Identifying Endophenotypes in Autism Spectrum Disorder and Fragile X Syndrome: A Multi-Method Approach

Background: Fragile X syndrome (FXS), caused by a mutation of the FMR1 gene, is the most common single-gene disorder associated with autism spectrum disorder (ASD), with up to 75% of males with FXS meeting diagnostic criteria for ASD (FXS-ASD; Klusek et al., 2014). However, in-depth studies of specific phenotypic domains in idiopathic ASD (ASD-O) and FXS suggest that while substantial overlap may exist in the social-communicative domain (Klusek et al., 2014, Losh et al., 2012), the expression of restricted and repetitive behaviors (RRB) is more distinct (Wolff et al., 2012), suggesting that FMR1 variation may play a role in selected ASD phenotypes. To address this question, this study directly compared profiles of social communication and RRBs in individuals with FXS-ASD, FXS without ASD (FXS-O), and ASD-O, applying multiple measures of symptoms in both domains.

Objectives:  To characterize social communication and restricted and repetitive behaviors in idiopathic ASD and individuals with FXS with and without ASD using a multi-method approach.

Methods:  Eighty-four males with ASD-O, 55 males with FXS-ASD, and 16 males with FXS-O were included. The Pragmatic Rating Scale-School Age (PRS-SA; Landa, 2011) and the communication and social items of the Autism Diagnostic Interview-Revised (ADI-R) were used to evaluate social communication. RRBs were assessed using the ADI-R RRB items and the Repetitive Behavior Scale-Revised (RBS-R). Principle Component Analyses were conducted to determine the factor structure of phenotypic profiles across FXS and ASD. Correlations were examined with FMR1-molecular variation, including CGG repeat length and levels of FMRP.

Results: On detailed clinical-behavioral ratings of pragmatic language (i.e., PRS-SA), individuals with ASD-O and FXS-ASD did not differ, and demonstrated significantly greater rates of impairment in domains of conversational reciprocity, redundancy, and disinhibition compared to males with FXS-O (ps<.05). For parent-reported difficulties in communication (ADI-R) a stepwise pattern emerged in which males with ASD-O demonstrated the greatest impairment, followed by males with FXS-ASD and then males with FXS-O (ps<.05). In contrast, on both the ADI-R and RBS-R, males with ASD-O demonstrated distinct profiles of RRBs, including significantly greater rates of behaviors associated with insistence on sameness, sensory interests and aversions, compulsions, and self-injury compared to males with FXS-ASD and FXS-O (ps<.05). Additionally, males with FXS-ASD demonstrated greater impairments in sensory seeking and aversive behaviors than those with FXS-O. Variation in FMR1 expression was related to greater impairments in social communication in all three groups, and greater rates of repetitive sensorimotor behaviors in FXS-O (ps<.05).

Conclusions:  Findings confirmed substantial overlap between ASD and FXS-ASD in the social-communicative domain, and revealed significant relationships between social communication and FMR1-related variation. In contrast, distinct profiles of RRBs were observed, indicating syndrome-specific areas of difficulty that are impacted by increased ASD symptomatology in FXS. Together, findings suggest that social communication impairments may constitute an important overlapping endophenotype in ASD and FXS, which may be used to clarify the complex etiology of ASD.