I’m sure you’re right, hankmanhub, but I’m not able to see it yet.
I don’t know where I put my notes, but when I sat down and calculated what outcomes would be necessary to separate the treated and untreated populations in a successful trial, the blinded data presented at least some potential for failure.
I think I worked the data a handful of different ways — separately calculating a 90 percent treatment arm at one extreme and 67 percent treatment at the other; and looking at the ranges across time for OS and PFS in comparison with assumptions provided by Smith on Stocks for typical placebo patients.
I don’t recall which numbers I used to represent expectations for placebo patients, but I think they were worse than the outcomes for the filtered Norwegian patients. So that would indicate a still achievable but more narrow path to success.
I still have hope, but other indicators seem to hint that this trial didn’t knock it out of the park. (But hey, if I turn out to be worrying for no reason I’ll have some consolation in the shares I have left.)
Has anyone else done similar calculations with the blinded data? If so, do you recall which assumptions you used for placebo patients?
Market Data 
Markets 
AI
