MannKind Corp.

Sanofi has licensing agreement for global commercialization.

Hayes Viewpoint

Afrezza is a rapid-acting insulin powder delivered via a proprietary breath-activated inhaler. The drug/device was approved by the FDA on June 27, 2014, after years of regulatory setbacks, including 2 previous FDA rejections and second-round pivotal trials. Afrezza was launched commercially in February 2015; uptake has been slow, and 2015 sales forecasts are conflicting. Afrezza is currently sold as single-use 4-unit and 8-unit cartridges. In April 2015, the FDA approved an Afrezza 12-unit cartridge, which will be commercially launched in the second half of 2015.

Afrezza is FDA approved to improve glycemic control in adult patients with type 1 or type 2 diabetes mellitus (DM) requiring mealtime insulin. Afrezza must be used with a long-acting insulin in patients with type 1 DM, and it should not be used in patients who smoke. Afrezza is contraindicated in patients with chronic lung disease such as asthma or chronic obstructive pulmonary disease, and it should not be prescribed for patients with active lung cancer or to treat patients for diabetic ketoacidosis. Pulmonary function should be assessed before initiating Afrezza therapy, after 6 months of therapy, and then annually (see link below to product label). The FDA approved Afrezza with a Risk Evaluation and Mitigation Strategy (REMS). The REMS addresses potential risk of acute bronchospasm in patients with chronic lung disease (see link below to Afrezza REMS).

FDA approval came 3 months after an advisory panel voted 13-1 and 14-0 to recommend Afrezza approval for type 1 and type 2 DM, respectively. Although some panel members expressed concerns about potential safety risks, there was an overall consensus that the drug was effective, offered an alternative to insulin injections, and that safety concerns could be addressed in the product’s labeling. Of note, FDA reviewers were more critical than advisory advisors of Afrezza NDA data (see link below to FDA Advisory Panel briefing document).

Two previous Afrezza FDA filings did not result in a final regulatory decision. The FDA instead issued Complete Response Letters (CRLs) to the manufacturer. A 2011 CRL requested 2 additional phase III trials using a second-generation inhaler, which was not the device used in pivotal trials submitted for FDA review. The newer inhaler was much smaller than its predecessor, and used 30% less insulin powder. The Afrezza NDA that was ultimately approved by the FDA included new data from these additional trials: Affinity 1 (NCT01445951; study 171) and Affinity 2 (NCT01451398; study 175), as well as data submitted in previous filings (see links below to information on Affinity trials).

Available published evidence is insufficient to draw definitive conclusions on the efficacy and safety of Afrezza. Affinity 1 and Affinity 2 pivotal trials are not published to date. The only published phase III trial found that mealtime Afrezza plus daily subcutaneous (SC) long-term insulin was non-inferior to twice-daily SC biaspart (mixture of rapid-onset and intermediate-acting insulin) in patients with type 2 DM. This trial used the now obsolete MedTone inhaler, permitted concomitant use of oral antidiabetic agents, and had high attrition rates in both treatment groups.

The FDA approved an inhaled insulin product once before, in January 2006. That product, Exubera (Pfizer Inc.), was withdrawn voluntarily from market less than 2 years after launch due to slow sales. Although not cited in the manufacturer’s withdrawal notice, postmarket concerns about potential increased risk for lung cancer may have contributed to Exubera’s demise. The prospect of poor market performance led Eli Lilly & Co. and Novo Nordisk Inc. to discontinue their inhaled insulin development programs.

Afrezza costs about $7.50 to $9.25 per day. However, cost does vary and the manufacturer has been issuing discount coupons.

Proposed Use 

An ultrarapid inhaled insulin to improve glycemic control in adult patients with type 1 or type 2 diabetes mellitus.


The best available published evidence is limited to 1 phase III randomized trial (NCT00309244) comparing Afrezza with biaspart insulin in insulin-experienced patients with type 2 DM (Rosenstock et al., 2010). Two phase II studies are also published: Rosenstock et al., 2008 and Tack et al., 2008. (See links below to all 3 trials.) All of these studies used the first-generation MedTone inhaler.


NCT00309244 Trial (Rosenstock et al., 2010):

This open-label multicenter trial randomized adult patients with poorly controlled type 2 DM to receive premeal Afrezza plus bedtime SC long-acting insulin (n=334) or twice-daily SC mixture of rapid-onset and intermediate-acting insulin (n=343). All patients were nonsmokers, and were receiving a standard insulin regimen of 2 or 3 SC per day. About half of the patients enrolled were also taking an oral antidiabetic drug. Patients were permitted to take metformin or thiazolidinedione oral antidiabetic agents throughout the study.


The trial’s primary efficacy endpoint of change in glycated hemoglobin (HbA1c) from baseline to week 52 did not statistically differ between the 2 treatment groups. The proportion of patients who achieved target HbA1c levels < 7% also did not significantly differ between the Afrezza plus long-acting SC insulin group and the premixed insulin group (22% and 27%, respectively). Patients who took Afrezza developed more cough and upper respiratory infections, deemed by investigators to be not clinically significant. Attrition was high in this trial, consisting of approximately 32% of patients in the Afrezza plus long-acting insulin group, and 25% in the premixed insulin group.


NCT00511602 Trial (Rosenstock et al., 2008):

This trial randomized insulin-naive patients with poorly controlled type 2 DM to receive either Afrezza (n=61) or an inhalable placebo powder (n=62) before each meal daily. All patients continued to take their usual oral antidiabetic drug regimen. Baseline HbA1c was 8.0% in the Afrezza group and 7.8% in the placebo group. By week 2 of treatment, mean HbA1c was decreased by 0.7% with Afrezza and by 0.3% with placebo. Between-group difference was statistically significant for this outcome but did not meet the study’s predetermined primary efficacy measure. No significant safety issues were reported.


NCT00511732 Trial (Tack et al., 2008):

This trial compared 4 different doses of Afrezza (n=181) with an inhalable placebo powder (n=46) in patients with type 2 DM. Afrezza alone was given 3 times daily before meals. Patients also received current care once-daily SC long-acting insulin. Compared with baseline measures, dose-dependent and significant reductions in mean HbA1c were observed at week 11 for all Afrezza dose groups, while a slight but nonsignificant increase in HbA1c was seen in the placebo group. Only HbA1c reductions seen in the lowest-dose Afrezza group met the primary efficacy endpoint.


In all the above published phase II or III trials, no significant pulmonary function issues were reported and body weight changes were similar in Afrezza and placebo groups.


FDA approved in June 2014.


Afrezza REMS

FDA Advisory Panel Briefing Document (April 2014)

FDA Label for Afrezza

Affinity 1 Trial

Affinity 2 Trial

Rosenstock et al. (2010) Study

Rosenstock et al. (2008) Study

Tack et al. (2008) Study

Last Updated 

May 2015


If you have a Hayes login, click here to view the full report on the Knowledge Center.