Replies to post #143408 on Biotech Values

[iwfal]

PS - Also note that if you found the survival curves for the Zytiga 302 trial disturbing because the curves didn't separate until 20% of the population had died then the ARRY OS data should be even more disturbing because, as their PR indicates in statisticalese, they probably have placebo patients doing better than treated for a substantial fraction of the population.

Note: It is amusing to me that our positions on Zytiga and ARRY are flipped - so, FWIW, my criteria are:

a) As long as the curves are never inverted then I do not have a problem with saying the treatment is good.

b) Even if they are substantially inverted I still think it should be approved if every effort was made to predict who benefits and who doesn't AND, in the event it isn't possible to predict who benefits and who is hurt, then there is a black box or other similar warning to patients that it is a roll of the dice as to whether they will be hurt or helped.

[hirogen]

Zytiga - Round 2 Clark? ;-)

To provide context for others the case you put forth is - if patient a) starts Z at time 0 as in 302 trial he has 20% chance of dying up to 18 months, thereafter the HR is about .4. Patient b) chooses to delay Z for 18 months. Still has the same 20% chance of dying for those 18 months at which point he starts Z. It's assumed his condition is similar to a patient starting in the 301 trial thus the HR he sees is .65. So your thesis is the sooner the better with Z.

My counterargument is that the survival benefit comparison isn't an entirely fair one because of the relatively short follow up and info fraction from 302. I think there are three scenarios that can explain the difference in HRs for patients a) and b).

1) earlier is better as you said. Perhaps this could explain the difference between median duration of treatment and start of survival curve separation and why median PFS, OS haven't been reached - Z was acting against the cancer but due to the relatively healthy population it's benefit is masked by background mortality.

2) at 18 months in 302 patients are not comparable to patients at 0 months in 301 as the latter were sicker. As I said on the other board, median time to first chemo use on the control arm was 16.8 months, while time to treatment discontinuation is about 8.2 months. So the typical patient felt well enough to delay start of chemo for almost 9 months.

3) the HR difference is an artifact of the short follow up in 302. Eventually Z's effect will dissipate as it did in 301 where the HR post interim (.65) to final (.75) analysis was roughly .95. In the 301 trial the curves start to converge around the median and then do so at about 30%.

As it relates to Provenge, cases 2) and 3) would indicate P before Z while relatively healthy with the theory one gets the added survival benefit. For case 1) it's more problematic for P because some fraction of patients will fall out of its label when they progress on Z. What isn't known is how sensitive the survival benefit is to starting treatment with Z - is a 6 week delay to take P going to offset the additional survival benefit? Is the best setting for Z even earlier in NM-CRPC?

I think the data indicates that there is space for P in the continuum of mCRPC. Whether DNDN is able to get that message across to docs and to size the business accordingly to the realities of the segment P occupies (under current label) is another matter.

[genisi]

JNJ submitted sNDA filings for Zytiga in chemotherapy-naive CRPC:

http://www.investor.jnj.com/releaseDetail.cfm?ReleaseID=683472&year=2012