Longitudinal trajectories in negative symptoms and changes in brain cortical thickness: 10-year follow-up study

Abstract

Background: Understanding the evolution of negative symptoms in first-episode psychosis (FEP) requires long-term longitudinal study designs that capture the progression of this condition and the associated brain changes.

Aims: To explore the factors underlying negative symptoms and their association with long-term abnormal brain trajectories.

Method: We followed up 357 people with FEP over a 10-year period. Factor analyses were conducted to explore negative symptom dimensionality. Latent growth mixture modelling (LGMM) was used to identify the latent classes. Analysis of variance (ANOVA) was conducted to investigate developmental trajectories of cortical thickness. Finally, the resulting ANOVA maps were correlated with a wide set of regional molecular profiles derived from public databases.

Results: Three trajectories (stable, decreasing and increasing) were found in each of the three factors (expressivity, experiential and attention) identified by the factor analyses. Patients with an increasing trajectory in the expressivity factor showed cortical thinning in caudal middle frontal, pars triangularis, rostral middle frontal and superior frontal regions from the third to the tenth year after the onset of the psychotic disorder. The F-statistic map of cortical thickness expressivity differences was associated with a receptor density map derived from positron emission tomography data.

Conclusions: Stable and decreasing were the most common trajectories. Additionally, cortical thickness abnormalities found at relatively late stages of FEP onset could be exploited as a biomarker of poor symptom outcome in the expressivity dimension. Finally, the brain areas with less density of receptors spatially overlap areas that discriminate the trajectories of the expressivity dimension.

Keywords: MRI; Negative symptoms; cortical thickness; factor analysis; first episode psychosis.

Fig. 1

Trajectories (stable, decreasing and increasing) of the three negative symptom dimensions as grouped by the latent class mixed model.

Fig. 2

Whole-brain cortical thickness changes for experiential and expressivity factors on the Scale for the Assessment of Negative Symptoms (SANS). Longitudinal cortical thickness changes across time for expressivity (a) and experiential (b) factors, with box plots showing the long-term cortical thickness changes across factors.

Fig. 3

Regional long-term cortical thickness changes for expressivity and experiential factors on the Scale for the Assessment of Negative Symptoms (SANS). Regions showing significant group effects (stable, decreasing, increasing trajectories) in long-term cortical thickness changes for expressivity (a) and experiential (b) factors. The scatter plots of longitudinal cortical thickness changes by region show a significant group effect for expressivity.

Fig. 4

Association between differential cortical thickness trajectories across expressivity dimensions and molecular cortical profiles. (a) Cortical map showing the statistic (F-value) associated with the differential cortical thickness trajectories across the three expressivity dimensions. (b) Associations between differential cortical thickness trajectories across expressivity and each of the six molecular cortical maps considered here.