Can Functional Connectivity Predict Longitudinal Clinical Outcomes in First-Episode Psychosis?
Poster No:
546
Submission Type:
Abstract Submission
Authors:
Isaac Pope1, Sidhant Chopra2, Priscila Levi3, Alexander Holmes4, Edwina Orchard5, Shona Francey6, Brian O'donoghue6, Vanessa Cropley7, Barnaby Nelson6, Jessica Graham6, Lara Baldwin6, Hok Pan Yuen6, Kelly Allott6, Mario Alvarez-Jimenez6, Susy Harrigan6, Christos Pantelis6, Stephen Wood8, Patrick McGorry6, Alex Fornito9
Institutions:
1Monash University, Melbourne, Victoria, 2Orygen, Preston, Victoria, 3Monash University, Melbourne, VIC, 4University of Oxford, Oxford, Oxfordshire, 5University of California, Santa Barbara, Preston, CA, 6Orygen, Melbourne, Victoria, 7Centre for Youth Mental Health, The University of Melbourne, Melbourne, Australia, 8University of Melbourne, Parkville, Victoria, 9Monash University, Clayton, Victoria
First Author:
Co-Author(s):
Introduction:
Clinical outcomes following a first episode of psychosis (FEP) vary substantially between patients in terms of both symptoms and functioning (Lally et al., 2017), with personalized treatment limited by a lack of reliable prognostic biomarkers (Coutts et al., 2023). Since FEP patients reliably display brain-wide disruption to functional connectivity (FC) (O'Neill et al., 2019), with some patterns linked to symptom dimensions (Wang et al., 2020), FC may also encode the capacity for clinical changes and thereby hold prognostic utility. Prior works have found group-level associations between FEP patients' baseline FC and their clinical outcomes (Dominicus et al., 2023), but these are not sufficient to determine whether FC is a reliable prognostic biomarker due to risks of overfitting models. This study aimed to determine whether resting-state FC could be used to predict the individual clinical outcomes of FEP patients (n=55, 15-25 years) who were randomised to receive antipsychotic or placebo tablets for 6 months, alongside psychosocial treatment (Francey et al., 2020). We also explored whether these outcomes could be predicted by patients' FC changes over the first 3 months.
Methods:
Functional magnetic resonance imaging data were acquired at baseline and after 3 months, then processed and correlated pair-wise across 328 regions to obtain subject-specific whole-brain FC matrices. Patients' clinical outcomes were defined as proportional changes in Brief Psychiatric Rating Scale or Social and Occupational Functioning Assessment Scale scores after 6 or 12 months. Three algorithms were used to make cross-validated predictions: (i) connectome-based predictive modelling (Shen et al., 2017), a simple method comprising feature selection, feature summarization, and linear modelling; (ii) kernel ridge regression (Li et al., 2019), a classical machine learning technique with hyperparameter optimization to mitigate overfitting within the training set; (iii) multilayer meta-matching (Chen et al., 2024), a transfer learning framework that leverages brain-behaviour predictions from much larger non-psychiatric datasets.
Results:
All three algorithms showed poor performance in predicting patients' 6- and 12-month changes in symptoms and functioning (all r_mean<0.3), and no model achieved statistical significance through permutation testing before or after family-wise error correction (all p>0.05).
Conclusions:
Our findings contrast those of prior FEP outcome prediction studies, where FC-based predictions achieve statistical significance or a classification accuracy exceeding 75% (Dominicus et al., 2023). These studies commonly involved diagnostic constraints (e.g: first-episode schizophrenia only), rigid treatment protocols (e.g: all patients received the same antipsychotic compound and dose with no psychosocial treatments), and outcomes measured earlier than 6 months post-baseline. We therefore propose that FC may only be able to predict FEP patients' clinical outcomes under narrow contexts, with insufficient generalisability to establish its clinical utility. However, our small sample size may have limited the detection of inherently weak associations between FC and clinical outcomes, meaning the reliability of our findings can only be assessed by performing cross-validation on a larger sample. Clinical outcome heterogeneity in FEP cohorts may also be explained by factors outside the brain, including demographic characteristics, social environments, and the duration of untreated psychosis, implying an upper limit to the performance of purely brain-based predictive models. We therefore recommend that future studies aim to develop multimodal predictive models for FEP by combining unique sources of prognostic information.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Classification and Predictive Modeling 2
Connectivity (eg. functional, effective, structural)
Keywords:
Computational Neuroscience
FUNCTIONAL MRI
Machine Learning
Modeling
Psychiatric Disorders
Other - First-episode psychosis
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Coutts, F. Psychotic disorders as a framework for precision psychiatry. Nature Reviews Neurology, 19(4), 221–234.
Dominicus, L. S. (2023). fMRI connectivity as a biomarker of antipsychotic treatment response: A systematic review. NeuroImage: Clinical, 40, 103515.
Francey, S. M. (2020). Psychosocial Intervention With or Without Antipsychotic Medication for First-Episode Psychosis: A Randomized Noninferiority Clinical Trial. Schizophrenia Bulletin Open, 1(1), sgaa015.
Gong, Q. (2017). Network-Level Dysconnectivity in Drug-Naïve First-Episode Psychosis: Dissociating Transdiagnostic and Diagnosis-Specific Alterations. Neuropsychopharmacology, 42(4), 933–940.
Lally, J. (2017). Remission and recovery from first-episode psychosis in adults: Systematic review and meta-analysis of long-term outcome studies. The British Journal of Psychiatry, 211(6), 350–358.
Li, J. (2019). Global signal regression strengthens association between resting-state functional connectivity and behavior. NeuroImage, 196, 126–141.
O’Neill, A. (2019). Dysconnectivity of Large-Scale Functional Networks in Early Psychosis: A Meta-analysis. Schizophrenia Bulletin, 45(3), 579–590.
Shen, X. (2017). Using connectome-based predictive modeling to predict individual behavior from brain connectivity. Nature Protocols, 12(3), Article 3.
Wang, D. (2020). Individual-specific functional connectivity markers track dimensional and categorical features of psychotic illness. Molecular Psychiatry, 25(9), 2119–2129.