Table 1: Studies investigating rTMS as a therapeutic tool in ADRD.

Studies investigating TMS for treatment of ADRD using clinical or biomarker diagnostic criteria. Age is shown as Mean±SD or Mean (SEM). Several studies followed the Neuro AD protocol, targeting 6 brain regions: R prefrontal, L prefrontal, R parietal, L parietal, Broca’s area, Wernicke’s area. ADAS-Cog = Alzheimer’s Disease Assessment Scale-cognitive; BDS = Blessed Dementia Scale; MMSE = Mini-Mental State Examination; CGIC = Clinical Global Impression of Change; GDS = Geriatric Depression Scale; MOCA = Montreal Cognitive Assessment; AVLT = Auditory-Verbal Learning Test; FAB = Frontal Assessment Battery; DSST = Digit Symbol Substitution Test; NPI = Neuropsychiatric Inventory; IDDD = Interview for Deterioration in Daily Living Activities in Dementia; CGI = Clinical Global Impression; 3MS = Modified Mini-Mental Status Exam; AES = Apathy Evaluation Scale; TMT = Trail Making Test; EXIT-25 = Executive Interview; ADLs = Activities of Daily Living; I-ADLS = Instrumental Activities of Daily Living; ZBS = Zarit Burden Scale; ACE = Addenbrooke Cognitive Examination.

TMS Studies 2016–2018	Criteria for AD/MCI and disease stage	No. of participants	Sham/Control	Age	Target area; localization method	Interleaved Cognitive Training	Intensity (%RMT)	TMS frequency and pattern	Stimulation duration; number of TMS trains; number of pulses/day	Length of Intervention; Number of Sessions	Cognitive Domain	Neuropsychological Tests – Primary Outcome	Neuropsychological Tests – Secondary Outcomes	Main Significant Neuropsychological Findings
Pilot Studies and Randomized Controlled Trials
Lee, et al., 2016	Probable AD based on DSM-IV criteria, CDR 1–2, MMSE 18–26 (mild-moderate AD)	26	2:1 Treatment:Sham	Treatment group age=71.2±7.6, Sham group age=70.3±4.8	6 brain regions; MRI guided	Yes	90–110% RMT	10 Hz rTMS; 20 trains applied with 2 seconds on, 20–40 seconds off; with interleaved cognitive task	1 hour session/day; 3 brain regions/day; 1200 pulses/day	6 weeks; 30 sessions	Global Cognition	ADAS-Cog	MMSE; CGIC; GDS	There was a significant improvement after the intervention in ADAS-Cog in the treatment group, but the between-group difference compared with sham was not significant. In both treatment and sham, the largest improvement was seen in mild AD compared to moderate AD.
Zhao, et al.; 2017.	Probable AD based on DSM-IV criteria, CDR 1–2, MMSE 18–26 (mild-moderate AD)	30	17:13 Treatment:Sham	Treatment group Age=69.3±5.8, Sham group Age=71.4±5.2	Treatment areas not clearly specified, but included parietal P3/P4, posterior temporal T5/T6; 10–20 system	Yes	Not Specified	20 Hz rTMS; 20 trains applied with 10 seconds on, 20 seconds off; with interleaved cognitive tasks	1 hour session/day; 3 brain regions/day; 12000 pulses	6 weeks; 30 sessions	Global Cognition & Verbal Memory	Not specified	ADAS-Cog, MMSE, MOCA, AVLT	There was a significant improvement in ADAS-Cog, MMSE, and AVLT in the treatment group, but there was no between-group difference compared with sham. Mild AD showed a larger improvement compared with moderate AD.
Nguyen, et al.; 2017	Probable AD by clinical diagnosis, severity ranging from MCI to moderate-to-severe AD	10	None	Age=70.3 (7.2)	6 brain areas plus L DLPFC and R DLPFC; MRI guided	Yes	100% RMT	6 Brain Region Treatment: 10 Hz rTMS; 20 trains applied with 2 seconds on over 10 minutes; with interleaved cognitive training. DLPFC Treatment: 10 Hz rTMS; 5 trains applied with 2 seconds on over 2.5 minutes; with interleaved cognitive training.	1 hour session/day; 4 brain regions/day; up to 1300 pulses/day	5 weeks; 25 sessions	Global Cognition	ADAS-Cog	performance on interleaved cognitive training tasks, MMSE, Dubois score, FAB, Stroop test, locomotor score, apathy score, caregiver burden interview, dependence score	Immediately after the treatment procedure, there was improvement in the ADAS-Cog, locomotor score, apathy score, and dependence score. Six months later, ADAS-Cog scores had returned to baseline, but apathy and dependence scores showed continued improvement.
