The FDA Cleared an Alzheimer's Blood Test. Here's Who It's Actually For.

The rain came in over the ridge on a Tuesday, the kind that arrives without ceremony and settles in for the day. I was on the phone with a woman I've known for years — she called about something else entirely, a form she couldn't make sense of — and near the end, in the voice people use when they've been carrying a thing a while, she asked me if I'd seen the news about the blood test.

She'd read a headline. A blood test for Alzheimer's. Ninety-seven percent.

"Eleanor," she said, "should I ask for it?"

I've been thinking about her question ever since. Not because the answer is complicated, exactly. But because the headline she read and the thing the FDA actually did are two different stories, and the distance between them is where a person can get hurt.

What the FDA Actually Cleared

On August 19, 2026, the FDA cleared a blood test called PrecivityAD2, made by a company called C2N Diagnostics.

Here is the sentence that matters, the one describing who the test is for. It is indicated for use in adults aged 40 years and older who present with signs or symptoms of cognitive impairment, undergoing evaluation for Alzheimer's disease or other forms of cognitive decline.

Read that twice. Not adults 40 and older. Adults 40 and older who already have symptoms.

The company says it plainly in its own announcement: the test "is not intended as a screening or stand-alone diagnostic test." It isn't something you take because you're worried. It's something a doctor may use when you've already come to them with a problem — when the forgetting has become the reason for the appointment, not a thought you have on the drive home.

There's a second thing worth understanding, and it's a word most of us skate past. The FDA cleared this test. It did not approve it. Clearance is a different, lighter process — the manufacturer shows the agency that the new test is substantially equivalent to one already on the market. It's a real review and it means something. It is not the FDA telling you this test is right for you, and it never was.

And one more, because it will save someone a phone call: the cleared version isn't for sale yet. C2N says it's "expected to be available later this year." What exists today is the older laboratory version the company has been running all along.

It Isn't the First. That's the Interesting Part.

The word "first" is doing a lot of work in the coverage of this, and it's worth slowing down on. The company's own announcement is careful — it says the first Alzheimer's blood test cleared for adults with cognitive symptoms as young as 40. That's true. But a qualified claim gets shortened every time it's retold, and by the time it reached my friend on the phone it had become simply: the first blood test for Alzheimer's.

It isn't. It's the fourth — and it wasn't even alone on the day it was cleared.

I went and looked at the FDA's own device records, because that's a thing anyone can do. The agency files these tests under a category it calls, in its unlovely way, "Immunoassay Blood Test For Amyloid Pathology Assessment." Two tests were already in it. And on August 19, the FDA cleared two more — PrecivityAD2 and a Roche test, Elecsys pTau217 — on the same day.

A note on dates, since I got this wrong myself when this article first ran: the announcements came four days apart, C2N on August 20 and Roche on August 24, and it is easy to read those as two separate clearances a few days apart. They are not. FDA's own records give both a decision date of August 19, 2026. A company's press release date is not the regulator's decision date.

Look down that last column. Fifty. Fifty-five. Fifty-five. Forty.

That's the actual news, and no headline carried it. The door didn't open — it opened wider, and it opened toward younger people. Early-onset Alzheimer's has always been the loneliest version of this disease, in part because a fifty-two-year-old who can't find the word for "colander" gets told she's tired, she's stressed, she's a woman of a certain age. For the first time there's a cleared blood test that doesn't age her out before the conversation starts.

The Number in the Headline Describes One Group of Patients

Now the part that made me want to write this.

The company validated the test in 1,142 people who came in with signs or symptoms of cognitive decline. It doesn't return a yes or a no. It uses what the company calls dual cutoffs, which is a technical way of saying the result lands in one of three places.

A Positive result had a 97.6 percent positive predictive value — the rule-in number, the one in the headline. When it says amyloid is there, it's almost always there. (Almost always compared to what, is the fair question. The company measured against the established methods: an amyloid PET scan read by a specialist, or a spinal fluid test.)

A Negative result had a 93.1 percent negative predictive value — the rule-out number. When it says amyloid isn't there, it's usually not. Usually. About seven people in a hundred with a negative result turn out to have it anyway.

And then there's the middle.

