Researchers say they have identified an objective biomarker of recovery in major depression that in a small cohort of treatment-resistant patients was able (on an ongoing basis) to capture whether each patient was in a “depressed” or “stable response” state while being treated with deep-brain stimulation (DBS). Unlike some other proposed depression biomarkers, this one was found to be present in all of the patients analyzed; its detection does not require individualization from patient to patient.

The result, which awaits validation in larger trials, would be a first, and could potentially have major implications for treating depressed patients, whose symptoms vary, often markedly, between individuals and also over time as treatments are being administered. The vast heterogeneity of depression often makes it hard for clinicians to know whether or how to adjust treatments over time—in the case of DBS, the level of current that is delivered by the pacemaker-like device implanted in the brain

DBS for treatment-resistant patients with severe major depression was pioneered in 2003-2005 by Helen S. Mayberg, M.D., and colleagues in Toronto, research supported by Dr. Mayberg’s 2002 BBRF Distinguished Investigator grant. It remains an experimental treatment for severe major depression that doesn’t respond to conventional treatments. It involves the surgical implantation of electrodes in the brain’s subcallosal cingulate, a small area behind and above the eyes which is also known as “area 25.” Contacts from the electrodes are placed with great precision using individualized “tractography guidance” at the intersection of four white matter tracts—bundles of nerve fibers that enable neurons in different brain areas to communicate and which are implicated in depression. In 2019, Dr. Mayberg and colleagues published the results of multi-year follow-ups of 28 DBS patients. “Robust and sustained” antidepressant responses were achieved in 21 of the 28 cases, one of which at that point had extended over 18 years.

The new study, published in Nature, reports on a new cohort of patients receiving subcallosal cingulate DBS for treatment-resistant depression. It was co-led by Dr. Mayberg, Patricio Riva Posse, M.D., and Christopher Rozell, Ph.D. Dr. Mayberg, a neurologist who heads the Nash Family Center for Advanced Circuit Therapeutics at the Icahn School of Medicine at Mount Sinai, is a member of BBRF’s Scientific Council, winner of the 2007 BBRF Falcone Prize for Outstanding Achievement in Affective Disorders Research, and the recipient of additional BBRF grants in 1995 and 1991. Ki Sueng Choi, Ph.D., a 2016 BBRF Young Investigator, and Allison C. Waters, Ph.D., a 2019 BBRF Young Investigator, were among members of the team.

Clinical management of DBS patients “is often complex,” Dr. Mayberg and colleagues on the new paper point out. The progress of the antidepressant response is “non-linear and different for each individual,” and commonly accompanied by periods of mood fluctuations despite overall improvement. “These intervals of transient but significant distress can be difficult for doctors to distinguish from the early return of depression symptoms,” Dr. Mayberg notes. Without objective markers of depression severity, clinicians rely on patients’ self-reports, questionnaire-based depression-score calculations, and their own clinical experience to decide whether to adjust the device in the brain that delivers DBS stimulation—or to adopt a watchful waiting approach.

In the new study, the team took advantage of a new DBS device that in addition to delivering stimulation to area 25 also collected brainwave data saved to the implanted device, which could be downloaded by the researchers at weekly intervals over the first 6 months of treatment following the implant. The DBS device delivered no stimulation for the first several weeks, enabling the team to carefully collect data on each patient while in the “depressed” state. This data was compared with readings taken in the final 4 weeks of the 24-week trial period.

Using standard assessment tools for depression symptoms, at the 24-week point, 9 of the 10 patients had demonstrated a “robust clinical response,” meaning a reduction in depression symptoms of at least 50%. Seven of the 10 had achieved a remission of symptoms—they no longer met the criteria for a depression diagnosis. Full data was obtained for 6 of the 10 participants. Five of those six had remissions; one, after responding for 4 months, then suffered a relapse.

Using artificial intelligence technology, the team was able to tease out subtle patterns in electrophysiological data which enabled them to discover a pattern in all six subjects that corresponded with positive response to the treatment. Interestingly, this response (as is vexingly typical in depression) occurred as early as the 8-week point for one participant and as late as the 20-week mark in another.

Just as important, the team was able to see a change in the “recovery” signal in the participant who relapsed. This signal, in fact, was found (in retrospective analysis) to have appeared a full month before the patient’s relapse occurred. Although it will have to be replicated in many other patients, such a result would in theory provide doctors with an advance warning that a patient who was showing signs of stable recovery was regressing and thus a candidate for an adjustment of his or her DBS device.

The study yielded two other important results. Videos taken of the participants as they progressed through the treatment period corroborated what many psychiatrists have long thought to be true: subtle changes in facial expression correlated with a reduction in depression symptoms. In this case, the changes were measured with great precision and patterns were identified, again using AI technology. In their paper the team offers potential explanations of the linkage between brain state and involuntary facial expressions.

Using MRI scanning technology, the team was also able to show that the white matter bundles in area 25 targeted by DBS treatment showed distinct irregularities (before the DBS treatments began) that turned out to correlate with both the number of past depression episodes and the time required to respond to DBS stimulation.

All of the team’s findings must be replicated. For their part, Dr. Mayberg and colleagues are already assessing results in another cohort of patients at Mount Sinai, again using a DBS system that can both deliver stimulation and sense signals such as the “depressed”/”recovery state” patterns reported in the current paper. The aim is to provide robust and reliable biomarker-based clinical-decision tools that can streamline and optimize DBS management and contribute to adoption of this treatment option in future.

There is also the possibility that the “recovery” signal found in the new research, if validated, could inform methods of treating depression that do not involve invasive surgery—such as TMS, or transcranial magnetic stimulation, in which magnets are used, non-invasively, to apply stimulation beneath the scalp in order to modify activity in underlying brain regions affected by depression pathology.