Journal: Neuromodulation (2026)
Authors: MacKillop J, McIntyre-Wood C, Vandehei E, Yaya H, Di Passa AM, Prokop-Millar S, Fein A, Yang A, Elsayed M, Schwartzmann B, Farzan F, MacKillop E, Duarte D
Background:
Cannabis use disorder (CUD) is a significant public health problem and no medical treatments have received regulatory approval to date. Deep repetitive transcranial magnetic stimulation (dTMS) provides brief powerful bursts of electromagnetic stimulation to brain networks implicated in psychiatric disorders and the Hesed (H)-4 coil, which targets the lateral prefrontal cortex and anterior insula, is approved for tobacco use disorder.
Objective:
This phase 1 open-label pilot study evaluated the feasibility and tolerability of H4 dTMS in conjunction with standard care for adults with moderate-to-severe CUD who were seeking treatment.
Methods:
Eleven participants received 18 sessions of H4 dTMS (three weeks of five sessions per week, one week of three sessions) in conjunction with standard care. Feasibility was operationalized as treatment completion and attainment of therapeutic dose (≥ 90% resting motor threshold). Tolerability was operationalized as adverse events. Exploratory clinical and mechanistic outcomes were 5-mg standard units of Δ9-tetrahydrocannabinol (THC), CUD symptoms, cravings, cannabis reinforcing value, self-efficacy, internalizing and somatic symptoms, and neuropsychological performance.
Results:
Of 11 participants (64% female, CUD symptoms mean ± SD = 9.27 ± 1.10), 92% completed all dTMS treatments, of whom 91% reached therapeutic dose. Tolerability was high, with the most common adverse events being low rates (2.4%–9.2%) of headache, drowsiness, and scalp pain, primarily during the first week. No participants experienced serious adverse events. Regarding exploratory clinical outcomes, significant reductions were observed in THC consumption, CUD symptoms, cravings, cannabis reinforcing value, and depressive, anxiety, and somatic symptoms (ps < 0.01–0.001), and significant increases were observed in self-efficacy (p < 0.001). No significant effects on neuropsychological performance were detected.
Conclusions:
These findings suggest this H4 dTMS protocol was feasible, tolerable, and associated with clinically meaningful reductions in clinical outcomes and related mechanisms, albeit in an uncontrolled design. The results support the prospect of a randomized controlled trial to test the efficacy of this novel intervention for CUD.
