Cognitive function in first-episode psychosis : trajectories and biological markers

Psychotic disorders, such as schizophrenia, are characterized by positive and negative psychotic symptoms, as well as cognitive symptoms. Cognitive impairment is present in a majority of individuals with a psychotic disorder and a predictor of functional outcome. Since current pharmacological treatments fail to alleviate cognitive symptoms, this area remains an unmet need in the treatment of psychotic disorders. Furthermore, recent data suggests that heterogeneity in level of cognitive ability and in cognitive trajectories could serve as important indicators of different underlying pathology in potential subtypes of psychotic disorders.

In order to improve functional ability in those with psychotic disorders, we need to understand more about what causes cognitive deficits, when these deficits occur and how they affect individuals differently. The studies in this thesis involves individuals who have recently experienced a first episode of psychosis (FEP) and who have not received anti-psychotic medication at all, or for a duration of less than four weeks. This approach minimizes the confounding effects of anti-psychotic medication and chronic illness.

The aim of this thesis was to characterize social (study I) and non-social (study III) cognition at psychosis onset, to investigate the relationship between cognition and the current target of pharmacological treatment (study II), the relationship between cognitive change and genetic risk scores and immune markers (study IV), and finally, to study variability in cognition at illness onset (study III) as well as over time (study IV).

In study I we compared a specific aspect of social cognition, facial affect recognition (FAR), in 67 individuals with FEP and 51 control subjects. Associations between FAR and psychotic symptoms were also tested. We found that FAR was impaired in individuals with FEP compared to controls (d = 0.84), where negative emotions were particularly difficult to identify. However, we found no support for a relationship between FAR ability and severity of psychotic symptoms.

Study II examined the neurotransmitter dopamine as a potential biological mechanism for cognitive symptoms in 18 drug-naïve FEP and 16 controls, by imaging dopamine D2 receptor availability using positron emission tomography (PET). We focused specifically on processing speed, working memory and verbal learning and how these related to D2 availability in the dorsolateral prefrontal cortex (DLPFC) and the hippocampus. No significant associations were found and instead, using Bayesian statistics, we found moderate support for the null hypothesis (BFH0:H1 = 3.3–8.2) – indicating that there is likely no meaningful relationship between these cognitive abilities and D2 availability in these brain regions.

Study III was an updated meta-analysis of cognitive function in 2625 drug-naïve FEP compared to 2917 controls. Additionally, we examined and quantified the amount of variability in the FEP group, compared to the control group. We were able to include 50 studies in the analysis and FEP individuals consistently across all cognitive domains displayed impairment compared to controls (Hedges g range -1.16 to -0.88). Individuals with FEP also had larger variability in cognitive function compared to controls (Coefficient of variation ratio (CVR) range 1.34- 1.92), which had not previously been demonstrated.

Study IV focused specifically on cognitive change within individuals and examined cognitive trajectories in 73 FEP and 53 controls from school age to 1,5 year after inclusion in study. How individuals varied in cognitive change over time was evaluated, and finally the amount of cognitive change was explored in relation to subsequent health care usage and potential biomarkers: polygenic risk scores (PRS) and the immune signaling protein complement component 4 A (C4A). In individuals with FEP, psychosis onset was associated with a large cognitive decline (-1.12 population standardized score). The degree to which individuals changed in cognitive capacity varied greatly. The amount of cognitive change was not associated to PRS scores or C4A levels but was related to later clinical outcome 5 years later: dispensation of antipsychotic medication and health care usage under a schizophrenia spectrum diagnosis.

Results from this thesis adds to existing evidence of cognitive deficits present already at illness onset, before pharmacological treatment. This motivates more studies even before psychosis onset, in individuals at clinical high risk for psychosis. No strong associations between the biomarkers investigated were evident, indicating the need for more sophisticated designs and more research into this topic. Results also demonstrate great variability between individuals with FEP in terms of cognition, both level of impairment and in the relative decline before psychosis diagnosis. This underscores the need for individual clinical assessment and prompts more research looking at cognitive differences rather than purely diagnostic groups.

List of scientific papers
I. Larsson, C.*, Lee, M.*, Lundgren, T., Erhardt, S., Sellgren, C M., Cervenka, S., Borg, J., Bölte, S., Fatouros-Bergman, H. Facial affect recognition in first-episode psychosis is impaired but not associated with psychotic symptoms. Heliyon 8, 9 (2022) https://doi.org/10.1016/j.heliyon.2022.e10424

II. Lee, M., Fatouros-Bergman, H., Plavén-Sigray, P., Ikonen Victorsson, P., Sellgren, C M., Erhardt, S., Flyckt, L., Farde, L., Cervenka, S. No association between cortical dopamine D2 receptor availability and cognition in antipsychotic-naive first-episode psychosis. npj Schizophr 7, 46 (2021) https://doi.org/10.1038/s41537-021-00176-x

III. Lee, M., Cernvall, M., Borg, J., Plavén-Sigray, P., Larsson, C., Erhardt, S., Sellgren, C M., Fatouros-Bergman, H., Cervenka, S. Cognitive function and cognitive heterogeneity in anti-psychotic drug-naïve psychosis patients - an updated meta-analysis. JAMA Psychiatry (2024) https://doi.org/10.1001/jamapsychiatry.2024.0016

IV. Lee, M., Matheson, G J., Cullen, A E., Lu, Z., Bergen, S., Sellgren, C M., Erhardt, S., Fatouros-Bergman, H., Cervenka, S. Cognitive trajectories in first-episode psychosis – associations with clinical outcomes and biological markers. [Manuscript]

*Authors contributed equally