Inhaled insulin (Afrezza)
MannKind
What it is
An inhaled mealtime human insulin. In a 30-person type 1 diabetes clamp study, glucose lowering began at about 12 minutes and returned to baseline after 90–270 minutes, depending on dose. The US indication includes ages 6+, with required lung-function assessment and important respiratory and ketoacidosis precautions.
Editorial review: .
Most recent recorded citation date: 2026-05 (month only). Only explicit date metadata is included; an undated citation may be newer. This does not mean every claim was reviewed on that date.
Trial status, labels and access can change between reviews. How we review the evidence · How to read the evidence
Evidence behind this assessment
Key evidence notes. Study results, product eligibility and access answer different questions.
- Who was studied?
- Study populations and analysis groups vary. Product age limits alone do not describe who was studied.See the linked studies and their populations →
- Benefit or performance
- Consistency: The label reports within-person variability of about 16% for insulin exposure and 21% for peak concentration; glucose-lowering variability was about 28% for total effect and 27% for peak effect. Single-use cartridges contain 4, 8 or 12 units; prescribed doses can require multiple cartridges.[1]
- Important harms and treatment burden
- Read the safety discussion and original sources. A missing summary does not establish safety.
- Approval and country access
- US indication ages 6+. Cipla announced an adult India launch in December 2025. Brazil has an adult label; current local supply and reimbursement were not verified.Approval, trial recruitment, local supply and funding are separate. Check the cited label or access source.
- Follow-up and remaining uncertainty
- Read the full discussion and original sources for follow-up duration and study limitations.
This record cites company or conference reports. Those sources may describe interim findings; their source type is shown in the source list.
Editorial score: calculation and evidence
A weighted editorial judgment on a 0–100 scale, not a probability of success or a measured treatment effect. Higher criterion scores mean more favorable assessments.
Score limitations. Scores are retained editorial judgments, not validated clinical measurements or probabilities. Clamp timing does not establish comparative exercise safety, and the access score is not a country-specific coverage assessment.
Default calculation: 92 × 20 + 90 × 25 + 88 × 15 + 50 × 20 + 72 × 10 + 32 × 10 = 7450; divide by total weight 100. Unrounded weighted result: 74.5.
Mealtime convention (earlier onset is preferred): the US label reports first measurable glucose lowering at about 12 minutes and peak serum insulin at 10–20 minutes in a 30-person type 1 diabetes clamp study. These data do not establish superiority over every current injectable.[1]
Mealtime convention (earlier peak is preferred): mean peak glucose-lowering effect occurred at about 35, 45 and 55 minutes after the studied 4-, 12- and 48-unit doses. This is a dose-dependent clamp result, not a guaranteed meal response.[1]
Mealtime convention (shorter tail is preferred): mean glucose-lowering effect returned to baseline at about 90, 180 and 270 minutes after the studied 4-, 12- and 48-unit doses. A shorter measured duration alone does not prove fewer late hypoglycemic events.[1]
The label reports within-person variability of about 16% for insulin exposure and 21% for peak concentration; glucose-lowering variability was about 28% for total effect and 27% for peak effect. Single-use cartridges contain 4, 8 or 12 units; prescribed doses can require multiple cartridges.[1]
The dose-dependent 90–270-minute glucose-lowering duration may be relevant when planning activity, but these clamp data do not test exercise outcomes. The label identifies changes in physical activity as a hypoglycemia risk factor; this score does not establish an exercise safety advantage.[1]
US access depends on prescription coverage and programme eligibility. The manufacturer advertises commercial savings, a time-limited pediatric bridge programme and a cash-pay pathway; these offers do not establish a universal patient price or availability outside the US.[4]
Insulins are scored relative to their role peers (see tags: rapid, ultra-rapid, basal, inhaled). A basal insulin's onset score compares it to other basals, not to mealtime insulins.
Editor’s take
A useful alternative for people who value inhaled mealtime dosing and can meet the respiratory requirements. Adult trial averages support it as an option, but individual HbA1c responses varied. The pediatric approval and the failed pediatric noninferiority endpoint both matter when discussing it with a clinician.
