Premorbid Deterioration in Psychosis Was Linked to Heavy Cannabis Use, Not Higher Genetic Risk

What the researchers found

Four clusters emerged from 802 first-episode psychosis patients: high-cognitive-functioning, low-cognitive-functioning, intermediate, and deteriorating. The deteriorating group (n=150) had normal-to-deteriorating premorbid adjustment and was uniquely characterized by lower schizophrenia polygenic risk scores but higher likelihood of daily high-potency cannabis use. Polygenic risk scores explained only 7.9% of between-group membership. All patient clusters had higher schizophrenia genetic risk than the 1,263 population controls.

Why it matters

This challenges the assumption that premorbid deterioration before psychosis reflects greater genetic loading. Instead, environmental factors — particularly high-potency cannabis — may drive a distinct pathway to psychosis in people with relatively lower genetic vulnerability.

The numbers in context

- 802 FEP patients, 1,263 controls

- 4 clusters: high-cognitive (n=205, IQ 106.1), low-cognitive (n=223, IQ 73.9), intermediate (n=224, IQ 80.8), deteriorating (n=150, IQ 80.6)

- PRS explained 7.9% of cluster membership

- Deteriorating cluster: lower SCZ_PRS, more daily high-potency cannabis

- All FEP clusters had higher SCZ_PRS than controls

How the study worked

Cluster analysis of EU-GEI multi-site data using premorbid social/academic functioning and IQ. Compared clusters on polygenic risk scores for schizophrenia, bipolar disorder, depression, and IQ. Assessed cannabis use patterns across clusters.

Who was studied

802 first-episode psychosis patients and 1,263 population-based controls from the EU-GEI study

What this study cannot tell us

Cross-sectional cluster analysis cannot establish causation. Cannabis use was self-reported. Premorbid adjustment assessed retrospectively. Polygenic risk scores explain only a fraction of genetic risk.

How to read the evidence

Large multi-site cluster analysis with genetic and environmental data. Strong methodology for identifying subtypes but cross-sectional design limits causal inference.

When this study was published

Published in 2023. Uses state-of-the-art polygenic risk scoring in a large European cohort.

The bigger picture

The existence of a cannabis-associated psychosis subtype — characterized by premorbid deterioration but lower genetic loading — suggests that different pathways lead to psychosis. This has implications for prevention: targeting cannabis use could prevent psychosis in people who would not otherwise develop it.

Questions still open

• Would reducing access to high-potency cannabis prevent the deteriorating trajectory?
• Do patients in the cannabis-associated cluster respond differently to antipsychotic treatment?
• Are there additional environmental factors that interact with cannabis to drive deterioration?

Read the original research

First-Episode Psychosis Patients Who Deteriorated in the Premorbid Period Do Not Have Higher Polygenic Risk Scores Than Others: A Cluster Analysis of EU-GEI Data.

Schizophrenia bulletin, 49(1), 218-227

Citation

Ferraro, Laura; Quattrone, Diego; La Barbera, Daniele; La Cascia, Caterina; Morgan, Craig; Kirkbride, James B; Cardno, Alastair G; Sham, Pak; Tripoli, Giada; Sideli, Lucia; Seminerio, Fabio; Sartorio, Crocettarachele; Szoke, Andrei; Tarricone, Ilaria; Bernardo, Miquel; Rodriguez, Victoria; Stilo, Simona A; Gayer-Anderson, Charlotte; de Haan, Lieuwe; Velthorst, Eva; Jongsma, Hannah; Bart, Rutten B P; Richards, Alexander; Arango, Celso; Menezez, Paulo Rossi; Lasalvia, Antonio; Tosato, Sarah; Tortelli, Andrea; Del Ben, Cristina Marta; Selten, Jean-Paul; Jones, Peter B; van Os, Jim; Di Forti, Marta; Vassos, Evangelos; Murray, Robin M. (2023). First-Episode Psychosis Patients Who Deteriorated in the Premorbid Period Do Not Have Higher Polygenic Risk Scores Than Others: A Cluster Analysis of EU-GEI Data.. Schizophrenia bulletin, 49(1), 218-227. https://doi.org/10.1093/schbul/sbac100