IMAGE: RAISSA MATHURA
improved over 1 year or more—beyond the
timeline of the initial study, says Stanford
University psychiatrist Mahendra Bhati,
a co-investigator. In October, he and col-
leagues published a follow-up study of
eight trial patients, most of whom con-
tinue to use their implant about 10 years
later. About one-half have had at least a
50% improvement over their pretreatment
score on a depression scale.
Many researchers say that because de-
pression can manifest differently in dif-
ferent brains, personalizing the treatment
could make it more effective. Several teams
are now exploring more precise stimula-
tion approaches informed by individual
brain anatomy and neural recordings.
Helen Mayberg, a neurologist at the
Icahn School of Medicine at Mount Si-
nai who has pioneered the use of DBS in
depression, has led one effort. She and
colleagues previously identified an area
near the front of the brain called the sub-
callosal cingulate (SCC) as a hub in a
network linked to the negative mood
component of depression. The halted ran-
domized St. Jude Medical trial, known as
BROADEN, targeted the SCC. Mayberg sus-
pects subtle differences in where surgeons
placed the electrodes account for much of
the variation in patient outcomes in that
trial. So her group and others now use an
MRI approach called tractography to visu-
alize locations of nerve bundles and target
electrodes more precisely.
Mayberg is also tracking how the brain
responds to this more precise stimula-
tion, hoping to identify neural responses
that predict whether and when a person’s
symptoms will improve. Last month in
Translational Psychiatry, she and col-
leagues describe a change in activity in the
SCC during brain stimulation in the oper-
ating room that correlates with treatment
response in the week after surgery in seven
studied patients.
Neuropsychologist Isidoor Bergfeld
and colleagues at Amsterdam University
Medical Center are testing tractography-
guided DBS implants in another part of
the brain implicated in depression known
as the superolateral branch of the me-
dial forebrain bundle. Bergfeld aims to
recruit 24 patients and expects results in
2 to 3 years. He hopes data from his group’s
trial and from a similar ongoing study in
Germany can combine to support regu-
latory approval in the European Union
for depression.
Other teams have more extensive person-
alization in mind. Psychiatrist Katherine
Scangos of the University of California, San
Francisco, and colleagues are using an in-
dividual’s neural data to decide which tar-get region to stimulate and exactly when
to deliver the stimulation. Before insert-
ing DBS electrodes, the team implants a
separate set of electrodes that can record
from and stimulate tissue across diverse
regions of the brain. Surgeons already rely
on the approach, known as stereotactic
electroencephalography (sEEG), to iden-
tify the source of seizures before epilepsy
surgery. Now, Scangos’s team has used it to
stimulate mood-related regions and select
a target where stimulation reliably relieves
depression symptoms.
For the first patient, of a planned 12 in an
ongoing trial, the ideal stimulation target
seemed to be the ventral capsule/ventral
striatum (VC/VS), already an established
target in DBS depression trials. The team
also found that activity in a second site in
the patient’s brain, the amygdala, stronglypredicted periods of severe symptoms.
So the patient’s implant—a device called
NeuroPace, which is approved to detect
and stop seizures—was programmed to
stimulate the VC/VS only in the presence
of those amygdala activity patterns. “Every
time that train sort of veers off the tracks,
we push it back on,” Scangos says. That
first patient has felt significant relief from
her symptoms for more than 1 year since
the surgery, Scangos and her colleagues re-
ported in October in Nature Medicine.
Other researchers are waiting for more
evidence that this automated, “closed-
loop” approach to the timing of stimula-
tion is necessary. “With severe depression,
it’s not like you’re alternating between re-
ally depressed and totally normal ... over
moments or hours,” Sheth says. His team is
using sEEG electrodes to find the best set-tings for continuous DBS in each patient.
During a 10-day hospital stay, the research-
ers use sEEG readouts to identify patterns
of brain activity associated with a posi-
tive mood and track how those readouts
change in response to DBS stimulation
from electrodes in both the SCC and VC/
VS. A computer algorithm then identifies
the ideal frequency, amplitude, and other
settings for improving mood.
Data from the first participant in an on-
going trial, published on 22 November in
Biological Psychiatry, show his symptoms
improved in the first few months after
surgery—and then steadily worsened dur-
ing a blinded discontinuation phase, when
stimulation at some of the sites was re-
duced gradually at a time point unknown
to the patient. (Researchers then turned
the implant back to full power.)Mayberg cautions that to be used widely
in the clinic, DBS will have to be cost effec-
tive and simple enough “that your average,
competent neurosurgeon and depression
expert can ... implement it.”
Sheth says that implanting the set of
sEEG electrodes to guide treatment is likely
too invasive and labor-intensive to be fea-
sible for every patient, though it might offer
a chance to adjust DBS in patients whose
symptoms aren’t responding. He also hopes
researchers can develop less invasive meth-
ods to inform stimulation settings.
“We’re the first to admit that time will
tell whether this approach or variants
of it can successfully be applied,” Sheth
says. But the variety of ambitious DBS ap-
proaches being tested is a good sign, he
says. “I’m just happy that groups are still
fighting to help all these patients.” jSCIENCE science.org 3 DECEMBER 2021 • VOL 374 ISSUE 6572 1183Recording sites on implanted electrodes (dots) defined optimal settings for a stimulation device (gray lines).