The Effectiveness of Gold-Standard ASD Diagnostic Tools in Phelan-Mcdermid Syndrome

Background: Phelan-McDermid syndrome (PMS), also known as 22q13 deletion syndrome, is a rare neurodevelopmental disorder caused by loss of one functional copy of SHANK3 on the terminal end of chromosome 22q. Individuals diagnosed with PMS frequently demonstrate global development delay, intellectual disability of varying degrees, hypotonia, significant speech and language delays (including absent speech), and behaviors characteristic of autism spectrum disorder (ASD) (De Rubeis et al., 2018; Kolevzon et al., 2014; Soorya et al., 2013). Recent studies using gold-standard diagnostic instruments have revealed high rates of clinical ASD diagnoses in individuals with PMS, ranging from 69-84% (De Rubeis et al., 2018; Laura et al., 2018; Soorya et al., 2013). However, due to severe intellectual impairments, language delays, and repetitive behaviors seen in many individuals with PMS, a greater proportion of patients within this population exceed the cut-off scores on gold-standard diagnostic tools (i.e., Autism Diagnostic Observation Schedule (ADOS) and Autism Diagnostic Interview-Revised (ADI-R)) than actually go on to receive a diagnosis of the disorder (De Rubeis et al., 2018; Laura et al., 2018). Thus, due to complex developmental and behavioral profiles, careful integration of standardized assessment, caregiver report, and structured observations are necessary components of diagnostic evaluations of individuals with PMS.

Objectives: This study aimed to investigate the sensitivity and specificity of the ADOS-2 and ADI-R with regard to detecting ASD in a sample of children and adolescents with PMS who have intellectual disability and limited language abilities.

Methods: Thirty-six participants between 2 and 18 years old (mean age=8 years, 2 months, SD=53.4 months; 58% male) with PMS underwent comprehensive diagnostic evaluations. Participants were administered the ADOS-2 (Toddler Module: 6%; Module 1: 78%; Module 2: 17%) by a research-reliable clinician as well as a developmentally-appropriate assessment of cognitive functioning (Mullen Scales of Early Learning: 78%; Differential Ability Scales-II: 8%; Stanford-Binet-5: 14%). Developmental quotients (DQs) were calculated based on Mullen age equivalents. Participants’ caregivers completed the Vineland Adaptive Behavior Scales-II (VABS-II) and the ADI-R with a research-reliable clinician. Consensus diagnoses of ASD accounted for test results, expert clinical judgment, and DSM-5 criteria.

Results: Cognitive assessment results revealed intellectual disability in all participants. Both the ADOS-2 and ADI-R (total scores) demonstrated high sensitivity in this population (100.0% and 82.8%, respectively); however, the specificity rate of both measures was poor (ADOS-2 specificity=28.6%; ADI-R specificity=57.1%). False positives on both measures had significantly higher IQs/DQs and VABS-II adaptive behavior composite scores than true positives (p<.05).

Conclusions: In a sample of children and adolescents with PMS, the ADOS-2 and ADI-R correctly identified ASD in most participants who ultimately received a clinical diagnosis. However, the rate at which the measures produced negative results for participants who did not receive clinical diagnoses was significantly lower than the specificity rates found in past studies of the measures with participants with idiopathic ASD (Lord et al., 1997; McCrimmon & Rostad, 2014). Thus, these tools must be used in conjunction with expert clinical judgement during diagnostic evaluations of individuals within this population.