The answer arrives before the advice does

Two to four years is a long time to hold a number nobody can act on.

That is the offer in the new Nature Medicine finding from Carlos Cruchaga's group at Washington University. A set of 34 circular RNAs in blood, measured across more than 1,200 people in independent cohorts, that diverge from normal about two to four years before symptoms appear and predict progression to symptomatic Alzheimer's more strongly than the pTau217 test that only reached family doctors this summer. The science keeps moving in one direction, and that direction is earlier.

The family's problem moves the other way, toward Tuesday. A caregiver handed a probability three years ahead of any symptom does not receive a treatment plan, because there isn't one. They receive a horizon. Then a decision about who else in the family gets told. Then a set of legal and financial conversations that were easier to postpone last week, and a parent who has to be asked whether they even want to know.

The honest caveat is that none of this is orderable. The researchers are working with commercial partners on an assay, which is the standard distance between a result and a product, and the gap is where families lose money to things that sound similar.

So the pattern holds, the same one the pTau217 rollout showed in June. Detection improves faster than anyone's ability to explain what a detection means. The test is not the hard part. The hard part is that prediction now outruns the infrastructure for making sense of it, and the person left holding an unexplained result is almost never the patient.