We’ve been very interested in the level of accuracy that’s needed for blood-based biomarkers to be used in clinical care and in general, we find that to be able to use these blood-based biomarkers, you really need very good tests that have very high agreement with CSF and amyloid-PET. There are many blood tests available right now and most of them are not performing at a level where we would want to use them in clinical care and it’s really the test for p-tau217 that seem to have extremely high accuracy in terms of predicting amyloid-PET status...
We’ve been very interested in the level of accuracy that’s needed for blood-based biomarkers to be used in clinical care and in general, we find that to be able to use these blood-based biomarkers, you really need very good tests that have very high agreement with CSF and amyloid-PET. There are many blood tests available right now and most of them are not performing at a level where we would want to use them in clinical care and it’s really the test for p-tau217 that seem to have extremely high accuracy in terms of predicting amyloid-PET status. And of course, this has become much more important now that we have a disease-modifying treatment that is becoming available for clinical care and so we are very interested in considering whether we can use these blood-based biomarkers in clinical care.
There is one test in particular that evaluates not just the concentration of p-tau217, but what we call the ratio of 217, the phosphorylated protein to the non-phosphorylated protein, and we are finding very high accuracies of this particular measure in predicting amyloid-PET status, both at the Knight ADRC and in additional cohorts, including with the Swedish BioFINDER cohort. So, we’re finding accuracies that are indistinguishable from what we see in terms of CSF prediction of amyloid-PET status. So, this is great.
Really, as a dementia specialist, the value of blood-based biomarkers is that we could potentially diagnose people much more rapidly. One issue with CSF biomarkers and amyloid-PET is there’s often capacity constraints, so they’re more expensive, it takes more time. It’s just a more laborious process and our capacity to do those kinds of tests is much less than just sending someone to the lab to have blood drawn. In the past, when it didn’t make much of a difference in some ways, at least with respect to treatments, whether someone was amyloid positive or amyloid negative, most people didn’t do biomarker testing because whether the test was positive or negative, you pretty much prescribe similar treatments and had a similar pathway. But now we’re going to have a really different pathway depending on the results of these tests and so that makes them much more important to perform and that means that we need to figure out how to do this at a much higher scale, like 10 to 100 times the amount of biomarker testing that we’ve done before and while the CSF and amyloid-PET tests are really good, I just don’t think they can reach that kind of capacity. So that means we really need to be able to start using these blood-based biomarker tests and I think they’re really approaching the level where we need them to be.
So in terms of the results that I showed at this meeting, we were just evaluating the p-tau217 ratio and how that corresponded to amyloid-PET and then also how the FDA-approved CSF tests, which are the Elecsys p-tau181/Aβ42 measure and then the Lumipulse Aβ1–42/Aβ1–40 measure, how that corresponded to amyloid-PET and so we found that this p-tau217 ratio was indistinguishable from the CSF test. We also looked at the concentration of p-tau217, I didn’t report on that. That also performed well, but we did not do a wide analysis of many different biomarkers. When other people have done that, we have consistently seen that p-tau217 is the winner. So I think the tests that we’re going to be using clinically will almost certainly be p-tau217 assays.
