NEW YORK—Biomarkers for disease—and companion diagnostics based on those markers—have proven a boon in a variety of disease states, helping clinicians to better predict disease progression and patient response to treatment. And while biomarkers have traditionally been used to look for physical markers of disease, a new approach is exploring the potential for imaging biomarkers for neurological conditions—specifically post-traumatic stress disorder (PTSD) or major depressive disorder (MDD).
This effort comes out of a study that found that a brain imaging biomarker could help stratify and guide treatment decisions for individuals with PTSD or MDD, conditions that lack tools to predict patient treatment response. The work, detailed in a paper titled “Identification of psychiatric disorder subtypes from functional connectivity patterns in resting-state electroencephalography (EEG),” was published in Nature Biomedical Engineering. The study was largely funded by Cohen Veterans Bioscience (CVB), and was led by Dr. Amit Etkin, Stanford University professor of psychiatry and behavioral sciences.
“There is a great need in psychiatry for objective tests that can inform diagnostic development and clinical treatment decisions for heterogeneous conditions such as PTSD and MDD,” said Etkin. “Our findings are exciting because they reflect progress towards identifying evidence-based biomarkers, and they also demonstrate the value of machine learning techniques for advancing a personalized approach to treatment—which are part of a tipping point in the field.”
By applying advanced machine learning techniques to high-density resting-state EEG signals, the team identified distinct functional connectivity patterns in patients’ brain circuits that enabled subtyping based on treatments. The researchers discovered the subtypes while working with “a PTSD discovery dataset involving 201 participants (106 with PTSD and 95 healthy controls),” as noted in the paper, and then performed replication analyses “using two independent PTSD datasets with 72 and 63 patients each (many with concurrent medication use); and second, using two MDD datasets (with 228 patients that were medication free and 179 patients, many with concurrent medication use, respectively) to determine the transdiagnostic potential of our subtyping results.”
“The traditional approach for studying the neurobiology of psychiatric conditions has followed this diagnostic framework through case-control studies whereby all patients with a given diagnosis are compared with healthy individuals. However, this approach has failed to deliver on hoped-for biomarkers due to high biological heterogeneity among patients with the same diagnosis and among healthy controls,” the authors explained.
“More importantly, such biological heterogeneity has substantial effects on treatment outcome, even while often being independent of pre-treatment clinical symptoms. For example, while antidepressants have only modest superiority over placebo, this is in part because the clinical diagnosis of MDD encompasses biologically heterogeneous conditions that relate differentially to treatment outcome,” they continued. “Likewise, even though psychotherapy is at present the most effective treatment for PTSD, many patients are nonetheless non-responsive and display differences in brain function relative to responsive patients. In neither case, however, are neurobiological differences related to clinical features, supporting the potential unique value of neurobiology in defining clinically relevant ‘disease subtypes.’ These subtypes may furthermore exist within or between traditional psychiatric diagnoses.”
They discovered that while subtypes found in PTSD and MDD didn’t differ in terms of the severity of symptoms before treatment, they did differ in their responses to treatment. Subtypes were determined based on functional connectivity patterns, neural signatures identified via EEG. According to a CVB press release, “similarly strong subtype-related connectivity differences” were found among PTSD patients and in MDD patients. One subtype had a poor response to both psychotherapy and antidepressants, but both subtypes had similar responses to noninvasive transcranial magnetic brain stimulation (TMS).
Moving forward, the authors noted that additional exploration would be necessary to eliminate limitations of the current study and to directly compare TMS to psychotherapy or antidepressant use based on subtype. Etkin is working to advance the research into development for clinical use, having founded Alto Neurosciences.
“These discoveries have significant implications as they help stratify individuals independent of clinical diagnosis based on what may represent a new transdiagnostic biomarker. This will enable discovery of a new generation of precision therapeutic discoveries and targeted treatments,” remarked Dr. Andreas Jeromin, chief scientific officer of CVB.
