Weekly IcoSema beat daily glargine in insulin-naive type 2 diabetes

Design

Phase 3b, randomised, open-label, treat-to-target, 97 sites in nine countries

Population

485 adults with type 2 diabetes, HbA1c 8.0% or more, on oral glucose-lowering drugs

Primary outcome

Change in HbA1c from baseline to week 40

Effect

-3.32 vs -2.44 percentage points; difference -0.88 (95% CI -1.12 to -0.63)

COMBINE 4 randomised 485 adults with type 2 diabetes and HbA1c of 8.0% or more on oral agents to once-weekly IcoSema, a fixed combination of insulin icodec and semaglutide, or once-daily glargine U100. Both arms titrated to a fasting target of 3.9-5.0 mmol/L over 40 weeks.

HbA1c fell by 3.32 percentage points with IcoSema and 2.44 with glargine, a difference of -0.88 (95% CI -1.12 to -0.63). Body weight fell by 0.79 kg with IcoSema and rose by 3.81 kg with glargine, a difference of -4.61 kg. Clinically significant or severe hypoglycaemia occurred at 0.29 versus 0.59 episodes per person-year (rate ratio 0.56, 95% CI 0.32 to 0.97). Gastrointestinal events were the most common adverse events with IcoSema.

The trial was open-label, the sponsor was the manufacturer, and mean baseline HbA1c was high at about 9.5%. The hypoglycaemia interval comes close to 1, so that finding is the least secure of the three. A weekly combination could ease the burden of starting insulin, but availability and cost in India are not established by this paper.

• Consider the trade-off if you start basal insulin in someone whose weight gain or hypoglycaemia is a concern.
• Counsel that nausea and other gastrointestinal effects were the main adverse events with the combination.
• Expect a larger HbA1c fall than glargine in people starting from about 9.5%, not from near target.
• Do not assume the weekly formulation is available or affordable locally; check before planning around it.

Why it matters

It tests the guideline idea of pairing a GLP-1 receptor agonist with basal insulin in a single weekly injection.

Don't overread it

This was a 40-week open-label trial against glargine, not against a separate GLP-1 receptor agonist plus basal insulin regimen.

The statistics, in plain English

The upper end of the hypoglycaemia confidence interval (0.97) sits just below 1, so the reduction is only borderline convincing. In an open-label trial, clinicians who knew the allocation may have titrated differently.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.