Examining connectomic heterogeneity and connectivity-derived subtypes in schizophrenia
Poster No:
454
Submission Type:
Abstract Submission
Authors:
Shuwan Zhao1, Longbiao Cui2, Yong Liu1, Yongbin Wei1
Institutions:
1School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, China, 2Fourth Military Medical University, Xi’an, China
First Author:
Shuwan Zhao
School of Artificial Intelligence, Beijing University of Posts and Telecommunications
Beijing, China
School of Artificial Intelligence, Beijing University of Posts and Telecommunications
Beijing, China
Co-Author(s):
Yong Liu
School of Artificial Intelligence, Beijing University of Posts and Telecommunications
Beijing, China
School of Artificial Intelligence, Beijing University of Posts and Telecommunications
Beijing, China
Yongbin Wei
School of Artificial Intelligence, Beijing University of Posts and Telecommunications
Beijing, China
School of Artificial Intelligence, Beijing University of Posts and Telecommunications
Beijing, China
Introduction:
Schizophrenia (SCZ) is a severe psychiatric disorder characterized by widespread dysconnectivity in the brain connectome (Jauhar et al., 2022). These disconnectivity patterns are early indicators of SCZ pathophysiology (Cui et al., 2019) and play a key role in the emergence of cognitive and negative symptoms (McCutcheon et al., 2020). However, SCZ display high heterogeneity in clinical features and treatment responses (Alnaes et al., 2019), but it remains largely unknown how dysconnectivity varied across individuals and correlate to clinical variability. Identifying individual variability in the brain connectome and the connectome-based subtypes are crucial for understanding the underlying neurobiological mechanisms and developing more personalized treatment strategies for SCZ.
Methods:
T1-weighted MRI and diffusion-weighted imaging (DWI) were collected from 105 SCZ patients and 97 healthy controls (HC). T1-weighted MRI data were processed using FreeSurfer (v.6.0) (Jenkinson et al., 2012), and the structural connectome were reconstructed using FSL (v.6.0) and CATO (v.3.1.2) from DWI images using deterministic tractography (de Lange et al., 2023). The reconstructed cortical ribbon was parcellated into 219 cortical regions according to a subdivision of the Desikan-Killiany atlas. A 219 × 219 structural connectivity matrix described the streamline volume density (i.e. the number of streamlines between brain regions divided by regional volume) was generated for each participant.
Double generalized linear models Gordon et al., 2016) was first applied to estimate both mean and dispersion of structural connectivity between SCZs and HCs, with age and gender taken as covariates. Network-based statistical analysis (Zalesky et al., 2010) was then performed to identify connectivity with significant mean or dispersion effect. Then, nonnegative matrix factorization (NMF) (Gaujoux et al., 2010) was performed to identify SCZ subtypes based on the altered connectivity. The SCZ subtypes were further linked to clinical assessment and medication sensitivity.
Double generalized linear models Gordon et al., 2016) was first applied to estimate both mean and dispersion of structural connectivity between SCZs and HCs, with age and gender taken as covariates. Network-based statistical analysis (Zalesky et al., 2010) was then performed to identify connectivity with significant mean or dispersion effect. Then, nonnegative matrix factorization (NMF) (Gaujoux et al., 2010) was performed to identify SCZ subtypes based on the altered connectivity. The SCZ subtypes were further linked to clinical assessment and medication sensitivity.
Results:
We found significant differences in the mean connectivity weight in connections of the posterior cingulate gyrus, orbital, medial, and inferior frontal regions, superior and middle temporal regions, and insula between SCZ and HC (Fig1. A). In parallel, more widespread connections showed significantly higher heterogeneity in SCZ compared to HC, involving connections of bilateral frontal, parietal, temporal, and insula regions (Fig1. B). The involved cortical regions revealed in the dispersion model were significantly enriched in the higher-order frontal-parietal, ventral-attention, default-mode networks (P = 0.025, 0.007, and 0.010, respectively; hypergeometric test; Fig1. C). as well as the rich club connectivity (P < 0.050 across degree thresholds 24 – 32, excluding 28) (Fig1. D).
NMF analysis then showed that SCZ were grouped into two distinct subtypes with different disconnectivity patterns, with the top 20% connections contributing to each subtype shown in Fig. 2A. The connections related to the second subtype were found to be significant enriched in hub connectivity (Fig2. B), but not for the first subtype. SCZ patients from the second subtype showed more severe positive symptoms than the first subtype (t = 2.532; P = 0.013; two-sample t-test; Fig2. C). This subtype also lower positive symptom reduction ratios with risperidone monotherapy compared to other treatments (P = 0.045; Fig2. D).
NMF analysis then showed that SCZ were grouped into two distinct subtypes with different disconnectivity patterns, with the top 20% connections contributing to each subtype shown in Fig. 2A. The connections related to the second subtype were found to be significant enriched in hub connectivity (Fig2. B), but not for the first subtype. SCZ patients from the second subtype showed more severe positive symptoms than the first subtype (t = 2.532; P = 0.013; two-sample t-test; Fig2. C). This subtype also lower positive symptom reduction ratios with risperidone monotherapy compared to other treatments (P = 0.045; Fig2. D).
Conclusions:
Our findings reveal that SCZ is characterized by substantial heterogeneity in the structural connectome, concentrated in central connections supporting higher-order cognitive functions. Based on the connectomic heterogeneity, biotype-guided approaches classify patients into distinct subtypes, providing new insights into the personalized treatment strategies in SCZ.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Classification and Predictive Modeling
Connectivity (eg. functional, effective, structural) 2
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
White Matter Anatomy, Fiber Pathways and Connectivity
Keywords:
MRI
Schizophrenia
Other - structural connectivity, heterogeneity, subtype
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
2. Cui, L. B., Wei, Y., Xi, Y. B., Griffa, A., De Lange, S. C., Kahn, R. S., Yin, H., & Van den Heuvel, M. P. (2019). Connectome-Based Patterns of First-Episode Medication-Naive Patients With Schizophrenia. Schizophr Bull, 45(6), 1291-1299.
3. de Lange, S. C., Helwegen, K., & van den Heuvel, M. P. (2023). Structural and functional connectivity reconstruction with CATO - A Connectivity Analysis TOolbox. Neuroimage, 273, 120108.
4. Gaujoux, R., & Seoighe, C. (2010). A flexible R package for nonnegative matrix factorization. BMC Bioinformatics, 11, 367.
5. Gordon, S., Peter, K. D., & Robert, W. C. (2016). dglm: Double Generalized Linear Models (v. 1.8.6). https://CRAN.R-project.org/package=dglm.
6. Jauhar, S., Johnstone, M., & McKenna, P. J. (2022). Schizophrenia. Lancet (London, England), 399(10323), 473-486.
7. Jenkinson, M., Beckmann, C. F., Behrens, T. E. J., Woolrich, M. W., & Smith, S. M. (2012). FSL. Neuroimage, 62(2), 782-790.
8. McCutcheon, R. A., Reis Marques, T., & Howes, O. D. (2020). Schizophrenia-An Overview. JAMA Psychiatry, 77(2), 201-210.
9. Zalesky, A., Fornito, A., & Bullmore, E. T. (2010). Network-based statistic: identifying differences in brain networks. Neuroimage, 53(4), 1197-1207.