Replies to post #5443 on Repros Therapeutics Inc (RPRX)

[iwfal]

IMHO, seams kind of OT here as few doubt the efficiacy aspect of Proellex

Yeah verily!!!!

[io_io]

<<<"Program Survival Bias for Dummies">>>


That is to completely miss the point.

Of course PSB exists, just as "weak gravitational forces" or light summer breezes exist.

However it has been trumpeted by the demagogues as the "cause" of various P3 failures. These people make the case that the specific biotechs should have known better not to trust their specific P2 results - and because of PSB. Inherently, they claim to bs "smarter" than the entire trial team.

Whereas the real truth is that the factors that go into P3 trial failures are vastly more broad.


Furthermore, as a Group, even if it was the industry that was running the P3 trials, there is NOTHING the industry can do about this slight PSB bias. Larger P3s are not the answer, if a stat-sig result on an inadequate HR or meaningless median survival benefit is worthless.

[io_io]

<<<"Clinical trials? Exactly the same.">>>


exwannabe, your example indeed generates well-known statistical bias inherent in random sampling.

but your extrapolation to Clinical Trials to support the PSB baloney has no basis in reality.


here are the analogies:


1. "McChich nuggets"

a) Model for: BULK MANUFACTURE

b) Product Samples Related ? YES - ALL MADE THE SAME WAY

c) Individually & Independently Tested ? NO, NOT POSSIBLE

d) Are characteristics known ? e.g. is failure an inevitable possibility due to make-up of product - YES, SOME FAILURES ARE INEVITABLE


2. RPRX (or DNDN) Trials

a) Model for: INDEPENDENT TRIAL OF DISCRETE ENTITY

b) Product Samples Related ? NO, ALL ARE DISCRETE

c) Individually & Independently Tested ? YES, BY DEFINITION

d) Are characteristics known ? (e.g. is failure an inevitable possibility due to make-up of product) - NO, AND FAILURE IS NOT INEVITABLE


"....on the biotech battle-field, you need some élan...."