ARTICLE IN BRIEF
Deep brain stimulation for depression has not been shown effective in several recent trials, but Helen Mayberg, MD, the investigator who pioneered the technique, is still focused in doing more research.
Patients with intractable depression treated with deep brain stimulation (DBS) describe the changes the treatment can produce as nothing less than life-altering. But results from the first published double-blind trial reported last December in Biological Psychiatry indicated it is no better than placebo at relieving depression.
A second trial by a different group of investigators was halted early for lack of benefit. In light of these results, device manufacturers are pulling back, said the lead investigator of the trial.
But Helen Mayberg, MD, who pioneered the technique, is undeterred, and is looking for answers in the brain connection signatures in those patients whose lives have been profoundly improved by deep brain stimulation. Dr. Mayberg, a professor of psychiatry, neurology, and radiology at Emory University in Atlanta, provided an update of her work here at the AAN Annual Meeting in April.
Dr. Mayberg treated her first depressed patients with DBS over a decade ago, the culmination of a theoretical and preclinical development program that first and foremost saw depression as a neurological disorder. The mood, motor, cognitive, and autonomic symptoms of depression were understood as the outputs of a circuit disorder, similar to the slowed movements and tremor of Parkinson's disease. The efficacy of DBS for Parkinson's disease “provided a roadmap for deconstructing depression,” and imaging of both healthy and depressed patients began to fill in the nodes in the circuit, she explained.
Over time the model evolved to include frontal and limbic areas “working in a choreographed and reciprocal manner,” with limbic areas responding to medication and driving changes in the cortex, while cognitive therapy induced cortical changes, which then drove limbic changes.
Dr. Mayberg then asked, “Is there a node in this widespread model, like the substantia nigra in Parkinson's disease, that is the leader rather than the follower?” While the psychiatric model of depression assigned equal weight to all symptoms, “when you listen to patients, the worst part of being depressed isn't that you aren't working or sleeping, it's that you are experiencing,” in the words of her patients, “an ‘active anguish,’ or a ‘gnawing agony,’ a ‘pain that makes you hate yourself.’”
On the hypothesis that negative mood was the “core symptom” that drove the rest of the disorder, Dr. Mayberg and colleagues began to focus on the subcallosal cingulate, Brodmann area 25, which is “activated in the setting of negative mood disproportionately to every other region. It drags with it other limbic regions, and is accompanied by a shutdown of other regions,” including prefrontal areas linked to intention. “We started to see that area 25 actually had a dominant role,” becoming activated even in healthy people confronted with sad memories or stressful situations, and becoming less active with successful treatment with antidepressants or even placebo.
Ultimately this led her to the hypothesis of DBS for depression: “If you have a patient who you can't talk, drug, or shock into recovery, maybe one can leverage the use of focal brain stimulation to actually downregulate the region directly. The hypothesis was that high-frequency stimulation would block the activity in area 25, and disinhibit those regions that were connected — the frontal cortex, basal ganglia, brainstem, and thalamus.”
Of the first six patients treated in 2003, half were in remission after six months, and one other patient did well. Imaging confirmed the disinhibition of the prefrontal cortex and other areas linked to area 25. That study was extended to 20 more patients, with similar results sustained for one year. “If you got better you stayed better,” Dr. Mayberg said. Patients spoke of “feeling lighter,” and commented that “the void in my head is gone,” and “I feel more connected.” When the batteries weakened, patients worsened, and when they were replaced, they improved again.
OTHER DOUBLE-BLIND TRIALS
Those extraordinary early successes led other groups to try DBS for treatment-resistant patients, and ultimately to two double-blind, sham-controlled clinical trials. “The problem is that as it has ramped up in clinical trials, it hasn't gone well,” Dr. Mayberg said.
One trial, called BROADEN (which stands for BROdmann Area 25 DEep brain Neuromodulation), in which electrodes were implanted into area 25, was halted in 2014 when an interim analysis indicated the chances were very slim that the final results would show any benefit. That trial was sponsored by St. Jude Medical Inc.
A second trial, sponsored by Medtronic, in which the target was the ventral capsule/ventral striatum, which had also shown striking benefit in open studies, also failed.
“It looks like, at least for depression, that open-loop DBS, using approaches comparable to those used in movement disorders, is in fairly bad shape,” said Darin Dougherty, MD, director of neurotherapeutics in the department of psychiatry at Massachusetts General Hospital, who led the trial sponsored by Medtronic. “Open-loop” refers to continuous stimulation, versus closed-loop stimulation, as is used in epilepsy, in which aberrant electrical activity in the brain is monitored and becomes the trigger for corrective stimulation.
The placebo response in the Medtronic trial was approximately 25 percent, “which isn't a lot,” Dr. Dougherty acknowledged, but he noted that it is higher than what was seen in the open studies. “The bottom line is that we can't separate out active treatment from placebo. Whether that is a dulled active or an overactive placebo, or a combination, it's bad news [for DBS as an intervention].
“It is confounding,” he continued, “given the degree of response in these patients who had failed to respond to other treatments. They had been so sick for so long, that to see them get better with open-loop DBS was astounding. I think [both Dr. Mayberg and I] were both shocked the controlled trials did not capture that signal.”
The results were not like the sham-controlled trials of DBS for Parkinson's disease, he said, in which the results, while more modest than in open trials, “were still statistically and clinically significant. Here, we fell off a cliff.”
Dr. Dougherty suspects that the results of these trials may mean an end for this iteration of the technology. “Given the investments in these pivotal trials, the manufacturers would have to have some awfully compelling reason to revisit these targets for open-loop stimulation from a regulatory standpoint. I think for some of the more complex disorders, like depression and substance abuse, we may have to move to more closed-loop technologies.”
Dr. Mayberg, meanwhile, hopes to supply that compelling reason, by looking more closely at her patients to ask what separates those who achieve remission from those who don't. “We went back to neurology to consider the source of variability,” she said, using diffusion tractography to obtain fine-scale maps of the individual variations in connections between brain regions in her patients.
Based on those studies, she said, “Everyone that was well at six months with the stimulation as it was originally done shared the same combination of targeted tracts. Reaching the medial frontal cortex makes the difference for getting well and having real effects.” Conversely, “non-responders were those who didn't have changes remote from the target,” which in turn appears to be due to not placing the lead deep enough into the tracts that lead out of area 25. Those results have led to inclusion of pre-surgical tractography and electrode placement based on the individual patient's wiring, with an increase in the success rate, albeit still with open-label treatment.