What´s so persuasive is that it fits (at the very least qualitatively) what everybody else (except for a few non-believers, mainly on message-boards) is saying: That there is a significantly higher rates of AE.
Yes, the French study might not have the depth to satisfy a statistician all by itself. But higher then expected SAEs were seen everywhere (P1, PACE, EPIC) and this was acknowledged by pretty much everybody (FDA, EMA, Dr Berger, Dr Cortez, etc.).
You can run across one busy street without first looking for 500 times to prove that this is dangerous. Or you can run across 500 different streets once each. The end result statistically is the same.
So if you take the French study as part of the bigger overall picture, the data fits quite well and is not contradictory. Therefore I tend to believe it.
Whether the rate of AEs is really 3x or 5x or 7x or 10x, I don´t know and I guess nobody really knows at this point in time. Yes, probably the French study is worse compared to what other PACE test-sites found.
But, Ariad will have to convince the FDA to get anywhere. And we know by now that the FDA takes a very critical look. So, maybe the FDA´s perception is close to what the French study shows.
So in order to convince the FDA it would be nice if Ariad could show that SAEs can be reduced by 7x through various measures (exclude certain patients, reduce dosing, add preventive measures, etc.). If they can show that side effects implementing these measures are similiar to competitor´s drugs, response rates higher and more durable, then there is nothing in the way of getting 2nd line approval.
IF they cannot do this, 2nd line mght not materialize.
And this is why I take the study serious.
AI
Market Data 
Markets 
