Whatever Happened to Eisai’s “Dementia Ecosystem”?
It's an old, old Eastern story, the earliest version found in a Buddhist text dated around 500 BC, about some blind men who encounter an elephant in a village. As the blind try to make sense of what each feels, it's also a memorable illustration of how our views and understanding are shaped by our fragmented analysis and, as is so often the case, why we don't see a whole picture.
In most versions the men argue over what this strange thing is, convinced they themselves must be right. In some tellings, they may even collaborate to identify the elephant together, although that's rare.
In no version of this story, however, a parable that has crossed many religious traditions and cultures and told over centuries, does anyone think to ask the elephant's opinion.
In May 2026 Eisai set a three-year plan in front of its investors and built the whole thing around a single drug. The revenue target, the focused pipeline, the cost cuts, the first bond issue in eighteen years — all of it radiates out from Leqembi, the Biogen-partnered Alzheimer's antibody Eisai is counting on to carry the company from 825 billion yen in revenue to a trillion by 2028 (roughly $5.5 billion to $6.7 billion). Leqembi is the center of gravity of the whole plan, the organizing idea for growth and the commercial model at once (in its investor presentation, Eisai's own word is "centered").
In strategy, center of gravity is a concept for competition, the source of a system's power, the point where its strength concentrates. Eisai has built a trillion-yen plan that stands or falls on a single molecule.
Which would be a fine bet if the point were strong.
Leqembi is a monoclonal antibody that, on paper, clears amyloid plaque, the prime suspect in Alzheimer's for thirty years. Calling it a suspect undersells what the amyloid hypothesis became, though. Somewhere along the way it stopped being a theory and turned into its own self-organizing economy. The bet that amyloid causes the disease sparked and organized a whole system of markets around itself: roughly $1.6 billion a year of federal research money, close to half the government's Alzheimer's budget, and with it the biomarkers, the trial designs, the diagnostic criteria, the careers, the regulatory path, every one of them built to find the protein and clear it. Any company that wanted into Alzheimer's came in through that door, competing less to cure dementia than to win inside a system that had already settled what winning meant.
In its pivotal trial Leqembi slowed cognitive decline by 27 percent over eighteen months. Twenty-seven percent is the number Eisai leads with. The other number is 0.45, the real distance between the drug and a placebo on the eighteen-point scale the trial used, against a threshold most researchers set a point or two higher before a doctor or a family would notice the difference. The benefit is real, statistically clean, and possibly too small to feel.
Set that against the machinery that stands between a patient and the first dose. The patient, or more often the family member beside him, has to find a specialist, confirm the disease with a PET scan or a spinal needle, test for the gene that says whether the drug will bleed his brain, sit in an infusion chair for an hour every two weeks, and return for the MRIs that catch the swelling and the bleeding before they do real harm. Three patients in the trial's open-label extension died after brain bleeds. The list price is $26,500 a year.
Regulators are split on the value proposition.
The FDA called the benefit meaningful and approved the drug. Europe refused it outright in July 2024, then relented only for the patients least likely to bleed; Germany, the continent's largest market, found no added benefit and restricted who may prescribe it. Britain approved it and declined to pay. Norway weighed the 0.45 points of benefit against three million kroner for every quality-adjusted year of life and walked away. And Medicare, which once braced for $3.5 billion a year in spending, has paid out a small fraction of that, because the patients never came.
The conventional read is that this is a launch problem. When Eisai trimmed its forecast for Leqembi sales in late 2024, Stifel's Paul Matteis wrote that investors were losing patience with how slowly it was moving: clear the prior authorizations, build more chairs, shorten the road to diagnosis, and the drug finds its patients. But a drug whose technical benefit is unclear does not have an access problem. It has a value problem, and value in the drug market is increasingly less a property of the molecule. It is a property of the system the molecule arrives inside, a system that has to produce and deliver the drug cheaply enough that the cost stops dwarfing the benefit, keep a patient on it long enough for the slowing to add up to something a family can see, and gather the evidence that positions its technical benefit into a broader story of value, told and sold at a system level.
