Chromosomal Microarray Analysis In Patients with Confirmed Diagnosis of Autism Spectrum Disorder: Diagnostic Yield and Clinical Characteristics

Background: Studies on Autism Spectrum Disorders (ASD) show a strong genetic predisposition, with multiple risk genes identified or postulated.  Currently, G-banded karyotype and fragile X DNA testing are recommended for all patients with ASD.  Guidelines for chromosomal microarray analysis (CMA) in the absence of dysmorphology and focal neurologic deficits have not been established; however recent reports suggest a high yield in ASD.

Objectives: We aimed to determine the yield of CMA, karyotype, and Fragile X testing in patients with a confirmed diagnosis of ASD and to determine whether any clinical characteristics separate patients with abnormal CMA.

Methods: Electronic Records from 2000 to 2010 from a single pediatrician autism specialty practice were reviewed.  All patients in this practice were offered genetic testing including karyoptype, Fragile X and CMA. CMA was recommended in place of karyotype to all patients after 2007. Abnormal CMA results were subclassified as clinically significant based upon the lab interpretation and literature review.

Patients with a diagnosis of autism confirmed by one or more psychological tests (Autism Diagnostic Observation Scale, Child Autism Rating Scale, Gilliam Autism Rating Scale) and DSM-IV clinical criteria were selected and clinical, cognitive, behavioral and developmental information was collected. Cognitive results were stratified into above, average, borderline and intellectual disability.  Contingency table analyses using Chi square  tested for significance in the relationship between CMA result and cognitive level, dysmorphology, seizures or congenital malformations.  

Domains from parental report scales, including the Parental Concerns Questionnaire (PCQ), Repetitive Behavior Scale (RBS), Child Behavior Checklist (CBCL), and psychologist administered Vineland were used to assess specific behavioral and developmental symptoms. T-tests compared CMA results (independent variable) and Vineland subdomains (Communication, Daily Living Skills, Socialization, Motor Skills, and Adaptive Behavior Composite); CBCL subdomains; RBS subdomains (Ritualistic/Sameness, Stereotypic, Self-injurious behaviors, Compulsive behaviors, and Restricted Interests); and Parental Concerns Questionnaire items (language, compulsion, anxiety, aggression, sleep, hyperactivity, attention, mood, social, self stimulatory, self injury)

Results: 90 CMA, 117 Karyotype and 174 Fragile X results were identified. There were 19 (21%) abnormal CMA results with 7 (8%) classified as clinically significant. 3 (2.5%) of 119 karyotypes were reported as abnormal, (2 translocation carriers and 1 gain). 1(0.6%) Fragile X positive was identified. There was no relationship between CMA result and cognitive level, presence of seizure disorder, dysmorphology, or congenital malformations (p>0.5) and developmental/ behavioral categories of obsession, compulsion, aggression, self-injury, repetitive behaviors, attention, hyperactivity, sleep disorders, mood, social interactions, language.  There were lower PCQ anxiety scores (p=.03) and less  CBCL externalizing  behavior (F_1,50=6.02, p=.02)  in the group with clinically significant CMA as compared to the other two groups.

Conclusions: CMA has a superior diagnostic yield as compared to Fragile X and karyotype in unselected patients with ASD. No clinical, intellectual   or developmental characteristics separate the groups of ASD patients with abnormal or clinically significant CMA results from the group with normal results.  Behavioral symptoms are identical or improved in the clinically significant CMA group.