Insulin Icodec

Novo Nordisk's once-weekly acylated basal insulin analog, approved as Awiqli in Switzerland, Canada, the EU and other markets in 2024 and by the FDA on March 26, 2026 for adults with type 2 diabetes.

Also referenced as: Insulin icodec-abae, Awiqli, Awiqli FlexTouch, NN1436, NNC0148-0287

Also appears in: Hormone

COA Public product evidence No certificate records are bound to a tracked Insulin Icodec product page A record counts here only when a vendor publishes a certificate against one of its own tracked Insulin Icodec product pages. Other certificates naming this compound may still exist — search the wider certificate database. →

---

What is Insulin Icodec?

Insulin icodec is a once-weekly basal insulin analog developed by Novo Nordisk under the codes NN1436 and NNC0148-0287 and sold as Awiqli. It is a peptide-based analog of human insulin, produced by recombinant expression in Saccharomyces cerevisiae and then chemically modified. Compared with human insulin, the threonine at position B30 is omitted, tyrosine A14 is replaced by glutamate, and tyrosine B16 and phenylalanine B25 are replaced by histidine; a side chain carrying a 1,20-icosanedioic acid (C20 fatty diacid) is attached through the side-chain amino group of lysine B29. The US label gives the molecular formula C280H435N71O87S6 and a molecular weight of 6,380.26 Da. The FDA-approved product carries the suffix insulin icodec-abae.

The design goal was a basal insulin whose plasma half-life of about one week allows a single subcutaneous injection every seven days in place of daily glargine, degludec or detemir. It was the first once-weekly basal insulin authorized anywhere, approved in Switzerland and Canada in March 2024, in the European Union in May 2024 and in Japan, Australia and China in June 2024. In the United States, a first application drew a complete response letter in July 2024, and Awiqli was approved on March 26, 2026 for adults with type 2 diabetes. Efsitora alfa, a competing once-weekly basal insulin built on an insulin-Fc fusion-protein scaffold, is a different molecule.

How it works

• Albumin-bound depot from a C20 fatty diacid. Icodec was re-engineered from OI338, an earlier ultra-long oral basal insulin with a 70-hour half-life, by lengthening the acyl side chain from a C18 to a C20 diacid and adding the B16 tyrosine-to-histidine substitution, which lowers insulin receptor affinity; candidates were screened for long intravenous half-life in dogs while retaining glucose-lowering potency in rats (Kjeldsen et al., Journal of Medicinal Chemistry, 2021;64(13):8942–8950). The diacid gives strong, reversible albumin binding, reported in the US label as greater than 99% plasma protein binding, creating a circulating reservoir from which active insulin is slowly released.
• Reduced receptor affinity and clearance. The A14E, B16H and B25H substitutions stabilize the molecule and attenuate insulin receptor binding and receptor-mediated clearance, further prolonging the half-life; in cell assays icodec triggers the same dose-dependent receptor signaling and metabolic responses as human insulin, with proportionately lower IGF-1 receptor affinity and low mitogenic activity. A multiple-dose trial in type 2 diabetes measured a mean half-life of 196 hours and a close to even distribution of glucose-lowering effect across the week (Nishimura et al., BMJ Open Diabetes Research and Care, 2021;9(1):e002301).
• Weekly pharmacodynamic profile. In 66 adults with type 1 diabetes, icodec reached steady state within 2 to 3 weeks, and model-predicted daily shares of the weekly glucose infusion rate were 14.3, 19.6, 18.3, 15.7, 13.1, 10.6 and 8.4% on days 1 through 7, so the effect peaks on days 2 and 3 and tapers toward the next injection (Hövelmann et al., Diabetes, Obesity and Metabolism, 2024;26(5):1941–1949). The US label adds that exposure falls as body weight rises across 40 to 160 kg.

Research status

Phase 2 (NCT03751657). A 26-week double-blind, double-dummy trial randomized 247 insulin-naive adults with type 2 diabetes on metformin to once-weekly icodec or once-daily glargine U100. Estimated HbA1c changes were −1.33 and −1.15 percentage points (difference −0.18; 95% CI −0.38 to 0.02; p=0.08), with level 2 or 3 hypoglycemia at 0.53 versus 0.46 events per patient-year (Rosenstock et al., New England Journal of Medicine, 2020;383(22):2107–2116).