Koch, et al.; 2018	prodromal AD by Dubois, 2016 criteria with positive CSF biomarker	14	Crossover design, with participants receiving both treatment and sham stimulation	Age=70.0±5.1	Precuneus; MRI guided, stimulation site confirmed with source localization	No	100% RMT	20 Hz rTMS; 40 trains applied with 2 seconds on, 28 seconds off.	20 minutes of rTMS; 1600 pulses/day.	2 weeks; 10 sessions	Verbal Memory, EEG, and TMS-EEG	Not specified	RAVLT, MMSE, FAB, DSST	RAVLT Delayed Recall showed a significant improvement after treatment compared with sham; other tests showed no main effect of treatment.
Alcalá-Lozano, et al.; 2018.	diagnosis of AD by DSM-V, MMSE >= 15, GDS-Reisberg level 2–4	19	1:1 randomization into 2 active treatment groups: “simple” vs “complex” stimulation protocol	Simple Group Age=73.3±6.0; Complex Group Age=71±4.3	Simple Protocol: singlesite DPLFC stimulation. Complex Protocol: 6 brain regions; 10–20 system	No	100% RMT	5 Hz rTMS; 30 trains applied with 10 seconds off, 60 seconds off.	In the Simple Protocol, single-site stimulation was applied to DLPFC daily, in the Complex Protocol, 3 brain regions were treated daily; 1500 pulses/day	3 weeks; 15 sessions	Global Cognition	ADAS-Cog	MMSE, NPI, GDS, IDDD, CGI	Both treatment groups showed an improvement in ADAS-Cog, MMSE, IDDD, NPI immediately after treatment, which persisted one month later. There was no significant difference between the two treatment groups.
Padala, et al.; 2018.	MCI diagnosis by Peterson’s criteria, MMSE >=23, with apathy (AES-C >=30).	8	Double-blind, randomized, Crossover design, with participants receiving both treatment and sham	Group 1 Age=68.0±10.0; Group 2 Age=64.0±9.0	L DLPFC; 5.5 cm anterior to motor hotspot location	No	120% RMT	10 Hz rTMS; 75 trains applied with 3 seconds on, 26 seconds off.	37.5 minutes of rTMS; 3000 pulses/day	2 weeks; 10 sessions	Apathy	AES-C	3MS, MMSE, TMT B, TMT A, EXIT-25, CGI, I-ADLS, ADLS, ZBS	There was a significant improvement in AES-C after the active treatment compared to the sham condition. There was also significant improvement in 3MS, MMSE, TMT A, and CGI-I in the treatment group compared with sham.
Case Reports and Clinical Case Series
Avirame, et al.;	moderate-severe AD by clinical diagnosis	11	None	Age=76±7	Bilateral Prefrontal Cortex using deep TMS; 6 cm anterior to motor hotspot location	No	120% RMT	10 Hz deep TMS, 42 trains applied with 2 seconds on, 20 seconds off	1 20-minute session/day, 2–3 times per week, with a minimum interval of 1 day between sessions	20 sessions	Global Cognition	n/a	Mindstreams and ACE	60% of patients improved on Mindstreams, and 77% showed improvement on the ACE compared to baseline. Treatment with deep TMS significantly improved ACE scores in a subset of the most progressed patients
Rabey and Dobronevsky; 2016.	mild-to-moderate AD clinical diagnosis	30	None; Patients treated in 2 private clinics offering commercial NeuroAD treatments	not reported	6 brain regions; MRI guided	Yes	90–110% RMT	3/4 Paradigms: 10 Hz rTMS; 20 trains applied with 2 seconds on over 10 minutes; with interleaved cognitive training. 1/4 Paradigm: 10 Hz rTMS; 5 trains applied with 2 seconds on over 2.5 minutes; with interleaved cognitive training.	1 hour session/day; 3 brain regions/day; 1300 pulses/day	6 weeks; 30 sessions	Global Cognition	n/a	ADAS-Cog and MMSE	ADAS-Cog and MMSE both improved after treatment compared to baseline scores.