Identification of two clusters within schizophrenia with different structural, functional and clinical characteristics

https://doi.org/10.1016/j.pnpbp.2015.06.015Get rights and content

Highlights

  • We identified two biologically separate clusters within schizophrenia.

  • High cortical curvature and low glucose metabolism characterized one schizophrenia cluster.

  • The other schizophrenia patients clustered with healthy controls and bipolar patients.

  • First episodes and chronic schizophrenia patients were equally distributed across clusters.

Abstract

Background

Several biologically distinct subgroups may coexist within schizophrenia, which may hamper the necessary replicability to translate research findings into clinical practice.

Methods

Cortical thickness, curvature and area values and subcortical volumes of 203 subjects (121 schizophrenia patients, out of which 64 were first episodes), 60 healthy controls and 22 bipolar patients were used to identify clusters using principal components and canonical discriminant analyses. Regional glucose metabolism using positron emission tomography, P300 event related potential, baseline clinical data and percentage of improvement with treatment were used to validate possible clusters based on MRI data.

Results

All the controls, the bipolar patients and most of the schizophrenia patients were grouped in a cluster (cluster A). A group of 24 schizophrenia patients (12 first episodes), characterized by large intrinsic curvature values, was identified (cluster B). These patients, but not those in cluster A, showed reduced thalamic and cingulate glucose metabolism in comparison to controls, as well as a worsening of negative symptoms at follow-up. Patients in cluster A showed a significant putaminal metabolic increase, which was not observed for those in cluster B. P300 amplitude was reduced in patients of both clusters, in comparison to controls.

Conclusions

Results of this study support the existence of a biologically distinct group within the schizophrenia syndrome, characterized by increased cortical curvature values, reduced thalamic and cingulate metabolism, lack of the expected increased putaminal metabolism with antipsychotics and persistent negative symptoms.

Introduction

More than 8000 genetic variants may contribute to the risk of suffering from schizophrenia (Ripke et al., 2013) and recent research supports schizophrenia subtypes characterized by both specific clinical presentation and clusters of genetic variants (Arnedo et al., 2014). Considering the likely effect of genetic variation upon brain structure (Papiol et al., 2005), distinct groups within schizophrenia might be characterized by specific patterns of cerebral alterations.

A possible approach to identify such groups is attempting data-driven (a priori independently of diagnosis) cluster segregation, starting from plausible cerebral variables: i) previously associated to schizophrenia as a group but not unanimously replicated across samples, ii) not primarily related to confounders, such as treatment or chronicity, iii) with likely genetic influence, iv) associated to neurodevelopment and v) likely related to relevant clinical and biological variables.

Cortical thickness may prove useful to this purpose, since cortical thinning has been reported in first-episode (FE) and chronic schizophrenia patients (Crespo-Facorro et al., 2011) (although not unanimously (van Haren et al., 2011)) and in antipsychotic naïve patients (Venkatasubramanian et al., 2008). Cortical thickness is highly heritable (Goldman et al., 2009), with a complex relation with genetic background in schizophrenia (Blasi et al., 2013) and early experience (Whittle et al., 2014). Moreover, cortical thickness increases during normal neurodevelopment (Schmitt et al., 2014) and v) has been associated to treatment response (Szeszko et al., 2012) and cognition (Cassidy et al., 2014) in FE patients.

Cortical curvature measures cortical folding and may reflect different neurobiological underpinnings (Ronan et al., 2011). It may contribute to the proposed clustering since higher (Falkai et al., 2007), normal (Fornito et al., 2008) and reduced (Cachia et al., 2008) gyrification indexes have been reported in schizophrenia and FE patients have shown altered gyrification (Harris et al., 2004). Furthermore, gyrification has genetic underpinnings (Piao et al., 2004) and cerebral gyrification takes place largely in the third trimester (White and Hilgetag, 2008), thus, altered gyrification may result from developmental events in this period. Finally, gyrification is associated to formation of proper cortico-cortical connections (Van Essen, 1997).

The present study has reanalyzed data from previous samples assessing cortical thickness and curvature in order to blindly investigate clustering of schizophrenia patients. Regional cortical area was also included in analyses, since it may convey different information than other structural data in schizophrenia (Rimol et al., 2012). In addition, subcortical volumes were included, given the caudate and thalamic volume association with poor-prognosis schizophrenia (Molina et al., 2010b). Regional cortical volumes were not included since they are largely explained by the corresponding area and thickness. In order to validate clusters, other data were collected, including clinical (baseline symptoms and variation with treatment) and biological (glucose metabolic rate, P300 amplitude and latency) parameters. Chronic bipolar subjects were introduced into the analyses to discard clusters being primarily related to chronicity and/or treatment. Therefore, rather than directly comparing anatomical values depending on clinical diagnosis (schizophrenia, bipolar, healthy controls), our approach was looking for biologically distinct clusters, not considering diagnosis a priori as the primary classifying factor. According to our hypothesis, clusters identified on basis of structural alteration would be expected to be treatment-independent, associated to clinical outcome and/or presentation and associated to other relevant biological markers in order to be considered as valid.

Section snippets

Sample description

Sample included 203 subjects: 121 schizophrenia (Sz) (DSM-IV criteria) patients (64 patients were FE), 22 chronic bipolar patients and 60 healthy controls (HC). Patients were recruited during a psychotic relapse (chronic patients) or first psychotic episode (FE), following their admission to a psychiatric short-term unit. After release they were treated and followed in an outpatient clinic.

