Low-dose IL-2 expanded Tregs but did not preserve C-peptide in new type 1 diabetes

Design

Phase 2b, multicentre, randomised, double-blind, placebo-controlled

Population

141 people aged 6 to 35 with type 1 diabetes of under 3 months

Primary outcome

Change in stimulated C-peptide AUC at month 12

Effect

Geometric mean ratio 1.06 (95% CI 0.56 to 2.01), p=0.85

DIABIL-2 was a double-blind, placebo-controlled phase 2b trial across 19 academic centres in five European countries. It enrolled 141 people aged 6 to 35 with type 1 diabetes, on insulin for under 3 months, with a peak C-peptide of at least 0.2 pmol/mL. Participants received low-dose interleukin-2 on one of two maintenance regimens, or placebo.

The primary endpoint was change in stimulated C-peptide area under the curve at 12 months. There was no statistically significant difference overall (geometric mean ratio 1.06, 95% CI 0.56 to 2.01; p=0.85), and results were similar for each regimen and by pubertal status.

The drug did what it was meant to biologically: regulatory T cells expanded by day 5. Serious adverse events were reported in 20% of the placebo group and 10% of the IL-2 group; one in each group was judged probably treatment related.

The authors suggest the inflammatory environment at diagnosis may be too intense for this single agent and propose testing it earlier in the disease or in combinations.

• Low-dose IL-2 is not an option for preserving residual beta-cell function in newly diagnosed type 1 diabetes.
• Target engagement (Treg expansion) did not translate into C-peptide preservation here.
• No safety signal separated IL-2 from placebo, so further trials earlier in the disease course remain plausible.
• Families asking about immunotherapy in new-onset type 1 diabetes can be told this approach has not shown benefit.

Why it matters

It is a clean negative result that separates showing a biological effect from showing a clinical one.

Don't overread it

This tested IL-2 as monotherapy at diagnosis; it does not rule out benefit earlier or in combination.

The statistics, in plain English

A confidence interval of 0.56 to 2.01 around a ratio of 1.06 is wide and includes both a halving and a doubling of C-peptide response. The trial shows no detectable benefit; it cannot exclude a small one.

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.