GLP-1 Agonists in Type 1 Diabetes
GLP-1 and Type 1 Diabetes
-4.9 kg
Liraglutide 1.8 mg reduced body weight by 4.9 kg versus placebo in overweight adults with type 1 diabetes in the Lira-1 trial.
Dejgaard et al., Lancet Diabetes Endocrinol, 2016
Dejgaard et al., Lancet Diabetes Endocrinol, 2016
View as imageGLP-1 receptor agonists are approved for type 2 diabetes and obesity. They are not approved for type 1 diabetes. This distinction matters because the two conditions have fundamentally different pathophysiology: type 2 involves insulin resistance with initially preserved beta-cell function, while type 1 involves autoimmune destruction of beta cells with absolute insulin dependence. GLP-1 agonists work partly by enhancing insulin secretion from functioning beta cells, a mechanism that should be irrelevant when those cells are destroyed.
Yet the clinical reality is more nuanced. People with type 1 diabetes who are overweight or obese, a growing population, face the same metabolic challenges as people with type 2 diabetes plus the additional burden of managing exogenous insulin. GLP-1 agonists offer glucagon suppression, gastric emptying delay, appetite reduction, and weight loss, effects that do not depend on beta-cell function and could theoretically benefit type 1 diabetes management. Several randomized controlled trials have tested this hypothesis. For broader context on autoimmune diabetes and GLP-1, see our pillar article.
Key Takeaways
- Liraglutide 1.8 mg added to insulin reduced HbA1c by 0.28-0.34% versus placebo and decreased body weight by 4.9 kg in overweight type 1 diabetes patients across three RCTs (Dejgaard et al., 2016; Mathieu et al., 2016; Ahren et al., 2016)
- The ADJUNCT ONE trial (n=1,398) found increased symptomatic hypoglycemia and hyperglycemia with ketosis in liraglutide groups, leading the manufacturer to not pursue a type 1 diabetes indication (Mathieu et al., 2016)
- A meta-analysis of 5 trials (n=2,445) confirmed modest HbA1c reduction and significant weight loss with liraglutide in type 1 diabetes, but the benefit-risk profile was unfavorable for regulatory approval (Dimitrios et al., 2020)
- In adolescents with type 1 diabetes, liraglutide reduced HbA1c by 0.6% at 12 weeks but the effect waned by 26 weeks (Danne et al., 2017)
- The Lira Pump trial showed liraglutide reduced time in hyperglycemia and body weight in insulin pump-treated type 1 diabetes patients (Dejgaard et al., 2020)
The ADJUNCT Trials: The Largest Evidence Base
The most substantial evidence comes from two industry-sponsored trials run by Novo Nordisk, the manufacturer of liraglutide.
ADJUNCT ONE (Mathieu et al., 2016) was a 52-week, double-blind, treat-to-target trial that randomized 1,398 adults with type 1 diabetes 3:1 to liraglutide (1.8, 1.2, or 0.6 mg daily) or placebo, added to individually optimized insulin therapy. Participants had a mean baseline HbA1c of approximately 8.0% and mean BMI of approximately 28 kg/m2.[1]
The glycemic results were modest. Liraglutide 1.8 mg reduced HbA1c by approximately 0.34% versus placebo at 52 weeks. Total daily insulin dose decreased, consistent with liraglutide's glucagon-suppressing and gastric-emptying effects reducing postprandial insulin requirements. Body weight decreased significantly across all liraglutide doses.
The safety findings were more consequential. Symptomatic hypoglycemia rates increased in the liraglutide groups compared to placebo. More concerning, hyperglycemia with ketosis (a precursor to diabetic ketoacidosis, DKA) also increased. This combination of increased hypoglycemia and increased ketosis risk, occurring simultaneously in the same treatment arm, reflected the challenge of adding a glucose-lowering agent to a system that already requires precise insulin dosing. When insulin doses were reduced to accommodate liraglutide's glucose-lowering effect, some patients reduced insulin too aggressively, triggering ketosis.[1]
ADJUNCT TWO (Ahren et al., 2016) used a different design: a 26-week trial with individually capped insulin doses (meaning insulin could not be increased to offset any glucose-lowering failures). This 835-patient trial showed HbA1c reductions of -0.33% (1.8 mg), -0.22% (1.2 mg), and -0.23% (0.6 mg) versus -0.01% for placebo. Body weight decreased by 3.5-5.1 kg across liraglutide doses. The insulin-capping design made the glycemic effects more visible but also introduced the same DKA risk concern.[2]
Based on the combined ADJUNCT data, Novo Nordisk decided not to pursue a type 1 diabetes indication for liraglutide. The benefit-risk profile, modest HbA1c improvement with increased hypoglycemia and ketosis risk, did not meet the threshold for regulatory approval.
