Biotech Values

[biomaven0]

All trial results are surrogates for how the drug will do in the "real world." Have to leave it to doctors to decide whether to use some of these marginal drugs with poor side-effect profiles (Sutent is even worse on that measure) on a particular patient.

OS is superficially appealing because it is a "hard" and undisputable endpoint. But even in a trial context it can get greatly muddled by post-trial treatments.

The paper goes on to suggest that the FDA force the trial population to be more "representative" - that's an untenable suggestion that would simply make drug development even slower and more costly than it is now.