ONWARDS 1 (NCT04460885). The largest phase 3a trial randomized 984 insulin-naive adults with type 2 diabetes to icodec or glargine U100 for 52 weeks plus a 26-week extension. HbA1c fell from 8.50% to 6.93% with icodec and from 8.44% to 7.12% with glargine (difference −0.19 percentage points; 95% CI −0.36 to −0.03), meeting noninferiority (p<0.001) and superiority (p=0.02); time in the 70 to 180 mg/dL range in weeks 48 to 52 was 71.9% versus 66.9%. Combined clinically significant or severe hypoglycemia occurred at 0.30 versus 0.16 events per person-year, a rate ratio of 1.63 (95% CI 1.02 to 2.61) at week 83 (Rosenstock et al., New England Journal of Medicine, 2023;389(4):297–308).

ONWARDS 2 (NCT04770532). In 526 adults already on basal insulin, switching to icodec versus degludec for 26 weeks gave HbA1c of 7.20% versus 7.42% (difference −0.22 points; 95% CI −0.37 to −0.08), confirming noninferiority and superiority (p=0.0028), with 1.70 kg more weight gain and level 2 or 3 hypoglycemia at 0.73 versus 0.27 events per patient-year (rate ratio 1.93, 95% CI 0.93 to 4.02) (Philis-Tsimikas et al., Lancet Diabetes and Endocrinology, 2023;11(6):414–425).

ONWARDS 3 (NCT04795531). The only double-masked trial in the ONWARDS program randomized 588 insulin-naive adults to icodec with daily placebo or degludec with weekly placebo for 26 weeks. HbA1c reached 7.0% versus 7.2% (difference −0.2 points, 95% CI −0.3 to −0.1), confirming noninferiority and superiority (p=0.002); weight change did not differ, but combined level 2 or 3 hypoglycemia was higher with icodec over weeks 0 to 26 (0.35 versus 0.12 events per patient-year; p=0.01) (Lingvay et al., JAMA, 2023;330(3):228–237).

ONWARDS 4 (NCT04880850). In 582 adults with long-standing type 2 diabetes on basal-bolus therapy, icodec plus mealtime aspart was noninferior to glargine U100 plus aspart at 26 weeks (HbA1c change −1.16 versus −1.18 points; difference 0.02, 95% CI −0.11 to 0.15), with similar level 2 and 3 hypoglycemia rates and a lower bolus insulin dose (Mathieu et al., Lancet, 2023;401(10392):1929–1940).

ONWARDS 5 (NCT04760626). A 52-week trial with real-world elements randomized 1,085 insulin-naive adults to icodec titrated with a dosing guide app or to once-daily degludec, glargine U100 or glargine U300 dosed per usual practice. The HbA1c difference was −0.38 points (95% CI −0.66 to −0.09), meeting superiority (p=0.009), with better treatment-satisfaction scores and similarly low hypoglycemia rates; the design cannot separate the insulin from the app (Bajaj et al., Annals of Internal Medicine, 2023;176(11):1476–1485).

ONWARDS 6 (NCT04848480), type 1 diabetes. Among 582 adults with type 1 diabetes on insulin aspart, icodec was noninferior to degludec for HbA1c at week 26 (−0.47 versus −0.51 points; difference 0.05, 95% CI −0.13 to 0.23), but combined clinically significant or severe hypoglycemia was 19.9 versus 10.4 events per patient-year, a rate ratio of 1.9 (95% CI 1.5 to 2.3; p<0.0001), and the excess persisted over 57 weeks (Russell-Jones et al., Lancet, 2023;402(10413):1636–1647). The US indication is limited to type 2 diabetes, and the EU label advises starting people with type 1 diabetes only when a weekly regimen offers clear benefit.

Pooled safety. Across ONWARDS 1 to 6 (2,170 participants per arm), adjusted rates per 100 patient-years were similar for all adverse events (248.49 versus 242.53), serious adverse events (14.51 versus 14.03), hypersensitivity and medication errors; injection-site reactions were more frequent with icodec (5.86 versus 3.66) and mostly mild, and the hazard ratio for a first major adverse cardiovascular event was 0.93 (95% CI 0.56 to 1.56) (Bajaj et al., Endocrine Practice, 2026;32(10):1475–1482).

