Newsletter · · Ashutosh Agarwal

Alzheimer's Blood Tests Win FDA Nod but Medicare Still Won't Pay - The Neuro & Alzheimer's Pipeline - Week of July 26, 2026

The Neuro & Alzheimer's Pipeline for the week of July 20 to 26, 2026. Two prescribing physicians confirmed the FDA-approved pTau-217 blood test is still not reimbursed by Medicare while the pricier amyloid PET scan is fully covered, favoring the imaging names near-term, and Biogen and Ionis's antisense tau drug became the first in its class to reverse the tau PET signal even as the trial missed its endpoint and roughly $2 billion came off Biogen's value.

The Neuro & Alzheimer's Pipeline

Week of July 26, 2026: Alzheimer's Blood Tests Win FDA Nod but Medicare Still Won't Pay

Last week the Alzheimer's field's big annual meeting in London handed us a firehose of data. This week was the hangover, quieter, more reflective, the kind of week where you find out what people actually think once the conference adrenaline wears off. And two separate podcasts, from two doctors who have never met, landed on the same uncomfortable fact at the same time: the cheap blood test everyone is excited about is now FDA-approved and works beautifully, and Medicare still won't pay a dime for it, while the expensive brain scan it was supposed to replace is fully covered. That plumbing problem, more than any drug headline, is the story of the week. We also got a smarter, calmer read on Biogen's confusing tau drug from last week, and a reminder that the neurologists actually prescribing these medicines are far more measured than the press releases.

(Quick vocabulary, used throughout: amyloid and tau are the two toxic proteins that build up in an Alzheimer's brain, amyloid clumps between nerve cells, tau tangles inside them. pTau-217 is a form of tau you can now measure in a simple blood draw. PET scan is a brain scan that lights up amyloid or tau directly. APOE4 is the most common gene that raises Alzheimer's risk. ARIA is the brain swelling-and-bleeding side effect of the anti-amyloid drugs.)

TL;DR

  • The reimbursement wall is the trade. Two prescribing physicians confirmed this week that the FDA-approved pTau-217 blood test is not covered by insurance or Medicare, patients pay cash, a couple hundred dollars, while an amyloid PET scan, which costs thousands, is "very well covered." Near-term that protects the PET imaging names; it's a real speed bump for the blood-diagnostics ramp everyone is modeling.
  • Biogen's tau drug did something no rival ever managed, it actually reversed the tau signal on brain scans, not just slowed its build-up. That reframes last week's "messy miss" as a genuine first. But the market still knocked ~$2 billion off Biogen's value, and one editor floated a scary idea: with this kind of broad tau-lowering, there may be a ceiling above which removing more tau stops helping.
  • The doctors are more sober than the headlines. A busy neurologist called today's anti-amyloid drugs "good, not great," roughly where multiple sclerosis drugs were in the 1990s, and stressed that plenty of amyloid-positive patients are not candidates for them.

What's new

The reimbursement asymmetry finally got named out loud, twice. The single most PM-relevant fact this week wasn't a drug; it was who pays. On The Empowering Neurologist Podcast (July 23), Dr. Jonathan Fellows, a practicing neurologist (Operator, he prescribes and orders these tests all day), put it bluntly: pTau-217 is "even FDA approved," has "the highest sensitivity and specificity," and "there's not a day that goes by that I don't order a pTau-217 on a patient." So it's used constantly. The catch: "the biggest problem is they're not paid for. So they're cash. You have to still, they're not covered by insurance... although it's FDA approved, Medicare is not reimbursing for these tests. So they could be a couple hundred dollars." Then the kicker that flips the whole diagnostics thesis on its head: "Contrast that with an amyloid PET scan, which is thousands of dollars, completely covered by Medicare." Two days earlier, on the Lilly-funded Keeping Current CME master class (July 21), Washington University's Dr. Tammy Benzinger (Expert/academic, note the Eli Lilly educational grant) said the same thing from the imaging side: amyloid PET "right now in the U.S. is very well covered by insurance. So the cost to the average patient, whether they have private insurance or Medicare, is actually relatively low." Why it moves numbers: the entire bull case for blood diagnostics assumes cheap tests flood the front door and pull diagnosis into primary care. But if Medicare pays for the $3,000 scan and not the $200 blood draw, the near-term economics run backwards. That's a tailwind for the PET vendors and a friction the blood-test bulls have to model honestly.

