Similarity between schizophrenia and dementia discovered for the first time

Summary: The study reveals striking similarities in behaviors and neuroanatomical changes between people with schizophrenia and behavioral variant frontotemporal dementia.

Source: Max Planck Institute

Researchers have, for the first time, compared schizophrenia and frontotemporal dementia, disorders that are both located in the frontal and temporal lobe regions of the brain.

The idea dates back to Emil Kraepelin, who coined the term “dementia praecox” in 1899 to describe the progressive mental and emotional decline of young patients. His approach was quickly called into question, since only 25% of those affected had this form of disease progression.

But now, with the help of imaging and machine learning, scientists have found the first valid indications of neuroanatomical patterns in the brain that resemble the signature of patients with frontotemporal dementia.

It is rare for basic scientists to revisit seemingly obsolete discoveries that are more than 120 years old. In the case of Nikolaos Koutsouleris and Matthias Schroeter, who are researchers and doctors, it was even training.

It is Emil Kraepelin, founder of the Max Planck Institute for Psychiatry (MPI) as well as the psychiatric hospital of the Ludwig Maximilian University of Munich (LMU), and his coined term “dementia praecox”. in 1899.

It was his definition for young adults who drift further and further from reality and fall into an irreversible, dementia-like state. Kraepelin lived long enough to see his concept disproved.

By the turn of the 20th century, experts were beginning to use the term “schizophrenia” for these patients, as the disease does not take such a bad course in everyone affected.

Kraepelin came up with the idea of ​​a frontotemporal disease, he speculated that the reason for the sometimes debilitating course of patients was in the frontal and temporal lobe areas of the brain. This is where personality, social behavior and empathy are controlled.

“But this idea was lost because no pathological evidence of the neurodegenerative processes seen in Alzheimer’s disease was found in the brains of these patients,” says Koutsouleris, who works at Kraepelin workplaces, MPI and the LMU.

He continues: “Since I became a psychiatrist, I wanted to work on this question.” Fifteen years later, with sufficiently large datasets, imaging techniques and machine learning algorithms, the professor had the tools at his fingertips to potentially find answers.

He had found the right partner in Matthias Schroeter, who studies neurodegenerative diseases, in particular frontotemporal dementias, at the Max Planck Institute for Human Cognitive and Brain Sciences.

Similarities Between Schizophrenia and Frontotemporal Dementia

Frontotemporal dementia (FTD), especially the behavioral variant (bvFTD), is difficult to recognize in its early stages because it is often confused with schizophrenia. Thus, the similarities are obvious: in sufferers of both groups, personality as well as behavioral changes occur.

An often dramatic development for those affected and their loved ones takes place. As both disorders are localized to the frontal, temporal and insular regions of the brain, it was obvious to also compare them directly.

“They seem to be on a similar spectrum of symptoms, so we wanted to look for common signatures or patterns in the brain,” says Koutsouleris, outlining his plan.

With an international team, Koutsouleris and Schroeter used artificial intelligence to train neuroanatomical classifiers of the two disorders, which they applied to brain data from different cohorts.

The result, which has just been published in the journal JAMA Psychiatrywas that 41% of schizophrenic patients met the classifier criteria for bvFTD.

“When we saw this in schizophrenic patients as well, it rang a bell, indicating a similarity between the two disorders,” Koutsouleris and Schroeter recall.

The research team found that the higher the patients’ bvFTD score, which measured the similarity between the two disorders, the more likely they were to have a “bvFTD-like” phenotype and the less likely they were to improve their symptoms. over two years.

A 23-year-old patient is not cured

“I just wanted to know why my 23-year-old patient with onset symptoms of schizophrenia, such as hallucinations, delusions, and cognitive deficits, had not improved at all even after two years, while another who had started just as badly was to continue his studies and had found a girlfriend. Again and again, I saw these young people who were not recovering at all,” says Koutsouleris.

When the researchers also checked the correlations in high-risk patients like the 23-year-old, they found confirmation at the neuroanatomical level of what Kraepelin had been the first to decisively describe: no improvement in the condition of some patients, quite the opposite.

