Early Identification of Clinical Predictors for Autism Spectrum Disorders in Infants with Tuberous Sclerosis Complex: Final Results from Epistop Project

Background: One of the most common genetic causes of syndromic Autism Spectrum Disorders (ASD) is Tuberous Sclerosis Complex (TSC), a single-gene syndrome often associated with autistic phenotype and developmental delay (DD) since very early stages of life. The possibility of prenatal/perinatal diagnosis of TSC has facilitated investigation of pathogenetic mechanisms of ASD, but there are still lacking evidences of predictive factors for ASD or DD occurrence. The EPISTOP project promoted the prospective investigation of clinical predictors that could be able to individuate early deviation of developmental trajectories before the appearance of symptoms of ASD or DD/cognitive impairment.

Objectives: The main objective of the work package 7 in EPISTOP was to identify early prognostic factors of neurodevelopmental outcome in TSC patients. The specific goals were to promptly detect deviation in developmental trajectories and in specific skills potentially predictive of ASD and DD/cognitive impairment.

Methods: 97 TSC subjects were prospectively followed from 6 to 24 months of age with detailed neuropsychological assessment for evaluation of developmental level with the Bayley Scales of Infant Development (BSID) and identification of ASD risk with Autism Diagnostic Observation Schedule (ADOS). Comparisons between groups were performed, as appropriate, with two-sample t test, ANOVA models, and Spearman’s correlations. An alpha level of 0.05 was used for all statistical analyses, which were performed using SPSS v.23.0 (IBM Corp., Armonk, NY, USA).

Results: Final data at 24 months were available for 82 children. BSID results have been carefully analyzed to identify early deviations in specific areas potentially predictive for DD/cognitive impairment. In particular, impairment in fine-motor, expressive and receptive language areas at 6 months were significantly correlated to DD diagnosis at 24 months (see Table 1). Infants with TSC who were diagnosed with ASD at age 24 months had impaired fine motor quotients on the BSID-III at age 6 months and their developmental trajectories differed significantly in all areas explored (cognitive: Spearman coefficient -,446; p < 0,001; language: Spearman coefficient -,438; p < 0,001; motor: Spearman coefficient -,540; p < 0,001) at age 12 months (see Table 1). ADOS score at 12 months of age showed a significant association with final ASD outcome (positive predictive value (PPV): 57,1%; negative predictive value (NPV): 85,7%). Specific ADOS items, particularly those representing social communication deficits related to visual impairment (e.g. integration of eye contact, joint attention) were more impaired in infants with ASD compared to infants without.

Conclusions: A deviation from the normal developmental trajectory can be detected in infants with TSC at high risk for ASD from as early as age 6 months, particularly in the motor area. Impairment in cognitive level and emerging atypical behaviors in visual-related social skills at 12 months can be significative predictors of subsequent ASD at 24 months in these high-risk children. Our findings strongly support the need for prospective neuropsychological assessments in infants with TSC, in order to early identify children at high risk for ASD and to develop treatment protocols targeting social communication function, as well as specific developmental domains, before the onset of autism symptoms.