The Alzheimer’s drug pipeline is littered with expensive failures. Cognito Therapeutics is betting it can skip the pipeline entirely.

The Cambridge, Massachusetts company just closed an oversubscribed $105 million Series C, led by Morningside Ventures, IAG Capital Partners, and Starbloom Capital. Total raised: $233 million. The pitch: a headset, not an infusion, that treats Alzheimer’s by flickering light and pulsing sound at exactly 40 cycles per second.

That frequency matters. Gamma oscillations at 40 Hz are tied to the brain’s housekeeping functions — clearing metabolic waste, consolidating memory, synchronizing the neural circuits that degrade as Alzheimer’s takes hold. The science originated in MIT’s Li-Huei Tsai lab, where researchers showed that driving gamma rhythms in mice reduced amyloid plaques and tau tangles. Cognito’s Spectris platform is the clinical bet that what worked in a mouse hippocampus will work in a human living room.

The FDA seems at least intrigued. Spectris AD holds a Breakthrough Device Designation, which fast-tracks the review process and puts the company in closer contact with regulators during development. The HOPE pivotal study — the trial that will make or break the commercial case — is now fully enrolled, with topline data expected later this year. If the numbers hold up, Cognito is targeting a 2027 market launch.

The competitive backdrop sharpens the opportunity. Lecanemab and donanemab, the two anti-amyloid antibodies that cleared FDA review in the past two years, deliver modest cognitive benefits at the cost of brain swelling, microbleeds, and regular IV infusions that run into the tens of thousands of dollars annually. Many patients can’t tolerate them. Many more won’t. A noninvasive device that patients use at home, daily, with few side effects would occupy a very different market niche — if the data cooperate.

The “if” is doing heavy lifting in that sentence. Noninvasive brain stimulation has a mixed track record. Transcranial direct current stimulation has spent years in a replication crisis. Other gamma-entrainment approaches have produced mixed results in small studies. Cognito’s HOPE trial is the first to test the concept at the scale and rigor that regulators and payers will demand. The $105 million is a bet that the answer is yes. The data readout will determine whether it was a good one.