Insulin icodec

Icodec is human insulin with A14E, B16H and B25H substitutions, deletion of threonine B30, and a C20 fatty diacid attached to lysine B29 through a linker. The diacid binds albumin reversibly and strongly, creating a large circulating depot that releases free insulin slowly; the substitutions simultaneously reduce receptor affinity and slow receptor-mediated clearance, and stabilise the molecule against enzymatic degradation. The result is a half-life of roughly a week and an almost flat pharmacodynamic profile over seven days. Once bound to the receptor, the biology is ordinary insulin biology: suppression of hepatic glucose output, promotion of peripheral glucose uptake and inhibition of lipolysis. The clinically important consequence of the long half-life is that any dosing error persists for days, which is why it was approved for type 2 diabetes only after the FDA and its advisory committee balked at the hypoglycaemia data in type 1.

Targets: Insulin receptor, Hepatic glucose output, Peripheral glucose uptake, Albumin as a circulating depot

• · Dosed in insulin units, not micrograms. Typical starting dose is 70 units once weekly, titrated weekly in steps of about 20 units against pre-breakfast glucose. Concentration is 700 units per mL, so injection volumes stay small.
• · Multiply the current total daily basal dose by seven. A one-time additional loading dose of around 50 percent is used in some switch scenarios per the label to reach steady state faster.

Work out your exact syringe units →

• very commonHypoglycaemia— The defining risk of any insulin, and here it cannot be reversed by skipping the next dose because the drug is already in you for a week.
• commonWeight gain— Typical of basal insulin therapy.
• commonInjection-site reactions and lipodystrophy— Rotate sites.
• uncommonHypokalaemia— Insulin drives potassium intracellularly; relevant in patients on diuretics or with cardiac disease.
• uncommonPeripheral oedema

• Episodes of hypoglycaemia
• Diabetic ketoacidosis, which needs short-acting insulin
• Type 1 diabetes, where it is not approved in the US because of the hypoglycaemia signal
• Hypersensitivity to icodec or excipients
• Situations with rapidly changing insulin requirements, such as acute illness or planned major surgery, where a week-long depot is a liability

• synergysemaglutide — Combining a weekly GLP-1 with a weekly basal insulin gives a genuinely once-weekly injectable regimen; basal dose usually needs reducing when the GLP-1 is started.
• cautionpasireotide — Pasireotide raises glucose substantially and insulin requirements will climb.
• cautioncorticotropin — Cortisol elevation raises insulin needs, and the week-long depot makes rapid adjustment impossible.

• · Fasting capillary glucose or continuous glucose monitoring, particularly in the first month
• · HbA1c every 3 months
• · Serum potassium in at-risk patients
• · Awareness of hypoglycaemia symptoms and a documented rescue plan

Prescription drug approved for type 2 diabetes in the US (March 2026), and previously in the EU, Canada, Japan, China and Australia. US launch is expected in the second half of 2026.

• Awiqli FDA prescribing information (label)
• ONWARDS 1 through 6 phase 3 trial programme of once-weekly insulin icodec (trial)
• FDA Endocrinologic and Metabolic Drugs Advisory Committee, May 2024, review of insulin icodec (guideline)

Once icodec finally reaches the insulin receptor the biology is entirely ordinary: receptor tyrosine kinase autophosphorylation, IRS-1 and IRS-2 recruitment, PI3K and Akt activation, GLUT4 translocation in muscle and fat, suppression of hepatic gluconeogenesis and glycogenolysis, and inhibition of hormone-sensitive lipase. Nothing about the downstream signalling is novel. The engineering is entirely about getting there slowly. The C20 diacid binds albumin reversibly and tightly, so the great majority of circulating drug is inert and bound at any moment, with a small free fraction in equilibrium. That alone would extend the half-life. What compounds it is the deliberate reduction in receptor affinity from the three substitutions: receptor-mediated endocytosis and degradation is normally a dominant clearance route for insulin, so an analogue that binds the receptor less avidly is cleared more slowly, at the cost of needing higher molar concentrations for the same effect. Combine albumin sequestration with slowed receptor clearance and you get a one-week half-life and a peak-to-trough ratio flat enough that day-to-day variation within the week is small. The clinical consequence that dominates everything else is irreversibility. With daily basal insulin, a dosing error is a bad day. With icodec, a dosing error is a bad week, and there is no way to remove the drug once it is in. That asymmetry is exactly why the FDA advisory committee balked at type 1 diabetes, where insulin requirements swing with exercise, illness and carbohydrate intake, and why the type 2 approval came through.