Biogen's tau drug, on second read: a genuine first, wrapped in a genuine puzzle. On BioCentury This Week (July 21), executive editors Selina Koch and Jeff Cranmer (both Pundit/analyst) gave the calmest, most useful autopsy of last week's confusing readout. The key detail that got lost in the noise: Biogen and Ionis's antisense drug, the one that lowers all forms of tau at the source rather than picking one, didn't just slow tau from accumulating, it reversed it. In Koch's words, earlier tau drugs "either failed to lower the tau PET signal, or... we saw a slowing of that accumulation. Well, [this drug] actually reversed tau PET. So it was lower than at baseline. So that felt like a breakthrough." That has never been done before in this class. On the cognition numbers, she lined them up against the marketed amyloid drugs: "a 26% slowing on CDR-sum of boxes... a 42% slowing on ADAS-COG13, 50% slowing of progression on the mini mental state exam. So if you just line those up with the approved anti-amyloid antibodies, those results look good, maybe even a little bit better."

"This therapy goes upstream... it's a bigger sort of sledgehammer, in a way." (Selina Koch, BioCentury This Week)

Why it moves numbers: this is the bull case for tau stated more clearly than at the conference itself. But Koch didn't dodge the two problems. First, the trial technically missed, the biomarkers showed a clean dose-response (higher dose, more tau lowered) while the cognitive scores did not, and, oddly, the lowest dose (60 mg every 24 weeks) worked best on cognition. Second, and more important for the whole field, she raised a possibility the bulls should sit with: "if you're taking this kind of a broader approach to lowering tau of all of its forms... there could be a therapeutic window above which getting rid of more tau isn't better." If that's true, it complicates every broad tau-lowering program, not just Biogen's. The market's verdict was already in: "the company's market cap fell over two billion that day," which Cranmer partly pinned on CEO Christopher Viehbacher's pledge to steer Biogen away from exactly these high-risk, expensive bets, and here they are launching "a very expensive, long, large Phase 3 in a high-risk situation."

The prescribing neurologist's reality check: "good, not great." Beyond the reimbursement point, Dr. Fellows on The Empowering Neurologist (July 23) offered the grounded operator view the sell-side rarely captures. His framing of the marketed anti-amyloid drugs: "Where we're at right now with anti-amyloid therapy, it's good, not great." His analogy, and it's a hopeful one if you're long the category for the long haul, is multiple sclerosis: "We're where we were in the '90s with MS, with the ABC drugs, Avonex, Betaseron, Copaxone... 20, 25, 30% reduction in relapses," and today MS infusion chairs sit largely empty because the modern drugs work so well. The unglamorous near-term truth, though: not everyone gets the drug. A patient who is amyloid-positive but has an A1C of 11, blood pressure in the 160s, and obesity? "I'm not putting that patient on an anti-amyloid drug." And on the gene that gates it all: one copy of APOE4 is roughly a 3x risk, two copies "10 to 12 times, sometimes 20," and, as he noted, Europe and Canada restrict the drugs in the two-copy patients over ARIA fears. Why it matters: this is the demand-side reality behind every Leqembi and Kisunla sales model, the eligible pool is narrower than the diagnosed pool, and the doctors are gatekeeping hard.

The debate

Does under-the-skin dosing plus cheap, accurate blood tests plus a widening pipeline finally turn anti-amyloid into a multi-billion-dollar franchise, or do modest benefit, brain-bleed risk, and diagnostic bottlenecks keep uptake weak while the tau bets stay unproven?

Bull: The science keeps validating. Tau just did something genuinely new, a drug that reversed the tau signal on brain scans and posted cognition numbers (26 to 50% slowing across measures) that stack up against the marketed drugs. The blood test works: FDA-approved, ordered daily by real neurologists, "highest sensitivity and specificity." Everyone in the field agrees the endgame is a cancer-style combination of amyloid plus tau drugs, which only expands the diagnostic and treatment market. And the doctors' own MS analogy is the bull case in one sentence: the '90s MS drugs were mediocre too, and look where that category went.

