Biotech Values

[tinkershaw]

Dew,

Looking at the 57% number, and I think this number will fall some more as I believe of the patients who achieved EVR something like 28% relapsed are otherwise left the trial by 12 week SVR. That is a prolific decline. In comparison telaprevir in Prove 1 had something like 12.8% or thereabout relapse and dropouts from EVR to 12 week SVR. Given this prolific fall off, I think the SVR 24 is going to go lower. 55%, 52%, I don't know, but we will see.

But given this, on the 20% increase basis, the control group is going to need to be extremely hard to treat and get a number perhaps as low as 32% SVR, and perhaps as high as 37%. However you cut it, that would be a very tough control group to treat.

Percentage wise boceprevir will need to show a 40-50% increase in SVR24 rate. That is what Prove 2 and Prove 1 respectively saw. From these numbers boceprevir would need a control group SVR percentage of 40 and 38 if SVR24 remains at 57 (which is quite realistic) to a control group of maybe as low as 37 and 34 is SVR24 falls to 52% for boceprevir.

Depending on the final SVR24 numbers boceprevir may be able to make such an argument if the control group is 37 or 38% SVR, but I do not see the number falling any further than this, so it is a bit of a long-shot for Schering. It will be much more difficult to get a control group that comes back low enough to show a straight 20 point advantage.

Further, when you look at it, RVR was very predictive. The pre-dosing group has an RVR of 62% and a SVR12 of 57%. I don't buy Schering's spin that RVR is not proven to be predictive of SVR where protease inhibitors are involved. Perhaps not proven, but in practice reality.

To make matters worse for Schering the previously treated but failed patients (however you want to categorize them), although Schering can clearly re-run these trials and obtain better results than they initially achieved, even those numbers appear to be 10 points or so less than the 35-45% numbers I'm anticipating out of Prove 3 (okay wild ass guess from 107, but I'm trying to extraprolate with greater adverse events due to the 107 exclusions of prior AEs, and the anticipated drop off from RVR to SVR). So across the board you are looking at boceprevir being an effective and competent treatment, but clearly inferior. The final SVR24 vs. control group is by all odds very likely to confirm this unless two things happen: the control groups were on the edge of the bell curve in regard to difficulty to treat and boceprevir, despite an ~ 28% fall off from EVR to SVR 12, that there is no further fall off from SVR 12 to SVR 24.

A long way to say, you have a point Dew, but the odds are not in any way in boceprevir's favor to produce the numbers needed to make the case.

Tinker