Risk Pathways to Autistic Traits and Autism Spectrum Disorder in Tuberous Sclerosis Complex (TSC)

Background: Tuberous sclerosis complex (TSC) is an autosomal dominant genetic condition caused by mutation of either the TSC1 or TSC2 gene, and characterised by tumour-like lesions in the skin, brain and other organs. The majority of children with TSC have epilepsy and around half have autism spectrum disorder (ASD) and/or intellectual disability. While the risk for ASD is elevated in TSC, there is marked variability in outcome, from autism, through the broader autism phenotype, to unaffected. It is not clear what causes this variability in outcome or what the underlying mechanisms might be. A number of risk factors have been associated with ASD, such as mutation, number of tubers in the cerebral cortex, and epilepsy. However, it is not known how these risk factors act together to influence the risk for ASD.  

Objectives: To investigate how multiple risk factors, including mutation (TSC1 vs. TSC2), cortical tuber load, and type and severity of epilepsy, act together to increase the risk for ASD.  

Methods: TS2000 is the first UK population-representative, prospective longitudinal study to chart the development of TSC throughout childhood. All cases identified as newly diagnosed during 2001-2006 were recruited (N=125) and have been followed for up to 14 years. Genotyping was carried out to determine the causal mutation; cortical tuber load was determined using brain MRI or CT scans; detailed seizure history was taken from parents and supplemented with information from medical records to rate seizure severity (Early Childhood Epilepsy Severity Scale, E-Chess); children were assessed with the Autism Diagnostic Interview-Revised (ADI-R) and the Autism Diagnostic Observation Schedule (ADOS-2) to characterise ASD outcomes.  

Results: Forty-three percent of participants met criteria for ASD on both ADI-R and ADOS-2, and a further 25% met criteria on one instrument. Thirteen percent showed subthreshold autistic traits. A minority (20%) did not meet criteria on either the ADI-R or ADOS-2. Mutation (TSC1 vs. TSC2) was not significantly associated with ADI-R or ADOS-2 scores (t(64)=0.34, p=.72; t(62)=0.51, p=.61). Higher ADI-R and ADOS-2 scores were associated with: higher cortical tuber load (ADI-R rho=.23, p=.048; ADOS-2 rho=.24, p=.038); epilepsy severity in the 1^st and 2^nd years of life, at age 3+ years and age 7+ years (rho .25-.48, all p<.03); and a history of infantile spasms (d=0.73, p<.001; d=0.81, p<.001). Structural equation modelling suggested a risk pathway from mutation to cortical tuber load (TSC2 predicted higher tuber load), to epilepsy severity in the first two years of life, to ASD risk.  

Conclusions: Children with TSC are at very high risk for ASD, as well as social/communication difficulties that fall short of diagnosis of ASD but might nonetheless be clinically important. Epilepsy in the first two years of life, especially that characterised by infantile spasms, early onset, and greater severity, leads to a high risk of ASD. The prompt diagnosis and treatment of seizures in infancy may improve the prognosis in these high risk children.