Whatever Happened to Eisai's Dementia Ecosystem?
None of this is a surprise to Eisai.
It is trying to build a dementia ecosystem, a wholly owned platform called Theoria, launched in 2023, built to carry a person from healthy through diagnosis through treatment and out the far side. Eisai even showed investors the diagram in an earnings call that year.
Since launching, it has been evolving and progressively integrating different features.
NouKNOW lets a person catch their own decline digitally, on a tablet. Sasaeru logs the small facts of a patient's day for the doctors and the family. EcoNaviSta, the Tokyo monitoring company Eisai bought outright in 2025 for around $107 million, reads the sleep and daily rhythms of people in senior facilities. Lifenet, a Tokyo online insurer Eisai allied with, sells the dementia coverage the two of them built. Two marketplaces Eisai runs through Theoria sit above the rest, Theo One for membership and data, Theotol for media and content, threaded across four stages of a life — healthy, at risk, diagnosed, declining — with a touchpoint or a partner at every step: a cognitive check, a meal-delivery deal with Benesse, a family-trust service, a sleep sensor, a care-facility locator.
Almost every piece does one job. It finds the patient, earlier and earlier, in the years before the diagnosis, while still only at risk, to close the test-to-treat window.
But finding an Alzheimer's patient has never been the bottleneck.
The bottleneck sits after the diagnosis: the infusion chair, the scans, the eighteen months, and the payer at the door asking what all of it actually buys. That is the question Norway answered no to, the question Britain answered by approving the drug and refusing to fund it, the question Medicare answered by declining to forecast the spend.
And at the May investor day, Eisai said the plan to make Leqembi grow revolves around three things: the subcutaneous autoinjector, so the drug can be started at home instead of in a chair; the blood test, so confirming the disease no longer takes a PET scanner or a spinal needle; and real-world evidence, the long accumulation of data meant to prove to payers that the slowing was worth the money. Delivery, diagnosis, proof. The long-term vision slide lists what Eisai means to weave together: biomarkers, data, technology, ecosystems.
So Eisai is aiming for a different direction, attempting to move to a new form, a place where the Standard Model no longer works. The molecule is one box, and the power is in everything around it. That is an ecosystem strategy, not a drug shoved into segments but a whole system of markets wired together and pointed in one direction, the kind of structure that stops competing inside a category and starts setting the category's terms. (Blue Spoon has published a rough mix of such a system for Alzheimer's and put a number on it, a new market on the order of $860 billion a year.)
But Eisai is building only the half it already knows, the half that orbits the drug: the finding, the touchpoints, the autoinjector it makes itself. The other half is the physical path that carries a weekly injection into people's homes and keeps them there.
Whatever Happened to Best Buy Health?
Best Buy spent the back half of the last decade deciding it was stepping into a new peer group, becoming a new species, a home health company.
It bought Lively for medical alerts and aging in place. In 2021 it paid nearly $400 million for Current Health, a hospital-at-home platform, and pointed its Geek Squad, the people who already walk into customers' living rooms to mount the television, at installing and explaining medical devices in those same homes. It built the home into a place medicine could be delivered, piece by piece, until the whole care-at-home operation sat under one roof. It was Best Buy making the same bet Eisai is now making, placed from the other side of the table.
It did not work, and it did not work for one reason: the standard economics of healthcare.
The rules that decide who gets paid for care, the codes and the waivers, had no settled line for what Best Buy was building. The federal waiver that pays for hospital-at-home kept getting renewed a few months at a time and never made permanent, and health systems running on thin margins will not build a service around a program that might disappear by September. The money never firmed up. In 2025 Best Buy wrote down its health unit by $475 million, spent another $109 million tearing it back apart, and sold Current Health back to the founder it had bought it from. Best Buy's chief executive called it the exit of the care-at-home business. The living room Best Buy had spent four years and half a billion dollars wiring up went dark.
