Biotech Values

[Summer2762]

$ARQL

[Rearranged the post after I answered your questions in the order you asked. However, I think the last question is where the misunderstanding lies. So I moved it to the top.]

If you meant the latter, why wouldn't you expect patients with no MET to fare worse when MET high consistently do well on OS endpoint and MET low fared much poorer. It would seem linear to me that continuing down from high to low then to no MET would result in even worse OS results.

I meant the first. I think met low covers no met as well.

Here is what the co said:

Of the 1048 patients in the trial, 445 were evaluable for MET status. Patients with MET-high NSCLC numbered 211, and the number with MET-low disease was 234.

--

The trial was stopped for futility on OS. That MET low data you cite on OS is not exactly inspiring, particularly with p=0.53. OS data in MET high was much, much better and has been in multiple trials (including stat sig in P2 HCC). So, don't really understand how you reach your conclusion here.

Assuming that we agree that Tivantinib is active even for met low patients (see, PFS and ORR data), let's just focus on the final OS data presented last September.

If we drop the 600 patients for which we do not know the met status, we have

Met high (211 patients): HR = 0.70, p = 0.03
Met low (234 patients): HR=0.9, p = 0.53

If we had two separate trials (One for met high and the other for met low patients), neither trial would have been stopped for futility at interim. (I know no futility analysis boundary that would exclude p=0.53 (z=0.07) at interim).

Thus, I'll say if we had a trial with just these two sets of patients (a trial that combines both sets of patients), with a very high probability, would not have been stopped for futility (I can image really extreme cases where that might be possible, that's why I used the term "high probability").

So my conclusion was that the patients who were not evaluable for MET status, did worse than either group and that's why we have HR=0.98 and p=0.53 and the trial got stopped at interim.

Do you mean in patients in which the MET status was not known (could be MET high or MET low patients) or when there was no MET at all?

Patients evaluable for MET status.