Autism Spectrum Disorder Characteristics in Smith-Lemli-Opitz Syndrome

Background: Smith-Lemli-Opitz syndrome (SLOS) is an autosomal recessive disorder of impaired cholesterol metabolism with an estimated incidence among individuals of European ancestry of 1 in 20,000 to 1 in 60,000 births [Kelley and Hennekam, 2001]. We demonstrated that approximately 50% of individuals with SLOS meet the Diagnostic and Statistical Manual for Mental Disorders, 4^th Edition (DSM-IV) criteria for autism [Tierney et al, 2001]. Furthermore, approximately three-fourths of the children with SLOS have some variant of ASD suggesting the most consistent relationship with ASD of any single gene disorder [Sikora et al, 2006].

Objectives: The purpose of this study was to determine (1) whether the individuals with SLOS+ASD have distinctive autistic profiles as measured by the Autism Diagnostic Interview-Revised (ADI-R) and the Autism Diagnostic Observation Schedule-G (ADOS-G), (2) whether the individuals with SLOS+ASD have distinctive profiles in terms of IQ and adaptive behavior, and (3) which components of the ADI-R and ADOS-G contribute most to the ASD diagnosis.

Methods: Twenty-three participants (mean age 7.8±3.3 years) with SLOS were assessed by measures of autism features, IQ (Stanford Binet, 4^th Edition (SB-IV) or Mullen Scales of Early Learning), and adaptive function [Vineland Adaptive Behavior Scales (VABS)]. We divided the SLOS subjects according to their summary diagnoses from the ADI-R/ADOS-G/DSM-IV and the Non-Verbal Mental Age (NVMA) into 3 groups: SLOS+ASD, NVMA >24 months (n=9), SLOS+ASD, NVMA<24 months (n=7),  and SLOS+None, NVMA>24 months (n=7). Descriptive statistics were used for overall assessment of the sample. Characterization of autistic behaviors was performed by non-parametric and regression analyses.

Results: 52% of youths with SLOS met criteria for autism and 70% met criteria for ASD. Cognitive and adaptive profiles of individuals with SLOS+ASD, NVMA<24 months were significantly more impaired  than SLOS+ASD, NVMA>24 months and SLOS+None groups. Despite differences in cognition, common patterns emerged among the two SLOS+ASD groups: (1) they both were significantly different from the SLOS+None group on ADOS-G Play/Imagination and (2) they were different at a trend level on Peer Relationships (A4) subdomain of the ADI-R Social. Logistic Regression analyses showed that ADOS-G Play/Imagination (p=0.0019) and A4 of the ADI-R  Social (p=0.0005) were the most significant predictors of the ASD diagnosis.

Conclusions: Our findings demonstrate distinct autistic patterns that separated individuals with ASD from the rest of the SLOS cohort. Furthermore, our results indicate that selective impairment in more complex behaviors (i.e., peer relationships) differentiated autism related behaviors in SLOS.  Further investigation of this phenomenon in a larger sample and its implications to ASD in general is suggested.