Replies to post #1025 on IDM Pharma, Inc. (IDMI)

[exwannabe]

OK, I am bashing w/o DD :-)

Quotes per FDA :

http://www.fda.gov/ohrms/dockets/ac/07/briefing/2007-4301b1-02-FDA-redacted.pdf

The experimental arm issue:

"INT 0133 was a multi-center randomized open label study. Two therapy-related experimental questions were investigated in this study.

1. A comparison of two chemotherapeutic regimens: standard arm (Regimen A: high-dose methotrexate, cisplatin, and doxorubicin) and an experimental arm (Regimen B: a modification of the standard arm with the addition of ifosfamide).

2. The contribution of MTP-PE given in combination with the chemotherapeutic regimens - Regimen A with or without MTP-PE and Regimen B with or without MTP-PE."

"There are three main issues regarding the DFS results:

1) The Applicant’s result is sensitive to changes based on FDA review of CRFs, patient’s eligibility and investigator’s additional follow-up data;

2) The Applicant’s result is driven by an experimental arm which performs worse than the control arm; and ..."

"Applicant’s result is driven by an experimental regimen performing worse than the control regimen"

[The experimental arm refers to the arm with ifosamide+SOC. Look at the KM curve on page 21. The Green and Blue lines show PFS data w/o ifosamide. They are IDENTICAL]

You: "Frequently trials do not go as planned and require revised statistical procedures and often more data gathering. IDMI had to do the latter"

Complete BS. So you are a stat guy? Then you should understand you can't just go changing the plan and gathering data until you hit stat sig. See for example DNDN where the FDA did not allow them to correct a simple clerical error in the data set. And here you want to just keep adding data to the post hoc (per FDA) OS analysis (after all agree the primamry endpoint failed).

I completely understand why there is no serious discussion on this board.