Reversible modulation of a deep white matter surgical target for depression with low-intensity focused ultrasound

Aki Tsuchiyagaito, Rayus Kuplicki, Masaya Misaki, Landon Edwards, Joan A. Camprodon, Kate Fitzgerald, Sahib S. Khalsa, Noah S. Philip, Martin P. Paulus, Salvador M. Guinjoan

Neuropsychopharmacology 2026, 51, 612-621 · 10.1038/s41386-025-02252-7

human patientdepressionfmriautonomic physiologyclinical scale

Abstract

Current efforts to establish anatomically precise and causal deep brain circuit-symptom relationships to predict surgical responses require the insertion of intracranial electrodes. In the present study (ClinicalTrials.gov NCT05697172) we examined whether low-intensity focused ultrasound (LIFU; FDA Q220192) could modulate the white matter tracts in the anterior limb of the internal capsule (ALIC), a common surgical target for the treatment of refractory depression. We based our human study on the in vitro observation that ultrasound attenuates conduction in myelinated axons by operating mechanosensitive potassium channels in nodes of Ranvier. We employed a cross-over, randomized, sham-controlled, double-blind design to establish the ability of LIFU (80 s; 0.5 MHz; estimated peak tissue energy 2.26 Watt/cm 2 ; 10% duty cycle) to engage a target in the ALIC in 21 patients with Major Depressive Disorder. Targets were defined as the area with highest density of thalamo-prefrontal probabilistic tractography streamlines. We measured changes in magnetic resonance functional connectivity, repetitive mentation, emotional state, and cardiac autonomic output. Compared with sham stimulation, active LIFU produced a functional disconnection of gray matter hubs connected by the sonicated tracts (d = -1.55; p < 0.001), an increase in positive emotion (d = 0.6; p = 0.04), and decreased cardiac sympathovagal balance (d = -0.57; p = 0.01). The observation that LIFU engages deep white matter tracts related to depression symptoms paves the way for rigorously investigating (1) mechanistic definitions of brain circuit-symptom relationships, (2) noninvasive, reversible, and anatomically precise probing of such circuits before surgery, and (3) potential use as a new therapy for depression.

Abstract via europepmc.

Specieshuman

Subjects21 participants

Sessions per subject1

Randomisedyes

Blindingdouble

Sham / controlinactive transducer

Auditory controlnot reported

Readout timingboth

Anaesthesianot applicable

Readoutsfmri, autonomic physiology, clinical scale — Resting-state fMRI seed-to-whole-brain connectivity (right thalamus seed); Brief State Rumination Inventory (BSRI); Positive and Negative Affect Schedule (PANAS-X); continuous ECG-derived heart rate variability (SDNN, LF/HF ratio); Columbia Suicide Severity Rating Scale and MADRS as safety/severity measures

Direction of effectinhibitoryActive LIFU of the ALIC produced a large decrease ('functional disconnection') in resting-state connectivity between the right thalamus and reciprocally connected orbitofrontal/ventromedial prefrontal/cingulate cortices compared with sham, consistent with LIFU-induced suppression of axonal conduction via mechanosensitive K+ channels; active LIFU also blunted the sham-session increase in cardiac sympathovagal balance and increased positive affect (PANAS-X Joviality).

Adverse eventsnone observedThere were no no focal neurological changes or any damage on clinical neuroimaging (data not shown). Participants also did not report adverse events when specifically questioned.

Exposures

Exposure 1: Active LIFU of right ALIC (0.5 MHz theta-burst pattern)

Target: internal capsule — “anterior limb of the internal capsule (ALIC)”
Device: NeuroFUS · Sonic Concepts (sold by Brainbox) · NeuroFUS Pro (CTX-500 transducer) ✓

Pulse timing
Waveform	theta burst
Fundamental frequency (kHz)	500	✓✓✓
Pulse duration (ms)	20	✓✓✓
Pulse repetition frequency (Hz)	5	✓✓✓⚑
Duty cycle (%)	10pulse duration × PRF gives 10%	✓✓✓
Sonication duration (s)	80	✓✓✓

Pressure and intensity, by domain
Free-field pressure (kPa)	not reported
Free-field Isppa (W/cm²)	9.06	✓✓✓
Free-field Ispta (W/cm²)	not reported
In-situ estimate	deratingsingle value
In-situ pressure (kPa)	not reported
In-situ Isppa (W/cm²)	2.26	✓✓✓
In-situ Ispta (W/cm²)	0.22	✓✓✓

Protocol, in the paper’s words

A single 80-s train of 20-ms ultrasound bursts (0.5 MHz), repeated every 200 ms (400 bursts total, 10% duty cycle), delivered once to the individually tractography-defined right ALIC target per session; total free-field energy 9.06 W/cm2 ISPPA, corresponding to estimated tissue values of 2.26 W/cm2 ISPPA and 0.22 W/cm2 ISPTA assuming 75% attenuation from scalp and bone transmission.

Flags from extraction

n_sessions_per_subject — Count of 2 sessions (one active, one sham) is inferred from the cross-over design description rather than a stated integer.
exposures[0].timing.pulse_repetition_frequency_hz — PRF (5 Hz) is derived from the stated repetition period ('repeated every 200 milliseconds'); not stated as a Hz value directly in this paper (contrast with the cited source protocol, Zeng et al. 2022, which does state 'PRF of 5Hz' explicitly for the same pattern).
conditions — Sample includes substantial psychiatric comorbidity (generalized anxiety disorder 33%, social anxiety disorder 43%, panic disorder 33%, PTSD 24%); primary enrollment diagnosis and target symptom was major depressive disorder, so conditions is recorded as depression only.
readout_timing — Classified as 'both': fMRI/BSRI/PANAS-X were measured offline (before/after sonication) while continuous ECG for heart-rate-variability analysis spanned before, during, and after the 80-s sonication.
auditory_control — Paper does not describe any specific auditory-masking or confound-control procedure for the LIFU sonication.

Notes: Single active exposure (right ALIC, 0.5 MHz theta-burst pattern identical to Zeng et al. 2022's tbTUS protocol); sham (sound-absorbing Sorbothane pad on the transducer) is captured via sham_type, not as a separate exposure.