The K-M curves might not be pretty, but Zytiga will be approved and it will be used in the pre-chemo setting. You’re kidding yourself, IMO, if you think it won’t adversely affect Provenge.
I am not impressed at all with Zytiga OS for the pre-chemo population.
Why not? These were early stage patients - unlikely to be dying directly of cancer early on (i.e. the deaths in the first 12 months, 18 months or ... ), so I wouldn't expect much benefit from any oncology treatment.
But the HR starting at 18 months is very good. The only question is does it become less good after 27 months or so (the end of the curves so far) - and only trial maturation will tell that story.
Or - put another way:
If you are at the zero point of the A302 trial you have two choices:
a) Start Zytiga now and get (guesstimated extrapolation of the curves) 9 months 'median' survival benefit.
b) Wait 18 months then start it at the same place as the earlier Zytiga trial started and get 4 months 'median' survival benefit.
(Note that I can make the same argument in HR space - but I am lazily telling it in median space because it is easier)
If you choose #a you take it for longer and well BEFORE you will see the benefit - but you get a bigger benefit down the road compared to delaying treatment. However there is a 20% chance you die before you can ever see the benefit of the treatment. Based upon the typical criteria of oncology trials (side effects are ignored) it is fairly likely that taking it earlier is better since it is fairly likely that taking it earlier gives a better total HR (I call it fairly likely only because it is possible the curves converge right after 27 months - but they certainly aren't doing so yet.)
PS I was having a similar discussion with Hirogen elsewhere and he took the same position as you did so he might chime in.
News 
Market Data 
Discover 
