Background

Sensory Stimulation Systems, and GENUS, refer to a noninvasive procedure, i.e., a "digital therapeutic," being developed by Cognito Therapeutics. The company developed a wearable GammaSense Stimulation device. It pairs headphones and opaque glasses to deliver light and sound stimuli at gamma band frequency in order to entrain gamma oscillations in the brain. An LED light flashing at 40 Hz for one hour was reported to boost gamma power in the visual (V1), prefrontal, and somatosensory cortices, and hippocampal CA1; it also increases synchronization between V1 and the other regions. A combination of light and sound at 40 Hz has been shown to strengthen gamma oscillations throughout the auditory cortex, hippocampus, and medial prefrontal cortex. Gamma oscillations play a role in sensory processing, cognition, and memory consolidation, and are weakened in AD.

In 5XFAD, APPPS1, and wild-type mice, one hour of light-based GENUS reportedly halved Aβ levels, and one week of GENUS in 6-month-old 5XFAD mice cut amyloid plaque load by two-thirds. In TauP301S, this regimen cut phosphorylated tau by 40 percent (Iaccarino et al., 2016).

Neuron numbers were maintained in TauP301S mice for three weeks with GENUS starting at 7.5 months of age, when neurodegeneration begins in this model. In the inducible p25 neurodegeneration mouse model, six weeks of GENUS preserved neuron number and cortical thickness. In both models, GENUS maintained neuronal expression of genes affecting synaptic transmission and vesicle trafficking, protected DNA, increased the density of mushroom spines, and improved performance in the Morris water maze. In all models tested, GENUS lowered microgliosis and inflammatory gene expression, while increasing internalization of Aβ (Adaikkan et al., 2019). The combination of light and sound at 40 Hz in 6-month-old 5XFAD mice had similar effects to light alone, but reached more brain areas (Martorell et al., 2019).

In other studies, 40 Hz light flicker or ultrasound was reported to improve cognitive function, enhance presynaptic plasticity, and activate microglia to reduce Aβ plaque load (Singer et al., 2018; Park et al., 2020; Zheng et al., 2020; Bobola et al., 2020). A separate study linked acoustic stimulation to lowered amyloid plaques and increased microglial numbers in the hippocampi of 5XFAD mice (Lee et al., 2018). Multisensory 40 Hz stimulation was reported to increase glymphatic clearance of Aβ in the same model (Murdock et al., 2024), and to restore hippocampal synaptic potentiation in rat models (Yang et al., 2023). Other reported effects of 40 Hz light flicker in AD preclinical models include increased expression of cytokines that regulate microglia (Garza et al., 2020; Prichard et al., 2023), normalization of circadian rhythms, altered APP processing, and improved effects of exercise on cognition (Yao et al., 2020; Shen et al., 2022; Park et al., 2022).

In one study using 40 Hz optogenetic stimulation, hippocampal gamma waves and spatial memory changed, but plaque load did not. In another, plaques grew; memory was not assessed (Etter et al., 2019; Wilson et al., 2020). In other negative studies, native gamma oscillation entrainment, amyloid lowering, and microglial activation were not replicated (Soula et al., 2023; Yang and Lai, 2023). Transgenic AD rats increased their activity, but did not improve cognition, after light and sound stimulation (Bentley and Broussad, 2024).

For reviews, see McDermott et al., 2018; Adaikkan and Tsai, 2020; Manippa et al., 2022; Blanco-Duque et al., 2024).

Devices that offer 40 Hz light and sound stimulation are sold directly to consumers. Evidence for their claimed uses is sparse.

Findings

In a brief pilot study conducted in Europe, without Cognito Therapeutics, 10 days of 40 Hz light therapy, administered for two hours per day, did not affect amyloid load, as measured by PiB PET, in 10 AD patients (Ismail et al., 2018).