The full picture
Afrezza is a dry-powder formulation of human insulin delivered through its own inhaler. It is a mealtime insulin, administered at the start of a meal. In type 1 diabetes it must be used with basal insulin; it does not replace basal insulin and is not a treatment for diabetic ketoacidosis (DKA). The US label covers adults and children aged 6 and older with diabetes.1
Timing depends on dose. The US label describes a clamp study in 30 people with type 1 diabetes. Mean first measurable glucose lowering was about 12 minutes after the studied 4-, 12- and 48-unit doses; mean peak effect was about 35, 45 and 55 minutes, and return to baseline was about 90, 180 and 270 minutes, respectively. That is a glucose-lowering duration of approximately 1.5–4.5 hours, not a fixed duration for everyone. Serum insulin peaked at 10–20 minutes; serum concentration and glucose-lowering effect are different measurements.1
Dosing and variability. Cartridges contain 4, 8 or 12 units, and a prescribed dose can require more than one cartridge. Dose conversion from injected insulin requires the labeled instructions and clinical supervision. The label reports within-person variability of about 16% for total insulin exposure and 21% for peak concentration, compared with about 28% for total glucose-lowering effect and 27% for peak effect. The inhaler is replaced after 15 days of use.1
Exercise remains a separate question. The clamp timing may help explain interest in a shorter mealtime insulin effect, but it does not establish fewer exercise-related lows. The US label lists changes in physical activity among factors that can increase hypoglycemia risk. A person’s activity plan still needs to account for their insulin regimen and glucose response.1
Adult efficacy. INHALE-3 randomized 123 adults with type 1 diabetes to inhaled insulin plus degludec (62) or usual care (61) for 17 weeks. Nearly half had used automated insulin delivery (AID) before enrollment; the intervention group switched to inhaled insulin plus injected basal insulin. The adjusted HbA1c difference was 0.11 percentage points (95% CI −0.10 to 0.33), meeting the prespecified 0.4-point noninferiority margin. Among participants with observed 17-week HbA1c values, a reduction greater than 0.5 points occurred in 12/57 (21%) versus 3/58 (5%), while an increase greater than 0.5 points occurred in 15/57 (26%) versus 2/58 (3%). The worsening comparison was added post hoc; these results do not establish which individual will benefit.2
Adult trial harms. In INHALE-3, cough occurred in 14/62 (23%) inhaled-insulin participants, and eight stopped inhaled insulin because of side effects. There was one severe hypoglycemic event in the inhaled-insulin group and none in usual care; the investigators related that event to excessive alcohol ingestion many hours after the last inhaled dose. No DKA occurred during this short trial. MannKind funded the trial and participated in protocol development and oversight. These results do not replace the broader safety warnings in the label.2
Meal-challenge evidence and AID. A companion randomized study compared one standardized meal in 122 adults, 61 per group. Inhaled insulin reduced the time-normalized 2-hour glucose area above 180 mg/dL by an adjusted 12 mg/dL (95% CI 2 to 22 lower) compared with injected rapid-acting analog insulin. Because the area was divided by time, this is reported in mg/dL, rather than raw glucose-by-time units. The study included pump and AID users following specific protocol instructions to avoid insulin accumulation; it does not establish routine long-term safety or effectiveness of adding inhaled insulin to every AID system. MannKind funded the study.3
Pediatric approval and evidence are distinct. The May 2026 US label permits use from age 6, but its 26-week pediatric study did not establish HbA1c noninferiority. Of 230 randomized participants, 117 received Afrezza and 113 an injected rapid-acting analog, both with basal insulin; 97.8% had type 1 diabetes. Ages were 6–17 in the Afrezza group and 4–17 in the comparator group. The adjusted HbA1c difference was 0.18 percentage points (95% CI −0.07 to 0.44), exceeding the prespecified 0.4-point upper margin. Severe hypoglycemia occurred in 2/117 (1.7%) versus 1/113 (0.9%); these small counts cannot establish comparative safety. The label reports cough in 21% versus 3% and cautions that the trial was not designed for meaningful adverse-reaction incidence comparisons.1
