> does this basically mean that afrezza with the next gen inhaler

> can perform .4% worse than the insulin aspart and still receive

> non-inferiority from the fda?

First, the estimate on the difference in a1c reduction is estimated along with 95% confidence interval. e.g., difference = 0.1%, 95% CI = (-0.3%, 0.5%)

If the lower boundary of the 95% confidence interval is above the non-inferiority margin [delta] (-.4% in AFFINITY I study), non-inferiority can be claimed.

Hence, the difference should be reasonably close to zero or better for Afrezza in order to show non-inferiority. I don't think this is a high huddle to overcome considering better kinetics with Afrezza.

Oncostat

Takeda may be another potential partner. They have recently lost market exclusivity for Actos and revenues fell down quite a bit. They have experience in diabetes sales & marketing.

Oncostat

This was discussed early this month.

http://agoracom.com/ir/Mannkind/forums/discussion/topics/575873-trial-results/?message_id=1807484#message_1807484

In pharma/biotech industry, top-line tables and figures (for key efficacy and safety) are shared with the senior leadership team once the database is cleaned, treatment assignment (randomization code) is unblinded, and eventually locked for final analysis.

Once the database is locked for final analysis, it does not take long to produce top-line results since the preparation work (programming, for example) is done and validated prior to database lock. In my experience, top-line results are shared with the senior management within 5 business days from the dabase lock.

Since the type 1 study was completed at the end of May, assmuing 3 weeks for database clean-up, query and locking, I anticipate the Mannkind management would see the top-line results by the end of this week or next week.

If Mannkind participates in a investor conference in July, we may be able to see/feel the confidence level from the management team - since they would have seen the outcome from Type 1 study by then...

Best,

Oncostat

I am sure that they are busy with study close-up activities for two studies and partnership discussions... I think there will be plenty to share in next year's ADA meeting though.

Oncostat

Once the last visit of last subject is completed, case report forms (that sites fill out) will be collected. There will be data checks, queries, and cleaning activities to make sure there are no errors in the data. Once these activities are completed, database is locked (i.e., you are no longer allowed to change the database).

Once the database is locked, the treatment assignment would be revealed and the treatment effect and safety results can be summarized.

Oncostat

If this is the case, top-line results from the study in patients with type I diabetes will be ready within two weeks from the database (DB) lock at the latest. Usually, top-line tables are generated within 5 business days from the DB lock. Hence, the management would know the key outcome of the results by end of June or early July.

Oncostat