Originally published 2021-10-08.
In October 2021, researchers at the University of California, San Francisco (UCSF) reported that a small brain implant had relieved the severe, long-standing depression of a 36-year-old woman known publicly only as Sarah. Nothing else had worked for her: years of antidepressant combinations and electroconvulsive therapy had failed. What made the device different was that it was tuned to her own brain. It watched for the specific pattern of activity linked to her worst symptoms and responded with a brief electrical pulse only when that pattern appeared.
The study, published in the journal Nature Medicine, involved one patient. It is a proof of concept, not a new treatment anyone can ask for, but it was a striking demonstration that depression might one day be treated by targeting the circuit involved in each individual.
Who the patient was
Sarah’s depression had become severe and unrelenting. Doctors call this treatment-resistant depression: the illness continues despite several adequate trials of medication and other standard treatments. Speaking at a press briefing, she said the illness had left her unable to function, and described the implant as having kept her depression at bay so that she could start rebuilding her life.
One detail caught the public’s attention. After the implant was in place, Sarah reported laughing spontaneously for the first time in about five years. Recalling the first test stimulation, she said she felt “the most intensely joyous sensation.”
How the implant works
The UCSF team, led by psychiatrist Katherine Scangos with neurosurgeon Edward Chang and psychiatrist Andrew Krystal, worked in two stages.
Step 1: mapping her depression
Surgeons first placed temporary electrodes at ten sites in Sarah’s brain and recorded activity for 10 days while she rated her mood. The team also tried brief stimulation at different sites to see which ones changed how she felt. They found that bursts of high-frequency activity in the amygdala, a structure involved in processing threat and emotion, lined up with her worst symptoms. Stimulating a region called the ventral capsule/ventral striatum, part of the brain’s reward circuitry, reliably lifted her mood.
Step 2: a device that responds only when needed
The researchers then implanted a commercially available device, the NeuroPace RNS System, normally used for epilepsy. One lead sat in the amygdala to listen; the other sat in the ventral striatum to stimulate. When the device detected the amygdala pattern, it delivered a six-second pulse, too weak for Sarah to feel. The number of pulses per day was capped, partly to avoid disturbing her sleep.
This is called closed-loop stimulation. Older deep brain stimulation for depression usually sends pulses continuously, whatever the brain is doing. Delivering stimulation only when the symptom signal appears is a more targeted approach.
What the results showed and what they do not
Sarah’s symptoms improved rapidly and she went into remission, and had stayed well for about a year of treatment at the time of reporting. Scangos described the work as a kind of personalised therapy that psychiatry had not been able to offer before.
Important limits apply:
- One person. A single case cannot show how often the approach works, or whether improvement might partly reflect expectation.
- Individual circuits. The team chose Sarah’s target based on her own brain map. Other people’s depression may involve different regions and patterns, and it is not known whether a usable biomarker exists for everyone.
- Brain surgery. Implants carry risks such as bleeding and infection, and the device needs long-term follow-up.
- Unknown duration. The researchers said it was unclear how long the stimulation would be needed.
At the time, two more patients had been enrolled, and the team aimed to recruit nine more to see whether the method works more widely.
Treatment-resistant depression in context
Depression is common and usually treatable, but a substantial minority of people do not get enough relief from the first medicines they try, and some do not improve after several. For these patients, established options include switching or combining medicines, psychotherapy, electroconvulsive therapy, transcranial magnetic stimulation and, in some countries, esketamine nasal spray.
Deep brain stimulation has been studied for depression since the early 2000s, with mixed results in larger trials. One reason may be that the same target was used for everyone. The UCSF case suggests that individual mapping could help explain why some people respond and others do not.
Anyone living with depression that has not responded to treatment should talk to a doctor or psychiatrist about the options. If you are in crisis or thinking about harming yourself, contact local emergency services or a crisis line straight away.
Common misconceptions
“It is a cure for depression.” It relieved one person’s severe depression in an early study. Larger trials are needed before anyone can say how well it works.
“The implant makes you constantly happy.” The goal is to interrupt a symptom pattern, not to produce euphoria. The joyful feeling Sarah described came during early testing; day-to-day, the pulses were set below the level she could feel.
“It shocks the brain all day long.” The often-quoted figure of 300 pulses a day was a daily limit set by the team. Each pulse lasts six seconds, is too weak to feel and is triggered only when the target pattern appears.
Why it matters
For decades psychiatry has diagnosed and treated depression mainly by symptoms, without a measurable brain signal to guide treatment. This study showed that, at least in one person, a specific signal could be identified, tracked and interrupted in real time. Scangos called it an advance in understanding the brain function behind mental illness. Whether that becomes a practical therapy depends on what happens when many more patients are studied.
Sources: The Guardian, 4 October 2021; Physics World, 5 October 2021: Personalized brain stimulation could treat untreatable depression; Futurism, 5 October 2021. Study: Scangos et al., “Closed-loop neuromodulation in an individual with treatment-resistant depression”, Nature Medicine, October 2021.

COMMENTS