The Lira-1 Trial: Targeting Overweight Type 1 Diabetes
Dejgaard et al. (2016) took a more targeted approach with the Lira-1 trial, enrolling only overweight adults (BMI greater than 25 kg/m2) with type 1 diabetes and HbA1c above 8%. This population better represents the type 1 diabetes patients most likely to benefit from GLP-1 agonist therapy, those whose metabolic profile overlaps with type 2 diabetes.[3]
In this 26-week randomized, double-blind, placebo-controlled trial, liraglutide 1.8 mg produced meaningful results. HbA1c decreased by 0.28% versus placebo. Body weight decreased by 4.9 kg. Total daily insulin dose decreased. These effects were clinically relevant for a population struggling with both glycemic control and weight management.
The Lira-1 trial did not report the same degree of ketosis risk as the ADJUNCT trials, possibly because the selected population (overweight, likely with some residual insulin resistance) was less vulnerable to insulin dose reduction-triggered ketosis. This suggests that patient selection may be critical for safe off-label use of GLP-1 agonists in type 1 diabetes.
The Lira Pump Trial: Insulin Pump Users
Dejgaard et al. (2020) extended the investigation to insulin pump-treated type 1 diabetes patients with the Lira Pump trial. This 26-week randomized, double-blind trial enrolled overweight, dysregulated pump users and added liraglutide 1.8 mg or placebo to ongoing pump therapy.[4]
Liraglutide reduced hyperglycemia and body weight in this population. The pump setting provided tighter basal insulin control than multiple daily injections, which may have reduced the ketosis risk associated with excessive insulin dose reduction. Continuous glucose monitoring data showed reduced time spent in hyperglycemic ranges, though the primary endpoint results were mixed.
The pump trial is relevant because insulin pump users represent a growing proportion of people with type 1 diabetes, and pump therapy provides the dosing flexibility needed to safely accommodate the glucose-lowering effects of an adjunct GLP-1 agonist.
Meta-Analysis: Pooling the Evidence
Dimitrios et al. (2020) conducted a systematic review and meta-analysis of liraglutide as adjunct to insulin in type 1 diabetes, pooling data from 5 randomized controlled trials with 2,445 participants. The analysis confirmed the pattern seen in individual trials: HbA1c reductions were statistically significant but modest (approximately -0.28% with liraglutide 1.8 mg), body weight reductions were substantial (up to 4.87 kg with 1.8 mg), and total daily insulin dose decreased.[5]
The meta-analysis also confirmed the safety concerns. Gastrointestinal side effects (nausea, vomiting) were more common with liraglutide, consistent with the GLP-1 agonist class effect. The signal for increased hyperglycemia with ketosis persisted across pooled analyses. The authors concluded that while liraglutide provided metabolic benefits, the benefit-risk ratio was insufficient to support a formal indication for type 1 diabetes.
This conclusion has shaped the regulatory landscape. No GLP-1 receptor agonist is approved for type 1 diabetes in any major market. The American Diabetes Association's 2023 Standards of Care acknowledges the potential for off-label GLP-1 agonist use in type 1 diabetes but does not make a formal recommendation.
Adolescents: A Different Equation
Danne et al. (2017) tested liraglutide in adolescents (10-17 years) with type 1 diabetes in a phase 2, double-blind, placebo-controlled trial. The results showed a pattern of early benefit that faded over time: HbA1c decreased by 0.6% at 12 weeks but the effect attenuated by 26 weeks. Body weight effects were not significant in this younger population.[6]
The attenuation of the glycemic effect over time is a concern. It may reflect compensatory mechanisms, dose tolerance, or the limitations of liraglutide's mechanism in a population with minimal residual beta-cell function. Whether higher doses or newer, more potent GLP-1 agonists like semaglutide would sustain the effect longer is unknown.
