Biotech Values

[iwfal]

EXEL -

Once you have corresponsing pain location and scintograph lesions established at baseline, it isn't too much of a leap to assume cessation of narcotics combined with bone scan resolution equates to clinical benefit.

As I said in a PM I wanted time to think through implementations and what they would mean. Below is one such implementation:

1) Randomize patients with pain that maps to a particular spot that shows hot on the bone scan

2) XL184 patient gets pain relief for that spot and bone scan will almost certainly show 'colding' of the hot spot. Great, no issues.

3) Dox patient gets equivalent (perhaps better) pain relief for that spot - but bone scan does not show 'colding' and in fact shows a hotter hot spot (flare indicating healing?).

Would the FDA really let a poorly understood test like a bone scan turn an equivalent clinical result (pain relief) into a very significant benefit for XL184? I don't think they will - nor do I think they should.

And I think that conceptually almost anything that takes significant advantage of XL184's bone scan clearing phenomenon is likely to run into the same issue (with one exception - see below). OTOH if they pair it with another imaging technique that can detect dead tumor then, maybe..., but of course it is a debatable question as to whether XL184 is doing better in that regard.

The one exception would be to take advantage of the FDA's weakness of allowing existing standards to be used even if they are clearly asinine. In this case taking advantage of the fact that a hot spot is counted as progression in RECIST. In a simple prostate cancer trial isolated to patients with a high ratio of bone mets to soft tissue mets they could just run a pfs trial and I'd bet most progression would be bone scan progression so XL184 would win hands down. The FDA might bite - I don't think they should, but they might. No need to even go after a composite.