Predictors of Early Intervention Outcomes in Autism Spectrum Disorder

Early detection and intervention are theorized to facilitate better outcomes in autistic children. However, response to early intervention varies considerably between individuals, some children show significant improvement, while others show minimal response to the intervention. Some pre-treatment individualized characteristics as well as intervention-specific factors were theorized to moderate outcomes, but literature revealed often mixed findings making difficult to understand whether and to what extent these factors facilitate learning during treatment. To parse the heterogeneity of the autistic population, recent studies have also attempted to investigate biological factors related to clinical and behavioral profiles. However, to date it is unknown whether and how biological information can provide insight into the variability of treatment outcomes. With this in mind, we attempted to expand the current knowledge by first investigating whether pre-treatment child’s characteristics and blood leukocyte gene expression patterns could predict developmental trajectories during treatment. Leveraging a cohort of 41 autistic toddlers who received the same early intervention and provided a blood sample, we were able to analyze the effect of starting treatment very early (i.e., <24 months) on treatment trajectories. We also provided for the first-time evidence that both pre-treatment blood leukocyte gene expression patterns and clinical-behavioral characteristics are important for predicting developmental change during intervention, and that pretreatment epigenetic mechanisms such as histone acetylation may be a key biological process that influence how a child respond to early intervention. Lastly, we carried out a mega-analysis of a large international consortium, isolating individual child characteristics and treatment related factors to examine their key role in moderating child developmental trajectories throughout the course of early intervention. The final dataset comprised 645 autistic toddlers who received two types of interventions either Early Start Denver Model (ESDM) or other treatment as usual/community interventions (TAU/COM). This mega-analysis provided strong evidence that individual factors, such as cognitive level and age at treatment start, predict most outcomes, and on-average, ESDM may promote some developmental skills better than COM/TAU. Taken together, these results advance our understanding of “what works, for whom, for what and why” questions, identifying new biological predictors and providing an alternative methodology that can effectively examine the individual variability of treatment outcomes.