When we use probability on something like this trial, it's a misuse of statistics to some extent. When a basketball player is said to have an 80% chance of making a free throw, it's based on many occurrences of the player doing an identical activity over and over.
This is a 10+ year trial of hundreds of unique factors, that were never assembled like this before, and never will be again. People can factor in all sorts of things, such as data from related studies or comments from those involved with the trial. Those things do add insight. I am amazed at the amount of research that people on this m.b. have done. Nonetheless, it still comes down to the trial itself and the data from the trial.
As for the spirit of your question, I have seen essentially 0% to 100% as being given as the odds of success. I think both of those can be tossed out w/o much comment. It's somewhere in between. And also there is the possibility that the trial fails on PFS but is given a path to approval on OS. Also the price reaction will depend on the data - barely making it on PFS will be different than a huge beat.
In my mind, I'm using 25% as the odds of success on PFS. I was concerned that LL thought they may have made a mistake using PFS as the primary measure - and statistically, most phase III trials fail. So I think the odds are less than 50-50, and I use 25% for convenience. But there are a number of positive things prevent me from being too pessimistic.
As for how the stock price will react on the outcome, stocks often run to round numbers, so a narrow margin could take the stock to $5 - medium results, $10 - through the roof numbers, $20. A failure, with no path for approval, .10 initially. Failure on PFS but with a path for approval, $1-$2. I can state with 100% certainty that I am guessing.
News 
Market Data 
Discover 
