Conatus Pharmaceuticals Inc. (CNAT)

[Ponch73]

http://ir.conatuspharma.com/releasedetail.cfm?releaseid=969213

The above links to the phase 2 results released in May for Emricasan. In perusing the tables, a few curious things jump out at me:

1. The p-values for the overall patient population seem pretty high (bad).
2. The p-values for the MELD score > 15 population seem pretty low (very good), but the sample sizes appear to be unusually small (N=9) for drawing conclusions
3. The p-values for the NASH patient population seem low (good), but, again, the sample sizes seem unusually small (N=11) for drawing conclusions

Does anyone have any informed thoughts on this?

Secondarily, what kind of track record does Novartis have in developmental biopharma in successfully identifying compounds at the phase 2 stage that go on to receive FDA approval?