Background and Rationale
Mild cognitive impairment (MCI) affects approximately 22% of adults over age 65 and represents a significant risk state for progression to dementia. Current pharmacological interventions, including cholinesterase inhibitors and memantine, demonstrate only modest efficacy (1-2.5 point improvements on cognitive scales), provide no established benefit specifically for MCI, and carry notable adverse effects including gastrointestinal symptoms, bradycardia, and falls. This has driven interest in non-pharmacological approaches, particularly repetitive transcranial magnetic stimulation (rTMS).
Recent meta-analyses demonstrate that rTMS produces medium-to-large effect sizes (standardized mean difference = 0.93) for immediate cognitive improvements in MCI and Alzheimer's disease, with sustained benefits at long-term follow-up (SMD = 0.42). High-frequency rTMS targeting the dorsolateral prefrontal cortex (DLPFC) shows particular efficacy in enhancing memory and executive function, with benefits lasting 4-12 weeks post-treatment. These improvements are thought to be mediated through upregulation of brain-derived neurotrophic factor, enhanced synaptic plasticity, and functional reorganization of large-scale brain networks.
Despite this evidence, critical gaps remain. Most existing studies employ conventional protocols with daily sessions over several weeks, which may limit accessibility and adherence. Furthermore, prior studies have relied primarily on global cognitive screening tools such as the Montreal Cognitive Assessment (MoCA), which may lack sensitivity to detect changes in specific cognitive domains. The neurophysiological trajectory of cognitive changes over extended follow-up periods also requires systematic characterization.
Study Design and Intervention
This is a single-center, open-label, prospective pilot study conducted at the San Francisco Neurology and Sleep Center. The study consists of three phases: a screening phase (up to 14 days), an open-label treatment phase (approximately 21 days), and a follow-up phase (90 days), for a total participation duration of up to 139 days.
During the treatment phase, participants receive 6 sessions of rTMS using the EXOMIND™ device (BTL-699-2), administered twice weekly over approximately 3 weeks. Each session delivers 6,300 pulses at alternating frequencies of 12, 15, and 18 Hz, with a total session duration of 24 minutes and 30 seconds. The stimulation target is the left DLPFC, localized using the standard 5-cm rule. Motor threshold is determined individually for each participant as the minimum stimulus intensity required to induce contraction of the right thumb.
Cognitive Assessment Strategy
A key methodological feature of this study is the dual-assessment approach combining global cognitive screening with domain-specific evaluation. The MoCA provides a validated measure of overall cognitive status, while the Creyos cognitive battery enables granular analysis of treatment effects across six specific domains: visual spatial working memory, episodic memory, deductive reasoning, mental rotation, verbal short-term memory, and attention.
To minimize practice effects associated with repeated MoCA administration, alternate validated versions are used at each time point: Version 8.1 (original) at baseline, Version 8.2 at post-treatment, Version 8.3 at 1-month follow-up, and Version 8.1 at 3-month follow-up. A Creyos cognitive battery training session is conducted during screening to familiarize participants with the testing procedures and reduce learning effects during the study.
Depressive symptoms are assessed using the Patient Health Questionnaire-9 (PHQ-9) at each time point, given the known relationship between mood and cognitive performance and the established effects of DLPFC stimulation on depressive symptoms.
Assessment Schedule
Cognitive and clinical assessments are performed at four time points: baseline (screening), post-treatment (Day 21, approximately), 1-month follow-up (Day 51 ± 7 days), and 3-month follow-up (Day 111 ± 7 days). Vital signs (blood pressure and heart rate) are recorded at every treatment session and study visit. Adverse events are monitored at each treatment session and follow-up visit using a standardized checklist derived from existing TMS safety literature.
Statistical Approach
Repeated measures ANOVA will be used to assess changes in MoCA scores and each Creyos cognitive domain across the four time points. Paired t-tests will evaluate pairwise changes between baseline and each subsequent time point. Cohen's d will be calculated for all primary and secondary endpoints to quantify effect sizes. The proportion of participants achieving a clinically meaningful improvement, defined as a ≥2-point increase in MoCA score, will be reported at each post-baseline time point.
Both intention-to-treat (ITT) and per-protocol (PP) analyses will be conducted. The last observation carried forward (LOCF) method will be used for sensitivity analyses in cases of missing data or early withdrawal. Exploratory subgroup analyses may be performed based on baseline cognitive severity (mild vs. moderate impairment), age, and sex. Adverse events will be summarized by frequency, severity, and relationship to treatment. A p-value of less than 0.05 will be considered statistically significant.
Significance
This pilot study addresses key gaps in the rTMS literature by evaluating a condensed treatment protocol (6 sessions over 3 weeks rather than daily sessions over several weeks), incorporating domain-specific cognitive assessment alongside global screening, and characterizing the trajectory of cognitive changes over an extended 3-month follow-up. Positive findings would support the design of larger, controlled trials and inform the potential integration of rTMS as a non-pharmacological therapeutic option for cognitive impairment.