Then, in April 2026, Best Buy said Corie Barry would step down as CEO on October 31. Growth had stagnated; revenue and earnings had drifted sideways for years; the big health market innovation never came. The board handed the company to Jason Bonfig, a thirty-year Best Buy man who built its retail-media and third-party marketplace businesses, the high-margin advertising and platform engines, the parts of the company that look least like a house call. Wall Street applauds those high-margin engines, and that applause is the market buying expert intuition, the safe, legible, pattern-matched bet. The engine they want him to throttle up is the one that sells ads against electronics. The engine quietly cut loose is the one that was being built to develop a new industry model.
A Crazy Hypothetical: Best Buy + Eisai = New Industry Narrative
What happened to Best Buy Health is the same thing that is happening to Eisai's dementia ecosystem, photographed from the other end. Two companies saw the same future and built real pieces of it. Neither can become the company that finishes it. Eisai is a drug maker thinking in molecules; Best Buy is an electronics retailer that could not stop thinking in screens and the rooms they sit in.
The FDA cleared a subcutaneous Leqembi in 2025, an autoinjector a patient can use at home for maintenance, and a decision on using it to start patients from scratch is due in August 2026. A shot, at home, once a week, in place of an hour on a hospital drip twice a month. The infusion chair, the most expensive and most limiting piece of the whole routine, is being engineered out of existence at the moment Best Buy, the company that owned the consumer's home, walked out of it.
The best ecosystems are products of novel combinations. What if Best Buy and Eisai collaborated as keystones to stand up a new industry model instead?
Eisai owns the molecule and the autoinjector. Best Buy owns the home, the Geek Squad (perhaps working with the Visiting Nurses Association) walking through the door, the monitoring stack, the lines into the health systems and the aging customer buying his fall sensor at the store. A payer, increasingly the self-funded employer, funds the home route because it is cheaper than the chair. None of the three, the drug maker, the retailer, the hospital, sells what the others sell. Make their markets and assets and capabilities interoperable and they lock into one commercial model, a new organism, none could build alone: easier diagnosis pulls patients in, the home shot keeps them past the eighteen-month mark where most quit, and the data does more than show the payer what its money bought; it creates the conditions for whole new categories of technologies and services. The loop closes and feeds itself.
That is not a partnership. It is collaborating on industry invention, intentionally crafting a better economic system organized around the production of affordable health in "dementia," with its own center of gravity, where two narrower ones used to be. And here is the question the Best Buy board did not put on the table: if Corie Barry had built it, if she had turned Best Buy from televisions and ad sales toward standing up a new dementia industry with a Japanese drug company, would she still have her job?
Making the Elephant Part of the Loop
The question everyone asks about Leqembi is whether it works. It is close to unanswerable by the molecule alone.
Whether the drug works is settled by the system around it or not at all. So the real question for Eisai is not about the drug. It is about identity. Is it in the business of making a molecule, or in the business of producing health, with the drug market as a keystone? The first is the company it has always been. The second is the company the system is demanding, and the distance between them is the distance between millions and billions. Best Buy faced the same question about identity and answered it the same way, in the language of screens and stores.
Years ago a conceptual magician named Derek DelGaudio built a show, In & Of Itself, around the very parable this essay opened with, the blind men and the elephant, and ran it more than five hundred nights on a single proposition: identity is an illusion. The audience filed in past a wall of cards that read "I am," each one finished a different way, and surrendered the half that named them. His own line fits either boardroom: "I don't know what I am. I just know what I do."
The show closed every night by folding each private "I am" in the room into a single "we are."
Finding strategic fit in a world gone upside down is not a drug market problem alone. The same ground is shifting under marketing, media, finance, automotive, defense, every business and every government built for a world that no longer holds still. The ones that come through will be the ones that stop firing all their force at the fragment and start building the system that carries it.
Can elephants dance? You would have to ask them first.
/ jgs
John G. Singer is the founder and Executive Director of Blue Spoon and the author of When Burning Man Comes to Washington: A Field Manual for Riding Chaos. Hardcore Zen is published weekly on Substack.
Disclosure: Blue Spoon has no financial relationship — equity, advisory, or commercial — with any company named in this essay. The analysis is independent.