Identifying the early signs of Autism Spectrum Disorder - integration of behavioral and genetic information for early autism detection in an at-risk population: final report

Autism Spectrum Disorder (ASD) is a common neurodevelopmental disorder (NDD) characterized by impairments in social interaction and reciprocal communication, as well as by patterns of repetitive and stereotyped behaviours. Educational interventions, language and behavioral therapies may significantly improve the patient’s prognosis, especially when initiated at a young age. A clinical diagnosis of autism, however, is a complex and lengthy process and, particularly for younger children, may be very difficult to attain even using complex diagnostic instruments by medical experts. Furthermore, diagnostic instruments generally perform well at discriminating between ASD and typically developing children, but not ASD from other neurodevelopmental disabilities. Distinguishing between ASD and other NDDs is frequently difficult through behavioural assessment at an early age, when children would most benefit from specific early intervention. Finding reliable methods for early and specific detection of ASD is thus a priority. Over the last decade much progress has been made regarding the genetic etiology of ASD. Recent studies from large research consortia showed that ASD in many cases results from a large number of highly penetrant rare variants that likely converge in a small number of affected pathways. The clinical significance of rare but recurrent submicroscopic deletions and duplications, or Copy Number Variants (CNVs), as well as single nucleotide variants (SNVs), is under study in large population samples, to establish frequencies in ASD vs control datasets, recurrence rates in ASD, segregation in families and gene content. This knowledge is slowly advancing towards translation into clinical practice, and CNV analysis is nowadays widely used for etiological diagnosis. However, questions frequently arise regarding the clinical significance of many of the variants identified, and there seems to be an extensive overlap between ASD and other NDDs. This issue has not been fully explored. A main objective of this project was therefore to understand whether there is a significant gain in diagnostic yield by adding genomic data, namely data from Copy Number Variant analysis, to the normally extensive clinical assessment required for the clinical diagnosis of Autism Spectrum Disorder. We further seek to understand whether genomic data could improve early differentiation between ASD and other NDDs, specifically targeting very young children. A final aim would be to define a screening method for early detection of ASD, integrating behavioural assessment instruments and genomic CNV analysis that could be a useful tool for clinicians.FCT - PTDC/SAL-SAP/119161/201