Identifying a neuroanatomical signature of schizophrenia, reproducible across sites and stages, using machine learning with structured sparsity

OBJECTIVE

Structural MRI (sMRI) increasingly offers insight into abnormalities inherent to schizophrenia. Previous machine learning applications suggest that individual classification is feasible and reliable and, however, is focused on the predictive performance of the clinical status in cross-sectional designs, which has limited biological perspectives. Moreover, most studies depend on relatively small cohorts or single recruiting site. Finally, no study controlled for disease stage or medication's effect. These elements cast doubt on previous findings' reproducibility.

METHOD

We propose a machine learning algorithm that provides an interpretable brain signature. Using large datasets collected from 4 sites (276 schizophrenia patients, 330 controls), we assessed cross-site prediction reproducibility and associated predictive signature. For the first time, we evaluated the predictive signature regarding medication and illness duration using an independent dataset of first-episode patients.

RESULTS

Machine learning classifiers based on neuroanatomical features yield significant intersite prediction accuracies (72%) together with an excellent predictive signature stability. This signature provides a neural score significantly correlated with symptom severity and the extent of cognitive impairments. Moreover, this signature demonstrates its efficiency on first-episode psychosis patients (73% accuracy).

CONCLUSION

These results highlight the existence of a common neuroanatomical signature for schizophrenia, shared by a majority of patients even from an early stage of the disorder.

Methylomic changes during conversion to psychosis

The onset of psychosis is the consequence of complex interactions between genetic vulnerability to psychosis and response to environmental and/or maturational changes. Epigenetics is hypothesized to mediate the interplay between genes and environment leading to the onset of psychosis. We believe we performed the first longitudinal prospective study of genomic DNA methylation during psychotic transition in help-seeking young individuals referred to a specialized outpatient unit for early detection of psychosis and enrolled in a 1-year follow-up. We used Infinium HumanMethylation450 BeadChip array after bisulfite conversion and analyzed longitudinal variations in methylation at 411 947 cytosine-phosphate-guanine (CpG) sites. Conversion to psychosis was associated with specific methylation changes. Changes in DNA methylation were significantly different between converters and non-converters in two regions: one located in 1q21.1 and a cluster of six CpG located in GSTM5 gene promoter. Methylation data were confirmed by pyrosequencing in the same population. The 100 top CpGs associated with conversion to psychosis were subjected to exploratory analyses regarding the related gene networks and their capacity to distinguish between converters and non-converters. Cluster analysis showed that the top CpG sites correctly distinguished between converters and non-converters. In this first study of methylation during conversion to psychosis, we found that alterations preferentially occurred in gene promoters and pathways relevant for psychosis, including oxidative stress regulation, axon guidance and inflammatory pathways. Although independent replications are warranted to reach definitive conclusions, these results already support that longitudinal variations in DNA methylation may reflect the biological mechanisms that precipitate some prodromal individuals into full-blown psychosis, under the influence of environmental factors and maturational processes at adolescence.

[Early detection of mental health disorders at school: The Fil Harmonie pilot program]

OBJECTIVE

Most psychiatric disorders arise during adolescence, a period of life during which school takes an important place. School in France has an official mission of health education and prevention, and early detection of mental disorders is part of these goals. The aim of this study is to describe an innovative service operating in Paris that helps educational staff to deal with students having psychological or psychiatric symptoms. The Fil Harmonie program was launched in 2011. It consists of a telephone line available to all educational staff working for high schools in Paris.

METHODS

When in need of assistance, a member of the educational staff can call the dedicated hotline and expose the situation of their student to a trained psychologist. Over the course of the study, data concerning these phone calls were collected such as: socio-demographic characteristics of the student, the reason behind the call, the caller's professional role within the school, and care pathway information. All data collected during the phone calls were anonymized and computerized. We performed an observational descriptive study based on this data by using mixed methods: we integrated quantitative analysis and qualitative research in order to provide a better understanding of the Fil Harmonie program.