MRI data were collected over a 15-year period, in the context of several research projects (identification

Cluster identification

The PCA analysis summary and the plots of the first two components for the four structural variables are shown in Fig. 1. Scatterplots in Fig. 1 clearly show two outliers, which were subsequently eliminated from remaining analyses. Also from these graphs, the PC1CURV (PCA scores in the first factor for curvature) was selected as initial index for generating separate clusters.

In the next step, PC1CURV was used to define three initial clusters: the first included 19 subjects with very large

Discussion

A function, mostly contributed by cortical curvature, segregated 20% of subjects within the schizophrenia group (cluster B), who also showed a more modest cortical thinning and hypometabolism in the left cingulate and thalamus, together with persistent negative symptoms. Cluster A schizophrenia patients showed normal structure but higher putaminal glucose metabolism than HC. Metabolism and clinical improvement were negatively related to curvature. All bipolar and HC subjects were grouped within

Financial disclosure

The authors have no conflicts of interest to declare.

Acknowledgements

This study was supported in part by grants PI08/0017 and PI011/02203 from the Carlos II Institute of the Ministry of Health (Spain), GRS 613/A/11 and 932/A/14 from the Gerencia Regional de Salud de Castilla y León and a predoctoral scholarship grant from the Consejería de Educación — Junta de Castilla y León and European Social Fund to A. Lubeiro.

References (40)

  • V. Molina et al.

    Subcortical and cortical gray matter differences between Kraepelinian and non-Kraepelinian schizophrenia patients identified using voxel-based morphometry

    Psychiatry Res.

    (2010)
  • S. Papiol et al.

    Ventricular enlargement in schizophrenia is associated with a genetic polymorphism at the interleukin-1 receptor antagonist gene

    NeuroImage

    (2005)
  • L.M. Rimol et al.

    Cortical volume, surface area, and thickness in schizophrenia and bipolar disorder

    Biol. Psychiatry

    (2012)
  • L. Ronan et al.

    Consistency and interpretation of changes in millimeter-scale cortical intrinsic curvature across three independent datasets in schizophrenia

    NeuroImage

    (2012)
  • D. Tkachev et al.

    Oligodendrocyte dysfunction in schizophrenia and bipolar disorder

    Lancet

    (2003)
  • S. Whittle et al.

    Positive parenting predicts the development of adolescent brain structure: a longitudinal study

    Dev. Cogn. Neurosci.

    (2014)
  • J. Arnedo et al.

    Uncovering the hidden risk architecture of the schizophrenias: confirmation in three independent genome-wide association studies

    Am. J. Psychiatry

    (2014)
  • E.J. Bartlett et al.

    Effects of haloperidol challenge on regional cerebral glucose utilization in normal human subjects

    Am. J. Psychiatry

    (1994)
  • G. Blasi et al.

    Association of GSK-3beta genetic variation with GSK-3beta expression, prefrontal cortical thickness, prefrontal physiology, and schizophrenia

    Am. J. Psychiatry

    (2013)
  • C. Cassidy et al.

    Association of a risk allele of ANK3 with cognitive performance and cortical thickness in patients with first-episode psychosis

    J. Psychiatry Neurosci.

    (2014)
  • Cited by (22)

    • Neurobiological underpinnings of cognitive subtypes in psychoses: A cross-diagnostic cluster analysis

      2021, Schizophrenia Research
      Citation Excerpt :

      Schizophrenia is likely biologically heterogeneous (Brugger and Howes, 2017; Clementz et al., 2016; Lubeiro et al., 2016), which may explain the lack of a consistent replication of cerebral findings in this disorder.

    • Identificacion of MRI-based psychosis subtypes: Replication and refinement

      2020, Progress in Neuro-Psychopharmacology and Biological Psychiatry
      Citation Excerpt :

      This indicates that illness duration may play a role in cluster distribution, but this would not be the only reason behind this distribution, in agreement with previous results (Dwyer et al., 2018). In the same direction, in our former study the proportion of FE patients was the same between clusters (Lubeiro et al., 2016). In this sample, PANSS scores were not different between clusters.

    • Specific patterns of brain alterations underlie distinct clinical profiles in Huntington's disease

      2019, NeuroImage: Clinical
      Citation Excerpt :

      Interestingly, cellular studies on post-mortem brain tissue of HD patients have found distinctive patterns of neuronal loss associated with motor, psychiatric and mixed (motor and psychiatric) profiles (Tippett et al., 2007; Thu et al., 2010; Nana et al., 2014; Mehrabi et al., 2016). Neuroimaging techniques have been employed in neurodegenerative, psychiatric and neurodevelopmental disorders in order to subdivide patients in different profiles (Hrdlicka et al., 2005; Fair et al., 2012; Gates et al., 2014; Costa Dias et al., 2015; Drysdale et al., 2016; Lubeiro et al., 2016; Park et al., 2017). However, this has never been applied to stratify HD profiles.

    • Deficit of entropy modulation of the EEG in schizophrenia associated to cognitive performance and symptoms. A replication study

      2018, Schizophrenia Research
      Citation Excerpt :

      Further developments may explore the relation between lack of modulation and structural connectivity and SE modulation. Recent reports suggest that structural connectivity may serve to identify biological subgroups (Lubeiro et al., 2016; Sun et al., 2015), in which can be worth assessing the characteristics of SE and MF modulation. This research has been supported by the following grants from Instituto de Salud Carlos III (PI11/02708, PI11/02203 and PI15/00299) and the Gerencia Regional de Salud de Castilla y León (GRS 932/A/14, GRS 1134/A/15 and GRS 1263/A/16) and predoctoral research fellowships to J. Pilar from University of Valladolid and to A. Lubeiro from the Consejería de Educación – Junta de Castilla y León.

    View all citing articles on Scopus
    View full text