A systematic review of molecular imaging (PET and SPECT) in autism spectrum disorder: Current state and future research opportunities

Highlights

• Prior use of PET and SPECT in autism is reviewed in detail.

• Disparity between autism prevalence and PET/SPECT studies to date is reported.

• Minority of PET/SPECT studies has assessed neurotransmitters in autism.

• We describe substantial opportunities for PET/SPECT in autism research.

• Molecular imaging may help uncover biochemical endophenotypes in autism.

Abstract

Non-invasive positron emission tomography (PET) and single-photon emission computed tomography (SPECT) are techniques used to quantify molecular interactions, biological processes and protein concentration and distribution. In the central nervous system, these molecular imaging techniques can provide critical insights into neurotransmitter receptors and their occupancy by neurotransmitters or drugs. In recent years, there has been an increase in the number of studies that have investigated neurotransmitters in autism spectrum disorder (ASD), while earlier studies mostly focused on cerebral blood flow and glucose metabolism. The underlying and contributing mechanisms of ASD are largely undetermined and ASD diagnosis relies on the behavioral phenotype. Discovery of biochemical endophenotypes would represent a milestone in autism research that could potentially lead to ASD subtype stratification and the development of novel therapeutic drugs. This review characterizes the prior use of molecular imaging by PET and SPECT in ASD, addresses methodological challenges and highlights areas of future opportunity for contributions from molecular imaging to understand ASD pathophysiology.

Introduction

Autism spectrum disorder (ASD) is a group of neurodevelopmental disorders with prevalence as high as 1 in 68 children (CDC, 2014). Although a number of genetic mutations appear to be associated with ASD or with an increased risk or vulnerability for ASD, to date the collection of genetic mutations account for only a small percentage of total ASD cases (Persico and Napolioni, 2013). Epigenetic and environmental factors are increasingly implicated in the disorder. To date, there are no biomarkers that can be used to diagnose subtypes of idiopathic ASD (i.e. without any known genetic or environmental cause). An ASD diagnosis is based on clinical criteria with the hallmarks of the disorder being deficits in social communication and interaction, as well as restricted and repetitive behavior (APA, 2013). As implied by the use of the term ‘spectrum’ to describe autism, different forms of ASD exist. Uncovering the pathophysiological differences among disorders on the autism spectrum could help define biological subtypes of ASD and may improve diagnostic precision and clinical management, potentially leading to the development of more effective treatments.

Although providing their own methodological strengths, animal models of ASD cannot reflect the complexity of the human disorder and may not represent atypical biology accurately. Thus, in order to better elucidate the underlying pathophysiology in ASD, it is critical to conduct non-invasive in vivo human neuroimaging studies. Nearly all neuroimaging studies in ASD share the common goal of explaining or stratifying ASD mechanisms by increasing our knowledge about structural, functional or neurochemical differences in the brains of individuals with ASD. In this review, we examine single-photon emission computed tomography (SPECT) and positron emission tomography (PET) studies that have been conducted in individuals with ASD to date. We will consider PET and SPECT together as molecular imaging (MI) although each of these modalities has distinct features.

Both PET and SPECT are nuclear imaging techniques in which a radioactive material, referred to as radiotracer, is administered (typically intravenously) into a participant. A radiotracer is often a molecule that binds specifically to a particular target protein, e.g. a receptor, thereby allowing for visualization of distribution and quantification of the protein of interest. However, in other cases the distribution of the tracer is determined by where it accumulates following biochemical modification, e.g. with radiolabelled glucose. A very small mass of the tracer is administered in order to allow specific binding to targets of interest without promoting pharmacological effects or interacting through self-competition. With PET, radiotracer concentration is measured through the detection of high energy (511 keV) anti-colinear gamma photon pairs that result from positron annihilation. SPECT, on the other hand, measures radiotracer concentration by detecting single gamma rays within a particular energy range, which depends on the isotope being used. In general, PET has about two to three times higher sensitivity than SPECT and while both methods have only moderate spatial resolution, PET has slightly higher spatial resolution than SPECT (Rahmim and Zaidi, 2008). Typically the isotopes used in SPECT studies have longer half-lives than those used in PET, which is why PET radiotracers are usually produced in an onsite cyclotron, and are thus often less accessible and more expensive.

Despite some additional differences, PET and SPECT share key strengths at their core: (1) PET and SPECT can be used to noninvasively visualize and quantify differences in density of essential proteins such as receptors, transporters and enzymes; (2) in many cases, these techniques can be used to assess neurotransmitter release and occupancy; (3) PET and SPECT can measure drug-target engagement; and (4) functional measures, such as glucose metabolism and oxygen consumption, can be evaluated (though this is mostly limited to PET).