How Frequent Is Loss of Skills in ASD Associated with Genetic Abnormalities?

Saturday, May 13, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
A. Thurm1, S. L. Bishop2, C. Farmer1, S. J. Sanders3, L. Soorya4, L. Berry5, C. P. Schaaf6, M. A. Gillentine7, A. R. Simon5 and R. P. Goin-Kochel8, (1)National Institute of Mental Health, Bethesda, MD, (2)Psychiatry, University of California San Francisco, San Francisco, CA, (3)UCSF, San Francisco, CA, (4)Department of Psychiatry, RUMC, Chicago, IL, (5)Baylor College of Medicine, Houston, TX, (6)Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, (7)Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, (8)Pediatrics, Baylor College of Medicine, Houston, TX
Background:  The rate of identification of genetic abnormalities in individuals diagnosed with ASD is rising. Functions of the involved genes include brain development and metabolic processes (de la Torre-Ubieta, Won, Stein, & Geschwind, 2016). The majority of extant literature on regression, or loss of skills, in ASD included samples considered “idiopathic,” and thus has not explored if and how genetic abnormality association relates to specific patterns of symptom onset or loss. While we do know that regression may be associated with specific genetic vulnerabilities related to ASD and/or intellectual disability (Molloy, Keddache, & Martin, 2005; Neul et al., 2014), full exploration requires both large data sets as well as data from cohorts of specific genetic disorders associated with ASD.

Objectives:  The purpose of this study was to investigate milestone attainment and loss of skills in individuals with genetic findings known to be associated with ASD, and in other specific genetic conditions.

Methods:  First, we explored findings from the Simons Simplex Collection (SSC), including 112 probands with at least one de novo loss of function mutation copy number variants in, or including, a high confidence (HC) ASD gene or locus (an age, IQ, and sex-matched control group was selected from the remaining non-HC [None] group). Second, we report data from several natural history studies of genetic disorders, including Phelan-McDermid Syndrome (22q13 deletion or mutation; N=39; mean age 7.5+4.5 years), 15q13.3 deletion/duplication syndromes (Ndel=18; mean age 14.5+1.7 years; Ndup=18; mean age 9.9+3.0 years respectively), 16p11.2 CNVs, and 1q21.1 CNVs. The ADI-R was used to determine rates and pattern of skill attainment and loss.

Results: In the SSC cohort, the HC group had a significantly later mean age of walking than the None group. Both language and other skill loss were more common in the None group than in the HC group. Table 1 shows loss of language and other skills from preliminary data in the SSC cohort, as well as other genetic conditions. A fuller description of onset patterns, areas of loss, and current functioning in these cohorts will be provided, including the 16p11.2 and 1q21.1 cases.

Conclusions:  While skill loss was more common in the SSC cohort when no genetic abnormality was present, it was present, along with a significant attainment lag in motor skills, in the HC group as well. Skill loss was relatively common in several other genetic conditions associated with ASD. These findings challenge the notion that regression is specifically associated with idiopathic or specific genetic conditions associated with ASD.