- Design
- Phase 3b, randomised, open-label, treat-to-target, 40 weeks
- Population
- 485 insulin-naive adults with type 2 diabetes, HbA1c at least 8% on oral agents
- Primary outcome
- Change in HbA1c at 40 weeks
- Effect
- -3.32 vs -2.44 percentage points; difference -0.88 (95% CI -1.12 to -0.63)
COMBINE 4 compared once-weekly IcoSema, a fixed combination of basal insulin icodec and semaglutide in a single injection, against once-daily insulin glargine U100 in 485 insulin-naive adults with type 2 diabetes and an HbA1c of at least 8% on oral agents. It was a 40-week, open-label, treat-to-target, phase 3b trial.
IcoSema was superior on both counts. HbA1c fell 3.32 versus 2.44 percentage points, an estimated treatment difference of 0.88 points in its favour. Weight fell 0.79 kg with IcoSema but rose 3.81 kg with glargine, a difference of 4.61 kg. Clinically significant or severe hypoglycaemia was about half as frequent. Gastrointestinal effects were the main adverse events with IcoSema.
A once-weekly injection that lowers HbA1c and weight more than daily basal insulin, with less hypoglycaemia, is a genuine change to how insulin might be started in type 2 diabetes. It is one open-label, industry-funded trial, and the combination's availability in India is not yet established, so it supports considering the option rather than mandating it.
- 485 insulin-naive adults with type 2 diabetes and HbA1c of 8% or above on oral agents.
- Once-weekly IcoSema combines basal insulin icodec with semaglutide in one injection.
- HbA1c fell 3.32 versus 2.44 percentage points with daily glargine over 40 weeks.
- Weight fell 0.79 kg with IcoSema but rose 3.81 kg with glargine.
- Clinically significant or severe hypoglycaemia was about half as frequent (rate ratio 0.56).
Why it matters
One weekly injection that lowers HbA1c and weight with less hypoglycaemia reshapes how insulin is started in type 2 diabetes.
Don't overread it
This is a single open-label, industry-funded trial, and glargine carries no weight or GLP-1 effect of its own, so some of the gap is expected.
The statistics, in plain English
The estimated treatment difference of -0.88 percentage points, with a confidence interval of -1.12 to -0.63, lies entirely below zero, so the extra HbA1c reduction is very unlikely to be chance. The hypoglycaemia rate ratio of 0.56 (0.32 to 0.97) sits just under 1.0, so the reduction is statistically significant but its upper bound is close to no difference.
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