Yes, but they aren't even trying.
If they are going to sell a reproducible product,
they ought to have some idea of what the important
parts of that product are. I'm not claiming it need be
perfect but certainly to support questionable stats it would help.
Consider this:
"it has become apparent from preclinical studies that the more important antigen-specific T-cell subsets to monitor may not be those directed to the antigen in the vaccine"
How can preclinical studies establish the [ clinical ] importance of somethings and if this is true, it is all DNDN has to show so far for biochemical results anyway. Add to that the possible antigen-independent mechanisms, blonde antigen response etc, and you have a lot of work to do and a lot of ways it would work even if they don't know the MOA. I don't see DNDN making any effor to find out.
No one ignores the validity of empirical clinical data, but then you have to have some decent criteria for discerning noise from reproducible results. If you don't know how it works then, even if a given trial result was due to a functioning vaccine and not statistical noise, there is no gurantee the next process will work the same way.
Certainly the gene expression or other array approaches would give them some idea what to look at but they aren't even doing this, AFAIK but maybe their CFO ( from AFFX) could be leading their science now. LOL.
Market Data 
Markets 
AI
