Theta-Burst Stimulation for MCI: Promise, Not Brain Cleaning

Intermittent theta-burst stimulation, or iTBS, may improve cognition in people with amnestic mild cognitive impairment (MCI) or very mild Alzheimer's disease. That is the encouraging part. The claim that it works by improving the brain's waste-clearance system is far less secure.

A 2025 sham-controlled trial found a delayed cognitive improvement signal after iTBS aimed at the left dorsolateral prefrontal cortex. But its "glymphatic" measure was DTI-ALPS, an MRI-derived proxy for diffusion near perivascular spaces—not a direct measurement of cerebrospinal fluid flow, amyloid removal, tau removal, or whole-brain waste clearance. The distinction is not academic. It separates a plausible clinical lead from a marketing-ready story that the data do not yet support.

The trial offers a real signal, with a small-study warning label

The central evidence is a 2025 randomized, double-blind, sham-controlled trial by Yang et al. The researchers assigned 60 adults, 1:1, to active or sham stimulation. Participants had amnestic MCI or very mild Alzheimer's disease, all with a Clinical Dementia Rating of 0.5 and average baseline Mini-Mental State Examination scores near 26. Their background medications remained stable.

Active treatment targeted the left dorsolateral prefrontal cortex over 10 sessions in two weeks. That target makes sense for a cognition study: the prefrontal cortex helps coordinate attention, working memory, planning, and other executive functions. The study assessed cognition and used DTI-ALPS as an exploratory imaging marker related to glymphatic function.

The reported result worth watching is delayed cognitive improvement after active iTBS compared with sham. A sham comparator matters because cognitive testing, clinic contact, expectations, and repeated assessments can all move scores. The design therefore gives this signal more weight than an uncontrolled before-and-after experiment.

Still, the published information available here does not provide the exact week-6 cognitive effect size, confidence interval, p-value, adverse-event counts, or arm-specific dropout pattern. Without those details, readers cannot judge the size of the benefit, its precision, or whether it reaches a threshold that patients would notice in daily life. Sixty participants is enough to justify a larger trial, not enough to settle treatment efficacy.

The study supports "possible delayed cognitive benefit from iTBS in very early Alzheimer's-spectrum impairment." It does not support "proven brain cleaning."

Why the glymphatic explanation remains a hypothesis

The glymphatic system is a proposed network through which fluid movement around brain blood vessels may help exchange cerebrospinal and interstitial fluid. It has become a popular explanation for sleep, aging, and neurodegeneration, partly because it offers an intuitively appealing image: the brain taking out its biochemical trash. Biology rarely makes the trash route that tidy.

Yang and colleagues inferred glymphatic activity with DTI-ALPS, short for diffusion tensor imaging along the perivascular space. This metric measures directional water diffusion in selected white-matter regions. It may reflect features related to perivascular fluid movement, but it does not directly track clearance of amyloid, tau, or any other solute.

A 2024 review by Sacchi et al. describes diffusion MRI and cerebrospinal-fluid aquaporin-4 measures as imperfect, indirect approaches to studying glymphatic dysfunction in dementia. Reviews published in 2026 by Fang et al. and Zahran et al. likewise treat DTI-ALPS as a biomarker candidate, not definitive proof of human clearance physiology.

That limitation becomes decisive here because the accessible report does not clearly quantify a treatment-linked ALPS improvement. A delayed cognitive signal without a clearly demonstrated, contemporaneous imaging shift cannot establish the sequence "stimulation improved clearance, which improved cognition." iTBS could affect cognition through cortical excitability, synaptic plasticity, or network-level changes without materially changing waste clearance.

Other iTBS studies support brain-network effects

Related research makes a conventional neuromodulation explanation more plausible. In amnestic MCI, Luo et al. reported in 2026 that accelerated iTBS combined with cognitive training modulated cortical plasticity and brain activation. A related 2025 study by Luo et al. found that the same general iTBS-plus-training approach affected brain function in amnestic MCI.

These studies do not prove that iTBS improves cognition on its own, because they combine stimulation with cognitive training and focus on physiological or functional brain measures. They do show that the intervention can alter systems relevant to cognition in the population of interest. That fits the Yang trial's delayed cognitive signal without requiring a clearance-based mechanism.

Glymphatic imaging tracks brain health—but tracking is not treatment proof

The broader literature gives researchers a reason to study this pathway. A 2025 cohort study by Qian Chen et al. linked glymphatic-related MRI measures in cognitively unimpaired older adults with Alzheimer's-signature regional brain volumes, plasma biomarkers, and white-matter-hyperintensity progression. Those associations suggest that these imaging metrics may track vulnerability relevant to preclinical Alzheimer's disease.

Similar disease associations appear elsewhere. Studies have reported altered DTI-ALPS or related MRI measures in major depressive disorder, restless legs syndrome, relapsing-remitting multiple sclerosis, Parkinson disease, frontotemporal dementia, and metabolic syndrome in older adults. For example, a 2023 study of restless legs syndrome compared 69 patients with 51 healthy controls using 3-Tesla MRI and DTI-ALPS. A 2025 depression study included 164 depressed participants and 46 healthy controls and found lower ALPS and higher white-matter free water in the depressed group after adjustment for several demographic and anatomical factors.

Association studies can identify a useful marker. They cannot show that raising the marker improves cognition, much less that it does so by clearing toxic proteins. White-matter anatomy, vascular injury, free water, and measurement choices can influence diffusion-based results. An ALPS change may be biologically interesting while remaining mechanistically ambiguous.

Cognition in MCI does not yield to one neat story

Other intervention trials provide useful perspective. The EXERT trial randomized 296 sedentary older adults with amnestic MCI to moderate-to-high-intensity aerobic training or low-intensity stretching, balance, and range-of-motion activity for 18 months. The interventions did not differ on the primary global cognitive composite (regression = -0.078, standard error = 0.074; p = 0.3), although the trial reported no cognitive decline at 12 months and examined Alzheimer's biomarkers.

That result does not argue against iTBS. It shows how difficult it is to isolate meaningful cognitive effects in MCI, especially when both study arms receive structured activity and attention. The 2024 Ornish lifestyle trial also tested a multidomain intervention in MCI or early Alzheimer's disease, using cognition and function outcomes after 20 weeks. Together, these trials point to a disorder shaped by many pathways rather than one switch waiting to be flipped.

What would make the case convincing?

Researchers need larger sham-controlled trials with prespecified cognitive outcomes, complete effect estimates, confidence intervals, adverse-event reporting, and follow-up long enough to show whether any benefit lasts. They also need mechanistic studies that pair DTI-ALPS with better-validated measures of glymphatic biology and test whether biomarker changes actually track cognitive change.

If future trials repeatedly find cognitive benefit but no consistent clearance-related biomarker change, iTBS could still become useful. The glymphatic explanation would simply lose ground. Conversely, replicated cognitive effects plus converging biomarker evidence would make the mechanism much more credible.

The practical verdict

iTBS deserves close attention as an early-stage research treatment for amnestic MCI and very mild Alzheimer's disease. The best human evidence cited here is a small, sham-controlled trial with a delayed cognitive signal, supported by related work showing that iTBS can alter plasticity and brain function.

For now, treat claims that theta-burst stimulation enhances glymphatic clearance as a hypothesis built on an indirect MRI surrogate. Do not infer that a short course of stimulation has been shown to wash amyloid or tau out of the human brain. The cognitive result is the lead. The cleanup story still needs proof.