Autism spectrum disorder symptomatology in children with neurofibromatosis type 1


  • Jonathan M Payne

    1. Institute for Neuroscience and Muscle Research, The Children's Hospital at Westmead, Sydney, NSW
    2. Discipline of Paediatrics and Child Health, Faculty of Medicine, University of Sydney, Sydney, NSW, Australia
    Search for more papers by this author

  • This commentary is on the original articles by Walsh et al. and Garg et al. on pages 131–138 and 139–145 of this issue.


This commentary is on the original articles by Walsh et al. and Garg et al. on pages 131–138 and 139–145 of this issue

The importance of genetic factors in autism spectrum disorder (ASD) is widely recognized, as indicated by the recurrence risk in families, twin studies, and the co-occurrence with chromosomal disorders and genetic syndromes. The genetic architecture of ASD is highly heterogeneous and although the cause of ASD is unclear in most cases, approximately 10–20% of individuals with ASD have an identifiable genetic aetiology.[1] The Mendelian condition neurofibromatosis type 1 (NF1) has been associated with an increased risk of ASD. NF1 is caused by a germline mutation in the NF1 gene (chromosome 17q11.2) which encodes neurofibromin, a negative regulator of the RAS/mitogen activated protein kinase (MAPK) signaling pathway. Epidemiological studies have reported the frequency of NF1 among patients with ASD at approximately 0.6% and the frequency of ASD in patients with NF1 has been estimated at 4% (3/74);[1] conflicting data, however, call these reports into question.[2] The greatest cause of lifetime morbidity in children with NF1 is cognitive, behavioural, and social impairment. While significant gains in understanding the cognitive phenotype have been made, investigations into the NF1 social phenotype are limited.

The studies by Walsh et al.[3] and Garg et al.[4] are among the first to have examined the prevalence of ASD symptomatology in children with NF1 and report that between 14–29% of their NF1 cohorts scored in the severe range (T score > 75) on a screening measure of ASD symptomatology; a range said to be associated with a clinical diagnosis of ASD.[3, 4] These data also indicate variation within NF1, with a further 26% of both cohorts displaying mild-moderate symptoms and 45–60% demonstrating normal social responsiveness.[3, 4] Both studies report a significant relationship between ASD symptomatology and attention-deficit–hyperactivity disorder (ADHD), an association that is well established in cohorts without NF1.[5]

When reading these papers, it is important to keep in mind the difference between symptom and disorder. Both studies screened for symptoms associated with ASD (using the Social Responsiveness Scale; SRS), but a diagnostic assessment was not performed. Although the SRS has commendable psychometric properties, the discrepancy between parent (29%) and teacher (5.7%) ratings of children within the severe range suggests the predictive validity of the questionnaire may be lower in this population.[4] Indeed, others have found the specificity of the SRS significantly reduced in children with previously identified social developmental problems.[6] It is possible that there is an increased risk of false positives in NF1 due to the range of impairments that overlap with pervasive developmental disorders, such as delayed social, executive, non-verbal, and verbal communication skills.

These studies should be of great interest to the NF1 research and clinical community for several reasons. First, they draw attention to a currently neglected area of need within the NF1 literature. Whereas previous studies have focussed on describing social outcomes (e.g. teasing, bullying, and awkwardness with peers), internalizing behaviors or basic social skills in NF1 cohorts, Walsh et al. and Garg et al. report a surprisingly high frequency of ASD traits in two separate NF1 populations – confirming an association between NF1 and features of ASD. Second, they create a platform for future research directed at identifying the neural and cognitive/psychological mechanisms underlying autistic traits in NF1. These data emphasize the need for investigations into theory of mind, nonliteral language comprehension, and eye gaze in children with NF1 – none of which have been examined to date. Interestingly, a number of structural brain abnormalities, including an increase in total brain, cerebellum and caudate nuclei volume are shared between NF1 and ASD and warrant further investigation.[7, 8] Third, these studies add to the mounting evidence that ASD may be associated with RAS/MAPK pathway dysregulation in human development. Like NF1, Noonan, Costello, and cardio-facio-cutaneous syndromes are caused by germline mutations in genes that encode protein components of the RAS/MAPK pathway and have also been associated with significantly increased rates of autistic features.[9] NF1 and these other RASopathies are likely to be important single-gene disorder models for ASD, in which gene × gene and gene × environment interactions can be explored. Finally, these studies highlight a need for longitudinal research to examine the developmental course of ASD symptoms in NF1 – including screening for symptoms in very young children that will possibly result in earlier intervention and improved outcomes.