Parcel-guided rTMS for treatment-resistant depression could offer a more accessible path toward personalized brain stimulation. A new pilot study suggests that mapping an individual patient’s cortical anatomy may help clinicians identify a more precise stimulation site without requiring functional MRI during treatment planning.
The findings add to a growing effort to move rTMS beyond standardized scalp coordinates and toward targets selected according to brain structure, network organization, and patient-specific anatomy.
Why Conventional rTMS Targeting Leaves Room For Improvement
Repetitive transcranial magnetic stimulation is an established treatment for major depressive disorder that has not responded adequately to antidepressant medication. Conventional protocols commonly position the coil over the left dorsolateral prefrontal cortex using a fixed distance from the motor cortex or the F3 location in the international EEG system.
These methods are practical, but they do not fully account for anatomical variation between patients. A scalp coordinate that reaches a therapeutically relevant network in one person may stimulate a somewhat different cortical region in another.
Previous research has linked stronger antidepressant outcomes with stimulation sites that are functionally connected, particularly through negative correlation, to the subgenual anterior cingulate cortex. This deeper brain region is closely associated with mood regulation but cannot be reached directly with standard rTMS coils.
How Parcel-Guided rTMS For Treatment-Resistant Depression Works
The investigators used structural MRI and tools developed through the Human Connectome Project to divide each participant’s cortex into standardized anatomical parcels. Their primary target was located near the junction of parcels 46, 9-46d, and a9-46v within the left dorsolateral prefrontal cortex.
This junction overlaps with a region previously identified as strongly anti-correlated with the subgenual anterior cingulate cortex. Eleven participants received stimulation at the junction target, while four received stimulation at the midpoint of parcel 46.
Importantly, the treatment target was generated from structural imaging. Resting-state functional MRI was collected for analysis, but it was not required to determine where the coil would be positioned.
A Small Trial With A Clinically Meaningful Contrast
The open-label study included 15 adults with moderate to severe major depressive disorder who had not responded to at least two adequate antidepressant trials. Participants completed 27 sessions of standard 10-Hz rTMS over six weeks.
Among the 11 patients treated at the parcel junction, average Montgomery-Åsberg Depression Rating Scale scores decreased by 57.8 percent. Eight participants met the threshold for response, representing a response rate of 72.7 percent. Six reached remission, producing a remission rate of 54.5 percent.
Results were less favorable among the four participants treated at the midpoint of parcel 46. That group experienced an average symptom reduction of 38.2 percent, with one response and no remissions.
Although the groups were very small and were not randomized, the difference suggests that modest changes in coil placement may have clinically important effects.
Connectivity May Help Predict Who Reaches Remission
Baseline connectivity between parcel 46 and the subgenual anterior cingulate cortex was significantly associated with treatment outcome. In this sample, patients who reached remission showed connectivity values closer to zero before treatment, while stronger baseline anti-correlation was associated with less improvement.
This result may appear counterintuitive because anti-correlated targets have often been associated with stronger rTMS effects. The authors suggest that baseline connectivity could function differently as a predictive patient biomarker than it does as a method for selecting a stimulation location.
Changes in connectivity during treatment were also related to symptom improvement, supporting the possibility that rTMS alters communication within depression-relevant networks.
Structural Mapping Could Expand Personalized Targeting
Highly individualized functional-connectivity targeting can require specialized scans, processing pipelines, and technical expertise. Parcel-guided targeting may provide a more standardized alternative by translating a validated cortical atlas onto each patient’s structural MRI.
This distinction matters for clinics that have access to structural imaging and neuronavigation but cannot routinely acquire or analyze resting-state functional MRI. Standardized parcels could also support more consistent targeting and biomarker development across research sites.
Promising Results That Still Require Controlled Testing
The study remains an early feasibility signal. It was open-label, included only 15 participants, and assigned just four patients to the parcel-midpoint target. The report is also a preprint that has not completed peer review.
Larger randomized trials will need to determine whether parcel-guided targeting reliably outperforms conventional methods and whether baseline connectivity can predict outcomes in independent samples. For now, the findings suggest that precision may depend not only on selecting the correct brain region, but also on identifying the most relevant location within it.
Citations
Berman J, Truong DQ, Murphy A, et al. “Individualized, Parcel-Guided rTMS to the Left Dorsolateral Prefrontal Cortex for Treatment Resistant Depression.” medRxiv. 2026. https://doi.org/10.64898/2026.09.03.26362155
Moreno-Ortega M, Kangarlu A, Lee S, et al. “Parcel-Guided rTMS for Depression.” Translational Psychiatry. 2020;10:283. https://doi.org/10.1038/s41398-020-00970-8
Explore more at https://www.interventionalpsychiatry.org/
