SURPASS-1 shows tirzepatide monotherapy cuts HbA1c by about two points

Results from SURPASS-1, the first phase 3 trial of tirzepatide used on its own, show that the weekly injection lowered average blood sugar by close to two percentage points over 40 weeks in adults with type 2 diabetes who had not yet started injectable diabetes treatment ^[1]. Tirzepatide is the dual GIP and GLP-1 receptor agonist later marketed as Mounjaro for type 2 diabetes and Zepbound for weight management.

What the trial did

SURPASS-1 was a 40-week, double-blind, randomized, placebo-controlled study run at 52 medical research centers and hospitals in India, Japan, Mexico and the United States ^[1]. Adults 18 and older with type 2 diabetes that was not controlled by diet and exercise alone, and who had never used an injectable diabetes therapy, were assigned in equal numbers to tirzepatide 5 mg, 10 mg or 15 mg once a week, or to placebo ^[1]. Participants, investigators and the sponsor were all masked to treatment assignment ^[1]. The main measure was the average change in HbA1c from the start of the study to week 40 ^[1]. The study is registered as NCT03954834 ^[2].

Of 705 people screened between June 3, 2019 and October 28, 2020, 478 were randomized: 121 to each tirzepatide dose and 115 to placebo ^[1]. At the start, average HbA1c was 7.9% (63 mmol/mol), average age was 54.1 years, 231 participants (48%) were women, average diabetes duration was 4.7 years, and average body-mass index was 31.9 kg/m² ^[1]. Sixty-six participants (14%) stopped the study drug and 50 (10%) left the study early ^[1].

The numbers

Average HbA1c fell by 1.87 percentage points with 5 mg, 1.89 points with 10 mg and 2.07 points with 15 mg, while it rose by 0.04 points on placebo ^[1]. That works out to differences versus placebo of −1.91, −1.93 and −2.11 percentage points, all with p values below 0.0001 ^[1].

More people on tirzepatide reached standard glucose targets: 87% to 92% got below 7.0%, compared with 20% on placebo, and 81% to 86% reached 6.5% or lower, versus 10% on placebo ^[1]. Between 31% and 52% of tirzepatide participants reached an HbA1c below 5.7% — a level below the usual threshold for diabetes — compared with 1% on placebo ^[1].

Weight loss was dose-dependent, ranging from 7.0 to 9.5 kg (roughly 15 to 21 pounds) ^[1]. The most common side effects were mild-to-moderate, temporary stomach and gut problems: nausea in 12% to 18% of tirzepatide users versus 6% on placebo, diarrhea in 12% to 14% versus 8%, and vomiting in 2% to 6% versus 2% ^[1]. No clinically significant low blood sugar (below 54 mg/dL) or severe hypoglycemia was reported with tirzepatide ^[1]. One death occurred, in the placebo group ^[1]. The trial was funded by Eli Lilly and Company ^[1].

Why it matters for patients

SURPASS-1 tested tirzepatide by itself, without metformin or other background diabetes drugs, in people relatively early in the disease — average diabetes duration under five years and a starting HbA1c near 7.9% ^[1]. That is a different situation from trials in people already on several medications, so the size of the effect here may not carry over to everyone.

The absence of clinically significant or severe hypoglycemia on tirzepatide matters because low blood sugar is a common limit on how aggressively diabetes is treated with insulin or sulfonylureas ^[1]. The stomach-related side effects, while more common than with placebo, were described as mild to moderate and temporary ^[1].

Some caveats are built into the design. The comparison was against placebo, not against an active drug like semaglutide, so this trial alone does not say how tirzepatide compares with existing GLP-1 medicines ^[1]. The study lasted 40 weeks, so longer-term durability and safety are not answered here ^[1]. And 14% of participants stopped the study drug ^[1].

What happens next

Enrollment ran from June 3, 2019 to October 28, 2020, and the results were published in The Lancet in 2021 ^[1]. A correction notice, "Department of Error," was published in The Lancet on July 17, 2021 ^[1]. The sources do not state what regulatory decisions followed or when.