I'm live-tweeting the afternoon discussion of the FDA's advisory committee meeting on $GRAL's blood test to detect multiple types of cancer.

This is a very interesting screening test. It definitely identifies cancers. But there are a good number of people who do not have cancer who will get a positive result, and a lot of patients who will get a negative result who do have cancer.

One really big question is whether it's really right to say this test identifies cancer "early." And even tougher one is how to think about a test that might make patients skip existing screening tests -- that could mean, for instance, that more breast or colorectal cancers might be missed.

The public hearing was very interesting. Some speakers, like Phil Castle from NCI, worried about the false negatives. There were also patients who felt they had been saved by the results of the test.

I'll be threading my notes from the afternoon here. Forgive me for typos -- these things move fast.