Neuroanatomical models of schizophrenia. 1 credit

Similar neural structures were affected, in particular the so-called default mode network and the salience network of the brain, responsible for attention control, empathy and social behavior, showed decreases in volume in the area of ​​gray matter that houses neurons. In bvFTD, some neurons (von Economo neurons) perish; in schizophrenia, these neurons are also impaired. This resulted in the neuroanatomical score: after a year, it had doubled in these severely affected people.

For comparison, the scientists had also calculated the Alzheimer’s score using a specific classifier and did not find these effects there.

“This means that the concept of dementia praecox can no longer be completely erased; we provide the first valid evidence that Kraepelin was right, at least in some patients,” Schroeter says.

Now, or in the near future, this means that experts will be able to predict which subgroup patients belong to.

“Then intensive therapeutic support can be initiated at an early stage to exploit any remaining recovery potential,” says Koutsouleris.

Additionally, novel personalized therapies could be developed for this subgroup that promote appropriate maturation and connectivity of affected neurons and prevent their progressive destruction as part of the disease process.

About this news about schizophrenia and dementia research

Author: Press office
Source: Max Planck Institute
Contact: Press Office – Max Planck Institute
Image: The image is credited to Koutsouleris

See also

This shows a female head model

Original research: Access closed.
“Exploring the Links Between Psychosis and Frontotemporal Dementia Using Multimodal Machine Learning” by Nikolaos Koutsouleris et al. JAMA Psychiatry


Exploring the Links Between Psychosis and Frontotemporal Dementia Using Multimodal Machine Learning


The behavioral and cognitive symptoms of severe psychotic disorders overlap with those seen in dementia. However, shared brain alterations remain contested and their relevance for patients at risk stages of the disease has not been explored so far.


Use machine learning to compare the expression of structural magnetic resonance imaging (MRI) patterns of behavioral variant frontotemporal dementia (bvFTD), Alzheimer’s disease (AD), and schizophrenia; estimate predictability in patients with bvDFT and schizophrenia based on sociodemographic, clinical and biological data; and to examine the prognostic value, genetic underpinnings, and progression in patients with high-risk clinical states (CHR) of psychosis or new-onset depression (ROD).

Design, framework and participants

This study included 1870 people from 5 cohorts, including (1) patients with bvFTD (n=108), established AD (n=44), mild cognitive impairment or early-stage AD (n=96), schizophrenia ( n=157), or major depression (n=102) to derive and compare diagnostic patterns and (2) patients with CHR (n=160) or ROD (n=161) to test prognostic relevance and progression of patterns. Healthy individuals (n=1042) were used for calibration of age- and cohort-related data. Data was collected from January 1996 to July 2019 and analyzed between April 2020 and April 2022.

Main results and measures

Case assignments based on diagnostic patterns; socio-demographic, clinical and biological data; Functional results at 2 years and genetic separability of patients with CHR and ROD with high or low pattern expression; and progression from baseline to follow-up MRIs in non-recovery patients versus preserved recovery.


Of 1870 patients included, 902 (48.2%) were women and the mean age (SD) was 38.0 (19.3) years. The bvFTD pattern comprising reductions in prefrontal, insular and limbic volume was more expressed in patients with schizophrenia (65 out of 157 [41.2%]) and major depression (22 out of 102 [21.6%]) than temporo-limbic AD patterns (28 of 157 [17.8%] and 3 out of 102 [2.9%], respectively). bvFTD expression was predicted by high body mass index, psychomotor retardation, affective disinhibition, and paranoid ideation (R2= 0.11). The schizophrenia pattern was expressed in 92 of 108 patients (85.5%) with bvFTD and was related to C9orf72 variant, oligoclonal bands in cerebrospinal fluid, cognitive impairment and younger age (R2= 0.29). Expressions of the bvFTD model and schizophrenia predicted psychosocial impairments at 2 years in patients with CHR and were predicted by polygenic risk scores for frontotemporal dementia, AD and schizophrenia. The findings were not associated with AD or accelerated brain aging. Finally, progression of the bvFTD/schizophrenia pattern over 1 year distinguished patients without recovery from those with preserved recovery.

Conclusions and relevance

Neurobiological links may exist between bvFTD and psychosis focusing on alterations of the prefrontal and salient system. Further transdiagnostic investigations are needed to identify common pathophysiological processes underlying the neuroanatomical interface between the 2 disease spectra.

Leave a Comment