The other Leerink Swann thread was getting hard to follow thanks to the new Yahoo format, especially since Yahoo seems to be filtering out all the posts except mine. Flattering, I suppose, but not very helpful.
I listened again to the part of the cc where Cormack says that if SYNERGY fails he expects the other 011 trials, AFFINITY and ENSPIRIT, to continue. As mentioned previously, he says Teva's only contractual outs are for safety and patent issues. (As Summer noted, the collaboration agreement also gives Teva an out for futility; the contract, which can be found in redacted form online in SEC filings, doesn't seem to clearly define futility, but my sense is that failure of the SYNERGY trial would not amount to futility of a different trial.) Cormack then says that the AFFINITY trial is in a sense "cleaner" than SYNERGY because it's not subject to as many potential confounding effects. He explains that in SYNERGY, after patients get either docetaxel or docetaxel + 011, they can go on to get newer drugs like Zytiga or MDV-3100. If there's an imbalance between the treatment and control groups in the percentage of patients who get one of the newer drugs, that could skew the results. (It's not clear to me whether even if there's no imbalance the results could be skewed simply because of the introduction of another variable that affects survival.) But in AFFINITY, by the time patients get cabazitaxel they're usually at a stage where they won't get any other treatments. Therefore, he said, if there's confounding of the results in SYNERGY or the margin of difference in the outcome is small (meaning, I guess, that there's a trend towards efficacy but it's not statistically significant), that would be an extra reason to continue AFFINITY. Only if SYNERGY showed that 011 was "detrimental," he said, would the other trials be abandoned.
What is "detrimental"? Is it a greater incidence of serious adverse events? Or would a shorter median overall survival in the treatment group than the controls be evidence that 011 is "detrimental"? If so, then that might mean that as long as the treatment group in SYNERGY demonstrates a trend towards longer survival, even if it falls short of statistical significance, the other trials are likely to continue (assuming there are no major adverse event issues).
There is an interesting plus side with the confounding though. This is a little bit difficult to explain So bear with me.
Let's say OGX-011 adds 2months to MOS. This by itself will probably be not sufficient to show statistical significance.
However, the timing of the approval of the new drugs is such that 2 months may be sufficient for the new drugs to be used by the patients. These are proven to add 4m+ (or even much longer if used consecutively).
So in one sense the timing of SYNERGY is unfortunate but it may be an advantage as well.
Hi Summer, I'm not sure I'm following. Are you saying that if 011 adds 2 months to MOS that will probably mean more 011 than control group patients will have received Zytiga and/or MDV-3100 because more of them will have still been alive when those drugs were approved/became available? If that's the case, would the confounding effects of the imbalance be factored out somehow by the FDA or would it simply compare the MOS of the two groups without regard to the imbalance?
Also, when you say in your previous post that the timing is the problem and that if SYNERGY fails AFFINITY should go on and SYNERGY should be repeated later, why do you think SYNERGY would be less subject to confounding if repeated? Is it that once Zytiga and MDV-3100 start being commonly administered pre-chemo patients won't receive them again post-chemo?
BTW, Cormack in the first few minutes of the cc tried to downplay the probability of confounding by saying that 2/3 of the SYNERGY patients are in Europe where the new agents were approved later, and only 1/3 are in North America. He also said that with 1,000 patients if there is some confounding it probably won't affect the statistical outcome very much. Of course, he later acknowledged that confounding could be a problem when he said AFFINITY is a cleaner trial.