Urche, I went to an FDA meeting on surrogate endpoints for oncology trials a while back, and one of the doctors there made an offhand remark that they weren't sure that PSA was a very good biomarker for prostate cancer.
Do you know anything further about this?
Basically, to be a good predictive biomarker, whatever you are testing for has to be associated with advancement of the disease, and reducing levels of the biomarker has to be associated with a better outcome for the disease.
I haven't followed prostate cancer treatments carefully, so can't speak to the biomarker issue.
However, the majority of men will develop at least benign prostate cancer if they live long enough, but often that has no affect on their quality of life, and trying to treat the disease surgically or otherwise can have more negative consequences than leaving it alone. The question is when do you try to do something? (And how do you know it helps?)
Market Data 
Markets 
AI