Everything that goes wrong flows from irreversibility. If a patient injects twice by mistake, or injects on the wrong day, or takes their weekly dose and then develops gastroenteritis and stops eating, there is nothing to do but feed them and wait, potentially for days. That is a genuinely different safety proposition from daily basal and it needs saying plainly at initiation. The second failure is over-eager titration: steady state takes three to four weekly injections, so a clinician who increases the dose weekly on the basis of a still-rising baseline is stacking, and the hypoglycaemia arrives two weeks later. The third is transferring the drug into a syringe: at 700 units per mL a volume error that would be trivial with U100 is catastrophic, and the pen must be used as supplied. The fourth is using it in situations with rapidly changing insulin requirements, acute illness, planned major surgery, or a hospital admission, where a week-long depot is a liability rather than a convenience. The fifth is type 1 diabetes, where it is not approved in the US: ONWARDS 6 showed significantly more level 2 and level 3 hypoglycaemia than degludec, and that is the reason. The sixth is forgetting which day is dose day, which sounds trivial and is in fact the most common real-world adherence failure for weekly injectables.

1. —Week 1, insulin-naive type 2 diabetes — Dosed in insulin units, not micrograms. Typical starting dose is 70 units once weekly on a fixed day, any time of day, into abdomen, thigh or upper arm with site rotation. The 700 units per mL concentration keeps the volume small.
2. —Weeks 2-4 — Adjust weekly in steps of about 20 units against the mean of the preceding three pre-breakfast glucose readings. Never titrate more than once a week: steady state takes 3 to 4 weekly injections and stacking increases is how people end up hypoglycaemic on day three of week two.
3. —Switching from a daily basal insulin — Multiply the current total daily basal dose by seven. In some switch scenarios the label uses a one-time additional loading dose of around 50 percent to reach steady state faster, which is a deliberate front-load rather than a permanent increase.
4. —Weeks 4 onward — Once at steady state, treat every change as a change that will not fully express for three weeks. Patience here is the single most important dosing skill with this drug.

• Insulin receptor — Deliberately reduced relative to human insulin by the A14E, B16H and B25H substitutions; numeric affinity not resolved this session.

  Standard receptor tyrosine kinase signalling through IRS-PI3K-Akt: GLUT4 translocation, suppression of hepatic glucose output, inhibition of lipolysis. Reduced affinity slows receptor-mediated clearance, which is a large part of the half-life extension.

• Serum albumin — Strong reversible binding via the C20 fatty diacid

  Not a pharmacological target but the depot. Most circulating drug is albumin-bound and inactive, releasing free insulin slowly and continuously across the week.

• IGF-1 receptor — Insulin analogues are screened for IGF-1R affinity and mitogenic potential; icodec's has been characterised in development but I did not resolve the figures this session.

  Relevant as a theoretical mitogenicity consideration for any modified insulin, not as a clinical effect.

• ONWARDS 1 Phase 3a, randomised, open-label, treat-to-target · 78 weeks · 2023

  Change in HbA1c at week 52 with once-weekly icodec versus once-daily glargine U100 in insulin-naive type 2 diabetes. Icodec was non-inferior and achieved a greater HbA1c reduction, with more time in range.

• ONWARDS 2 Phase 3a, randomised, open-label · 26 weeks · 2023

  Switching to once-weekly icodec versus once-daily degludec in basal insulin-treated type 2 diabetes. Icodec was non-inferior on HbA1c change.

• ONWARDS 3 Phase 3a, randomised, double-blind, double-dummy · 26 weeks · 2023

  Once-weekly icodec versus once-daily degludec in insulin-naive type 2 diabetes. Non-inferior HbA1c reduction. The double-dummy design makes this the most methodologically rigorous trial in the programme.

• ONWARDS 4 Phase 3a, randomised, open-label · 26 weeks · 2023

  Once-weekly icodec versus once-daily glargine U100 in basal-bolus treated type 2 diabetes. Non-inferior HbA1c reduction with comparable hypoglycaemia.

• ONWARDS 5 Phase 3a, randomised, open-label, real-world design · 52 weeks · 2023

  Once-weekly icodec with a dosing guide app versus once-daily basal analogues in insulin-naive type 2 diabetes, in a pragmatic real-world setting. Icodec achieved greater HbA1c reduction.

Glucose lowering begins within the first day of the first injection, but full steady-state effect takes three to four weekly injections, roughly three to four weeks. That gap between apparent onset and true steady state is the single most important thing to understand about dosing this drug, because a dose that looks correct in week one may be too high by week four. Hypoglycaemia risk, when it occurs, is typically in the first few days after an injection when free insulin concentrations are relatively higher. Weight gain follows the usual basal insulin pattern over months.

