New Blood Test Offers Early Warning for Alzheimer's Onset

Georgetown researcher (and executive dean of Georgetown's medical school) Howard Federoff has taken a "systems" approach to diagnostics for certain chronic diseases. By comparing blood samples taken from patients who subsequently developed Alzheimer's to blood samples after the disease has manifested, Federoff has identified markers and created a blood test that is described as "90 percent accurate" (the BBC article does not delve into the ratio of false positives to false negatives) in predicting whether a currently healthy patient is likely to develop Alzheimer's in the following three years. Understandably, this raises some ethical and practical questions. What would you do differently if this test came back positive for yourself? Or for a parent? Here's the (paywalled) paper, at Nature Medicine.

Comments

As a neurologist.

[MPAB]

The main symptom that brings people to the neurologist is forgetfulness. Most of the time it's subjective (ie. I know someone with Alzheimer's and I begin to notice and worry about the times I meet people and the names won't come to my head). We look for signs of cognitive impairment, with tests that include memory and other mind processes. Of course, YMMV depending on your previous performance, career, educational level, etc.
Once we get proof of MCI, we can make some tests because Alzheimer's isn't the only thing that can cause it. The usual stuff ranges from depression or unfelt strokes to syphillis. The CAT scan/MRI only tells us if the brain is intact, somewhat like trying to work out if a car works by just opening the hood.
Alzheimer's itself can only be diagnosed under the microscope right now. Not a thing we'd agree to to do a live brain.
Other than this blood test, there are radioactive tracer tests and CSF tests. In all of them the result is a chance or ratio telling the possibility of the MCI to be a sign of Alzheimer's against something else.
So, it's a disease for which there is no prevention nor a cure and the current tests just tell us "yes your worries about that time you left the keys on the toilet are related to a 75% propability of having Alzheimer's". We should get into positive and negative predictive values here.
As I tell my patients: "No: there is no sign of cognitive impairment right now. If I knew you were to develop a demence, I'd suggest you settle your pending issues right away, but I don't see a reason not to do that, anyway, You don't know what awaits you at the turn of the corner."