Skip links

MannKind Breathes Easier – Inhaled Insulin Finally Approved

MannKind’s Afrezza Receives FDA Approval

In June of this year, MannKind Corporation announced that they received FDA approval for Afrezza®, their rapid-acting inhaled insulin product. MannKind is currently working to identify a pharma partner to manufacture and distribute Afrezza, and the product could be available as soon as January 2015. This approval has been a long time coming, following years of testing and two prior applications that were rejected due to concerns about safety and effectiveness.

The main issue that delayed the approval of Afrezza is the lack of demonstration of bioequivalence of insulin delivery from various delivery devices used in the clinical program compared to the proposed commercial delivery device. The demonstration of bioequivalence of an insulin product can be difficult even under the best of circumstances; in the case of Afrezza, the complication of the use of various drug delivery devices was significant. There are some lessons that can be learned from the Afrezza regulatory development pathway, which could be useful for other drug companies as they prepare bioequivalence protocols and assemble data packages for FDA review.

Mechanism of action and drug development

Afrezza is a rapid-acting powder insulin product, which uses a novel and proprietary drug delivery system to administer the drug to improve glycemic control in adult patients with diabetes mellitus. It is aimed at post-meal blood sugar spikes and is intended to be used with an existing long-acting insulin treatment, not as a standalone. The powdered insulin is delivered via a small inhaler called the DreamBoat. The powder dissolves immediately once it is inhaled into the lungs, and then is quickly released into the bloodstream. Afrezza peaks at about 15 minutes and is out of systemic circulation in about an hour; other current short-term insulin products take longer to kick in and peak at 2-3 hours, sometimes staying in the system for as long as 5 hours. Some obvious advantages to delivering insulin with this device are that the inhaler uses no needles, and the insulin can be stored at room temperature. The inhaler is disposed of after 2 weeks to prevent any powder buildup that could clog the device.

During the development of Afrezza, MannKind switched to the DreamBoat inhaler from the initial design (the MedTone inhaler). They cited advantages such as a significantly smaller size as well as smaller dose delivery capabilities, cutting both the cost of the device as well as respiratory side effects.  Ultimately, this switch resulted in significant delays in the drug approval process. Inhalers come with common side effects, such as cough, and can produce a measurable reduction in lung function, albeit reversible and clinically insignificant. Overall, the inhaler was redesigned three times over the course of the clinical development program, and the FDA demanded a battery of additional clinical studies proving that the final device was safe and effective.

Regulatory history

In March of 2009, MannKind announced that it had submitted an NDA to the FDA to request approval for Afrezza (initially called Afresa, until the agency requested a change to avoid confusion with another drug). The initial NDA submission was based on an extensive clinical program, with 44 completed studies and 5 ongoing trials at the time of submission; over 2400 patients with diabetes took part in Phase 2 and Phase 3 clinical trials, and more than 450 patients were involved in single-dose studies, meeting FDA guidelines for extent of exposure. One year later, MannKind was issued a complete response letter from the agency, indicating that the review of the application was complete and questions remained, precluding approval of the NDA in its current form. According to MannKind, this letter raised a number of points, including concern about the comparability of the DreamBoat inhaler to the earlier version of the device, which was employed in several pivotal clinical trials. In a subsequent briefing document, the FDA mentioned that inspection of the study site employed for the initial trials had revealed “multiple deficiencies affecting reliability of the data”, and concluded that this data could not be used to support the new inhaler. Other insufficiencies highlighted by the Agency were related to efficacy, pulmonary safety, and inhaler device-related issues.

The Afrezza NDA was resubmitted as a Class 2 resubmission in July of 2010. In this application, MannKind provided data from a recently completed efficacy study in patients with type 1 diabetes, as well as an updated pool of safety data and information on the comparability of the new inhalation device to that used in the pivotal studies. In a response dated January 2011, the FDA reiterated their initial apprehension, questioning MannKind’s claim of bioequivalence and raising concerns about the in vitro performance and clinical pharmacology data used to bridge the next generation inhaler to that acquired with the original design. Specifically, the response stated that due to the changes in the device and differences in exposure found between the two dosing regimens, a single BE study would not be sufficient to bridge efficacy and safety between the devices. The agency requested that MannKind conduct 2 clinical trials (in both types of diabetes mellitus) with the next generation inhaler, with at least one trial including a treatment group using the MedTone inhaler to obtain complete comparative data on the two devices. Additionally, the FDA requested information characterizing the performance, usage, and storage of the device.

New pivotal trials lead to approval

MannKind finally determined that conducting new trials with the next generation device would be necessary to compare prior results with the original device, and gain FDA approval. In August of 2011, MannKind announced the design of a pivotal open label study in which subjects would be randomized into one of three arms: a control arm utilizing rapid-acting injectable insulin, or one of two Afrezza arms for either the DreamBoat or MedTone devices. Positive preliminary results were reported in August of 2013, with studies showing both noninferiority of Afrezza to injected insulin, and changes in pulmonary function that were no different between the two inhaler designs. These findings were crucial in establishing bioequivalence and facilitating the bridging of the new device to the pulmonary safety data collected in the initial clinical trials with the MedTone inhaler. The Afrezza NDA was resubmitted as a Class 2 resubmission in October of 2013, and the FDA’s Endocrinologic and Metabolic Drugs Advisory Committee convened and subsequently recommended the drug for approval in April of 2014. Official FDA approval of Afrezza came in June 2014. According to Jean-Marc Guettier, director of the Division of Metabolism and Endocrinology Products, “Today’s approval broadens the options available for delivering mealtime insulin in the overall management of patients with diabetes who require it to control blood sugar levels.”

A risky venture?

All told, the cost of Afrezza development has reportedly reached $1.8 billion dollars, most of which was used to pay for over 60 clinical trials involving ~6500 patients. Meanwhile, the future success of Afrezza in the market is still very much up in the air: while inhaled insulin has clear advantages, there is still lingering uncertainty as to a potential link between inhaled insulin and lung cancer. This uncertainty has resulted in label warnings on risks and contraindications in the approved Afrezza prescribing information. To add further speculation on the potential market success of Afrezza, Pfizer’s Exubera®, the first inhaled insulin to receive FDA approval in 2006, was later pulled from the market due to unpopularity with patients, who found the delivery device to be large and awkward.  Thus, all eyes will be on MannKind in the coming months to see how the Afrezza story plays out.

This post was prepared by Jillian Orans, Ph.D., a Camargo Research Scientist.