Hey Charles!
Thanks for the data link - here's the money quote - but keep reading below!!!!! :
Mutations Suggest New Prostate Cancer Treatment Options
"Almost 12% of men with metasta c prostate cancer (PCa) had germline DNA- repair muta ons, a nding that could help guide therapy selec on toward nonstand- ard agents for selected men, gene c sequencing data for 700 men suggested.
The frequency of inherited DNA-repair muta ons was almost three mes higher than rates observed in men with localized PCa."
This reference as you say includes BRCA and other DNA repair mutations, not just the dMMR mutations that lead to the FDA Keytruda approval.
Both your references (looks like the same data actually) are for inherited=germline mutations. But what REALLY matters anyway is the somatic rate of those mutations i.e. the rate in the cancer itself (preferably the rate in metastatic sites). This is REALLY IMPORTANT and if this point slid by you, I think it's probably slid by most other readers, because I've found your comments to be quite acute.
So, I'll say it again: Dave did not have all these mutations inherited. He was tested (blood or saliva) and was explicitly negative for some of them (I don't know for sure whether the earlier test was for all the ones the later test detected). His cancer developed these mutations over time. That's why I keep saying we don't have good estimates for the actual frequency for our advanced cancer guys.
Here's a small number study from 2014 that puts the dMMR estimate at (coincidently) also 12%:
/www.ncbi.nlm.nih.gov/pubmed/25255306 -- Carolyn