A result the company calls Likely Positive occurred in 17.3 percent of the people tested — about one in six — and it had a 77.3 percent positive predictive value.

Sit with that a moment. Roughly one person in six lands in a category where the number is not 97.6 percent. It's 77.3. Which means more than one in five people who get that result do not have the amyloid the test is pointing at.

Both of those figures are positive predictive values. Both are accurate. They describe different people — and a headline that says "97 percent accurate" has quietly folded the second group into the first and thrown away the fold. If you are the woman who lands in the middle band, the headline number was never about you.

I don't say this to frighten anyone away from a good test. It is a good test, and the middle result still carries real information — the company's own point is that every result category tells the doctor something useful about what to do next. I say it because the woman on the phone with me was about to make a decision based on a number, and the number had a shape she couldn't see.

What a Result Can and Cannot Tell You

These tests look for amyloid — a protein that builds up in the brain in Alzheimer's disease.

Amyloid is not the same thing as dementia. A test that finds it has not told you that you have Alzheimer's, and a test that misses it has not told you your memory is fine. The FDA's own summary for the Roche test says a negative result means a person "should be investigated for other causes of cognitive decline" — because there are many, and a good number of them are treatable. Thyroid trouble. A B12 deficiency. An infection. Medications working against each other.

That's why none of these tests stands alone, and why all three are written for a doctor who already knows how to evaluate cognitive impairment. The blood draw is one piece of evidence, laid beside the history and the cognitive testing and imaging that were always the heart of the thing.

What It Costs, and the Honest Answer About Coverage

I want to be straight with you here, including about the limits of what I could find out.

The test bills under a code called 0503U. I went looking for what you'd actually pay and what Medicare would actually cover, and here is what I found: C2N's own cost and billing page publishes no price and takes no position on whether Medicare, Medicaid, or private insurance covers it.

There are numbers circulating out there. I found a price quoted on pages dating to early 2024, and flat statements that Medicare doesn't cover any of it. I'm not going to repeat them to you. A two-year-old price is not a price, and a coverage claim that turns out to be stale is worse than no claim at all — it's the kind of thing that makes a person decide not to bother asking.

What the company does say on that page is that it runs a Patient Financial Assistance Program "intended to enable access to tests and reduce cost barriers for eligible patients," available to people who have no insurance or whose plan doesn't cover the test, and that it offers payment plan options, subject to application.

So here is the only advice I'll give on money, and it works no matter what the answer turns out to be:

Call and ask what you will pay before the blood is drawn. Client services is 1-877-C2N-DIAG — that's 1-877-226-3424 — or billingservices@c2n.com. Ask the price. Ask whether your plan covers it. Ask about the assistance program by name. Get it before, not after.

If You're the One Who's Been Wondering

I keep coming back to that Tuesday phone call, because I don't think she was really asking about a test.

She was asking whether she was allowed to take her own worry seriously.

If that's where you are — and if you've read this far, it may be — then the useful step isn't a test. Not first. The useful step is the appointment, and the sentence you say when you get there. Not "I've been a little forgetful lately," which every doctor hears ten times a day and which invites the answer you least want and least deserve. Something with an edge on it. In March I could still manage the checkbook. I can't now. Specific. Dated. Two points on a line, so that someone can see the direction.

Bring someone with you if you can. Not to speak for you — to remember the middle of the conversation, which is the part everyone loses.

And if the word "amyloid" comes up, you might ask what a middle-band result would change. It's a fair question, and a doctor who works in this area will have a real answer.

What I'd gently ask you not to do is call and ask for the blood test by name because of a headline. Not because you shouldn't advocate for yourself — you should, always — but because a test cleared for people with symptoms cannot answer the question most of us are actually carrying, which is will this happen to me. That test doesn't exist. If it did, I'd tell you.

Breathe

The dogwoods will bloom again in April whether or not any of us get answers this year. I've sat with more families than I can count in the long, unlit hallway between noticing something and knowing something, and I can tell you the hallway is survivable, and that almost nobody walks it as alone as they feel.

Something did change this August. Not a cure, and not a crystal ball. A door opened a little wider, for people a little younger, and that is worth knowing about.

You're allowed to take your worry seriously. That was always true, and it didn't take a blood test to make it so.