How does TMS work in the brain to treat depression?
TMS appears to work by normalizing overactive connectivity in the default mode network, the brain system tied to rumination and self-referential thinking. This was a small resting state fMRI study of 17 depressed patients, scanned before and after a five week course of TMS and compared with 35 healthy control subjects, so it is a mechanistic imaging study rather than a clinical efficacy trial.
The default mode network is a collection of brain regions that are active when a person is not focused on an external task. It has been implicated in rumination, self-referential processing, and episodic memory retrieval, and it includes areas of medial prefrontal cortex, posterior cingulate cortex, and multiple, mostly medial, areas of posterior parietal cortex.
What the data show:
- Baseline abnormality: Functional connectivity in the depressed patients was abnormally elevated within the default mode network and diminished within the frontoparietal central executive network, with altered connectivity between the two networks
- Network normalization: TMS normalized the depression related subgenual cingulate hyperconnectivity in the default mode network
- Selectivity: TMS did not alter connectivity within the central executive network, so the effect was not a generalized shift across the whole brain
- Connectivity changes: TMS induced anticorrelated connectivity between the dorsolateral prefrontal cortex and medial prefrontal default mode network nodes
- Prediction: Baseline subgenual connectivity predicted subsequent clinical improvement, which points toward a possible imaging biomarker of treatment response
This mechanistic study published in Biological Psychiatry used resting state functional MRI to measure connectivity within and between the default mode network and the central executive network, seeded from the dorsolateral prefrontal cortex and from the subgenual cingulate, a region closely aligned with the default mode network in depression.
Dr. Kumar’s Take
Repetitive TMS of the dorsolateral prefrontal cortex has been an established depression treatment for years, and its mechanism of action has remained unknown. I find this study useful because it gives a specific circuit level account rather than a vague one: stimulation over the dorsolateral prefrontal cortex was followed by normalization of subgenual cingulate hyperconnectivity in the default mode network, while the central executive network was left unchanged. That selectivity matters to me clinically. It argues that TMS is acting on a particular abnormal circuit rather than raising activity everywhere. The finding that baseline subgenual connectivity predicted later improvement is the part I would most want replicated, because a pretreatment scan that sorts likely responders from unlikely ones would change how I counsel patients about a five week commitment. This is 17 patients. It is a mechanism study, not a reason to change practice on its own.
How This Works (Biological Rationale)
Acutely, TMS produces transient current flow and neuronal depolarization in the cortical tissue directly beneath the stimulation site and in interconnected downstream circuits. Over a longer course, repetitive TMS has more durable effects on neural function. In healthy subjects, high frequency TMS to the motor cortex causes long lasting changes in electrophysiological measures of cortical excitability at the stimulation site, and those changes depend on NMDA receptor signaling, which suggests they involve long term potentiation like plasticity.
Stimulation of the left dorsolateral prefrontal cortex also modulates activity in more distant regions that function abnormally in depression. That combination, local plasticity plus distant modulation, is the proposed route by which TMS changes functional connectivity in cortical networks.
The default mode network and the central executive network are two of the intrinsic networks whose activity patterns are consistently abnormal in depression, which makes both of them plausible therapeutic targets. In this study the subgenual cingulate, closely aligned with the default mode network in depression, was the region where TMS produced the clearest normalization.
Results in Real Numbers
Seventeen depressed patients were scanned with resting state functional MRI before and after a five week course of TMS. Their connectivity was compared with a cohort of 35 healthy controls.
Before treatment, connectivity was abnormally elevated within the default mode network and diminished within the central executive network, and connectivity between the two networks was altered. After the five week course, subgenual hyperconnectivity within the default mode network had normalized, connectivity within the central executive network was unchanged, and connectivity between the dorsolateral prefrontal cortex and medial prefrontal default mode network nodes had become anticorrelated. Baseline subgenual connectivity predicted subsequent clinical improvement.
