Semaglutide 2.0 mg lowers blood sugar more than 1.0 mg in 40-week trial

A 40-week trial found that a 2.0 mg weekly dose of semaglutide lowered HbA1c more than the 1.0 mg dose in adults with type 2 diabetes that was not well controlled on metformin, with slightly greater weight loss and a similar safety profile ^[1].

Semaglutide 2.0 mg lowered HbA1c by an average of 2.2 percentage points from baseline to week 40, compared with 1.9 percentage points for 1.0 mg ^[1]. The estimated treatment difference was 0.23 percentage points (95% CI 0.36 to 0.11; p=0.0003) under the trial product estimand, an analysis that counts only participants who stayed on treatment without rescue medication ^[1]. Under the treatment policy estimand, which includes everyone regardless of whether they stopped treatment or needed rescue medication, the reductions were 2.1 and 1.9 percentage points (treatment difference 0.18 points; 95% CI 0.31 to 0.04; p=0.0098) ^[1].

For bodyweight, the trial product estimand showed an average loss of 6.9 kg with 2.0 mg and 6.0 kg with 1.0 mg (treatment difference 0.93 kg; 95% CI 1.68 to 0.18; p=0.015) ^[1]. Under the treatment policy estimand, the losses were 6.4 kg and 5.6 kg (difference 0.77 kg; 95% CI 1.55 to 0.01; p=0.054), which did not reach the threshold the trial set for statistical significance ^[1].

How the trial was run

The study, called SUSTAIN FORTE, was a phase 3B, double-blind, randomized, active-controlled, parallel-group trial at 125 outpatient clinics in ten countries ^[1]. Researchers assessed 1,515 adults and enrolled 961, randomly assigning 480 to semaglutide 2.0 mg and 481 to 1.0 mg once weekly ^[1]. Participants had a mean age of 58.0 years (SD 10.0), and 398 of them (41%) were women ^[1]. At the start, their average HbA1c was 8.9% (SD 0.6) and average BMI was 34.6 kg/m2 (SD 7.0) ^[1]. To qualify, participants had to be 18 or older with an HbA1c between 8.0% and 10.0% while taking a stable dose of metformin, with or without a sulfonylurea ^[1]. Enrollment ran from June 19 to Nov 28, 2019 ^[1]. By the end, 462 (96%) of the 2.0 mg group and 471 (98%) of the 1.0 mg group had completed the trial ^[1].

Gastrointestinal disorders were the most commonly reported adverse events, affecting 163 participants (34%) in the 2.0 mg group and 148 (31%) in the 1.0 mg group ^[1]. Serious adverse events were similar between the groups: 21 (4%) with 2.0 mg and 25 (5%) with 1.0 mg ^[1]. Three deaths were reported during the trial, one in the 1.0 mg group and two in the 2.0 mg group ^[1]. The paper does not report the causes of those deaths or whether investigators judged them related to treatment.

The authors concluded that 2.0 mg was superior to 1.0 mg for lowering HbA1c, with additional bodyweight loss and a similar safety profile, and described the higher dose as an option for intensifying treatment in patients who need more blood sugar control ^[1]. The trial was funded by Novo Nordisk, and three of the listed authors are full-time employees of the company who own shares in it ^[1]. The study is registered as NCT03989232 ^[1].

Why it matters for patients

About 20% to 30% of patients given semaglutide 1.0 mg do not reach their blood sugar goals, according to the paper's background section ^[1]. This trial was designed to test whether a higher dose helps in that situation, and it found a real but modest additional effect: roughly two-tenths of a percentage point more HbA1c reduction and about nine-tenths of a kilogram (about 2 pounds) more weight loss on the main analysis ^[1].

The extra weight benefit is less certain than the blood sugar benefit. When everyone was counted, including people who stopped treatment or took rescue medication, the difference in weight loss was small enough that it was not statistically significant ^[1]. That means the added weight effect may be smaller in real-world use than the main analysis suggests.

Side effects are a practical consideration with this drug class, and in this trial the higher dose did not produce notably more gastrointestinal complaints: about one-third of each group reported them ^[1].

The trial ran for 40 weeks and compared two doses of the same drug. It did not include a placebo group, so it does not show how either dose compares with no semaglutide, and it does not compare semaglutide with other diabetes drugs. Long-term safety data, effects on heart and kidney outcomes, and results beyond 40 weeks have not been reported in this paper ^[1]. Participants were also taking metformin, with or without a sulfonylurea, so the findings may not apply to people on other treatment regimens ^[1].

One limitation worth noting: the two analysis methods used in the trial gave somewhat different answers, particularly for weight, and the trial's conclusions rest mainly on the primary HbA1c outcome ^[1].