MannKind Is a Buy, Despite Roller Coaster FDA News

May.11.11 | About: MannKind Corporation (MNKD)
Want a good case of high blood pressure, acid reflux and clinical depression all rolled into one? This has been the historical prescription offered by Mannkind (NASDAQ:MNKD) to its shareholders. Over the past 3 years this rollercoaster has delivered drastic price plunges 4 times (April, ’08, October, ’09, March, ’10 and January, ’11). The company, led by the serial entrepreneur, Al Mann, has been trying to obtain FDA market clearance for their lead product, AFREZZA, an ultra fast-acting insulin powder delivered to the lungs with an inhaler.
It seems that every time there is optimism anticipating good news (and share price appreciation) they get the opposite. The last dive off the cliff occurred on January 19th of this year when the Company received a second complete response letter (CRL) the day before from the FDA stating the reasons that AFREZZA could not again be cleared.
The major reason for the CRL was the Agency’s desire to see in vitro performance and pharmacology data to bridge the Company’s next generation inhaler, currently called “Dreamboat” with the inhaler used in the Phase III trial, the MedTone® (Model C). The Dreamboat inhaler is smaller, cheaper to produce and more efficient than the older device. A Dreamboat user inhales 1/3 less powder with the same amount of recombinant insulin so this has significant patient acceptance implications considering that 25 – 30% of clinical trial patients using the MedTone experienced a mild transient cough after inhaling. The Company is well justified in wanting to launch AFREZZA with this new device.
The problem, as I see it, has been the Company’s management of the different inhaler models vis a vis the NDA submission. It looks like an 11th hour insertion in an attempt to swing for the fence hoping to hit a grand slam but instead they’ve flied out to shallow center. Here’s the sequence of events per Mannkind’s most recent 10-K:
  1. NDA for AFREZZA submitted to FDA in March, ’09, based on Phase III data generated with the MedTone Model C inhaler. Prior to the conclusion of the clinical trial the Company developed a more robust less expensive commercial version of the Model C that they called Model D. In order to include the Model D in their NDA submission they ran a bioequivalence study (study 138) on 75 Type 1 patients comparing C and D.
  2. At approximately the same time as the NDA submission, the Company was developing their next generation inhaler called the Dreamboat. In August, ’09, the Company requested a meeting with the FDA to clarify how they could transition to the next generation inhaler. In November, ’09, the FDA sent a letter to Mannkind advising them about this transition. Their advice included generating bioequivalence data between the Model C (Phase III clinical trial) and Dreamboat.
  3. On March 12, 2010, the Company received a CRL from the FDA stating, among other items, that they would require comparability data for Models C and D. Management believed that the reason the CRL was issued was because of the method used by their contract laboratory to analyze blood samples for insulin and glucose. The FDA preferred a different analytical method but the samples could not be reanalyzed because too much time had elapsed and the samples degraded.
  4. In response to the CRL and the November, ’09, letter the Company ran an additional bioequivalence study (Study 142) comparing Dreamboat to the Model C in 66 healthy volunteers. They apparently did no follow-on studies with the Model D. The study demonstrated statistical bioequivalence despite the fact that the Dreamboat inhaler delivers 1/3 less powder for the same insulin dose as the Model C.
  5. In June, ’10, the FDA held an EOR meeting with Mannkind concerning the CRL issues. Among other items, the FDA asked about the comparability data between the Model C and Model D devices. Management apparently didn’t have this information but instead presented data from Study 142 comparing the Model C to Dreamboat. I’ll speculate that they were vigorously selling the virtues of the next generation device. After this meeting Management was apparently confident enough with how their presentation was received that they decided to submit the data from Study 142 to address the Agency’s inhaler-related questions. Their response was submitted in June.
  6. On January 18, 2011, they received their second CRL.
  7. On May 4, 2011, they had another EOR meeting with the FDA.
This sequence of events doesn’t reflect well on Mannkind’s regulatory staff or on Management. The delivery device was totally new and different and the amount of inhaled powder changed significantly but they only ran an equivalency study on 66 healthy people. These aren’t the people who would be using the Dreamboat. Thinking that this would be sufficient for approval was a serious miscalculation. I’m surprised that Mr. Mann would have authorized this. He clearly knows the regulatory process and the bureaucratic mentality of the FDA. The June meeting must have been a real love fest for them to have tried this.
After the earnings call on Monday afternoon, the share price gapped down Tuesday morning after company management failed to inspire anyone with news of a shorter timeline to approval subsequent to their May 4 EOR meeting with the FDA or a significant partnership deal.
Some key points of the call were:
  1. The 2 additional studies requested by the FDA in the CRL, the Affinity 1 trial for Type 1 diabetics and Affinity 2 for Type 2, remain the major prerequisites to approval. These studies will compare the bioequivalence between the Model C inhaler and the Dreamboat inhaler. The Affinity 1 parameters are unchanged, will include ~450 patients and will commence shortly after Management receives the written meeting minutes from the FDA in ~3 weeks. Affinity 2, however, has a new direction. The Agency informed Management that the basal bolus therapy data can be used for late stage Type 2 patients as well as Type 1. They apparently recommended that the Company refocus Affinity 2 on early stage Type 2 patients to include a possible preventative indication. Management was enthusiastic about this expansion of potential patients for AFREZZA. Affinity 2, though, will start later than Affinity 1. Management stated that they would communicate additional information about A2 after they receive the follow up letter from the FDA and they have clarified the study parameters and objectives.
  2. They forecast that they will conclude both Affinity 1 and 2 by late 2012. They base this projection on the feedback from the CRO that will be managing the studies. They believe that they can achieve this optimistic timeline despite the higher number of patients (may be as high as ~450 for each) required and the uncertainty of the specific parameters of Affinity 2.
  3. They wouldn’t speculate on an approval date but I imagine it would be late 2013 at best.
  4. Mr. Mann stated that the he wants to have another EOR meeting with the FDA before the end of 2011 to resolve all remaining issues for AFREZZA except the two clinical trials. These issues include more information about the performance characteristics, usage, handling, shipment and storage of the Dreamboat inhaler as well as updated safety information, user training guidelines and proposed product labeling. He indicated that a human factors study with Dreamboat was almost complete.
  5. The FDA had no questions about the clinical utility of AFREZZA.
  6. The Company currently has $47.5M in cash and a total source of funds of $144.9M. This will be sufficient to fund operations through Q1, 2012.
  7. Management is currently exploring additional funding mechanisms since spending is forecasted to accelerate in preparation for the AFREZZA launch and the eventual ramp up of the cancer immunotherapy business.
So what do we make of all this? Although the Company’s regulatory gamble didn’t pay off, I believe that AFREZZA will eventually be approved. The FDA is apparently comfortable with the ultra fast-acting insulin. The performance advantages are substantial. It reaches peak levels in only 12-14 minutes and the hypoglycemic episodes have been virtually eliminated. This is huge. No other insulin comes close to it. Throw in the allure of needle-free diabetes management and you have a blockbuster.
The FDA principally wants to see data comparing the inhalation devices. It should be a straightforward effort, in my opinion, to demonstrate bioequivalence. Ideally, they would be able to do this with a smaller patient set but I guess it’s not to be. The other areas that they identified in the January CRL are more of the routine variety, e.g., labeling, updated data, etc. Mr. Mann’s intent to resolve these by year-end seems entirely reasonable.
There is no reason to expect the ride to get any smoother between now and approval but the payoff will be there. I rarely go long with a stock priced below $15 but this is an exception. I’m in.



Disclosure: I am long MNKD.