Dementia research has produced real breakthroughs — a blood test for Alzheimer's pathology, two approved drugs that slow decline, and a home-injectable version of one of them — but none of it has become routine care. Three things must happen first: the benefit has to grow larger than it is today, the brain-swelling risk has to come down further, and the system has to be able to diagnose and monitor millions of people it currently cannot reach. This page explains what the approved tools actually do, what the trial numbers mean for one person rather than a study population, and which specific gaps stand between today's specialist-only treatment and something a general practice could offer.
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
- How big is the benefit, really?
- The blood test changed diagnosis — within limits
- Brain swelling is the safety gate
- Who can actually get treated
- What the recent failures teach
- Practical steps for families right now
- Frequently Asked Questions
How big is the benefit, really?
The headline result comes from Clarity AD, the 1,795-patient Phase 3 trial of lecanemab — an antibody that clears amyloid plaque from the brain. According to Eisai's announcement of the Clarity AD results, lecanemab slowed decline on the CDR-SB scale by 0.45 points versus placebo over 18 months, a 27% slowing with a P-value of 0.00005. CDR-SB is an 18-point clinician rating covering memory, judgment, home life, and personal care. A 0.45-point difference is statistically solid and clinically modest.
It does not stop the disease, reverse damage, or restore a lost ability. What it buys is time: patients on the drug were, on average, a few months behind where they would otherwise have been. That distinction matters for a family deciding whether to pursue treatment. The honest framing is not "a drug that treats Alzheimer's" but "a drug that measurably slows early Alzheimer's, with real side effects and heavy monitoring." Routine use will likely require either a bigger effect or a much cheaper, safer way to deliver the current one.
The blood test changed diagnosis — within limits
Until recently, confirming amyloid pathology meant a PET scan or a spinal tap. That changed when the FDA cleared the Lumipulse G pTau217/β-Amyloid 1-42 Plasma Ratio, described in the FDA's clearance announcement as the first blood test for Alzheimer's amyloid pathology. In 499 patients, 91.7% of positive results and 97.3% of negative results agreed with PET or spinal fluid testing. The clearance is narrow on purpose.
It covers symptomatic adults aged 55 and older, and the FDA states it is not a standalone test and not a screening test. A person with no cognitive symptoms should not be handed this test to find out whether they are "going to get" Alzheimer's — the clearance does not cover that use, and there is no approved treatment for that situation. What the test does change is the bottleneck. A primary care office that could never order a PET scan can now draw blood, which is the first genuinely scalable step in the whole pathway. The confirmation, interpretation, and treatment decision still sit with a specialist.
Brain swelling is the safety gate
ARIA — amyloid-related imaging abnormalities — is the defining risk of these drugs. Eisai's Clarity AD data recorded ARIA-E, meaning brain swelling, in 12.6% of lecanemab patients versus 1.7% on placebo, and ARIA-H, meaning small brain bleeds or iron deposits, in 17.3% versus 9.0%. Most cases are found on a scan rather than felt by the patient, but some are serious. That is why treatment comes with serial MRI scans and APOE4 genotyping.
APOE4 is a gene variant that raises both Alzheimer's risk and ARIA risk, and carriers — especially those with two copies — need a frank conversation about that trade-off before starting. Safety is being engineered down rather than eliminated. Eli Lilly reports that the FDA approved an updated Kisunla (donanemab) label with a modified titration schedule after the TRAILBLAZER-ALZ 6 trial showed significantly lower ARIA-E at 24 and 52 weeks than the original regimen, with similar plaque removal. Slower ramp-up, same clearance, fewer swelling events — a meaningful improvement, and still not a drug you could prescribe without imaging.
Who can actually get treated
Access is governed as tightly as safety. Medicare covers anti-amyloid antibodies only under Coverage with Evidence Development, which sets conditions most families do not expect: Then there is capacity. Alzheimer's Association 2026 Facts and Figures estimates 7.4 million Americans live with Alzheimer's, while 55% of primary care physicians say their communities lack enough dementia specialists, and between 34% and 59% of people 65 and older live in areas facing specialist shortfalls.
Approving a drug does not create a neurologist, an MRI slot, or an infusion chair. Delivery is improving at one point in the chain. Eisai and Biogen report that the FDA approved LEQEMBI IQLIK, a once-weekly subcutaneous autoinjector giving 500 mg as two 250 mg injections, as an initiation dose on July 14, 2026, with US availability in late August 2026. That removes the infusion center from the start of therapy — but not the MRI monitoring, the amyloid confirmation, or the registry paperwork.
- The patient must have mild cognitive impairment or mild Alzheimer's dementia — not moderate or advanced disease.
- Amyloid pathology must be documented, not assumed from symptoms.
- The prescribing clinician must enter data into a CMS-approved registry, such as the Alzheimer's National Registry for Treatment and Diagnostics, approved January 29, 2024.
What the recent failures teach
Two 2025–2026 results are as informative as the approvals. Novo Nordisk announced on November 24, 2025 that the evoke and evoke+ Phase 3 trials of oral semaglutide in 3,808 people with early Alzheimer's failed to beat placebo on CDR-SB, as reported through Alzheimer Europe, and the one-year extensions were discontinued. The drug improved disease biomarkers and patients still declined at the same rate. That is the single most useful lesson for reading dementia headlines: a biomarker moving is not a person doing better.
The same pattern appeared in tau research. The Alzheimer's Association reports that Biogen's antisense drug diranersen missed its primary endpoint in the Phase 2 CELIA trial in May 2026, despite cutting spinal fluid total tau by 52–65% and slowing decline by 9–26% at 72 weeks — with the lowest dose performing best and a side effect emerging at high doses. Prevention remains untested. TRAILBLAZER-ALZ 3, Lilly's Phase 3 trial of whether donanemab can delay progression in people who have amyloid but no symptoms — using a plasma p-tau217 assay for eligibility — was active and not recruiting as of January 19, 2026. Until it reads out, treating a symptom-free person with a positive amyloid result is not supported by evidence.
Practical steps for families right now
For anyone weighing the decision this year, the concrete arithmetic is Eisai's 0.45 CDR-SB points against 12.6% ARIA-E, under a registry requirement, at a clinic with MRI access.
- Act early on symptoms. Eligibility is limited to mild cognitive impairment and mild dementia, so a delayed evaluation can close the door entirely.
- Ask specifically whether amyloid status has been documented, and by which method. Without it, Medicare coverage does not apply.
- Ask whether the clinic participates in a CMS-approved registry. If it does not, coverage will not follow the prescription.
- Ask about APOE4 testing and the MRI schedule before agreeing to treatment, not after the first dose.
- Treat biomarker-only results in news coverage with caution — semaglutide moved biomarkers and changed nothing clinically.
Frequently Asked Questions
Can a blood test tell me if I will develop Alzheimer's later?
No. The FDA cleared the Lumipulse plasma ratio for symptomatic adults 55 and older, and it is neither a standalone nor a screening test.
Why does treatment require repeated MRI scans?
To catch ARIA. Clarity AD found brain swelling in 12.6% of treated patients, usually detected on imaging before symptoms appear.
Is the home autoinjector a replacement for infusions?
LEQEMBI IQLIK was approved as a once-weekly initiation dose, removing the infusion center at the start of therapy. Monitoring requirements are unchanged.
Does clearing amyloid or tau guarantee slower decline?
No. Oral semaglutide improved biomarkers without beating placebo, and diranersen cut spinal fluid tau by 52–65% while missing its primary endpoint.
