Understanding the Comorbidity of Autism Spectrum Disorder and Fragile X Syndrome: Moving Beyond a Categorical Approach

Background: Fragile X syndrome (FXS), the leading inherited form of intellectual disability, results from a mutation in the FMR1 gene on the X chromosome, which leads to a reduction in the associated protein (FMRP). Autism spectrum disorder (ASD) is a common comorbid condition in FXS. The prevalence of ASD in FXS has been estimated at >50%, with FXS accounting for 3-6% of ASD cases. It has been assumed that ASD symptoms reflect the same underlying psychological and neurobiological impairments in FXS and idiopathic ASD and, therefore, drugs showing benefit for FXS will also be beneficial for idiopathic ASD.

Objectives: In contrast, we have previously presented data documenting important differences between individuals with comorbid FXS and ASD and those with idiopathic ASD in terms of ASD symptoms, behavioral and psychiatric correlates, and developmental trajectories. In this presentation, we will discuss three new studies, each involving a different cohort, designed to further clarify the comorbidity of FXS and ASD. The studies entail comparisons of FXS and idiopathic ASD in terms of profiles of language impairments and predictors of impairment (Study 1), analysis of predictors of variation in the developmental trajectories of social-communication impairments within FXS (Study 2), and analysis of predictors of variation in ASD symptoms within FXS (Study 3).

Methods: In Study 1, we examined scores on measures of structural language (i.e., receptive and expressive vocabulary and grammar) in 4- to 10-year-old males with FXS (n = 51) or idiopathic ASD (n= 36), as well as the predictors of those scores, which included age, IQ, and ASD symptom severity. In Study 2, we examined predictors of within-syndrome variation in the developmental trajectory of social-communication using an experimental measure of the skills involved in correcting communicative misunderstandings. Participants in Study 2 were males with FXS (n = 36), ages 10-16 years and followed longitudinally. In Study 3, we examined the prediction of social-affective and restricted and repetitive behavior symptom severity (measured by the ADOS) from a variety of factors, including language ability, nonverbal cognitive ability, and FMRP. Males with FXS (n = 58), ages 15-23 years, participated in Study 3.

Results:  In Study 1, we found a different profile of language impairments and different predictors of language impairment in FXS and idiopathic ASD, even when the FXS sample was limited to those with an ASD diagnosis. In Study 2, we found that individual differences in this measure of social-communication were related to differences in nonverbal cognitive ability and FMRP, but not to ASD symptoms (contrary to expectations for idiopathic ASD). In Study 3, we found that language ability and nonverbal cognitive ability predicted social-affective and restricted and repetitive behavior symptom severity, respectively, which is unlike the pattern reported for idiopathic ASD.

Conclusions: These studies reinforce the notion that there are clinically and mechanistically important differences between FXS and idiopathic ASD that are masked by the categorical diagnosis of ASD. Thus, we argue for a symptom-based approach rather than a categorical diagnosis-based approach future research, including in studies of treatment efficacy.