Normative Modelling of Molecular-Functional Circuits Captures Transdiagnostic Heterogeneity
Poster No:
541
Submission Type:
Abstract Submission
Authors:
Timothy Lawn1, Alessio Giacomel1, Daniel Martins2, Mattia Veronese3, Matthew Howard1, Federico Turkheimer1, Ottavia Dipasquale2
Institutions:
1King's College London, London, London, 2King's College London, London, United Kingdom, 3University of Padua, Padua, Italy
First Author:
Co-Author(s):
Introduction:
Clinical neuroscience aims to delineate the neurobiology underpinning the symptoms of various disorders, with the ultimate goal of developing mechanistically informed treatments for these conditions. This has been hindered by the hierarchical organisation of the brain and heterogeneity of psychiatric disorders1,2. However, advances in multimodal analytic techniques – such as Receptor Enriched Analysis of Connectivity by Targets(REACT) – have allowed integration of functional dynamics from fMRI with the brain's receptor landscape, providing novel trans-hierarchical insights2,3. Similarly, normative modelling of brain features has allowed translational neuroscience to move beyond group-average differences between patients and controls and characterise deviations from health at an individual level4. Here, we bring these novel methods together in order to address these two longstanding translational barriers in clinical neuroscience.
Methods:
We utilise healthy data from the CamCAN5 and UCLA phenomics6 datasets(N=607), and patients suffering from Schizophrenia(SCHZ), Bipolar-disorder(BPD), and ADHD from the latter(N=119). Transdiagnostic symptom scores were dimensionally reduced using principal components analysis. REACT3 was used create functional networks enriched with group average receptor/transporter distributions of the main modulatory (noradrenaline, dopamine, serotonin, acetylcholine), inhibitory(GABA), and excitatory(glutamate) neurotransmitter systems, which were parcellated using a custom atlas. Next, using hierarchical Bayesian regression within the predictive clinical neuroscience toolbox7, we generated normative models of these molecular-enriched networks. Using these models, we characterised deviations from normality within a held out subset of healthy controls(30%) as well the patients. This produced a deviation score for each subject, molecular system, and brain region. We analysed these deviation scores to examine between group differences using ROI-wise ANOVAs and summary brain-wide deviation metrics; between-subject similarity and how this relates to transdiagnostic symptomatology; and transdiagnostic network-symptom relationships through mass-univariate permutation based correlation analyses.
Results:
We identified four principal components which explained 68.9% of the total variance in symptom scores. Broadly, our results align with accounts of excitatory-inhibitory imbalance in schizophrenia and bipolar disorder, with between group analyses showing significantly greater negative deviations in glutamatergic and GABAergic systems, driven primarily by SCHZ and BPD. Between subject similarity analyses emphasised the substantial overlap in symptoms and deviations across these disorders transdiagnostically, with levels of within-group similarity significantly correlating with the brain-wide negative glutamatergic and GABAergic network deviations as well as principal component 2(PC2: primarily capturing psychotic symptoms) across multiple receptor systems(fig-1). Finally, across all patients, both the cholinergic and glutamatergic deviations across widespread regions including cingulate, insular, and opercular cortices correlated with PC2, such that those with more negative deviations had greater symptom scores(fig-2)
·fig-1
·fig-2
Conclusions:
The integration of novel functional-molecular neuroimaging techniques, normative modelling, and a transdiagnostic perspective utilised here offers methodological and theoretical progress towards an understanding of the shared neurobiological foundations that underpin psychiatric conditions. Our transdiagnostic approach moves away from case-control analyses and offers an interesting way to situate clinical groups or individuals within between-subject similarity and deviation-symptom landscapes, which when scaled up across diagnoses, symptomatology, and molecular systems may offer novel perspectives on how complex aberrations of affect and cognition map onto dysfunction spanning molecular and systems level readouts.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Transmitter Systems 2
Keywords:
Acetylcholine
Attention Deficit Disorder
Data analysis
FUNCTIONAL MRI
Glutamate
Modeling
Neurotransmitter
Psychiatric Disorders
Schizophrenia
1|2Indicates the priority used for review
Provide references using author date format
2. Lawn, T. et al. From neurotransmitters to networks: Transcending organisational hierarchies with molecular-informed functional imaging. Neuroscience & Biobehavioral Reviews 150, 105193 (2023).
3. Dipasquale, O. et al. Receptor-Enriched Analysis of functional connectivity by targets (REACT): A novel, multimodal analytical approach informed by PET to study the pharmacodynamic response of the brain under MDMA. Neuroimage 195, 252–260 (2019).
4. Marquand, A. F. et al. Conceptualizing mental disorders as deviations from normative functioning. Mol Psychiatry 24, 1415–1424 (2019).
5. Taylor, J. R. et al. The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data repository: Structural and functional MRI, MEG, and cognitive data from a cross-sectional adult lifespan sample. Neuroimage 144, 262–269 (2017).
6. Poldrack, R. A. et al. A phenome-wide examination of neural and cognitive function. Sci Data 3, 160110 (2016).
7. Rutherford, S. et al. The normative modeling framework for computational psychiatry. Nat Protoc 17, 1711–1734 (2022).