Tau PET reimbursement is the next diagnostic access question

Two related facts frame this signal.

First, amyloid PET has moved from a clinical-trial tool to routine clinical care over the past three years. It is available at most major health systems in the US, Medicare-covered under a defined NCD framework, and increasingly used as the confirmatory step before anti-amyloid therapy initiation. The access system around amyloid PET is not perfect - the biomarker access equity gap Signal covers where it still fails - but it is in place.

Second, tau PET is not. There are FDA-approved tau PET tracers. The imaging is clinically informative. But tau PET is available at meaningfully fewer sites, the Medicare coverage framework is narrower, commercial payer coverage is inconsistent, and the infrastructure required to expand it overlaps with - and competes against - the already-constrained amyloid PET capacity.

The reason this matters now is that tau-directed therapy is advancing toward approval. Several phase 3 programs use tau PET for trial enrollment and, if approved, would plausibly require tau PET for routine-care eligibility confirmation. See our Signal on tau-targeting programs for the pipeline context.

The operational pattern that gated anti-amyloid rollout is about to repeat, one mechanism later:

• Scanner supply. Tau PET uses different radiotracers and interpretation than amyloid PET, and not every site that does amyloid PET also does tau PET. The sites that do are concentrated at academic medical centers, with the same geographic skew the anti-amyloid rollout has already produced.
• Reimbursement. Medicare coverage for diagnostic tau PET is narrower than for amyloid PET, and the pathway to broader coverage is likely to resemble the amyloid PET coverage arc - including the possibility of CED-style coverage-with-evidence-development as a transition.
• Workforce. Nuclear medicine physicians and radiologists trained in tau PET interpretation are in shorter supply than for amyloid PET. Tau PET reads are more nuanced than amyloid reads, and the error bars on interpretation are different.
• Patient selection. Tau PET findings have prognostic implications beyond simple positive/negative. How much tau, in which regions, matters - and payers have not yet settled on what counts as a confirmed eligible population.

What we are watching: tau-tracer sponsor filings that address broader indications than trial enrollment; Medicare NCD activity on tau PET as tau-directed therapy programs read out; payer coverage pilots at large commercial carriers; and early academic-to-community diffusion patterns that will indicate whether tau PET follows the same academic-center skew as amyloid PET did - or finds a different distribution through plasma-screening workflows that triage patients more efficiently before imaging.

The signal is that the access question about to define the next wave of Alzheimer's treatment is not pharmacologic. It is about whether the diagnostic infrastructure to identify tau-eligible patients expands fast enough to match the therapy that is coming. That is exactly the problem anti-amyloid access has been working through - and the solution for tau PET is likely to be more, not less, operationally demanding.