I am sure you know of issues with Provenge/DNDN and the politicization of approval (sci advisory panel got death threats (no kidding) on their first pass on it. Angry men with funky prostates going nuts I guess!) 1. I don't believe in approval by referendum/politicization as it may not coincide with standards of scientific investigation that give us clear results...I only want proven drugs/biologics on the market UNLESS-and I agree with you here--: burden of proof should be modified SOC is dismal and prognosis is dismal AND and if the public/prescribers are willing to be part of what is essentially a very large clinical trial post-approval AND if they won't be pixxed off if it then gets pulled on lack of efficacy OR kills a few people along the way due to something unexpected.
I think that adaptive trial design will help get biologics approved faster and it helps to get conditional approval. The later has risks as we recently saw with Avastin.
So: FDA should have a track just like they have for orphan drug status. How about a track for deady diseases (such as CA) with only poor SOCs available?