FDA Advisory Panel Endorses Inhaled-Insulin Product Afrezza
A US Food and Drug Administration (FDA) advisory panel has voted in favor of MannKind's Afrezza (Technosphere insulin-inhalation system) for the treatment of both type 1 and type 2 diabetes in adults.
Votes of the Endocrinologic and Metabolic Drugs Advisory Committee in favor of Afrezza were 13-1 for type 1 diabetes and 14-0 for type 2 diabetes. One panel member left early and did not vote.
Afrezza consists of a premeal insulin powder loaded into a cartridge for oral inhalation. The company is seeking an indication as an ultrarapid-acting insulin for adults with type 1 or 2 diabetes. In type 1 patients, the indication would be for use with injected basal insulin.
Panel members cited the advantages of more rapid onset of insulin action and shorter duration, resulting in a lower risk for hypoglycemia, as well as the potential for greater acceptance of inhaled-insulin therapy for patients with type 2 diabetes who might otherwise refuse or be unable to self-inject insulin.
However, the panel did express concern about data suggesting lower glucose-lowering efficacy compared with injected short-acting insulin (aspart) in type 1 patients and about lack of long-term lung-function data beyond 2 years, as well as a possible signal for lung cancer.
Panel members presented the agency with a long list of recommendations for postmarketing studies and labeling requirements to address those issues, among others.
Risks vs Benefits
Technosphere insulin begins working within 12 to 15 minutes, peaks by 30 minutes, and is essentially cleared by 180 minutes. Current rapid-acting insulins, in contrast, peak at about 1 hour and remain in the circulation up to 5 hours.
In a pivotal trial involving 353 patients with type 2 diabetes inadequately controlled on 1 or more oral agents, Afrezza was superior to placebo in lowering HbA1c at 24 weeks, with a week-24 treatment difference of -0.40 (P < .0001). However, in a pivotal study of 344 patients with type 1 diabetes, Afrezza had significantly inferior HbA1c reduction compared with premeal injections of insulin aspart, although the difference remained within the prespecified noninferiority margin.
Rates of severe hypoglycemia were 43% lower among the type 1 patients with premeal Afrezza compared with aspart injections.
Cough was the most common adverse event, occurring in 27% of all the study subjects and leading to discontinuation in 3%. Small decrements in forced expiratory volume in 1 second (FEV1) of about 40 to 60 mL were noted within the first 3 months and persisted for the 2 years of follow-up, although there wasn't further progression.
Because of the risk for bronchospasm in patients with underlying lung disease, including chronic obstructive pulmonary disease (COPD) and asthma, the product would be contraindicated in those groups, and the label would include recommendations for screening of patients for these conditions before starting them on Afrezza and possibly for periodic ongoing monitoring of lung function as well.
A total of 4 cases of lung cancer occurred in patients who had used Afrezza, 2 during the trial in smokers, and 2 others, both squamous-cell tumors, at 2.6 and 3.8 years after the end of the trial in nonsmokers. An oncologist on the panel, Liz Szabo, MD, from the National Cancer Institute, called that "unusual" and advised that more data be collected.
Pulmonologist James K. Stoller, MD, of the Cleveland Clinic Foundation, voted yes for both type 1 and type 2 because he was "extremely impressed by the unmet need and the advantages of this drug for subsets of individuals." Nonetheless, he said he remained concerned about the potential for the product's use in patients with unrecognized COPD and expressed skepticism about the extent of adherence to a label recommendation for spirometry.
"I think we sometimes take solace that the labeling indications will inform against risk, but I am highly concerned that would not be the case in clinical practice, regardless of what it says in the label." Because of that, he advised that the contraindication include ever-smokers as well as current smokers.
Acting committee chair Robert J. Smith, MD, an endocrinologist at Brown University, Providence, Rhode Island, raised the concern that MannKind had not done a study in which Afrezza was added to basal insulin in type 2 patients, whereas basal insulin is typically introduced first in clinical practice.
Nonetheless, he also voted yes for both type 1 and type 2 patients. "Perhaps the strongest motivation for me is that an inhaled form of insulin represents a drug that will serve some patients who are not effectively served by currently available forms of insulin. I am convinced enough by the efficacy data to believe that there are at least some subgroups of patients that will substantially benefit."
However, Dr. Smith added, "I think it's quite possible that there are other subgroups, and perhaps even a majority of type 1 diabetes patients, who would do better in terms of glucose control on injected rather than this form of insulin."
He also said he voted yes because "the data about potentially serious adverse events are not strong enough that I feel it's imperative to resolve those questions before the drug might be marketed, but I think there are a number of potential concerns about adverse effects that are very, very important to follow up postmarket."
FDA advisory panel members are vetted for conflicts of interest and are given temporary waivers for participation if necessary. No waivers were issued for this meeting.