What Biogen and AriBio’s New Data Mean for Slowing Cognitive Decline

Two November 2026 Alzheimer's readouts, what their 31% and 38% slowing figures really mean, and why neither drug is prescribable yet.

Two Alzheimer's drug readouts landed on November 17, 2026, and both pointed the same direction: measurable but partial slowing of cognitive decline, in trials that each fell short of a clean win. Biogen's diranersen slowed decline by 38% at two years in a Phase 2 extension, and AriBio's oral AR1001 slowed it by 31% — but only in patients taking no other Alzheimer's medication. Neither drug is available to patients, and neither result is the kind that settles a question. What they do change is the map of what might work: diranersen attacks tau rather than amyloid, and AR1001 is a pill rather than an infusion.

Medical information disclaimer: This article is for general educational purposes only and does not provide medical advice, diagnosis, or treatment. Always consult a physician or other qualified health professional about symptoms, medications, tests, or treatment decisions.

Table of Contents

What Biogen's diranersen data actually showed

Diranersen (also called BIIB080) is an antisense therapy — a drug designed to switch off production of a specific protein. Its target is tau, the protein that forms tangles inside brain cells as Alzheimer's advances. Every approved Alzheimer's drug that modifies the disease targets amyloid instead, the protein that forms plaques between cells. At the CTAD conference in Boston, Biogen presented 96-week extension data from its Phase 2 CELIA trial. According to the Ionis announcement, the 60 mg twice-yearly regimen (60 patients) slowed decline on the CDR-SB scale by 38% against the original placebo group (115 patients) — a 1.09-point difference.

CDR-SB stands for Clinical Dementia Rating–Sum of Boxes. A clinician scores six areas — memory, orientation, judgment, community affairs, home life, and personal care — and adds them up, from 0 to 18. Higher means worse. A 1.09-point difference after two years is small on paper but sits in the range regulators have accepted before. The slowing also grew with time: 26% at week 76, then 38% at week 96. Biogen reads that as evidence that longer tau lowering produces more benefit — the pattern you would hope for if the drug is changing the disease rather than masking symptoms.

Why CELIA still counts as a mixed result

CELIA missed its primary endpoint. That endpoint was a statistically significant dose-response relationship on CDR-SB at week 76 — meaning higher doses should have produced more slowing. Per Biogen's investor release, all doses beat placebo, but the higher ones did not beat the lower ones. That is not what a working dose-response curve looks like. The two-year comparison has a second problem.

By week 96, some participants originally assigned to placebo had crossed over and received diranersen. The control group was no longer a clean control. Fierce Biotech reported that BMO Capital Markets called the effect clinically meaningful but the comparison flawed, and noted the update included no tau imaging — so there is no picture showing tangles shrinking alongside the clinical numbers. On safety, the drug was generally well tolerated through two years with no new signals. That matters for an intrathecal therapy — one injected into the spinal fluid — where tolerability is a real barrier to use.

AriBio's AR1001 and the monotherapy split

AR1001 is a 30 mg oral pill, a PDE5 inhibitor — the same drug class as sildenafil, originally developed for blood flow. AriBio ran POLARIS-AD, a 1,535-patient, 52-week Phase 3 trial in early Alzheimer's. In the overall population, AR1001 did not separate from placebo. In a prespecified subgroup — patients taking no acetylcholinesterase inhibitor (donepezil, rivastigmine, galantamine) and no memantine — it did. AriBio reported a 0.44-point CDR-SB advantage in that monotherapy group (p=0.0266), which it describes as 31% slowing.

The biomarker data split the same way. CSF p-tau217 — a spinal-fluid marker that tracks Alzheimer's disease activity closely — fell 5.34 pg/mL versus placebo in the monotherapy group (p=0.0402), but only 3.52 pg/mL in the overall population, which was not statistically significant (p=0.1784). Two measures agreeing on the same subgroup is more persuasive than a lone clinical number, though a subgroup finding in a trial that missed overall remains a subgroup finding. Side effects were more common on the drug than on placebo: diarrhea 7.7% versus 4.4%, falls 7.3% versus 5.9%, dizziness 6.1% versus 3.1%. Discontinuation for adverse events ran 5.2% on AR1001 against 1.7% on placebo. The falls number deserves attention in a population already at fracture risk.

How these numbers compare to drugs you can get today

Both effects land in roughly the same 25–38% band as the approved anti-amyloid antibodies Leqembi (lecanemab) and Kisunla (donanemab), per Fierce Biotech's coverage. That is the honest framing: these are not breakthroughs that dwarf current options, they are additional routes to a similar size of benefit. The practical difference is in delivery and mechanism rather than magnitude. Anti-amyloid antibodies require regular infusions and MRI monitoring for brain swelling and bleeding.

A pill with gastrointestinal side effects is a different proposition for a family managing appointments and transport. And diranersen is the first therapy to suggest that lowering tau — not clearing amyloid — can slow progression, which opens a second target if it holds up. Neither is prescribable. AriBio plans to discuss POLARIS-AD with the FDA. Biogen said in May that it would advance diranersen into Phase 3, which means years before any approval decision.

What a patient or caregiver can do with this now

These results do not change treatment today. They change what is worth watching and asking about.

  • Ask whether early-stage diagnosis is documented. Both trials enrolled early Alzheimer's; every therapy in this class depends on being caught early, and that depends on biomarker confirmation, not just a memory complaint.
  • Raise clinical trials with the treating neurologist. Diranersen's Phase 3 will need participants, and trial entry is the only route to these drugs for the foreseeable future.
  • Do not stop donepezil or memantine on the strength of AR1001's monotherapy finding. The subgroup was patients who were already not taking them, which is not the same as evidence that stopping helps.
  • Treat falls as a tracked risk if any PDE5 inhibitor is ever prescribed for cognition. The 7.3% rate in POLARIS-AD is the kind of number that outweighs a modest cognitive gain in a frail patient.
  • Read percentage claims as slowing, not improvement. A 31% or 38% slowing means decline continues more slowly — no one in these trials got better.

Frequently Asked Questions

Does a 38% slowing of decline mean symptoms improve?

No. It means the measured decline over two years was 38% smaller than in the comparison group — a 1.09-point difference on an 18-point clinician scale. Function still worsens, more slowly.

Why does it matter that diranersen targets tau instead of amyloid?

Every disease-modifying Alzheimer's drug approved so far clears amyloid plaques. Diranersen is the first to suggest that reducing tau, the protein forming tangles inside neurons, can also slow progression — a second mechanism to pursue if larger trials confirm it.

Can I ask my doctor for AR1001 since it is an oral drug?

No. AR1001 is investigational. AriBio has said it plans to discuss the POLARIS-AD results with the FDA, and the trial did not succeed in its overall population.

What is CSF p-tau217?

A protein fragment measured in spinal fluid that reflects Alzheimer's disease activity. AR1001 lowered it significantly in the monotherapy subgroup (5.34 pg/mL versus placebo) but not in the full trial population.


You Might Also Like

HelpDementia.com

Dementia, Alzheimer's, Caregiving & Healthy Aging Guidance

© 2026 HelpDementia.com. All rights reserved.

Educational information only. It is not medical advice and does not replace care from a qualified clinician.