[Skip to Navigation]
Original Investigation
May 18, 2022

Clinical, Brain, and Multilevel Clustering in Early Psychosis and Affective Stages

Author Affiliations
  • 1Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany
  • 2Centre for Youth Mental Health, University of Melbourne, Melbourne, Victoria, Australia
  • 3International Max-Planck Research School for Translational Psychiatry, Munich, Germany
  • 4Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany
  • 5Max-Planck School of Cognition, Leipzig, Germany
  • 6Center for Adaptive Rationality, Max Planck Institute for Human Development, Berlin, Germany
  • 7Institute of Human Genetics, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany
  • 8Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
  • 9Max-Planck Institute of Psychiatry, Munich, Germany
  • 10Department of Education, Psychology, Communication, University of Bari Aldo Moro, Bari, Italy
  • 11Institute for Mental Health and Centre for Brain Health, University of Birmingham, Birmingham, United Kingdom
  • 12Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York
  • 13Department of Psychiatry (Psychiatric University Hospital, UPK), University of Basel, Basel, Switzerland
  • 14Early Intervention Service, Birmingham Women’s and Children’s NHS Foundation Trust, Birmingham, United Kingdom
  • 15Department of Mathematics and Computer Science, University of Basel, Basel, Switzerland
  • 16Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany
  • 17Department of Psychology, Aston University, Birmingham, United Kingdom
  • 18Faculty of Medicine, University of Basel, Basel, Switzerland
  • 19Department of Psychiatry, University of Turku, Turku, Finland
  • 20GE Healthcare GmbH (previously GE Global Research GmbH), Munich, Germany
  • 21Department of Child and Adolescent Psychiatry, University of Münster, Münster, Germany
  • 22University Hospital of Child and Adolescent Psychiatry and Psychotherapy, University of Bern, Bern, Switzerland
  • 23Department of Psychiatry, Psychotherapy and Psychosomatics, University Hospital of Psychiatry Zurich, Zurich, Switzerland
  • 24Institute of Genomic Statistics and Bioinformatics, University of Bonn, Bonn, Germany
  • 25Melbourne Neuropsychiatry Centre, University of Melbourne & Melbourne Health, Melbourne, Victoria, Australia
  • 26Department of Psychiatry and Psychotherapy, University of Münster, Münster, Germany
  • 27Department of Psychiatry and Psychotherapy, University of Lübeck, Lübeck, Germany
  • 28Department of Basic Medical Science, Neuroscience and Sense Organs, University of Bari Aldo Moro, Bari, Italy
  • 29Department of Neurosciences and Mental Health, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy
  • 30Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy
  • 31Department of Psychology, Faculty of Psychology, Airlangga University, Surabaya, Indonesia
  • 32Heinrich Heine University, Düsseldorf, Germany
  • 33Orygen, the National Centre of Excellence for Youth Mental Health, Melbourne, Victoria, Australia
  • 34Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia
  • 35Center for Biomedical Image Computing and Analytics, Perelman School of Medicine, University of Pennsylvania, Philadelphia
  • 36Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, United Kingdom
JAMA Psychiatry. 2022;79(7):677-689. doi:10.1001/jamapsychiatry.2022.1163
Key Points

Question  Can data-driven techniques redraw boundaries within early psychosis stages using clinical, brain, and multilevel approaches, and where do affective illnesses fit into the solutions?

Findings  In this cohort study of 749 individuals in the discovery sample and 610 individuals in the validation sample, unsupervised machine learning techniques revealed transdiagnostic clinical subgroups associated with cardinal symptom and functioning axes validated by premorbid, longitudinal, and genetic associations. A transdiagnostic reduced brain volume subgroup was detected; multilevel analyses demonstrated an association with generally higher psychopathology but not specific stages or symptoms.

Meaning  Early psychosis stages were reclassified beyond positive symptom severity at clinical and biological levels; specific symptom profiles, general brain phenotypes, and overlaps with affective illness were highlighted.


Importance  Approaches are needed to stratify individuals in early psychosis stages beyond positive symptom severity to investigate specificity related to affective and normative variation and to validate solutions with premorbid, longitudinal, and genetic risk measures.

Objective  To use machine learning techniques to cluster, compare, and combine subgroup solutions using clinical and brain structural imaging data from early psychosis and depression stages.

Design, Setting, and Participants  A multisite, naturalistic, longitudinal cohort study (10 sites in 5 European countries; including major follow-up intervals at 9 and 18 months) with a referred patient sample of those with clinical high risk for psychosis (CHR-P), recent-onset psychosis (ROP), recent-onset depression (ROD), and healthy controls were recruited between February 1, 2014, to July 1, 2019. Data were analyzed between January 2020 and January 2022.

Main Outcomes and Measures  A nonnegative matrix factorization technique separately decomposed clinical (287 variables) and parcellated brain structural volume (204 gray, white, and cerebrospinal fluid regions) data across CHR-P, ROP, ROD, and healthy controls study groups. Stability criteria determined cluster number using nested cross-validation. Validation targets were compared across subgroup solutions (premorbid, longitudinal, and schizophrenia polygenic risk scores). Multiclass supervised machine learning produced a transferable solution to the validation sample.

Results  There were a total of 749 individuals in the discovery group and 610 individuals in the validation group. Individuals included those with CHR-P (n = 287), ROP (n = 323), ROD (n = 285), and healthy controls (n = 464), The mean (SD) age was 25.1 (5.9) years, and 702 (51.7%) were female. A clinical 4-dimensional solution separated individuals based on positive symptoms, negative symptoms, depression, and functioning, demonstrating associations with all validation targets. Brain clustering revealed a subgroup with distributed brain volume reductions associated with negative symptoms, reduced performance IQ, and increased schizophrenia polygenic risk scores. Multilevel results distinguished between normative and illness-related brain differences. Subgroup results were largely validated in the external sample.

Conclusions and Relevance  The results of this longitudinal cohort study provide stratifications beyond the expression of positive symptoms that cut across illness stages and diagnoses. Clinical results suggest the importance of negative symptoms, depression, and functioning. Brain results suggest substantial overlap across illness stages and normative variation, which may highlight a vulnerability signature independent from specific presentations. Premorbid, longitudinal, and genetic risk validation suggested clinical importance of the subgroups to preventive treatments.

Add or change institution
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words