Like with the format compared to the analysis of plasma p-tau, amyloid, GFAP and NFL, to predict Aβ and tau positivity, we separate our groups by cognitively unimpaired and cognitively impaired because we want to evaluate if this was disease stage-specific. And one thing that we were very interested in was the addition of these biomarkers to the demographics that we use, age and sex. Will it improve the model that we use to predict Aβ positivity and tau positivity? Because sometimes, the addition of the biomarker will not predict better Aβ positivity, for example, than if you only use the demographics...
Like with the format compared to the analysis of plasma p-tau, amyloid, GFAP and NFL, to predict Aβ and tau positivity, we separate our groups by cognitively unimpaired and cognitively impaired because we want to evaluate if this was disease stage-specific. And one thing that we were very interested in was the addition of these biomarkers to the demographics that we use, age and sex. Will it improve the model that we use to predict Aβ positivity and tau positivity? Because sometimes, the addition of the biomarker will not predict better Aβ positivity, for example, than if you only use the demographics. Like, don’t have a significant difference.
So, we think about how in clinic is this relevant. So okay. So if I add NfL, for example, on the model, it’s better than if I use only the demographics? So, we did this and we compared the models, and what we found was that, for the cognitively unimpaired, if we add p-tau231 to our models, we can predict Aβ and tau positivity with a significant improvement on the model when we compare if we only use age and sex.
And for the cognitively impaired, we have this improvement with the GFAP, which is very interesting that we have the same biomarker for amyloid and tau positivity in the cognitively unimpaired and the same biomarker in the cognitive impaired. So, we think that it’s the biomarker going with the disease, and it’s more disease stage, then we use a different biomarker.