In April 2018, Cognito Therapeutics began a Phase 1 trial called Overture to evaluate its CogTx-001 GammaSense stimulation device in 60 people with AD or mild cognitive impairment due to AD. Devices were calibrated for each participant and EEG-verified to elicit 40 Hz steady-state gamma oscillations. Participants received one hour of combined 40 Hz auditory and visual stimulation, or a sham treatment, daily for six months at home, with a follow-up visit one month later to assess safety. The primary outcome was ADAS-Cog, CDR-SB, and MADCOMS scores, with ADCS-ADL and MMSE and amyloid imaging as secondary outcomes. The trial took place at five sites in the U.S. and ran until August 2021.

In March 2021, results were presented at AD/PD. Forty-six participants received active stimulation, 28 sham. There were no serious adverse events, but more participants on active stimulation complained of tinnitus. In each group, 28 percent of participants dropped out. The trial missed all three primary outcomes. Of the 53 people who finished the six-month regimen, 44 continued to a one-year, open-label extension (Apr 2021 conference news). Published results on the first 22 participants reported a decrease in nighttime activity. Active treatment was reported to result in stabilization on the ADCS-ADL, compared to a decline in the sham group (Cimenser et al., 2021). Complete trial results were published, and showed no statistically significant changes in any of the primary or secondary endpoints, nor any change in amyloid levels on PET (Hajós et al., 2024). Additional analysis of MRI data suggested reduced white matter atrophy and myelin content loss, especially in the entorhinal cortex (Da et al., 2024). Results of the open-label extension were presented at the March 2024 AD/PD conference. Adherence was 80 percent, with no new adverse events of tinnitus or headache reported. ADCS-ADL scores were stable in the 15 continuously treated patients. The seven patients who started on placebo were reported to have shown a trend toward slowing of decline in the ADCS-ADL and loss of brain volume after switching to active treatment. Progression modeling implied a disease-modifying effect.

In May 2018, a second Phase 1 study called Etude began comparing dosing paradigms for auditory and visual stimulation with the GammaSense device in 20 people with AD or mild cognitive impairment due to AD. These included one hour once per day, one hour twice per day, one hour every other day, and 30 or 120 minutes twice per day. The primary outcomes were amyloid PET and safety; secondary, ADAS-Cog. The trial was to be completed by March 2022; no results have been reported.

From November 2018 to February 2020, a third study called Flicker evaluated the tolerability of a combined audio-visual stimulation for one hour per day in 10 people with mild cognitive impairment due to AD. Five participants received eight weeks of GammaSense stimulation; five received four weeks of no intervention followed by four weeks of intervention. There was no placebo/sham. This investigator-initiated trial was conducted at Emory University, and sponsored by Cognito. As reported at AD/PD 2021, the study first tested escalating brightness and loudness. Most people tolerated full-intensity stimulation. Overall, participants completed 95 percent of sessions. No severe treatment-related adverse events were reported. Some participants reported dizziness, headaches, or tinnitus; one person’s hearing loss worsened. In exploratory measures, EEG detected gamma entrainment in many brain areas. Gamma power was the same after four weeks treatment, and weakened after eight. Strengthened brain connectivity was detected at eight weeks between the posterior cingulate and precuneus, two regions in the default mode network that is disrupted in AD. Brain amyloid and CSF Aβ or tau did not change after eight weeks, although CSF levels of many cytokines changed (Apr 2021 conference news, He et al., 2021). Nine of the 10 participants chose to enter a one-year, open-label extension.

Two other studies of gamma entrainment were conducted at MIT, with results also presented at AD/PD (April 2021 conference news). These studies used an easel-mounted array of lights and speakers, not the GammaSense device. One began in April 2019 and involved 46 healthy adults and 26 older people with mild AD. EEG monitoring was used to optimize stimulation parameters during 30-60 minute sessions. The 40 Hz light/sound stimulation increased gamma waves in most brain areas, while triggering no headaches, vision or hearing changes, or seizures.

