- Design
- Phase 3b, randomised, open-label, treat-to-target, 40 weeks
- Population
- 485 insulin-naive adults with type 2 diabetes, HbA1c ≥8% on oral agents
- Primary outcome
- Change in HbA1c at week 40
- Effect
- −3.32 vs −2.44 points; difference −0.88 (95% CI −1.12 to −0.63)
COMBINE 4 randomised 485 insulin-naive adults with type 2 diabetes and HbA1c of 8% or more on oral agents to once-weekly IcoSema (insulin icodec with semaglutide) or once-daily glargine U100, both titrated to a fasting target of 3.9–5.0 mmol/L, for 40 weeks across nine countries. Baseline HbA1c was about 9.5%. The trial was open label and funded by the manufacturer.
HbA1c fell 3.32 points with IcoSema and 2.44 with glargine, a difference of 0.88 points. Weight fell by 0.8 kg with IcoSema and rose by 3.8 kg with glargine. Clinically significant or severe hypoglycaemia occurred at 0.29 versus 0.59 episodes per person-year. Gastrointestinal effects were the most common adverse events with the combination.
This fits what guidelines already favour: when basal insulin is needed, pairing it with a GLP-1 receptor agonist limits weight gain and hypoglycaemia. The new element is one injection a week for both. The comparator was basal insulin alone, not glargine plus a GLP-1 receptor agonist, so the trial does not show the fixed combination beats the two drugs given separately.
In practice, for a patient on oral agents with a high HbA1c who needs an injectable and struggles with daily injections, a weekly combination is a reasonable option to discuss where it is available. Expect nausea early, and review the fasting glucose target in older or frail patients.
- Consider a once-weekly basal insulin–GLP-1 combination for insulin-naive patients who need an injectable and want fewer injections
- Warn patients about nausea and other gastrointestinal effects in the first weeks
- Check whether the patient is already on a GLP-1 receptor agonist before switching to a fixed combination
- Relax the fasting target in older or frail patients; the trial titrated to 3.9–5.0 mmol/L
- Weigh the weekly convenience against the loss of separate dose control for each component
Why it matters
Starting insulin no longer has to mean daily injections and weight gain.
Don't overread it
Glargine was given without a GLP-1 receptor agonist, so this is not a test against the separate two-drug regimen.
The statistics, in plain English
The HbA1c difference of 0.88 points has a confidence interval of 0.63 to 1.12, so even the smallest plausible gap is clinically meaningful. The hypoglycaemia rate ratio of 0.56 has an interval reaching 0.97, close to no difference, so the hypoglycaemia advantage is real but its size is uncertain.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for diabetes & endocrinology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free