Biogen's Alzheimer's Drug Failed Its Big Test. They're Going to Phase 3 Anyway.

Why Tau Matters (and Why It's So Hard)

Alzheimer's research has two main villains: amyloid plaques and tau tangles. For decades, the field obsessed over amyloid. That bet eventually paid off with drugs like lecanemab (which Biogen co-developed with Eisai), but the benefits have been modest, and amyloid clearance alone doesn't stop cognitive decline.

Tau is the other half of the equation, and in many ways the more important one. Tau pathology correlates more tightly with how fast patients actually get worse. If amyloid is the match that starts the fire, tau is the fire itself spreading through the brain.

But tau has been a graveyard for drug developers. AbbVie's tilavonemab? Missed its Phase 2 endpoints and got discontinued. Roche's semorinemab? Failed its co-primary endpoints in Phase 2, though it did show a significant reduction in cognitive decline on the ADAS-Cog11 measure. J&J's posdinemab flopped in November 2025. The antibody approach to tau has produced a string of expensive disappointments.

Diranersen's ASO mechanism is different from all of these. Instead of trying to catch tau protein after it's already been made (the antibody strategy), it goes upstream and blocks the MAPT gene's instructions before tau gets produced at all. It's the difference between mopping up a flood and turning off the faucet.

The Risky Precedent of Pushing Past a Miss

Biogen knows this playbook. They've run it before, and it's a bit of a mixed bag.

Aducanumab, their controversial amyloid drug, jumped from a Phase 1b study to twin Phase 3 trials without a proper Phase 2. Both Phase 3 trials were stopped for futility in 2019 before one arm showed a marginal signal, leading to a conditional FDA approval in 2021 that satisfied almost nobody. Roche did something similar with gantenerumab: the first Phase 3 failed, so they cranked up the dose and tried again. That also failed, in November 2022.

Alzheimer's drug development has a roughly 99% failure rate for disease-modifying therapies. The history of this disease is littered with companies that talked themselves into continuing when the data was whispering "stop."

So is Biogen fooling itself again? The analysts don't seem to think so. Jefferies had flagged a 5-15% stock upside scenario if tau PET reductions exceeded 50%, and the Phase 1b data had already shown 50-60% CSF soluble tau reductions. Stifel called the tau target a favorite among key opinion leaders. H.C. Wainwright noted the data could "validate tau knockdown as a therapeutic strategy" even without a clean primary win.

What Makes This Time Different (Maybe)

The case for advancing diranersen is stronger than the typical "we failed but found something in the subgroups" story for a few reasons.

First, the miss was technical, not biological. The drug clearly engaged its target and moved biomarkers in the right direction. The primary endpoint asked a specific statistical question about dose-response; the drug's actual effect on cognition showed up in other pre-planned analyses. That's a design problem, not an efficacy problem.

Second, the biomarker-clinical alignment is new for tau. No previous tau therapy has shown both sustained tau reduction (soluble and aggregated) and cognitive slowing in the same trial. That's what Singhal was getting at with "unprecedented confluence." Previous tau antibodies could lower one measure or the other, but never both alongside a clinical signal.

Third, the mechanism is genuinely different. Every major tau failure has been an antibody trying to intercept or clear tau protein after the fact. Diranersen is the first tau ASO to reach Phase 2 in Alzheimer's, attacking the problem at the genetic level. The failures of antibodies don't necessarily predict the failure of gene-targeted approaches.

Biogen plans to share full CELIA data at the Alzheimer's Association International Conference (AAIC) in 2026 and begin regulatory discussions about a registrational (Phase 3) program. Those conversations will determine whether the FDA agrees that the secondary signals justify the cost and time of a larger trial.

The Bigger Picture

This decision ripples beyond Biogen. The tau field is watching. Eisai has E2814 in a Phase 2/3 adaptive trial. If Biogen's advancement signals that regulators and investors still believe in tau, it could encourage the entire class.

If it flops in Phase 3, it could set the field back years.

Alzheimer's affects tens of millions of people worldwide, and current treatments only modestly slow the disease. The field desperately needs something that works on the tau side of the equation. Diranersen's CELIA results aren't a home run; they're more like a deep fly ball that might clear the fence depending on the wind. Biogen is betting the wind is blowing out.

We'll know soon enough whether they're right.

The Bidding War That Turned a $18 Stock Into a $23.50 Payday

Two companies got into a bidding war over a single-product pharma company with shrinking revenue and no pipeline. The final price: a 76% premium that says a lot about how the market values small-cap pharma right now.