A second study began in August 2019 to evaluate the effects of daily stimulation in 15 people with mild AD. Participants were randomized to active or sham stimulation for one hour daily using the same light/speaker array at home for six or nine months. The primary outcomes measured device usage and adverse events. Other outcomes included cognition, function, neuropsychiatric symptoms, blood levels of amyloid, tau, and neurofilament light, EEG changes, structural and functional MRI, sleep time and efficiency, and physical activity. After the first period, all participants had the option to continue on active settings for six or nine months. In this study, data collection was paused at three months due to the COVID pandemic. Preliminary results found daily stimulation resulted in better sleep by one month. At three months, active treatment strengthened functional MRI connectivity in brain circuits involved in vision and memory, improved performance on a face-name-matching test, and may have slowed loss of hippocampal volume. Participants reportedly could not tell if they received active or sham stimulation (Apr 2021 conference news). Results of both studies were subsequently published (Chan et al., 2022).

Seven participants in these studies entered the open-label extension. At the 2024 AD/PD conference, results were reported on three late-onset and two early onset AD patients, who completed treatment for a total of 30 months. Induction of 40 Hz neural activity was sustained, but brain connectivity was reduced. The properties of circadian rhythms improved compared to baseline; hippocampal volume loss slowed compared to historical controls. In the late-onset patients only, cognitive scores stabilized, with a preservation of performance on the MMSE, CDR-SB compared to historical controls.

In January 2021, Cognito's GammaSense stimulation device received breakthrough status from the FDA. 

In July 2022, the University of Tennessee Medical Center in Knoxville started enrolling 20 people with Alzheimer's disease into an open-label study of an eight-week course of one hour per day of treatment with Cognito’s GammaSense device. The primary outcome assesses cortical visual processing; the study will run until December 2025.

In December 2022, Cognito Therapeutics began enrolling for a Phase 3 trial in 345 people with AD. They are to be randomized to either active or sham treatment lasting one hour per day for up to a year, using a sensory stimulation device called GS120. Primary outcomes are change from baseline at 12 months on the ADCS-ADL and on the combined statistical test (CST) of the ADCS-ADL and MMSE. In mid-2023, the trial increased enrollment to 530, and later added a one-year open-label extension. It is being run at 68 sites across the U.S., and is expected to be complete in 2025.

In December 2022, a new feasibility study began at MIT, enrolling 60 people with mild AD for six months of one hour daily, at-home stimulation, with a sham control. This study uses the easel-mounted GENUS device developed at MIT, and also will test a vibrating device that delivers 40 Hz tactile stimulation (see Suk et al., 2023). Primary outcomes are 40 Hz brain entrainment, safety, functional connectivity by fMRI, and AD-related inflammation biomarkers based on RNA sequencing in blood. Other outcomes include changes in the microbiome, sleep/wake patterns, and a cognitive testing battery. Participants will have the option of continuing the regimen for one year in an open-label extension. Completion is anticipated in September 2026.

In January 2024, MIT and MGH researchers began a prevention trial. It plans to enroll 50 cognitively normal people with cerebral amyloid deposits to a one-year, one-hour daily at-home treatment with the easel-mounted GENUS device, against a primary outcome of change in brain amyloid on PET. Other outcomes are to be change in tau-PET, structural and functional MRI, EEG, night-time activity, blood biomarkers of AD, and compliance. A substudy will assess CSF biomarkers of amyloid and tau. The study also plans to measure changes in CSF flow, white matter and myelination by diffusion MRI, neuropsychiatric symptoms, gamma oscillations, sleep, activity levels and adverse events. Completion is planned for May 2026. This trial offers a one-year, open-label extension.

At Emory University, a Parkinson’s trial is set to begin in August 2024. This study will enroll 28 PD patients who have problems walking, for six months treatment with a commercially available stimulation device from the Canadian company MindAlive Inc. Primary outcomes are compliance and adverse events. The study will also assess general symptoms and specifically freezing of gait, as well as CSF biomarkers of amyloid and inflammation after active or sham stimulation. 

MIT investigators are also running safety, feasibility, and optimization trials of GENUS in people with PD, and with Down's syndrome.

For details on trials of this approach, see clinicaltrials.gov.