Bear: The plumbing is broken in a way this week made concrete. The blood test that's supposed to democratize diagnosis isn't reimbursed, so patients pay cash while Medicare happily funds the $3,000 scan, which is backwards for volume. The one fresh tau data point missed its endpoint, showed no cognitive dose-response, and may point to a ceiling where more tau-lowering stops helping, and investors erased $2 billion of Biogen anyway. The prescribing neurologist calls the marketed drugs "good, not great" and won't even give them to metabolically unhealthy or high-genetic-risk patients. Narrow eligibility plus an unpaid front-door test plus modest efficacy is not the shape of a runaway franchise.

Net: a quiet, sobering week that tilted the balance back toward the bears at the margin, not on the science (tau's reversal is real, the blood test is real) but on the access. The most important unlock for this whole category isn't the next molecule; it's a Medicare decision to pay for the blood test. Until that lands, the cheap test stays a cash-pay curiosity and the expensive scan keeps the volume.

Stocks in play

  • BIIB (Biogen). Bull: its tau drug (diranersin, the antisense drug partnered with Ionis) reversed tau on PET scans, a first for the class, with cognition slowing of 26 to 50% that rivals the marketed amyloid drugs; combination therapy is the field's consensus future and Biogen is out front. Bear: the trial missed its primary endpoint, showed no cognitive dose-response, the lowest dose worked best, and there may be a ceiling on how much tau-lowering helps; the stock lost ~$2B on the readout, and a CEO who promised less risk-taking is funding an expensive Phase 3. Next catalyst: the Phase 3 design and go-forward decision. (BioCentury This Week, July 21)
  • IONS (Ionis). Bull: Biogen's antisense partner on the tau drug that just posted the class's first tau-PET reversal, the technology platform is the story. Bear: still an early, single, messy Phase 2 signal inside a volatile name; no near-term revenue from it. Next catalyst: Phase 3 progression on the tau program. (BioCentury This Week, July 21)
  • LLY (Lilly). Bull: Lilly quietly sits on both sides of the diagnostics debate that's currently favoring imaging, it makes the amyloid PET tracer Amyvid (florbetapir) and the tau PET tracer Tauvid (flortaucipir), the FDA-cleared scans that Medicare fully reimburses; its Kisunla anti-amyloid drug remains the early-treatment share leader. Bear: this week's Lilly podcast chatter was all about its GLP-1 obesity fight (Cramer dismissing the Novo Nordisk false-advertising lawsuit as noise on Squawk on the Street, July 21), with zero fresh Kisunla launch color; and if blood tests eventually win reimbursement, Lilly's own PET franchise faces long-run substitution. Next catalyst: the next Kisunla sales print. (Keeping Current CME, July 21; Squawk on the Street, July 21)
  • GEHC (GE HealthCare). Bull: directly makes Vizamyl (flutemetamol), one of the three FDA-cleared amyloid PET tracers, all of which are Medicare-covered while the competing blood test is not, a real near-term reimbursement moat. Bear: PET availability is concentrated in big metros ("if you're living in a rural area, you may have to travel to a university"), and blood tests are the long-run cheaper substitute if coverage flips. Next catalyst: any CMS move on blood-biomarker reimbursement. (Keeping Current CME, July 21)
  • LNTH (Lantheus). Bull: a pure-play radiopharmaceutical name levered to the same PET-is-reimbursed, blood-is-not dynamic that dominated this week's diagnostics conversation. Bear: no direct Lantheus commentary this week, the read is thematic, not company-specific, and the same long-run blood-test substitution risk applies. Next catalyst: none flagged in podcasts this week. (Keeping Current CME, July 21)
  • QTRX (Quanterix), DGX (Quest), LH (LabCorp), C2N, Roche diagnostics. Bull: the underlying test is excellent and in daily clinical use, a neurologist orders pTau-217 every single day. Bear: the whole volume thesis runs into a wall this week, "although it's FDA approved, Medicare is not reimbursing for these tests," so patients pay cash while the rival PET scan is fully covered. Reimbursement, not accuracy, is now the binding constraint. Next catalyst: a Medicare/CMS coverage decision for blood-based AD testing. (The Empowering Neurologist, July 23)
  • Sensorium (private, CNS pipeline watch, no ticker). Bull: on Talking Precision Medicine (July 21), the CEO (Operator) described a clinical-stage CNS pipeline, lead drug Sens01 for anxiety with Phase 1 top-line data "in a couple months," plus programs in drug-resistant epilepsy and cognitive enhancement that "could be positioned... [in] neurodegenerative disorders." A reminder the CNS pipeline beyond Alzheimer's is broadening; the anxiety pitch is sharp (SSRIs "work for about a third of the people," take six weeks, and carry a black-box suicidality warning in adolescents). Bear: private, single Phase 1 asset, natural-product-derived, early and unproven. Next catalyst: Sens01 Phase 1 top-line, expected within a couple of months. (Talking Precision Medicine, July 21)

