Heterogeneity in ASD: Biological Mechanisms, Trajectories, and Treatment Response

Parent-mediated interventions often target social communication in young children with ASD, although to date studies yield inconsistent effects. One reason for the limited evidence may be the considerable heterogeneity in both parent and child characteristics that affect the fit of intervention to family and ultimately influence treatment outcome. For parents, these factors might include stress associated with the uncertainty of their child's diagnosis, caregiver expectations for the intervention itself, and a parent's own style of interaction that may be influenced by milder but qualitatively similar ASD characteristics, known as the broad autism phenotype (BAP). For children, these factors might include nonverbal DQ, language, or sensory impairment. The fit between type of intervention and optimal outcome for parent and child is an understudied, yet essential component of early intervention that may be susceptible to the influence of heterogeneity in the parent and child. One approach to addressing this variability is to implement an adaptive intervention approach that seeks to capitalize on heterogeneity among children and parents. Utilizing an adaptive treatment design, the current study tests the optimal sequence of intervention delivery and specific parent and child characteristics that may moderate treatment success in three 10-week stages of intervention. The first phase will randomize parents and children to a parent education condition, consisting of a parent support and education group focused on social communication development, or to a parent mediated and therapist delivered condition involving coaching of the parent with their child in social communication strategies. Phase 2 involves re-randomizing parents and children to maintain the same treatment arm, or change to the opposite arm to test the optimal sequence of intervention delivery and specific parent and child characteristics that may moderate treatment success. In the final phase, dyads are randomized to different maintenance arms, each comprised of 5 sessions with one involving skype and text contact, the other in -home visits, to explore how best to maintain treatment gains once the active intervention phase is complete. This study has the potential to dramatically improve child social communication outcomes by individualizing and personalizing parent intervention approaches with very young children, a high priority need of the Interagency Autism Coordinating Council and NIH.

While social communication is a core developmental deficit that characterizes children with ASD, there is great heterogeneity in both presentation and gains with treatment. Long term outcomes of children with ASD vary with nearly 40% remaining minimally verbal by school age. Beginning early and providing high doses of intervention appears critical to child outcomes, and one cost-effective, efficient way to accomplish these goals is to involve parents. Particularly, for very young children (under the age of 3 years) who may not have a confirmed ASD diagnosis, parent mediated interventions have several advantages. One is that there is potential for earlier access to evidence based interventions. Parents can intervene with their child immediately without waiting for access through a lengthy diagnostic process or for therapist -delivered interventions that can have long wait times. Another is that parents can increase the dose of an intervention since they are with their child for many more hours than non-family members.

Despite the increasing numbers of parent mediated interventions, inconsistent results are noted. A few studies have found significant differences in both child and parent outcomes but others find limited to no effects on parent and/or child. This heterogeneity in outcomes is often attributed to child characteristics or to the intervention itself. However, the fit between type of intervention-focused on the parent-(education about social communication in their child, or stress reduction interventions), or focused on the child-(parent coached to work directly with their child)-has rarely been examined in parents engaged in these interventions. Fit may be influenced by parent characteristics that can affect their ability to implement the interventions.

These include stress associated with the uncertainty of their child's diagnosis, caregiver expectations for the intervention itself, and parent's own style of interaction that may be influenced by milder but qualitatively similar ASD characteristics, known as the broad autism phenotype.

It is widely recognized that a single intervention is not effective for all parents and children. One approach to addressing this variability is to implement an adaptive intervention approach that seeks to capitalize on heterogeneity and evolving status among children and parents. An adaptive intervention is a replicable, sequence of treatment decision rules designed to help guide clinicians concerning whether, how or when-and based on which measures-to provide certain intervention components. This type of intervention design provides information on the most effective intervention for children and parents who need it (leading to individualized and personalized sequences of treatment). Using a novel experimental design, the proposed study will develop a more effective adaptive intervention by addressing the following specific aims:

Primary Aim (Best Initial Strategy): To determine the effect of PARENT focused intervention vs CHILD focused intervention on change in child initiated joint engagement (primary outcome), play, and joint attention (secondary outcomes) from baseline to end of phase 1. Hypothesis: Children will improve more, on average, when offered CHILD focused intervention.

Secondary Aim 1 (Best Sequence): To compare and contrast the four pre-specified adaptive interventions from baseline to end of phase 2 on primary and secondary outcomes. Hypothesis: The sequential intervention beginning with CHILD focused interventions followed by PARENT education sessions will lead to the most improved outcomes.

Secondary Aim 2 (Toward More Personalized Intervention Sequences): To develop a more individually- tailored adaptive intervention by examining whether parental factors including BAP, parental stress, or parent's expectancies for the intervention, and child factors including non-verbal DQ, language or sensory impairment at baseline moderates the effect of Phase 1 interventions from baseline to end of Phase 1 on primary and secondary outcomes. Hypotheses: (i) Children with parents who report greater BAP and parenting stress will benefit more from PARENT focused intervention and children with parents who report greater expectancy will benefit more from CHILD focused intervention. (ii) Children with low non-verbal DQ and high levels of sensory impairment at baseline will benefit more from the CHILD focused intervention from baseline to end of phase 1. Secondary Aim 3 (Maintenance Protocol): To determine the maintenance effect of in person home visits versus technology supported maintenance protocol on change in primary and secondary outcomes from (i) baseline to end of phase 3 and (ii) from end of phase 2 to end of phase 3. Hypothesis: Children in the home visit maintenance protocol will have more improved outcomes from baseline to end of phase 3 compared to the children in the technology supported maintenance protocol.

Secondary Aim 4 (Exploratory: Biomarkers and Genetics): To characterize children at risk of ASD in terms of electrophysiological biomarkers (neural synchrony measured by EEG power, peak alpha frequency and coherence) and genetic risk (as measured by the presence of CNVs and polygenetic risk score) and to explore each's role as a biological moderator of treatment effects.