Finding biomarkers for improved patient stratification

When developing new drugs for autism, its large neurobiological heterogeneity poses a challenge. Thus, it is important to identify and validate biomarkers that allow stratifying autism into biologically more homogeneous subgroups. This was the aim of the various biomarker analyses by the AIMS-2-TRIALS consortium presented at the ECNP 2021 Congress.

The overarching aim of AIMS-2-TRIALS, which stands for Autism Innovative Medicine Studies-2-Trials is to explore the biology of autism to tailor treatments and develop new medication.

Recent treatment approaches have focused on the neuropeptide oxytocin, which regulates aspects of social behaviour in mammals. Dr Hanna Hörnberg (Max Delbruck Center for Molecular Medicine, Germany) presented research on the impact of genetic factors associated with autism on oxytocin signalling [1]. The results showed that mice carrying an autism-associated mutation in neuroligin-3 (NlGN3) have impaired oxytocin responsiveness in dopaminergic neurons and altered social recognition. Unexpectedly, the loss of NlGN3 was accompanied by a disruption of translation homeostasis in the ventral tegmental area (VTA). NlGN3 knockout mice were treated with ETC-168, a novel, highly specific, brain penetrant inhibitor of mitogen-activated protein kinase interacting kinases ½ (MNK ½). MNK inhibition was found to restore protein synthesis rate and social behaviour. It also improved learning in a fragile-X mouse model. Side effects were modest. “Targeting protein synthesis by inhibiting MNK kinases restored part of the translation machinery, the response to oxytocin, as well as social behaviour,” Dr Hörnberg added.

In another presentation, Dr Emily Jones (Birkbeck University of London, UK) presented the results of a study of neural measures of social expertise that may stratify biomarkers for autism [2]. The N170 is a negative deflection in the event-related potential (ERP) waveform over occipitotemporal electrode sites approximately 130–200 ms after perception of a visual event that marks face structural encoding (i.e. recognising a face as such) and is differentially sensitive to individual facial features. The expertise-based neural response to faces is slower in children and adolescents with autism and is related to autism-linked genetic variation. The results of the study showed that these neural responses can predict individual variation in later social functioning. According to Dr Jones, N170 may yield fruitful potential stratification and prognostic biomarkers.

Other research illustrated the impact that genetic heterogeneity in neurodevelopmental conditions has on brain connectivity [3]. Dr Clara Moreau (Institut Pasteur, France) described that effect sizes of copy number variation syndromes (CNVs) associated with autism and schizophrenia on functional connectivity are consistent with their effects on cognitive ability; however, effects on psychiatric conditions in terms of polygenic risk score (PRS) were almost undetectable. “This could mean PRS are too heterogenous for neuroimaging studies,” Dr Moreau said. Still, genetic risks (both common and rare) for autism and schizophrenia did converge on the thalamus and the somatomotor network.

Prof. Louise Gallagher (Trinity College Dublin, Ireland) presented some of the work her group has done, identifying translational biomarkers in rare genetic synaptopathies associated with autism [4]. She concentrated on 2 CNVs: Phelan-McDermid syndrome and neurexin 1 (NRXN1) deletion, which are both synaptopathies. This means impacted genes are directly expressed at the synapse, and the rates of autism and other neuropsychiatric disorders is increased in both conditions. A first exploratory neuroimaging study of a cohort of NRXN1 deletion carriers revealed a significantly higher number of errors in the Spatial Working Memory (SWM) and a numerically lower score on the Emotion Recognition Test (ERT). Significant differences were detected in within-network functional connectivity in the visual and ventral attention networks, but these were not correlated to brain-behaviour relationships. A poorer segregation of networks in NRXN1 deletion carriers that was found may suggest a disconnectivity phenotype.

1. Hörnberg H. Pharmacological rescue of social behaviour in a mouse model of autism. S.20.02, ECNP 2021 Congress, 2–5 October.
2. Jones E. Towards stratification biomarkers for ASD: neural measures of social expertise. S.20.03, ECNP 2021 Congress, 2–5 October.
3. Moreau CA. The genetics of autism from risk to resilience. S.20.01, ENCP 2021 Congress, 2–5 October.
4. Gallagher L. Translational biomarkers in rare genetic synaptopathies associated with autism. S.20.04, ENCP 2021 Congress, 2–5 October.

Copyright ©2021 Medicom Medical Publishers



Posted on 26 November 202126 November 2021