Biomarker discovery for disease status and symptom severity in children with autism

Abstract

Autism spectrum disorder (ASD) is characterized by social impairments and repetitive behaviors, and affects 1 in 68 US children. Despite ASD's societal impact, its disease mechanisms remain poorly understood. Recent preclinical ASD biomarker discovery research has implicated the neuropeptides oxytocin (OXT) and arginine vasopressin (AVP), and their receptors, OXTR and AVPR1A, in animal models. Efforts to translate these findings to individuals with ASD have typically involved evaluating single neuropeptide measures as biomarkers of ASD and/or behavioral functioning. Given that ASD is a heterogeneous disorder, and unidimensional ASD biomarker studies have been challenging to reproduce, here we employed a multidimensional neuropeptide biomarker analysis to more powerfully interrogate disease status and symptom severity in a well characterized child cohort comprised of ASD patients and neurotypical controls. These blood-based neuropeptide measures, considered as a whole, correctly predicted disease status for 57 out of 68 (i.e., 84%) participants. Further analysis revealed that a composite measure of OXTR and AVPR1A gene expression was the key driver of group classification, and that children with ASD had lower neuropeptide receptor mRNA levels compared to controls. Lower neuropeptide receptor mRNA levels also predicted greater symptom severity for core ASD features (i.e., social impairments and stereotyped behaviors), but were unrelated to intellectual impairment, an associated feature of ASD. Findings from this research highlight the value of assessing multiple related biological measures, and their relative contributions, in the same study, and suggest that low blood neuropeptide receptor availability may be a promising biomarker of disease presence and symptom severity in ASD.

Keywords: Arginine vasopressin receptor 1A; Autism spectrum disorder; Blood biomarkers; Children; Oxytocin receptor.

Figure 1. Total neuropeptide receptor gene expression predicts disease status in children with and without ASD

The effect of total neuropeptide receptor gene expression (i.e., the sum of the OXTR -ACT and the AVPR1A -ACT) on predicted (line) and observed (colored circles) group is plotted, adjusted for the other terms in the model, and normalized to its original mean and standard deviation. Children with ASD plotted above, and Control children plotted beneath, the dashed line are correctly classified.

Figure 2. Group comparisons are presented for the blood neuropeptide measures

Each of the four neuropeptide measures was evaluated in turn, controlling for the same blocking variables as well as the other three neuropeptide measures, in the model. Only total neuropeptide receptor gene expression differed significantly between the ASD and control groups. Data are plotted as LSM +/- SE (i.e., adjusted for the other variables in the model).

Figure 3. Total neuropeptide receptor gene expression predicts symptom severity for core, but not associated, features of ASD

a) Social impairments, as measured by the SRS Total (Raw) Score, and b) Stereotypies, as measured by the RBS-R Stereotyped Behavior Subscale, are most severe in ASD children with the lowest levels of total neuropeptide receptor gene expression. c) Cognitive ability, as measured by the Stanford Binet IQ test, is unrelated to total neuropeptide receptor gene expression. Data are plotted adjusted for the other variables in the analysis.