The obvious pairing is with a weekly GLP-1 receptor agonist, which turns a daily-injection regimen into two injections a week total, and the adherence argument for that is strong. When a GLP-1 is added to an existing icodec dose, the basal dose usually needs reducing, and with a week-long depot that reduction has to be anticipated rather than made reactively. Metformin, SGLT2 inhibitors and pioglitazone all continue unchanged. Sulfonylureas are the combination to be most careful with, since they add an insulin-secretagogue hypoglycaemia risk on top of an irreversible basal depot, and most clinicians stop them at initiation. Corticosteroids, pasireotide and anything else that acutely raises insulin requirements are difficult on a weekly basal, because you cannot ramp up quickly and you cannot ramp down at all; a short course of daily basal or prandial cover alongside is the usual workaround. Beta-blockers mask hypoglycaemia awareness, which matters more when the hypoglycaemia cannot be withdrawn.

Against daily basal analogues, glargine U100, glargine U300 and degludec: the ONWARDS programme showed HbA1c reduction at least equal to daily basal across five trials in type 2 diabetes, with more time in range in the insulin-naive settings. The trade is 52 injections a year instead of 365, against a dosing error you cannot take back. For a patient whose barrier to insulin is the daily injection itself, that is a real advantage; for a patient with erratic intake or frequent illness, it is a real liability. Against insulin efsitora alfa: the two are direct competitors in the weekly basal category with different engineering, albumin binding versus Fc receptor recycling, and efsitora's roughly 17-day half-life gives an even flatter profile at the cost of an even longer commitment. Against continuing with no basal insulin at all: the honest framing is that weekly basal makes insulin initiation psychologically easier, and clinical inertia around starting insulin is a genuine problem in type 2 diabetes, so the adherence and acceptance argument may matter more than any pharmacodynamic difference.

Not established at the time of writing. Approved for type 2 diabetes in the US in March 2026 with launch expected in the second half of 2026, and previously approved in the EU, Canada, Japan, China and Australia. Pricing relative to daily basal analogues is the key open commercial question and I did not source figures this session.

• ?I did not resolve the Awiqli label in this session, so tmax, absolute bioavailability, volume of distribution, protein binding percentage and numeric clearance are all null rather than estimated.
• ?The 24-day time to steady state is derived from the widely reported figure of three to four weekly injections rather than from a label statement.
• ?The insulin receptor and IGF-1 receptor affinity ratios for icodec have been published in development literature but I did not resolve them this session.
• ?Blood-brain barrier penetration is marked partial on the general basis that insulin crosses by a saturable transport mechanism; whether icodec does so meaningfully given its albumin binding is not established.
• ?Trial enrolments were not confirmed against the papers, so all participants fields are null. ONWARDS 6, the type 1 trial, is described in the Core record but I did not resolve its publication this session and it is therefore not listed as a verified trial here.
• ?No pricing information exists yet.

• Weekly Icodec versus Daily Glargine U100 in Type 2 Diabetes without Previous Insulin Rosenstock J, et al., New England Journal of Medicine, 2023 · PMID 37356066

  ONWARDS 1, the flagship insulin-naive trial.

• Once-Weekly Insulin Icodec vs Once-Daily Insulin Degludec in Adults With Insulin-Naive Type 2 Diabetes: The ONWARDS 3 Randomized Clinical Trial Lingvay I, et al., JAMA, 2023 · PMID 37354562

  The double-blind, double-dummy trial, and therefore the cleanest comparison in the programme.

• Switching to once-weekly insulin icodec versus once-daily insulin degludec in individuals with basal insulin-treated type 2 diabetes (ONWARDS 2): a phase 3a, randomised, open label, multicentre, treat-to-target trial Philis-Tsimikas A, et al., Lancet Diabetes & Endocrinology, 2023 · PMID 37148899

  The switch scenario most clinicians will actually face.

• Switching to once-weekly insulin icodec versus once-daily insulin glargine U100 in individuals with basal-bolus insulin-treated type 2 diabetes (ONWARDS 4): a phase 3a, randomised, open-label, multicentre, treat-to-target, non-inferiority trial Mathieu C, et al., Lancet, 2023 · PMID 37156252

  Shows the weekly basal works even alongside mealtime insulin.

• Once-Weekly Insulin Icodec With Dosing Guide App Versus Once-Daily Basal Insulin Analogues in Insulin-Naive Type 2 Diabetes (ONWARDS 5): A Randomized Clinical Trial Bajaj HS, et al., Annals of Internal Medicine, 2023 · PMID 37748181

  The pragmatic real-world trial, and the one that best reflects ordinary primary care use.