SAN FRANCISCO —The glucagon-like peptide-1 (GLP-1) agonist liraglutide (Victoza, Novo Nordisk) taken every day preserved postprandial insulin secretion for 1 year after type 1 diabetes diagnosis in patients in the NewLira trial, while the effects disappeared 6 weeks after treatment stopped.
Thomas Dejgaard, MD, from the Steno Diabetes Center, Gentofte, Denmark, presented the work June 8 here at the American Diabetes Association (ADA) 2019 Scientific Sessions.
“After 52 weeks, in patients with newly diagnosed type 1 diabetes, we found that liraglutide 1.8 mg SC once daily, added on to insulin, preserved postprandial insulin secretion, reduced insulin dose, and reduced events of hypoglycemia, compared to placebo. No changes were seen in body weight, HbA1c, or 7-point blood glucose profile,” he reported.
Session moderator Uma Gunasekaran, MD, from the South Western Medical Center, University of Texas, Dallas, commented on the work.
“There have been multiple studies over the years about the use of GLP-1 receptor agonists in type 1 diabetes. It is always somewhat promising but not a big enough impact. But what was really interesting here was the preservation of C-peptide levels and potentially the delay in needing to use insulin,” she asserted.
In type 1 diabetes, GLP-1 agonists have shown a modest effect on glycemic control and insulin dose, and in C-peptide positive patients there is evidence of a greater reduction in HbA1c compared with C-peptide negative patients, said Dejgaard, explaining the motivation behind the study.
“Also, lower rates of symptomatic hypoglycemia and very few episodes of hyperglycemia with ketosis have been seen with GLP-1 agonist treatment. In vitro and rodent studies suggest GLP-1 agonists also increase beta-cell mass and prevent beta-cell apoptosis,” he added.
C-Peptide Level Significantly Higher With Liraglutide After 1 Year
In light of this background research, the investigators decided to examine the effect of liraglutide on residual beta-cell function as an adjunctive therapy to insulin in patients with newly diagnosed type 1 diabetes.
Patients were required to have received a diagnosis of type 1 diabetes 6 weeks prior to study start and all were aged 18-40 years. C-peptide level was assessed postprandially and required to be ≥ 200 pmol/L.
In total, 68 patients were randomized (1:1) to liraglutide 1.8 mg plus insulin or placebo plus insulin for 52 weeks followed by 6 weeks on insulin only. Upon randomization, end of treatment, and follow-up, a test was conducted using Boost liquid mixed meal to evaluate C-peptide response.
Patients were well-matched, with a fasting C-peptide of 100 pmol/L in both groups, almost all were antibody positive, and mean diabetes duration was 4.5 weeks. Five patients discontinued treatment, three in the liraglutide arm.
The primary endpoint was the ratio of the area under the curve (AUC) for C-peptide divided by plasma glucose.
Those randomized to liraglutide had a 44% higher AUC C-peptide than the placebo group (P = .04).
Six weeks after study end (and discontinuation of liraglutide), this effect disappeared.
“Patients treated with liraglutide had a significantly higher C-peptide value 1 year after diagnosis of type 1 diabetes compared to those on placebo,” Dejgaard highlighted.
Secondary endpoints included daily insulin dose needed, body weight, self-monitored blood glucose profile, hypoglycemia, and adverse events.
There was a significant reduction in insulin dose in patients taking liraglutide of almost 50%, Dejgaard noted. Over the treatment period, patients treated with liraglutide showed a change of –9 IU compared to +12 IU in patients taking placebo (P < .001).
Again, 6 weeks after liraglutide treatment ended, no difference was seen.
There was no difference between the groups in HbA1c overall, but patients taking liraglutide had a lower HbA1c at 12 and 36 weeks. At the end of treatment, there was no difference between groups.
For body weight, those on liraglutide showed a nonsignificant decrease soon after initiation of study drug; however, overall, there was no difference between groups.
“In fact, after 52 weeks, both groups increased their body weight,” Dejgaard pointed out.
Over 58 weeks, episodes of self-reported mild hypoglycemia (blood glucose < 3.9 mmol/L) were 1860 events with liraglutide vs 2147 events with placebo (P < .05). There were no severe hypoglycemic events.
“Other adverse events were similar to those already known, so nausea and dyspepsia, but actually nausea was quite high in the liraglutide patients at nearly 50%,” reported Dejgaard. “There were five severe adverse events but none related to study drug, and there was no diabetic ketoacidosis.”
But Is There Added Benefit to a GLP-1 Agonist in Type 1 Diabetes?
“This is an important paper because it confirms data on dulaglutide [another GLP-1 agonist] in the past in latent autoimmune diabetes,” an audience member commented after the presentation.
“Substantial residual beta-cell function can take advantage of GLP-1 at diagnosis. This gives us a new way to look at newly diagnosed type 1 diabetes.”
“It would be good to do a subanalysis of patients according to antibody titer,” the audience member added.
Dejgaard said his group plan to do such analyses despite the study being small.
Reflecting on the long-term view and providing some perspective, Gunasekaran pointed out, “Ultimately, I think the issue is that we’ve been doing this for years but it doesn’t get to the fundamental problem that everyone will end up on insulin anyway, so is there any added benefit to using a medication like liraglutide? You could use it, but in the long term that might not be a feasible solution.”
Dejgaard has reported being a consultant for Novo Nordisk, AstraZeneca, and Boehringer Ingelheim. He has received research support from Novo Nordisk and AstraZeneca. Gunasekaran has reported no relevant financial relationships.
ADA 2019 Scientific Sessions. Presented June 8, 2019. Abstract 59-OR