Pramlintide Comparison: Different Adjunct Strategies
Galderisi et al. (2018) compared pramlintide (an amylin analog) and liraglutide as adjuncts to insulin in type 1 diabetes, providing a head-to-head comparison of two different peptide-based strategies. Pramlintide and liraglutide both slow gastric emptying and suppress glucagon, but through different receptors and mechanisms.[7]
This comparison is relevant because pramlintide (Symlin) is the only non-insulin injectable approved for type 1 diabetes in the United States. If GLP-1 agonists are to establish a role in type 1 diabetes management, they need to offer advantages over the already-available pramlintide option, whether through greater glycemic efficacy, better weight loss, or more favorable dosing (once weekly vs. multiple times daily).
For context on whether GLP-1 drugs can reduce insulin requirements in type 1 diabetes, see our dedicated analysis.
The DKA Risk: Why Off-Label Use Requires Caution
The most clinically significant safety concern with GLP-1 agonists in type 1 diabetes is diabetic ketoacidosis. DKA occurs when insulin levels are insufficient to suppress lipolysis and ketogenesis. In type 1 diabetes, this can happen rapidly with even modest insulin dose reductions.
GLP-1 agonists create a specific DKA vulnerability: by lowering blood glucose through non-insulin mechanisms (glucagon suppression, delayed gastric emptying), they can mask the need for insulin. A patient seeing improved glucose readings may reduce insulin doses, not realizing that the insulin is still needed to suppress ketone production regardless of the glucose level. This "euglycemic DKA," ketoacidosis with normal or only mildly elevated blood glucose, is particularly dangerous because the normal glucose reading delays recognition and treatment.
The ADJUNCT ONE trial documented this pattern with its finding of increased "hyperglycemia with ketosis" in liraglutide groups. This was not frank DKA in most cases but represented a shift toward ketone production that could progress to DKA with further insulin reduction or intercurrent illness.
The Semaglutide Question
All major randomized trials in type 1 diabetes used liraglutide, a once-daily GLP-1 agonist that is less potent for weight loss than semaglutide or tirzepatide. Whether the newer, more potent GLP-1 agonists would produce better results in type 1 diabetes is an open question. Preliminary case series and small studies suggest that weekly semaglutide may produce more substantial weight loss and glycemic improvement in type 1 diabetes, with some patients discontinuing prandial insulin entirely while maintaining basal insulin. These reports are uncontrolled and cannot establish safety or efficacy.
The weight loss potential of semaglutide (15-20% body weight reduction in some type 2 diabetes trials) is substantially greater than liraglutide's (5-8%). For overweight and obese type 1 diabetes patients, this degree of weight loss could fundamentally change their metabolic profile, reducing insulin resistance and potentially improving glycemic control more than the modest HbA1c reductions seen with liraglutide. Whether the DKA risk scales proportionally with the greater potency remains the key safety question. Large, controlled trials of semaglutide in type 1 diabetes are needed but have not yet been completed. For context on semaglutide's weight loss evidence in non-diabetic populations, see our dedicated article.
Cross-cluster links are also relevant: C-peptide as a diabetes biomarker is particularly useful for identifying type 1 diabetes patients with residual beta-cell function, who may respond differently to GLP-1 agonist therapy.
The Bottom Line
Liraglutide added to insulin in type 1 diabetes produces modest HbA1c reductions (approximately 0.3%), substantial weight loss (up to 5 kg), and reduced insulin requirements across multiple RCTs. The benefit-risk profile was deemed insufficient for regulatory approval due to increased hypoglycemia and ketosis risk. Off-label use may be most appropriate for overweight type 1 diabetes patients with insulin resistance features, where the weight loss and glucagon suppression benefits are most relevant. Whether newer, more potent GLP-1 agonists like semaglutide change this equation remains to be determined by controlled trials.