manifestations. Chromosomal regions 1q21.1, 3q29, 15q11.2, 15q13.3, 16p11.2, 16p13.1 and 22q11 harbour rare but recurrent CNVs that have been uncovered as being important risk factors for several of these disorders. These rearrangements may underlie a broad phenotypical spectrum, ranging from normal development, to learning problems, intellectual disability (ID), epilepsy and psychiatric diseases, such as autism spectrum disorders (ASDs) and schizophrenia (SZ). The highly increased risk of developing neurodevelopmental phenotypes associated with some of these CNVs makes them an unavoidable element in the clinical context in paediatrics, neurology and psychiatry. However, and although finding these risk loci has been the goal of neuropsychiatric genetics for many years, the translation of this recent knowledge into clinical practice has not been trivial. In this article, we will: (1) review the state of the art on recurrent CNVs associated with NDs, namely ASD, ID, epilepsy and SZ; (2) discuss the models used to dissect the underlying neurobiology of disease, (3) discuss how this knowledge can be used in clinical practice. INTRODUCTION Neurodevelopmental disorders (NDs) are a large group of clinical entities encompassing a spectrum of neuropsychiatric manifestations caused by dis- ruption of brain development, including autism spectrum disorders (ASD), intellectual disability (ID), communication disorders, attention deficit and hyperactivity disorder (ADHD), specific learn- ing disorders and motor disorders. 1 Schizophrenia (SZ) has also been proposed to result from neuro- developmental disturbances, usually manifesting only in the adult stage. 2 The majority of NDs do not fit the Mendelian disease model where one gene is responsible for a given trait. 3 Most of them are polygenic or multifactorial and their clustering in families is believed to be influenced by genetic and environmental factors. 4 Two contrasting hypotheses have been advanced to explain the nature of this complexity: the common variant common disease (CVCD) and the rare variant common disease (RVCD) models. 5 According to the CVCD model, the genetic risk in an individual is attributable to many high frequency variants, each one having a modest effect on risk. In contrast, the RVCD model states that genetic risk in a given individual can be explained by rare mutations that confer significant risk. 56 Most likely, both types of contribution are important; the narrow-sense heritability in autism is ∼52.4%, most being due to common variants. 7 Rare CNVs—DNA segments larger than 1 Kb that present a copy number different from that of the reference genome 8 —contribute to a substantial pro- portion of the genetic variability in humans 9 but can also contribute for risk of developing a neuro- developmental disturbance. Its association with a range of NDs 5 10 was only possible because advancements in chromosomal microarray (CMA) technology have allowed for CNV analysis in very large case-control cohorts. 11 A significant proportion of risk for ID, ASD, SZ, epilepsy, bipolar disease (BD) and ADHD can be explained by these rare variants. 12–20 The estimated risk, or OR, for most common disease-associated single nucleotide polymorphisms will be of—at most—up to 2 (with many between 1.1 and 1.4); in contrast, many—if not most—rare variants have been associated with ORs greater than 2, in some cases considerably larger. 6 Most of the experiments to study the impact of CNVs in dosage-sensitive gene expression in normal brain development have used lymphoblas- toid cell lines, suggesting a functional impact of CNVs via transcriptome alterations. 21–23 Two studies on postmortem brain samples have shown that 1q21.1 and 22q11.2 CNVs influence gene expression in the dorsolateral prefrontal cortex; 24 25 interestingly, a significant proportion of CNVs influencing gene expression in the human prefrontal cortex were located in chromosomal regions implicated in psychiatric disorders, namely those in 1q21.1, 3q29, 15q11.2, 16p11.2, 16p13.1, 17q12 and 22q11.2. 25 Most recurrent pathogenic CNVs are large (>400 kb), typically involving dozens of genes, and are individually rare (frequency <0.1%). 11 Their dis- covery emphasised the importance of de novo and essentially private mutations in NDs, and indicated that the distinction between milder neuropsychiatric conditions and severe developmental impairment may be a consequence of increased mutational burden affecting multiple genes in the latter case. 3 Although finding such risk loci has been the goal of neuropsychiatric genetics for many years, the translation of this recent knowledge into clinical practice has not been trivial. In this article, we will: (1) review the state of the art on recurrent CNVs associated with NDs, namely ASD, ID, epilepsy and SZ; (2) discuss the models used to dissect the underlying mechanism of disease, (3) discuss how this knowledge can be used in clinical practice. Torres F, et al. J Med Genet 2015;0:1–18. doi:10.1136/jmedgenet-2015-103366 1 Review JMG Online First, published on November 7, 2015 as 10.1136/jmedgenet-2015-103366 Copyright Article author (or their employer) 2015. Produced by BMJ Publishing Group Ltd under licence.