Article

MR Imaging helps Unlock the Mysteries Behind Autism

Autism Spectrum Disorder (ASD) is a life-long developmental condition that impairs the ability to communicate and interact socially. The prevalence of autism recently increased to 1.7 percent1, or one in 59 children aged eight years, as more children are being diagnosed. However, other factors may be contributing to the higher prevalence, such as increased awareness, screening, diagnostic services, treatment and intervention services, better documentation of ASD behaviors and changes in diagnostic criteria (2).

The European Autism Interventions – A Multicentre Study for Developing New Medications (EU-AIMS), supported by a five-year grant from the Innovative Medicines Initiative (IMI), was the first European-wide collaboration that encompassed 14 institutions to study this neurodevelopmental condition. The public-private partnership, led by King’s College London and F. Hoffmann-La Roche AG, aimed to identify markers of autism that could be used for earlier and more accurate diagnosis and prognosis to inform the development of potential new therapies.

“Not everyone with autism is the same, there is a large heterogeneity in ASD” says Christine Ecker, BSc, MSc, PhD, a Professor at Johann Wolfgang Goethe-University of Frankfurt am Main and formerly a lecturer and researcher at King’s College London. Professor Ecker is involved in the EU-AIMS study and the recently launched AIMS-2-TRIALS designed to expand on the work and findings of the initial study.

“The focus now is to examine if we can use imaging techniques and other data modalities to stratify individuals,” she explains. Professor Ecker’s research group is examining brain anatomy using MRI measures such as cortical thickness and/or cortical folding, as well as functional measures of theory of mind in combination with resting-state functional MRI (rs-fMRI).

Based on multi-modal imaging data and clinical evaluations, the researchers hope to further determine different genetic subgroups of individuals with ASD. ASD ranges from high functioning autism, e.g., Asperger’s syndrome, to more severely intellectually disabled individuals who require long-term and more extensive care. Professor Ecker expects that differences detected in both the structural and functional MR data will enable the biologically driven stratification of individuals that also aligns with neurocognitive, behavioral and clinical profiles.

Several published studies have reported interesting neuroimaging results based on different imaging modalities. An excitatory/inhibitory imbalance in ASD, which is linked to learning, memory, cognitive, sensory, motor deficits and seizures, has been reported (3,4). Initial findings using MR spectroscopy (MRS) to compare excitatory glutamate and inhibitory Gamma-Aminobutyric Acid (GABA) levels in adults with ASD and a panel of six diverse rodent ASD models showed that a reduction in striatal glutamate is related to core ASD symptom severity such as social communication ability and restricted, repetitive behaviors. The study also found that high-functioning adults with idiopathic ASD have reduced glutamate in the striatum but no change in GABA(5). The authors conclude that these findings support the concept that glutamatergic dysfunction in the corticostriatal pathway is an underlying core pathophysiological mechanism of ASD, and their study along with Nelson, et al,(6) indicate that ASD may be rooted in a region-specific imbalance leaning toward reduced neuronal excitation.

“Serotonin is one neurotransmitter that is also of significant interest in ASD research,” Professor Ecker adds. “It is estimated that 30 percent of autistic individuals may have hyper serotonin, or too much serotonin, while another 30 percent don’t have enough.” The Institute of Psychiatry (IOP) at King’s College is examining serotonin modulation and the effect on brain functioning. Led by Eileen Daly, PhD, Professor Ecker and researchers at the IOP conducted a double-blind, placebo-controlled, crossover trial of acute tryptophan depletion, which in turn alters Serotonin levels, using fMRI. They found that in the autistic participants the depletion “normalized” fronto-cerebellar dysfunctions and the severity of ASD was connected to depletion within the frontal, striatal and thalamic regions (7).

“Although the differences are small, we see the same cortical networks being highlighted across neuroimaging studies in regions that are linked to autistic symptoms and traits, and that fit into three different symptom domains: social communication and interaction; social reciprocity; and repetitive and stereotyped behaviors and interests,” Professor Ecker says.

Professor Ecker further explains that due to the clinical heterogeneity and potentially different biological subgroups of ASD, rather than focus on isolated diagnostic biomarkers they are developing potential stratification markers using multimodal data derived from a combination of EEG and MR imaging data.

Another goal of the EU-AIMS study is to use big data approaches such as machine learning (ML) to help develop a normative modeling approach — e.g., to model the typical development of brain maturation in neurotypical controls. This data, along with the structural, functional and connectivity imaging data from 700-plus individuals scanned through the EU-AIMS consortium and data coming from other existing samples of 1,500 children scanned as part of the ABIDE initiative, is extremely valuable for stratification purposes, Professor Ecker adds.

Much of the MR scanning in younger age groups is conducted during natural sleep. Therefore, the development of silent MR imaging sequences, such as Zero TE (ZTE) for structural MR and Looping Star for rs-fMRI, (8) are increasingly important. Functional MR may also help in probing the efficacy of non-pharmacological treatments, such as talking therapies. Currently, however, the primary focus of research is to understand the neurobiological mechanisms underlying ASD, and the validation of stratification biomarkers.

“It is likely that different co-morbidities, such as OCD, affective disorders and ADHD reflect different subgroups of ASD individuals that may guide the development of stratification markers,” says Professor Ecker. “While there is currently no effective pharmacotherapy to treat the core symptoms of ASD, what we can treat relatively well is co-morbidities. We will, therefore, attempt to break down clinical phenotypes into different subgroups to see if they are also biologically different. This will also help us to establish whether these co-morbidities are separate or an integral part of ASD.”

There is some promise that multi-modal imaging using different MR sequences to obtain structural, functional and connectivity data on patients may help stratify patients according to the severity of their disease. Further, progress in detecting the neurobiological differences across the autism spectrum may guide a greater understanding of associated co-morbidities and treatments.

 

References

  1. CDC report. https://www.cdc.gov/ncbddd/autism/data.html
  2. https://www.sciencedaily.com/releases/2018/04/180426141604.htm
  3. Uzunova et al, World J Biol Psychiatry.2016 Apr;17(3):174-86.
  4. Selten M, van Bokhoven H, Nadif Kasri N. Inhibitory control of the excitatory/inhibitory balance in psychiatric disorders. F1000 Res. 2018;7:23. Published 2018 Jan 8. doi:10.12688/f1000research.12155.1.
  5. Horder J, Petrinovic MM, Mendez MA, et al.Glutamate and GABA in autism spectrum disorder—a translational magnetic resonance spectroscopy study in man and rodent models. Transl Psychiatry 8, 106 (2018) doi:10.1038/s41398-018-0155-1
  6. Nelson SB, Valakh V. Excitatory/inhibitory balance and circuit homeostasis in autism spectrum disorders. Neuron87, 684–698 (2015).
  7. Daly E, Ecker C, Hallahan B, et al. Response inhibition and serotonin in autism: a functional MRI study using acute tryptophan depletion. Brain. 2014 Sep; 137(9): 2600–2610.
  8. Wiesinger F, Menini A, Solana AB. Looping Star. Magn Reson Med. 2019 Jan;81(1):57-68. doi: 10.1002/mrm.27440. Epub 2018 Aug 14. PubMed PMID: 30106186.