For decades, brain imaging has generated compelling evidence for brain atrophy in people diagnosed with schizophrenia. Brain atrophy involves the loss of brain tissue—the loss of both neurons and the connections they form—and can affect both grey and white matter.

But, note authors of a new paper in Nature Mental Health, “there remains substantial heterogeneity across different studies, with some reporting decreases in regional grey matter volume in the prefrontal cortex…while others have also found reductions in regions of the temporal lobe, occipital cortex, and subcortical structures.” Similarly, some studies have found increased functional connectivity within certain brain networks in schizophrenia patients (including the default mode network, the salience network, and frontoparietal networks) while other studies have reported decreased connectivity.

The researchers, led by Ahmed T. Makhlouf, M.D., and Shan H. Siddiqi, M.D., both of Mass General Brigham and Harvard Medical School, stress that in spite of the inconsistent data about brain atrophy in schizophrenia patients, “when different methods yield different answers in a similar patient population, it may not be necessary to discard this information,” particularly if these varying results “share something in common.”

Dr. Siddiqi, winner of the 2022 BBRF Klerman Prize for Exceptional Clinical Research and a 2019 BBRF Young Investigator, and his colleagues report they have found a potentially significant commonality. They describe how varying patterns of brain atrophy in schizophrenia patients “localize to a common brain network” that they believe is fairly stable across the course of schizophrenia and which may be useful in diagnosing the illness as well as a source of potential targets for future treatments.

To date, the heterogeneity in neuroimaging results has made it difficult to develop reliable biomarkers for schizophrenia, as well as targeted interventions. Dr. Siddiqi and colleagues note, for example, the difficulty so far in finding places in the brain that might be therapeutically targeted by various brain stimulation technologies.

The team used a method called coordinate network mapping (CNM) which Dr. Siddiqi and others have used to discover common networks in a range of illnesses, including depression, addiction, Alzheimer’s, and Parkinson’s dementia. The networks identified in each case are interconnected brain regions defined according to their functional connectivity. The identified networks are shared across a wide range of patients with each diagnosis, and are specific to each disorder.

The process of trying to identify a commonly affected neural network in schizophrenia patients with brain atrophy began with utilizing neuroimaging data from 113 peer-reviewed scientific articles comprising 11,270 participants in four clinical groups, as well as 6,007 healthy controls. The final analysis in the team’s paper incorporated results from 90 published studies of atrophy in schizophrenia involving over 8,000 individuals. CNM is a complicated methodology that involves leveraging an already existing map of the human connectome based on scans of 1,000 healthy people—a consensus wiring diagram of the brain—and mapping specific coordinates derived from the neuroimaging scans of the 8,000+ patients with atrophy onto these connectome data, “providing a more comprehensive understanding of connectivity patterns of atrophy locations across the brain.” The analysis yielded a functional connectivity map of each study’s reported atrophy patterns. Connectivity maps of atrophy patterns across the dataset were then compared with 10 control groups that showed atrophy patterns in normal aging and in nine brain disorders.

The study yielded several findings about patterns of brain activity in schizophrenia, the team said. First, the heterogenous published atrophy coordinates from the 90 studies “were united [using CNM] by a specific pattern of connectivity to one common network.” This common network was “relatively stable at different stages” of schizophrenia, and was “specific to schizophrenia, compared with [data from scans of] normal aging and patients with other brain disorders.” Different clusters of symptoms seen in different patients with schizophrenia still “localized to similar brain networks.”

The study broadly supported the “interconnected nature of brain regions involved in schizophrenia,” the team noted, and enables useful comparison with data gathered by a large international imaging consortium on schizophrenia called ENIGMA. “Our results replicate several regions identified in ENIGMA studies, including the bilateral insula, hippocampus, temporal gyri, and fusiform cortex. This congruence demonstrates the value of integrating diverse neuroimaging research to converge on a more unified understanding of schizophrenia.”

The team remarked on “the singular nature of the schizophrenia network” they identified. The distinctiveness of the network, they said, “highlights the potential of using network-based approaches as a therapeutic target and potentially a diagnostic tool for schizophrenia as a whole.”

The study generated another important result that bears on the longstanding problem of trying to identify which individuals deemed at high risk for schizophrenia will progress to a diagnosis. This happens only in about 1 in 3 high-risk individuals, but which ones are most at risk is still impossible to determine. (“Risk” is typically assessed using genetic and/or behavioral information.) Dr. Siddiqi and colleagues reported that “a similar network [to the common network they identified in the atrophy scans] was found in imaging data from people at high risk for schizophrenia.”

Importantly, “atrophy patterns in high-risk patients who progressed to schizophrenia” had a distinct characteristic: these scans “showed more connectivity [in the pattern of atrophy] to the brain’s medial temporal lobe and anterior cingulate cortex.” This could mean that people with this pattern are more likely to progress to schizophrenia, the team said, or, alternatively, that progression to schizophrenia may lead to atrophy in these specific regions. The result, whatever the explanation, is a prompt for additional research.

The team also included Joseph J. Taylor M.D., Ph.D., 2022 BBRF Young Investigator; and David Silbersweig, M.D., 1996 BBRF Young Investigator.