Respiratory precautions. Afrezza has a boxed warning for acute bronchospasm and is contraindicated in chronic lung disease such as asthma or COPD. Spirometry is required before treatment, after six months and annually thereafter, even without symptoms. It is not recommended for people who smoke or recently stopped; the US medication guide specifies less than six months since quitting. The label reports approximately 40 mL greater FEV1 decline than comparators in trials lasting up to two years, evident within the first three months and persisting during observation. Effects beyond two years and reversibility after stopping are not established. Reported lung cancers were too few to determine whether Afrezza affects lung or respiratory-tract tumor risk.1
DKA and other insulin risks. Across the label’s type 1 diabetes trials, DKA was reported in 13 Afrezza-treated patients (0.43%) versus 3 comparator-treated patients (0.14%). Basal insulin remains necessary, with closer assessment during illness or infection. Hypoglycemia, serious allergic reactions and low potassium are also labeled risks. This is a summary of important issues, not a replacement for the full prescribing information.1
Country access. The age 6+ indication above is specifically the US label. Cipla announced an adult India launch on 22 December 2025; this is a company launch report, not a current stock check. The Brazilian patient leaflet hosted by Biomm specifies adult use and says that Anvisa approved that leaflet on 12 June 2025. Present Brazilian supply and reimbursement were not confirmed. US manufacturer savings and cash-pay programmes have eligibility conditions; a programme advertisement does not establish an individual’s coverage or price.1564
Sources
- [1]FDA: Afrezza US prescribing information, revised May 2026 · Regulatory decision · 2026-05 (month only) — Sections 1–6, 12.2–12.3 and 14.4. US indication ages 6+; clamp timing in 30 people with type 1 diabetes; pediatric trial 117 versus 113 randomized, failed HbA1c noninferiority. Boxed bronchospasm warning, pulmonary monitoring, lung-cancer uncertainty and increased DKA frequency in pooled type 1 diabetes trials.
FDA. Afrezza US prescribing information, revised May 2026: sections 1–6, 12.2–12.3, 14.4 and Medication Guide.
- [2]Hirsch et al. A Randomized Trial Comparing Inhaled Insulin Plus Basal Insulin Versus Usual Care in Adults With Type 1 Diabetes · Peer-reviewed study · 2024-12-06 — Full original article, Tables 2–3 and Outcomes/Safety/Funding. 123 adults randomized for 17 weeks; observed HbA1c response counts use 57 and 58 participants. Worsening thresholds were post hoc. Funded by MannKind, which participated in protocol development and oversight and reviewed the manuscript.
Hirsch et al. INHALE-3 original trial report, Tables 2–3, Outcomes, Safety and Funding.
- [3]Hirsch et al. A Randomized Comparison of Postprandial Glucose Excursion Using Inhaled Insulin Versus Rapid-Acting Analog Insulin in Adults With Type 1 Diabetes Using Multiple Daily Injections or Automated Insulin Delivery · Peer-reviewed study — Original single-meal study of 122 adults, 61 per group, within INHALE-3. The 2-hour area above 180 mg/dL was divided by time, giving mg/dL units; adjusted difference −12 mg/dL (95% CI −22 to −2). Pump users followed study-specific instructions. MannKind funded the study.
Hirsch et al. Original randomized meal-challenge report, Methods, Table 2 and Funding.
- [4]MannKind: Paying for Afrezza — US programme terms · Manufacturer — Commercial savings, pediatric bridge and cash-pay pathways have eligibility conditions. The advertised pediatric bridge window ends 31 December 2026. Current individual coverage, price and pharmacy stock were not checked.
MannKind. Paying for Afrezza, US programme conditions.
- [5]Cipla launches Afrezza in India · Manufacturer · 2025-12-22 — Original distributor announcement describes the launch for adults with type 1 and type 2 diabetes and reports earlier CDSCO approval. The underlying Indian regulatory decision, current local stock and reimbursement were not independently retrieved.
Cipla. Original India launch announcement, 22 December 2025, page 1.
- [6]Biomm: Afrezza Brazilian patient leaflet · Regulatory decision — Manufacturer-hosted patient leaflet states adult use and Anvisa approval of this leaflet on 12 June 2025. Sections 1 and 6 describe diabetes treatment and basal-insulin combination in type 1 diabetes. This document does not establish present supply or reimbursement.
Biomm. Brazilian Afrezza patient leaflet, sections 1 and 6 and leaflet-approval statement.