Yes, you've got it. That's the point I'm trying to make.

009 A1C reductions weren't good enough for approval and the first CRL was issued.

117 data was used as a response to the first CRL and A1C reductions shown in 117 were good enough for approval.

But we just received word that A1C reductions from the most recent trial (171) were similar to the reductions seen in 009.

If the above is true, this puts the company in an uncomfortable position.

"171 met the non-inferior threshold."

Absolutely true. But 009 met the non-inferior threshold as well and the FDA still had questions.

http://www.news.mannkindcorp.com/phoenix.zhtml?c=147953&p=irol-newsArticle&ID=1197424&highlight

In May 2010 at the Bank of America Merrill Lynch Healthcare Conference, Al Mann stated that “the 117 trial is our response to the agency relative to the A1C issue.”

I agree that the FDA was about to approve this the second time around (before "the letter") but wouldn't the FDA have been looking at the more favorable 117 data at that point? If this is the case, will 117 data plus non-inferior 171 data be good enough this time around?

Re: 175: I don't believe weight gain is a big issue and neither are hypos. This is because the comparator arm was not receiving insulin at all (orals only + Technopowder without insulin loaded on). Weight gain and hypos are an unfortunate side effect of insulin so this was probably expected.

The only concern I have is regarding the 009, 117, 171 data you mentioned.

009 A1C results weren't too hot (or at least the FDA thought something along those lines which had something to do with the first CRL). These issues were addressed in 117 which showed much better A1C reductions. Now A1C data from 171 comes in and it's about the same as it was in 009...

I suppose it's possible that even if the FDA has further questions about 009,117,171 they could still approve for T2. Anyone have any thoughts around this scenario? Would this be unprecedented?

Corrections welcomed as this story gets more complicated by the day and I'm always looking for someone to poke holes in my reasoning.