Multifocal Noninvasive Deep Brain Stimulation to Enhance Cognition in Mild Cognitive Impairment

Importance: Impairment of working memory (WM), which relies on distributed cortical and subcortical structures, including cerebello-striatal pathways, is a core contributor to functional decline in patients with mild cognitive impairment (MCI) and dementia. Because the striatum is affected early in dementia, whereas the cerebellum remains relatively preserved, the cerebellum may offer support for maintaining striatal function in neurodegenerative diseases. Objective: To evaluate whether sequential cerebellar transcranial magnetic stimulation (TMS) and striatal transcranial temporal interference stimulation (tTIS) enhances WM in patients with MCI, especially those with early striatal degeneration, like MCI with Lewy bodies (MCI-LB). Design, Setting, and Participants: This randomized, quadruple-blind, placebo-controlled, crossover trial was conducted from August 2023 to October 2024 at 1 site in Switzerland and 1 in Czech Republic. Participants included individuals with MCI-LB or amnestic MCI (aMCI) and age-matched healthy controls (HCs). Intervention: Three pseudorandomized single stimulation sessions: (1) cerebellar TMS followed by striatal tTIS, (2) control TMS followed by striatal tTIS, and (3) an active control combining control TMS with control striatal tTIS. Main Outcomes and Measures: The primary outcomes were changes in accuracy and correct answer reaction times in a visual WM task with varying memory loads and distractors. Modulators included resting-state functional connectivity and gray-matter volume in WM-related regions. Linear mixed-effects models were used to examine the main effects of stimulation condition on accuracy and reaction times. Results: Forty-one patients with MCI, including 21 with MCI-LB (16 female [76%]; mean [SD] age, 71.05 [6.84] years) and 20 with aMCI (11 female [55%]; mean [SD] age, 72.30 [7.50] years), and 20 HCs (10 female [50%]; mean [SD] age, 69.5 [4.42] years) were analyzed. In patients with MCI-LB, cerebellar TMS plus striatal tTIS increased accuracy in distractor trials compared with the active control (control TMS plus control tTIS, P < .001; d = -0.77) and striatal tTIS alone (control TMS plus striatal tTIS, P = .004; d = -0.60). Greater accuracy gains were associated with lower putaminal connectivity (r = -0.66; P = .03). In patients with aMCI, cerebellar volumes showed associations with task performance in both distractor trials (cerebellar TMS plus striatal tTIS, R2 = 0.27; P = .02) and high-load trials (control TMS plus striatal tTIS, R2 = 0.24; P = .03), and larger volumes were associated with faster responses; however, significant behavioral differences between active and control stimulation conditions were not observed. No serious adverse events occurred. Conclusions and Relevance: Synergistic dual-target cerebellar TMS and striatal tTIS improved WM in MCI-LB, particularly in patients with lower striatal connectivity, suggesting dual-targeted neuromodulation as a potential therapeutic avenue in neurodegenerative disorders. Trial Registration: ClinicalTrials.gov Identifier: NCT07090681.