An investigation of clinical and neurobiological subtypes of autism

Abstract

Autism is a heterogeneous neurodevelopmental condition. There are a range of mixed findings across the literature which indicate that autism might actually consist of several homogenous subtypes. Longitudinal and intervention studies have shown that individuals on the spectrum can diverge into a range of trajectories and outcomes across the lifespan. The identification of subtypes could therefore lead to the development of tailored and targeted supports and help improve the outcomes of individuals on the spectrum. Ideally, gaining a better understanding of what works for whom, can help ensure that all individuals on the spectrum are given the resources they need to achieve their full potential. The aim of this thesis was to explore the clinical and neurobiological heterogeneity in autism. Further, it sought to determine which factors were related to good outcomes for individuals on the spectrum. Unsupervised clustering algorithms were used to investigate whether subtypes of autism exist by examining the behavioural phenotype, neurobiology and longitudinal outcomes of individuals on the spectrum. The results of this thesis showed that subtypes with distinct neurobiological and phenotypic profiles could be identified when examining the neuroanatomical presentation and longitudinal outcomes of individuals on the spectrum, but not when examining clinical presentation on its own. It also showed that psychiatric comorbidities and white matter connectivity in the brain may have an impact on the ability of children to make improvements over time. Overall, this thesis generated empirical evidence that the current diagnostic criteria of for autism describes a condition that consists of several homogenous subtypes. In addition, it was able identify groups of individuals that showed better outcomes compared to the rest of the autism population. These findings call into question the characterisation of autism as ‘one’ condition and instead provide insights into subtypes of autism with unique aetiologies and needs. The identification of individuals with unique clinical, neurobiological and outcome profiles opens up a new realm of targets for therapies and intervention.