Turner Syndrome: A New Research Model for ASD in Girls?

Background: Autism Spectrum Disorders (ASD) are rarely diagnosed in high-functioning females with normal/high verbal IQ. Consequently, there have been few studies on this female phenotype across the life-span.

Turner Syndrome (45,X; TS) is a relatively common sex chromosome aneuploidy (4:10,000 live births). Affected females often experience social interaction difficulties that resemble those of idiopathic ASD. Yet most have normal-range verbal intelligence and they typically possess good formal language skills.

Objectives: We compared and contrasted the phenotypic profiles of TS females (5-18 years of age) with an age/VIQ matched sample of females diagnosed with idiopathic ASD.

Methods: Participants with TS (N=70) were recruited nationally from support groups and paediatric clinics. They were participants in the Social Skills and Relationships in Turner Syndrome study (SOAR), but they were not selected because of any previously diagnosed social communication disorder. Participants with idiopathic ASD (N=35) were recruited from the National Centre for High Functioning Autism (NCHFA-London, UK). The idiopathic-ASD and TS groups were matched for chronological age (t(54)=-0.71, p=0.48) and mental age (t(50)=-0.38, p=0.7). All had verbal IQ in the normal range. ASD was diagnosed in all cases by clinicians, using standardized procedures applying DSM-5 criteria. Participants were also rated according to the Strengths and Difficulties Questionnaire (SDQ) and the Social Responsiveness Scale 2 (SRS-2).

Results: 22.6% of TS females were diagnosed with ASD in the course of the SOAR study. There were no significant group differences on SDQ total scores, a broad ranging measure of emotional and behavioural adjustment, but the ASD-TS group were rated as having greater peer difficulties than the idiopathic-ASD group (t(52)=-3.95, p<0.0001). There were no significant group differences on the SRS-2 total score; only 14% of the TS participants obtained a ‘normal-range’ score. Overall, the TS group with ASD were more impaired than the idiopathic ASD females, in terms of social awareness, social cognition, social communication and autistic mannerisms but they scored similarly for social motivation. Group differences were not statistically significant but effects sizes ranged from small to medium.

Conclusions: 86% of a large and putatively representative sample of TS females, recruited from paediatric endocrinology clinics, experienced autistic-like social interaction difficulties of considerable severity, likely to impact their everyday life. One in five met DSM-5 criteria for an ASD. Phenotypic comparisons with females matched for age and IQ showed a similar range and severity of symptomatology. We provide preliminary evidence indicating females with TS are at high risk of ASD despite normal verbal intelligence. They represent a suitable model syndrome for interventions aimed at remediating social interaction difficulties in females with high-functioning ASD.