Skip to main content
Premium Trial:

Request an Annual Quote

Indiana University Researchers Develop Gene Expression-Based Test for Schizophrenia


NEW YORK – Building on more than two decades of work, a team of researchers led by faculty from Indiana University School of Medicine has developed a blood-based test to diagnose and predict psychotic disorders such as schizophrenia. 

The team has discovered multiple gene expression biomarkers found in blood after conducting long-term studies that included a cohort of psychiatric patients that they followed, both when they were in a highly symptomatic and severe clinical stage and when they were doing better, said Alexander Niculescu, a professor of psychiatry at Indiana University School of Medicine and one of the main developers of the test. 

The researchers followed them when they were hospitalized and when they were out of the hospital, he noted, looking for peripheral biomarkers that could indicate psychotic disorders, either in cerebrospinal fluid or blood, Niculescu said. 

"Patients are not super enthusiastic about getting spinal taps, so we wanted to see if we could find something in blood and develop a liquid biopsy," he said. 

Described in a paper published last month in Molecular Psychiatry, the test begins with whole-transcriptome sequencing. The team's first step was "looking within individuals at what changes in their transcriptome when they're in high-psychosis states versus when they're … asymptomatic," Niculescu said. They then inputted prior evidence in the field, along with their experimental evidence, into an algorithm to prioritize their findings from the discovery step and see if there was "any other converging evidence for the markers that we discovered in the first step," he said. 

Niculescu and his colleagues then took the top-performing RNA gene expression biomarkers from their research to test in an independent cohort of people with severe schizophrenia to determine if the candidate biomarkers were changed even more in that cohort. The biomarkers were also tested in additional individual cohorts to see if they could predict high hallucination and high delusion states and whether a patient would be hospitalized for psychosis in the future. The biomarkers that are predictive of high hallucination states include PPP3CB, DLG1, ENPP2, ZEB2, and RTN4, while the biomarkers predictive of high delusion states include AUTS2, MACROD2, NR4A2, PDE4D, PDP1, and RORA. 

The researchers focused on hallucinations and delusions because "it's very important to have some sort of quantitative phenotypes that you focus on, as opposed to just broadly schizophrenia, which is a very heterogeneous disorder," Niculescu said. 

In addition to tracking psychosis and predicting future severity and hospitalization, the biomarkers can also be used to match patients with medications for their disorders, according to the researchers, who by using bioinformatics can match a patient's gene expression signature with medication gene expression effects. The researchers can rank medications by the percentage they match with a patient's gene expression signature, providing a list for physicians to choose from, Niculescu said, adding the biomarkers may also be of use in drug development clinical trials or for repurposing existing medications, he added. 

Because the RNA markers change with treatment and the disease state, they can also be used to monitor a patient's response to treatment and their disease progression, he said, though he recommends the test be repeated at least once a year for monitoring. "These are dynamic biomarkers; it's not one and done like DNA," he said. 

The panel consists of three parts: one to measure the current severity of disease, one to measure the future risk of severe psychosis, and one to determine potential medication matches. The final test report also includes a risk assessment score. 

Entering the market 

According to Niculescu, the test will be commercially available in about two to three months through MindX Sciences, an Indiana University spinout founded by him and his colleagues four years ago. It already offers tests for mood disorders and suicide risk, and physicians can combine the new test with other panels in the same patient, since the patient's entire transcriptome is being sequenced, he added. "You can assess all the different risk factors for mental health in a single blood draw," Niculescu said. 

MindX works with a national lab, Q2 Solutions, that performs the RNA sequencing for its tests. The sequencing data is then analyzed by MindX, and a report is sent to the physician. Right now, the assays are only available in the US as laboratory-developed tests due to the difficult nature of shipping samples globally, but Niculescu said the company has a long-term plan where the sequencing and gene expression assessment could be done locally in other countries with the data sent to MindX via the cloud for analysis, which would remove "the need for shipping our kits to different far-flung geographic locations." 

Niculescu believes that these tests can help with the clinical assessment of psychiatric patients, allowing physicians to arrive at a diagnosis earlier than with current standards of care. The longer it takes to diagnose these conditions, delaying proper treatment, the "more cumulative damage" occurs in the brain, he added, so getting patients on the right treatment early is imperative. 

While there are pharmacogenetic tests that look at polymorphisms to determine whether certain treatments may be right for a specific patient — Saladax Biomedical, for example, offers clinical chemistry tests that measure levels of antipsychotic prescription drugs — the development of blood-based biomarker testing for psychosis has lagged behind tests for diseases such as cancer. 

However, the field is making some progress in that area. Danish startup GLX Analytix has developed immunoassays to monitor chronic disease and treatment response through changing volumes of glycocalyx in blood and is investigating the use of its technology for schizophrenia, while a research team from the University of California, Irvine, in 2018 described the development of an 18-gene panel to differentiate between patients with bipolar disorder and schizophrenia. 

Adrian Preda, a professor of psychiatry and human behavior at the University of California, Irvine, and the medical editor in chief of the American Psychiatric Association's Psychiatric News, said that schizophrenia is a heterogeneous disorder and that the diagnosis "continues to present conceptual limitations and inconsistencies," which "explains the difficulties we have in detecting valid diagnostic biomarkers." 

With those limitations, "a blood test could be helpful in improving an overall risk assessment, as opposed to a diagnostic test that would rule in or out the diagnosis," he said. In Preda's view, a test like the MindX Sciences assay could be implemented clinically as a risk estimation tool, but further validation in larger independent cohorts would be needed. 

Preda also noted that "'automatizing' the diagnosis of [schizophrenia] via a blood test carries the risk of minimizing the contribution of psychosocial determinants, including the need to consider psychosocial interventions as part of a comprehensive treatment plan for our patients with schizophrenia." 

Currently, the researchers are planning larger clinical utility trials to determine if the use of their test changes medical practice and outcomes, which would help make the case for Medicare and private payor reimbursement of the assay. Because the tests are offered as self-pay and one test costs $1,999, access is currently limited, he added. However, Niculescu noted that the cost of one test is still cheaper than a day of hospitalization in a psychiatric ward. 

The firm is also working on porting its test to a quantitative PCR platform to make it more accessible. The researchers would select the top predictive markers for a smaller panel test that could eventually be performed at the point of care, he said. Right now, the test has a three-to-four-week turnaround time, but if a point-of-care version could be developed it would encourage broader use in hospitals and doctors' offices and significantly cut down that time, he said.