http://www.thestreet.mobi/story/12016463/1/biotech-stock-mailbag-mannkind-osiris-biocryst-vical-amarin.html?puc=yahoo&cm_ven=YAHOO The Afrezza data, as disclosed by MannKind last week, are: 1. Mediocre. 2. Incomplete. 3. "Positive" if one defines the term only as a drug that technically meets a study's primary endpoints but with data that are neither exciting nor groundbreaking. (See: Mediocre.) 4. Barely approvable, but only if 1) MannKind isn't hiding additional negative data, or 2) FDA doesn't give a crap and simply wants Al Mann to go away. (See: Incomplete.) 5. Insufficient to support MannKind's $1.7 billion market value. In the AFFINITY-1 study of Type 1 diabetics (also known as Study 171), Afrezza delivered via the second-generation "Dreamboat" inhaler was numerically inferior to Novo Nordisk's Novolog in terms of A1c reduction, the study's primary endpoint. (-0.21% vs. -0.40% A1c reduction.) The upper end of the confidence interval for the difference between Afrezza and Novolog was 0.36%, which didn't exceed the 0.40% difference that would have caused the study to fail. Therefore, MannKind can claim Afrezza Dreamboat is statistically non-inferior to Novolog -- hence a "positive" study -- but just barely. Afrezza Dreamboat performed worse than Novolog in terms of the percentage of Type 1 patients achieving A1c levels of less than 7% or 6.5% -- important markers of glucose control. Afrezza demonstrated a numerically lower severe hypoglycemic event rate than Novolog but the difference was not statistically significant. MannKind didn't disclose exact numbers. An important goal of the AFFINITY-1 study was to demonstrate equivalence between the new Dreamboat inhaler and the older Medtone inhaler. Here, MannKind is being very opaque about results, stating only that the two devices were "comparable" with respect to lung function, as measured by FEV1. MannKind did not disclose any efficacy results from the Medtone arm of the study. The company clearly has the data, so why is it being withheld? Both Afrezza inhalers cause decreased lung function in patients. MannKind says the lower FEV1 is not clinically significant, but of course, this will be a safety issue only settled by FDA and/or an advisory panel. Some 30% of the patients using the Afrezza Dreamboat inhaler reported cough as an adverse event. Switching to the AFFINITY-2 study (Study 175) in type 2 diabetics, Afrezza's A1c reduction of 0.82% was superior to placebo's A1c reduction of 0.42%. (All patients were also treated with oral diabetes medications.) The between-group difference in A1c of 0.4%, while statistically significant, was lower than the 0.5 to 1.0% difference MannKind hoped to achieve in this study. Again, MannKind has every right to declare the study positive because it met the primary endpoint, but Afrezza's performance still underwhelmed. Afrezza also caused more weight gain among type 2 diabetics and the difference in fasting blood glucose between Afrezza and placebo was not statistically significant. All in, Afrezza is mediocre insulin, with lots of data, particularly around lung function, still undisclosed. I give MannKind 50-50 odds, at best, for an FDA approval this time around. (FDA has already rejected Afrezza twice.) Even if approved, Afrezza is far from the blockbuster product MannKind supporters believe it is, making MannKind a very over-valued stock today.