Transdiagnostic and Illness-Specific Functional Dysconnectivity Across Schizophrenia, Bipolar Disorder and Major Depression and Relationships with Working Memory

Abstract

Background: Mental disorders are typically defined as distinct diagnostic entities but similar patterns of clinical impairments are frequently found in practice across the diagnostic groups. We investigate whether the transdiagnostic deficits result from a common neurocognitive mechanism across disorders, or various illness-specific mechanisms, or a combination of both. Methods: Functional MRI data were collected from the clinically stable patients with major depressive disorder (MDD; n=53), bipolar disorder (BIP; n=78), or schizophrenia (SCZ; n=100), and the matched healthy control subjects (n=109) using a single scanner. Group comparisons were conducted to identify both the transdiagnostic and the illness-specific features. A multivariate approach with cross-validation was used to assess the association between the brain functional dysconnectivity and the cognitive deficits. The confounding effect of medication on the findings was also tested. Findings: Compared with the healthy controls, the patients displayed shared working memory deficits [Cohen's d (d) in MDD/BIP/SCZ=0·73/0·65/0·89], and functional dysconnectivity within brain networks [somatomotor (d:0·50- 0·58) and salience (d:0·52-0·58)] and between brain networks [subcortical-limbic (d:0·55-0·69) and subcortical-dorsal attention (d:0·56-0·61)]. The illness-specific dysconnectivity were found between the executive control and default-mode networks in SCZ (d:0·46-0·56), between the salience and subcortical networks in BIP (d:0·47-0·48), and between the salience and default-mode networks in MDD (d:0·53-0·55). Working memory deficits were associated with a linear combination of 10 transdiagnostic and 4 illness-specific dysconnectivity patterns (r = 0·368, p = 2·45e-12, n = 340). The associations between the dysconnectivity patterns and the medication dosage did not reach statistical significance. Interpretation This study emphasizes that transdiagnostic deficits in cognition may be mediated by multiple interactions between many brain networks, and including diagnostic-specific patterns of functional dysconnectivity in different psychiatric disorders. Funding Statement: Ministry of Science and Technology, National Health Research Institutes (Taiwan). National Natural Science Foundation, National Key Research and Development Program, Shanghai Municipal Science and Technology Major Project, Shanghai Science and Technology Innovation Plan, Shanghai AI Platform for Diagnosis and Treatment of Brain Diseases, the Project of Zhangjiang Hi-Tech District Management, and Zhangjiang Lab (China). Academic Medical Organsiation of Southwest Ontario and the Bucke Family Funds. Declaration of Interests: Dr Palaniyappan reports educational grants and personal fees from Otsuka Canada and Janssen Canada, educational grants from Sunovion, personal fees from SPMM Course (UK) and Canadian Psychiatric Association, and royalties from Oxford University Press outside the submitted work. Dr. Robbins reports personal fees from Cambridge Cognition, Unilever, Mundipharma, Greenfiled Inc, grants from Shionogi, Lundbeck, Small Pharma, personal fees from Elsevier, Springer Verlag, outside the submitted work. No other disclosures were reported Ethics Approval Statement: The study was approved by the Institutional Review Board of Taipei Veterans General Hospital. Written informed consent was obtained from all participants before commencement of the study.