Later trials. ONWARDS 10 (NCT06340854), a 26-week phase 3b trial, randomized 412 basal-insulin-treated adults to icodec started without the one-time additional first dose or to glargine U100; the HbA1c difference was −0.2 points (95% CI −0.4 to −0.1) with a numerically lower hypoglycemia rate (0.45 versus 0.59 episodes per patient-year) (Rosenstock et al., Diabetes, Obesity and Metabolism, 2026;28(9):7791–7801). A fixed-ratio combination of icodec with semaglutide, IcoSema, is in the COMBINE program; in COMBINE 4 (NCT06269107), 485 adults on oral agents randomized to IcoSema or glargine U100 for 40 weeks showed HbA1c and body-weight differences of −0.88 percentage points and −4.61 kg in favor of the combination (Ji et al., Lancet Diabetes and Endocrinology, 2026;14(10):831–840). IcoSema is a separate product outside the Awiqli approvals.

Regulatory path. Swissmedic granted the first authorization on March 7, 2024 through the Access Consortium work-sharing review, and Health Canada issued its Notice of Compliance on March 12, 2024. The European Commission’s marketing authorization followed on May 17, 2024 after a CHMP opinion of March 21, 2024, for treatment of diabetes mellitus in adults, covering both type 1 and type 2. Approvals in Japan and Australia came in June 2024, and China approved the product for adults with type 2 diabetes the same month. In the United States, an FDA advisory committee voted 7 to 4 in May 2024 against the type 1 diabetes indication, and the FDA issued a complete response letter in July 2024 requesting more information on the manufacturing process and that indication. After resubmission, BLA 761326 was approved on March 26, 2026 for adults with type 2 diabetes only. As of November 2026, Drugs@FDA lists no US application for type 1 diabetes.

Common dosage forms

• Awiqli FlexTouch prefilled pens (US). A 700 units/mL (U-700) clear solution in single-patient-use pens of 3 mL (2,100 units), 1.5 mL (1,050 units) and 1 mL (700 units), for subcutaneous self-injection by the patient; the 1 mL pen is labeled as a sample.
• Awiqli FlexTouch pens (other markets). The same 700 units/mL solution in prefilled pens under national licenses.
• No vial, cartridge or syringe presentation. The product is supplied only in the FlexTouch pen.

Key considerations

• Regulatory status. The FDA Approved status reflects the March 26, 2026 approval of Awiqli (insulin icodec-abae) injection under BLA 761326 as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes. The approval followed a July 2024 complete response letter, covers type 2 diabetes only, and came after 2024 approvals in Switzerland, Canada, the European Union, Japan, Australia and China, several of which include adults with type 1 diabetes.
• Hypoglycemia is the central trade-off. In the insulin-naive type 2 trials, clinically significant or severe hypoglycemia was numerically or statistically higher with icodec than with daily comparators at absolute rates below one event per patient-year, while in type 1 diabetes the rate roughly doubled. Because the half-life is about one week, the label warns that recovery from hypoglycemia may be delayed compared with shorter-acting insulins.
• Dosing-error risk is a named concern. The U-700 concentration and weekly schedule differ from every daily insulin, and the label reports serious hypoglycemia requiring hospitalization from mix-ups with other insulins or once-weekly injectable antidiabetic medicines, incorrect dose selection and dosing-frequency errors.
• Glycemic gains are modest and partly convenience-driven. HbA1c advantages over daily basal insulins were 0.2 to 0.4 percentage points in treat-to-target trials, and the largest margin, in ONWARDS 5, cannot be attributed to the molecule alone because app-guided titration was part of the intervention. The main benefit is 52 instead of 365 basal injections a year.
• Naming. Insulin icodec, insulin icodec-abae, NN1436, NNC0148-0287, Awiqli and Awiqli FlexTouch denote the same molecule. It is structurally unrelated to efsitora alfa, and IcoSema is a distinct combination product.