VALENCIA, Calif. ( TheStreet) -- MannKind ( MNKD) founder and CEO Al Mann has assured investors that positive results from two phase III studies of his inhaled, fast-acting insulin Afrezza due in August will lead quickly to an FDA approval filing. But no one at MannKind, including Mann, talks at all about a third, ongoing phase III lung safety study of Afrezza. Strange. This 510-patient study, currently open and enrolling, is titled "Evaluate Safety of Technosphere Insulin Inhalation Powder on Diabetic Subjects With Mild Obstructive Pulmonary Disease." Technosphere Insulin Inhalation Powder is Afrezza. According to the ClinicalTrials.gov registry, MannKind last updated and verified this large Afrezza lung safety study in February with scheduled completion in 2015. Remember, MannKind is promising to re-file Afrezza with the FDA before the end of the year on the basis of two efficacy studies known as Affinity-1 and Affinity-2. But if FDA doesn't also want to review data from this separate and equally large lung-safety study set to complete in 2015, why is MannKind spending the money to run it? One possibility, of course, is that FDA wants more lung safety data on Afrezza. That's not surprising, given the poor tolerability and depressed lung function attributable to Afrezza in a previous, large safety study conducted by MannKind and reported first in 2009. Results from this safety study were included in MannKind's original approval application for Afrezza rejected twice by FDA already. The two-year study enrolled 2053 patients with Type 1 and Type 2 diabetes and compared changes in lung function between Afrezza or usual diabetic therapy (Usual Care.) MannKind announced results of this Afrezza lung safety study in 2008 and data were presented at a European diabetes conference in 2009. The study was also published in the journal Diabetes, Obesity and Metabolism in February 2012. MannKind has long claimed the safety study showed no difference in adverse events or lung function between Afrezza-treated patients and those on usual care. The actual data from the study tell a different story. The drop-out rate for Afrezza patients was 49.4% compared to 30.4% in the usual car arm. Cough was the leading cause of discontinuation for Afrezza patients and was reported as an adverse event by 28% of Afrezza patients overall. Only 4% of patients treated with usual care reported cough as an adverse event.
A 19% difference in the discontinuation rate for Afrezza compared to usual care is not the sign of a therapy that is well liked or tolerated by diabetic patients. Adverse events related to Afrezza caused the sharply higher discontinuation rate, according to the version of the study published in Diabetes, Obesity and Metabolism. The study authors explain Afrezza patients might have abandoned the study in larger numbers because they were forced to use the inhaled insulin instead of injectable insulin, even if it was making them cough. This is hardly a ringing endorsement for the tolerability of Afrezza. Three outside authors on the paper received consulting fees or other payments from MannKind. Four additional authors on the study paper were MannKind employees. Afrezza caused depressed lung function over two years compared to usual care. Despite the finding, the study authors (all with financial ties to MannKind) conclude, "We do not consider the observed small, non-progressive lung function changes to be clinically significant." Of course, FDA is the only arbiter of safety that matters, which is why the ongoing lung-safety study MannKind is conducting but never discusses is so interesting. MannKind CFO Matt Pfeffer did not respond to an email seeking comment.
The chart above is reprinted from the lung-safety study results published in Diabetes, Obesity & Metabolism. It shows a statistically significant, 34 ml relative decrease in forced vital capacity for Afrezza compared to usual care after 24 months. The same chart shows a 37 ml drop in FEV1 (forced expiratory volume at 1 second) for Afrezza relative to usual care, again statistically significant. The study also showed 5.75% of Afrezza-treated patients with a 15% or greater loss in lung function (as measured by FEV1) at 24 months compared to baseline versus to 3.28% of Usual Care patients. Statistically significant reductions in lung function attributable to a chronic, lifetime drug like Afrezza raise serious and troubling safety questions. MannKind may be able to assuage these lung-safety concerns when it reports data from the ongoing study in 2015. Until then, it's hard to imagine FDA approving Afrezza. -- Reported by Adam Feuerstein in Boston. Follow Adam Feuerstein on Twitter.