Semaglutide in type 1 diabetes: ADJUST-T1D

The 26-week ADJUST-T1D trial found that adults with type 1 diabetes who had obesity and used automated insulin delivery were more likely to meet a combined glucose-and-weight target when randomized to semaglutide than placebo. The result was 36% versus 0%. Semaglutide also improved time in range and body weight, but the study was small and short. It was also limited to a selected group. It did not test whether semaglutide prevents cardiovascular events in type 1 diabetes^[1].

What ADJUST-T1D tested

Investigators randomly assigned 72 adults in a double-blind trial to once-weekly semaglutide, titrated up to 1 mg, or placebo. Every participant had type 1 diabetes, used an automated insulin delivery system, and had a body mass index of at least 30. Treatment lasted 26 weeks^[1]. The trial registration identifies the study as randomized, parallel-group, and masked for participants, care providers, investigators, and outcome assessors^[2]. This was adjunct treatment. Semaglutide did not replace insulin or the automated delivery system.

The primary endpoint deliberately combined glucose control, avoidance of excess low-glucose time, and weight loss. To count as a responder, a participant needed more than 70% of continuous glucose monitor readings between 70 and 180 mg/dL, less than 4% below 70 mg/dL, and at least a 5% reduction in body weight^[1]. A composite can capture several goals at once. It also means the headline percentage should not be read as the rate of any one outcome on its own.

The glucose and weight results

At week 26, 36% of participants assigned to semaglutide met every part of the composite endpoint, compared with none assigned to placebo. The between-group difference was 36 percentage points, with a 95% confidence interval from 20.6 to 52.2 percentage points. The reported P value was below 0.001^[1]. This is a clear difference within the trial population. It does not show that every adult with type 1 diabetes and obesity would meet the same target.

The estimated between-group change in glycated haemoglobin favoured semaglutide by 0.3 percentage points. Time with glucose between 70 and 180 mg/dL improved by 8.8 percentage points relative to placebo, and the between-group body-weight change was 8.8 kg in favour of semaglutide^[1]. These outcomes address glycaemia and weight over six months. They are not evidence of fewer heart attacks, strokes, kidney failures, or deaths.

What the safety report does and does not show

The published abstract reports two severe hypoglycaemia events in each group and no diabetic ketoacidosis during the trial^[1]. Equal event counts in groups this small do not establish equal risk, and zero observed ketoacidosis events does not prove zero risk. Rare outcomes need more participants and longer follow-up. The registration also excluded people with recent severe hypoglycaemia or recent diabetic ketoacidosis requiring hospital admission^[2], which narrows how far the safety result can be generalized.

The study also excluded people using glucose-lowering medicines other than insulin at screening, those with recent GLP-1 use, and several groups with conditions that could raise treatment risk. Participants had to use an approved hybrid closed-loop system and have an HbA1c above 7.0% but below 10.0%^[2]. Results from this selected population should not be carried over automatically to people using injections without automated delivery, children, pregnancy, or those with recent acute diabetes complications. Our semaglutide safety section provides broader context.

Why the wider question remains open

The review that prompted this brief argues that people with type 1 diabetes continue to face elevated cardiovascular risk despite a century of insulin therapy. It points to glycaemic variability, insulin-related weight gain, insulin resistance, and hormonal disturbances as possible contributors. Its authors also emphasize that long-term outcomes and hard cardiovascular endpoints remain sparsely studied for adjunct treatments^[3]. ADJUST-T1D offers randomized evidence for glucose and weight measures. It does not close that cardiovascular evidence gap.

That distinction matters when discussing treatment beyond insulin. Better continuous-glucose-monitor readings and lower weight may be clinically useful, but surrogate and intermediate measures cannot substitute for a cardiovascular outcomes trial. Nor does a positive adjunct trial make insulin optional in type 1 diabetes. The trial compared semaglutide plus automated insulin delivery with placebo plus automated insulin delivery^[1]. Readers can review the established indication and evidence boundaries on our semaglutide evidence page.

What we do not yet know

The 26-week result cannot tell us whether the glucose and weight differences persist for years, what happens after semaglutide is stopped, or whether treatment changes cardiovascular and kidney outcomes in type 1 diabetes. The 72-person publication is also too small to define uncommon harms. It studied adults with obesity using automated insulin delivery, so separate evidence is needed for other type 1 diabetes populations^[1].

Larger and longer trials need to preserve the basics that made ADJUST-T1D informative: randomized comparison and clearly defined glucose outcomes. They also need enough follow-up for severe hypoglycaemia, ketoacidosis, cardiovascular events, and treatment discontinuation. For now, this trial supports a limited conclusion. In selected adults with type 1 diabetes, obesity, and automated insulin delivery, semaglutide improved a combined glucose-and-weight endpoint over 26 weeks. Any off-label adjunct decision still belongs with a diabetes specialist who can adjust insulin safely. Our semaglutide overview is educational, not a treatment plan.