Three-year SCALE trial finds liraglutide 3.0 mg reduced diabetes diagnoses in adults with prediabetes

Adults with prediabetes who took liraglutide 3.0 mg for three years were diagnosed with type 2 diabetes at roughly one-fifth the rate of those taking a placebo, according to results from the SCALE Obesity and Prediabetes trial published in The Lancet on Feb. 23, 2017 ^[1].

The trial ran at 191 research sites across 27 countries and enrolled 2,254 adults with prediabetes. Participants had a body-mass index of at least 30, or at least 27 if they had other health conditions. They were assigned 2-to-1: 1,505 people received a once-daily liraglutide 3.0 mg injection under the skin, and 749 received a matching placebo shot. Both groups also followed a reduced-calorie diet and increased their physical activity. Enrollment ran from June 1, 2011, to March 2, 2015 ^[1].

By week 160, type 2 diabetes had been diagnosed while on treatment in 26 of 1,472 people in the liraglutide group, or 2%, and in 46 of 738 people on placebo, or 6% ^[1]. The average stretch from joining the study to a diagnosis was 99 weeks for the liraglutide group and 87 weeks for the placebo group ^[1]. Taking into account how differently the two groups were diagnosed, the researchers calculated that the time to diabetes onset over 160 weeks was 2.7 times longer with liraglutide than with placebo (95% CI 1.9 to 3.9, p<0.0001), which they reported as a hazard ratio of 0.21 (95% CI 0.13 to 0.34) ^[1].

Weight fell more sharply with the drug. At week 160, the liraglutide group had lost an average of 6.1% of body weight, against 1.9% for placebo, a difference of 4.3 percentage points (95% CI 4.9 to 3.7, p<0.0001) ^[1]. Serious adverse events were reported by 227 of 1,501 treated liraglutide participants (15%) and 96 of 747 placebo participants (13%) ^[1].

Not everyone stayed. Only 1,128 participants, half of those enrolled, completed the full 160 weeks. Withdrawals reached 714 people (47%) in the liraglutide group and 412 (55%) in the placebo group. The investigators said that people who quit were not followed after they stopped treatment, and they called that a limitation of the study ^[1]. Their conclusion was cautious: liraglutide 3.0 mg "might provide health benefits in terms of reduced risk of diabetes in individuals with obesity and prediabetes" ^[1].

Novo Nordisk, the Danish company that makes liraglutide, funded the trial ^[1]. A Department of Error correction notice for the paper appeared in The Lancet on April 8, 2017 ^[1]. The journal also carried commentary asking whether GLP-1 drugs are the answer for prediabetes in people with obesity, and whether liraglutide prevents a diabetes diagnosis or only puts it off ^[1].

Why it matters for patients

The medicine tested here is liraglutide 3.0 mg, an older GLP-1 drug given as a daily injection. Nothing in this paper describes semaglutide (sold as Ozempic, Wegovy and Rybelsus), tirzepatide (Mounjaro and Zepbound), orforglipron or any other drug in the class, so these percentages cannot be transferred to them ^[1].

The diagnoses counted were made while people were still taking the drug, and every participant was also eating less and moving more. The 2% and 6% figures therefore reflect a whole treatment program rather than the injection by itself, and they describe only the people still enrolled — by week 160, half the original group had left ^[1].

Because no one who withdrew was followed afterward, the trial cannot show whether liraglutide delayed diabetes, prevented it, or produced a benefit that fades once the injections stop ^[1].

What is not known

Whether the lower diagnosis rate held up after treatment ended has not been reported, since the paper offers no follow-up data past the treatment period ^[1]. The authors also did not report whether the effect tracked with how much weight each person lost ^[1]. No results in this publication address semaglutide, tirzepatide, orforglipron or any other GLP-1 or related medicine for preventing type 2 diabetes ^[1].