EBS,
1) I agree 1st look-in has little chance for stop of trial given that we have eveneted patients in the table and good results at that time are represented by NON-evented patients which are not in the tables (YET).
So only a COMPARISON with the placebo arm can be done (because both arms get patients randomized into it on a 1 by 1 and therefore have equal amount of patients (+/-1) during the complete trial). We would be completely dependant on a very populated placedo median table with a low MOS compared to an under populated Bavi median table with a much better MOS to have any chance on early stops. I don't think that is a realistic expectation and if indeed that is what would happen you are correct that would b unseen and huge.
2) I think the TWO launch scenario's have nothing to do with the early stop. IT IS A PERSONAL OPINION AND IT CAN BE WRONG!!!!! I think these scenarios are the original ones (not specifically made for the trial). They were made not knowing if a Fast Track would be granted or not. PPHM will in no case have anticipate anything else then the normal procedure elements in its estimation, GIVEN the DELICATE NATURE for PPHM to make over-optimistic projections. They have ALWAYS played things down and I think they keep doing so.
3) I think therefore the Launch JAN 2017 is the launch with Fast Track and that launching BEFORE that date would INDEED depend on a successful early stop (which would then rather be the 2nd then the 1st look-in).
And each date JAN 2017 and JAN 2018 will have had some margin included because also the End of Enrolment will have margins.
4) I understand your MARCH 1st argument based on 2nd look-in, illustrated with the Bayer case. But it has a flaw. Counting backwards is OK but not if one is not 100% sure that JAN 2017 is projected by PPHM based on Fast Track or not or on Early stop or not. And that is where we differ from opinion, I think PPHM based it on Fast Track or not. So you CANNOT use the JAN 2017 to subtract 9 months to end-up on the expected 2nd look-in. Furthermore PPHM has clearly stated that the look-ins are EVENT DRIVEN and that is why I think they would NEVER make a launch scenario based on "estimated" look-in date that depends on "estimated" eventing, that depends again on "estimated" end of enrolment dates that depend again on "estimated" centre opening success. That is no basis to base a launch scenario on. The the binary 'We get the Fast Track or we Don't get the Fast Track' however is.
5) They want to have Avid extension ready on MID 2015 (that is in 4-5 months!). For a launch in 2017 that is 22 months UPFRONT. The PROVED that they can extend Avid in 6 months time (time between announcement and extension ready). So they would have ALL THE TIME to extend after the 2nd look-in since with your Bayer case you show they have 9 mounts. Certainly if you have the preparative work done, as is the case. So EITHER they expect a real big increase in Avid 3rd party production, of such extend that the current facilities could not deliver it OR they envisage a possibility that they will need the capacity for bavituximab (COMMERCIAL PRODUCTION of Bavi was the other reason they gave for the extension with single-use technology). The only other option I can think of is that for the mentioned collaborations (24+) they need to ship such large amounts of Bavi to all kind of research partners that they need to produce them in the extended Avid facilities ASAP. I also remember someone posting that all Bavi needed for SUNRISE is already produced, so it is not that the current Avid would be able to produce more.
My conclusion is that we miss, again, a number of infos to nail it. I keep JAN 2017 in mind as an ULTIMATE launch date but I would not be surprised if PPHM starts producing Bavi earlier then that. When they do they know the OUTCOME because NO WAY they start loading the refrigeration with large quantities of Bavituximab if they have no short term destination for it. It would lock-up cash!
Market Data 
Markets 
AI
