A1 Refereed original research article in a scientific journal
Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach
Authors: Lalousis Paris Alexandros, Wood Stephen J.,Schmaal Lianne, Chisholm Katharine, Griffiths Sian Lowri, Reniers Renate L. E. P., Bertolino Alessandro, Borgwardt Stefan, Brambilla Paolo, Kambeitz Joseph, Lencer Rebekka, Pantelis Christos, Ruhrmann Stephan, Salokangas Raimo K. R., Schultze-Lutter Frauke, Bonivento Carolina, Dwyer Dominic,Ferro Adele, Haidl Theresa, Rosen Marlene, Schmidt Andre, Meisenzahl Eva, Koutsouleris Nikolaos, Upthegrove Rachel, CA PRONIA Consortium
Publisher: OXFORD UNIV PRESS
Publishing place: Oxford
Publication year: 2021
Journal: Schizophrenia Bulletin
Journal name in source: SCHIZOPHRENIA BULLETIN
Journal acronym: SCHIZOPHRENIA BULL
Volume: 47
First page : 1130
Last page: 1140
Number of pages: 11
ISSN: 0586-7614
eISSN: 1745-1701
DOI: https://doi.org/10.1093/schbul/sbaa185
Web address : https://doi.org/10.1093/schbul/sbaa185
Self-archived copy’s web address: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8266654/
Diagnostic heterogeneity within and across psychotic and affective disorders challenges accurate treatment selection, particularly in the early stages. Delineation of shared and distinct illness features at the phenotypic and brain levels may inform the development of more precise differential diagnostic tools. We aimed to identify prototypes of depression and psychosis to investigate their heterogeneity, with common, comorbid transdiagnostic symptoms. Analyzing clinical/neurocognitive and grey matter volume (GMV) data from the PRONIA database, we generated prototypic models of recent-onset depression (ROD) vs. recent-onset psychosis (ROP) by training support-vector machines to separate patients with ROD from patients with ROP, who were selected for absent comorbid features (pure groups). Then, models were applied to patients with comorbidity, ie, ROP with depressive symptoms (ROP+D) and ROD participants with sub-threshold psychosis-like features (ROD+P), to measure their positions within the affective-psychotic continuum. All models were independently validated in a replication sample. Comorbid patients were positioned between pure groups, with ROP+D patients being more frequently classified as ROD compared to pure ROP patients (clinical/neurocognitive model: chi 2 = 14.874; P < .001; GMV model: chi 2 = 4.933; P = .026). ROD+P patient classification did not differ from ROD (clinical/neurocognitive model: chi 2 = 1.956; P = 0.162; GMV model: chi 2 = 0.005; P = .943). Clinical/neurocognitive and neuroanatomical models demonstrated separability of prototypic depression from psychosis. The shift of comorbid patients toward the depression prototype, observed at the clinical and biological levels, suggests that psychosis with affective comorbidity aligns more strongly to depressive rather than psychotic disease processes. Future studies should assess how these quantitative measures of comorbidity predict outcomes and individual responses to stratified therapeutic interventions.
