Tom Fagan: I am Tom Fagan from the Alzheimer Research Forum and I will be moderating today. John Trojanowski, would you like to start the ball rolling? John Trojanowski: Okay, Tom. Briefly, we will focus here on chemical analyte biomarkers in various bodily fluids such as cerebrospinal fluid (CSF), urine, blood, etc., and while genetics may come up, too, we will save imaging biomarkers for another time. Concerning apolipoprotein E (ApoE): It does seem to be critical to consider this for cohort assignment as demonstrated in the recent NEJM paper from R. Peterson [1] on mild cognitive impairment (MCI), and we will do this as well in the Alzheimer Disease Neuroimaging Initiative (ADNI) study. Anne Fagan: John, what is your gut feeling about the potential utility of urine showing decent biomarkers? John Trojanowski: So far, urine has seemed fruitful for studies of isoprostanes, although it depends on which one is measured and the method, but I am not certain if other analytes in urine are informative [2-4]. 1 Note: The transcript has been edited for clarity and accuracy. Anne Fagan: We are already seeing some differences in levels of Aβ42 in ApoE4-positive versus ApoE4negative folks in our antecedent study, but not other CSF markers, although the data are very preliminary. It suggests that perhaps Aβ42 is the first thing to change. We must verify with follow-up, of course. Tom Fagan: Anne, do you have indications that urine analysis might prove useful? Anne Fagan: Tom, we are not collecting urine. It may be too far down the line (i.e., processed) to be meaningful, but the idea is very clinically appealing. John Trojanowski: Les, do you have information on other urine analytes? Les Shaw: John, for AD, no. John Trojanowski: Anne, what do the receiver operating curves (ROCs) look like in the studies you mention, and perhaps the group would benefit from a definition of "antecedent," as this is not on everyone's radar screen? [ROCs can provide useful information on the sensitivity and specificity of clinical measurements.] Anne Fagan: True, John. We have not done formal ROC analysis yet. I use "antecedent" to mean biomark