RESULTS

Between 18 September 2013 and 12 May 2014, the Fil Harmonie program handled 68 calls from educational staff. Students concerned by the calls were aged between 11 and 22 and the average age was 17.3 years. Over half (52.5%) of the pupils concerned had never seen a mental health professional before the call. In more than 70% of cases, the caller was a school nurse while other professionals such as teachers or headmasters represented only a minority of the callers. Approximately two thirds (67.2%) of students were described by the caller as socially isolated and 48.2% were described as sad or anhedonic. One out of four (26.7%) had repeated a school year at least once, and 55.9% of young people for whom a member of staff contacted Fil Harmonie had been missing class. In 56.7% of cases, there had been no contact with the student's family about the psychological situation. The qualitative analysis particularly highlighted the complexity of the collaboration between the family and the educational staff.

CONCLUSION

Schooling is an important opportunity to seize in mental health regarding early detection and access to care. By fostering collaboration between educational professionals and mental health services, Fil Harmonie meets a public health objective of prevention and should contribute to the reduction of care delays thus leading to better treatment outcome. Our study shows that such programs are feasible and answer a real need in our current health care system.

White matter changes in microstructure associated with a maladaptive response to stress in rats

In today's society, every individual is subjected to stressful stimuli with different intensities and duration. This exposure can be a key trigger in several mental illnesses greatly affecting one's quality of life. Yet not all subjects respond equally to the same stimulus and some are able to better adapt to them delaying the onset of its negative consequences. The neural specificities of this adaptation can be essential to understand the true dynamics of stress as well as to design new approaches to reduce its consequences. In the current work, we employed ex vivo high field diffusion magnetic resonance imaging (MRI) to uncover the differences in white matter properties in the entire brain between Fisher 344 (F344) and Sprague-Dawley (SD) rats, known to present different responses to stress, and to examine the effects of a 2-week repeated inescapable stress paradigm. We applied a tract-based spatial statistics (TBSS) analysis approach to a total of 25 animals. After exposure to stress, SD rats were found to have lower values of corticosterone when compared with F344 rats. Overall, stress was found to lead to an overall increase in fractional anisotropy (FA), on top of a reduction in mean and radial diffusivity (MD and RD) in several white matter bundles of the brain. No effect of strain on the white matter diffusion properties was observed. The strain-by-stress interaction revealed an effect on SD rats in MD, RD and axial diffusivity (AD), with lower diffusion metric levels on stressed animals. These effects were localized on the left side of the brain on the external capsule, corpus callosum, deep cerebral white matter, anterior commissure, endopiriform nucleus, dorsal hippocampus and amygdala fibers. The results possibly reveal an adaptation of the SD strain to the stressful stimuli through synaptic and structural plasticity processes, possibly reflecting learning processes.

Hyper-responsivity to stress in rats is associated with a large increase in amygdala volume. A 7T MRI study

Stress is known to precipitate psychiatric disorders in vulnerable people. Individual differences in the stress responsivity can dramatically affect the onset of these illnesses. Animal models of repeated stress represent valuable tools to identify region-specific volumetric changes in the brain. Here, using high resolution 7T MRI, we found that amygdala is the most significant parameter for distinction between F344 and SD rats known to have differential response to stress. A significant substantial increase (45%) was found in the amygdala volume of rats that do not habituate to the repeated stress procedure (F344 rats) compared to SD rats. This strain-specific effect of stress was evidenced by a significant strain-by-stress interaction. There were no significant strain differences in the volumes of hippocampi and prefrontal cortices though stress produces significant reductions of smaller amplitude in the medial prefrontal cortex (mPFC) (9% and 12%) and dorsal hippocampus (5% and 6%) in both strains. Our data further demonstrate the feasibility and relevance of high isotropic resolution structural ex vivo 7T MRI in the study of the brain effects of stress in small animals. Neuroimaging is a valuable tool to follow up brain volumetric reorganization during the stress response and could also be easily used to test pharmacological interventions to prevent the deleterious effects of stress.