For context on how much benefit to expect from TMS overall, the authors note that recent meta-analyses put the effect size, Cohen’s d, in the 0.39 to 0.55 range, and that most treatment refractory patients will not achieve full remission. That sits against a medication backdrop in which only about one third of patients achieve full remission during acute phase treatment, and fewer than half maintain sustained remission after multiple medication trials.
What the Research Shows
This study used resting state functional MRI to measure functional connectivity within and between the default mode network and the central executive network in 17 depressed patients, before and after a five week course of TMS. Connectivity was seeded from the dorsolateral prefrontal cortex and from the subgenual cingulate.
Connectivity in the patients was also compared with a cohort of 35 healthy controls, which is what allowed the authors to describe the baseline pattern as abnormal and the post treatment subgenual pattern as normalized.
Who Benefits Most
The one predictive signal in this study is baseline subgenual cingulate connectivity, which predicted subsequent clinical improvement. That is a pretreatment imaging measure, not a symptom profile, and the authors present it as a potential neuroimaging biomarker for predicting treatment response rather than a validated selection tool.
More broadly, TMS has proven efficacy in depression, including in patients otherwise resistant to antidepressant pharmacotherapy, and it is a reasonable alternative for patients who either fail to respond to an antidepressant or experience intolerable side effects.
Safety, Limits, and Caveats
This study was designed to examine mechanism, not safety. TMS is described by the authors as well tolerated, and the connectivity changes observed here moved the abnormal subgenual pattern toward the pattern seen in healthy controls.
The limits are real. This was 17 patients scanned at two timepoints, with no sham comparison described in the abstract, so the connectivity changes cannot be separated from the natural course of illness on the strength of this design alone. TMS also left central executive network connectivity untouched even though it was abnormal at baseline, so network normalization was partial. And most treatment refractory patients will not achieve full remission with TMS, which is why the authors frame the work as a step toward understanding how and for whom it works.
Practical Takeaways
- TMS was followed by normalization of subgenual hyperconnectivity in the default mode network, not by a broad change across every network
- Central executive network connectivity did not change, so the effect was selective
- Connectivity between the dorsolateral prefrontal cortex and medial prefrontal default mode nodes became anticorrelated after treatment
- Baseline subgenual connectivity predicted later clinical improvement and is a candidate biomarker for response
- The sample was 17 patients and 35 controls, so treat this as mechanistic evidence rather than practice changing
- Optimizing treatment protocols, particularly the anatomical target for stimulation, is one route to better TMS outcomes
Related Studies and Research
Episode 31: Depression Explained, The Biology Behind the Darkness
Episode 32: Depression Recovery Roadmap: A Step-by-Step, Evidence-Based Plan
Accelerated TMS - moving quickly into the future of depression treatment
FAQs
What is the default mode network?
The default mode network is a set of brain regions, including medial prefrontal cortex, posterior cingulate cortex, and mostly medial areas of posterior parietal cortex, that has been implicated in rumination, self-referential processing, and episodic memory retrieval.
Why is the default mode network important in depression?
Activity patterns in the default mode network are consistently abnormal in depression. In this study, connectivity within the network was abnormally elevated in depressed patients compared with healthy controls before treatment.
What did TMS change in the brain?
TMS normalized the depression related subgenual cingulate hyperconnectivity in the default mode network and induced anticorrelated connectivity between the dorsolateral prefrontal cortex and medial prefrontal default mode nodes. It did not change connectivity within the central executive network.
Do brain changes predict who will respond to TMS?
Baseline subgenual connectivity, measured before treatment started, predicted subsequent clinical improvement in this sample. The authors put it forward as a potential neuroimaging biomarker for predicting treatment response.
Bottom Line
In 17 depressed patients scanned before and after five weeks of TMS and compared with 35 healthy controls, TMS selectively modulated connectivity within and between the default mode and central executive networks, normalizing subgenual cingulate hyperconnectivity while leaving the central executive network unchanged. Baseline subgenual connectivity predicted who improved. That gives a specific mechanistic account of how TMS may relieve depression and a candidate marker for predicting response, in a sample small enough that it needs replication before it guides treatment decisions.