Read-throughs

  • PET imaging (LNTH, GEHC, LLY), the surprise winner of the week. The reimbursement math is currently pro-imaging: three FDA-cleared amyloid tracers (florbetapir/Amyvid, flutemetamol/Vizamyl, florbetaben/Neuraceq) plus the tau tracer flortaucipir/Tauvid, all validated against autopsy at 88 to 98% accuracy and, critically, "very well covered" by Medicare, while the cheaper blood test is not. Benzinger also flagged an under-appreciated upgrade: visual PET reads are now FDA-approved for quantification (a numeric centiloid score, not just an eyeball read), which tightens their tie to the clinical-trial gold standard. Near-term tailwind; long-term the blood test is still the cheaper substitute the day coverage flips. (Keeping Current CME, July 21)
  • Blood diagnostics (QTRX, DGX, LH, C2N, Roche), great test, broken payment. Last week the story was "the accuracy is now good enough" (Quest's 91%/91%). This week the story is "and it still doesn't get paid for." The technology risk is largely gone; the reimbursement risk just moved to center stage. Watch CMS. (The Empowering Neurologist, July 23)
  • Payers / CMS, the whole ballgame. The one policy fact that would re-rate the entire diagnostics complex is a Medicare decision to cover blood-based testing. Two independent physicians this week confirmed it hasn't happened. Amyloid PET is also not covered for asymptomatic people, "because those patients are not currently approved for these therapies," which caps the pre-symptomatic screening dream on both the blood and imaging sides for now. (The Empowering Neurologist, July 23; Keeping Current CME, July 21)
  • Combination therapy (BIIB, Eisai), still the consensus endgame. BioCentury reinforced the field's belief that Alzheimer's ends up like cancer: "everybody thinks this is probably a combination play," amyloid plus tau, with "greater efficacy from combining" if the safety allows. Bullish for the whole measure-then-treat stack, because combinations need more diagnostics, not fewer, which makes the reimbursement bottleneck above even more consequential. (BioCentury This Week, July 21)
  • Broader CNS pipeline (beyond AD). The neuro pipeline keeps widening past Alzheimer's, this week it was anxiety, drug-resistant epilepsy, and cognitive enhancement from a private developer. The recurring theme: the incumbent CNS drugs (SSRIs, benzodiazepines) are old and only partly effective, leaving a large unmet-need opening across psychiatry and neurology. (Talking Precision Medicine, July 21)
  • Infusion centers / specialty pharmacy (OPCH). No direct coverage this week. The structural picture is unchanged from last week: fully at-home Leqembi is a slow headwind to infusion volume, while Kisunla's IV-only model is the offsetting tailwind, for now.

What changed vs last week

Last week we got the AAIC box score; this week we got the second read, and it was more sober. On Biogen's tau drug, the framing improved even as the price action didn't. The detail that got buried in the conference noise, that this drug reversed tau on brain scans, something no prior tau therapy achieved, is a real scientific first, and BioCentury made the bull case more coherently than the conference did. But the same editors added a genuinely new bear brick: the possibility of a "therapeutic window" above which lowering more tau stops helping, which would be a problem for the entire broad-tau-lowering field, not just Biogen. And the market's ~$2 billion haircut stood.

The bigger shift is that the bottleneck moved from science to payment. For weeks the diagnostics debate was "is the blood test accurate enough?" Last week Quest answered yes (91%/91%). This week two prescribing physicians answered the next question, "will anyone pay for it?", and the answer is no: FDA-approved, ordered daily, still cash-pay, while the pricier PET scan is fully covered by Medicare. That's a real, fresh, tradeable inversion of the usual bull narrative, and it quietly favors the imaging names over the blood-test names in the near term.