SURPASS-2
Study design
- Randomized, open-label, phase 3 trial
- 1,879 patients randomized 1:1:1:1
- 40-week treatment period
- Patients remained on background metformin
- Primary outcome: change in A1c from baseline to week 40
- Noninferiority margin 0.3 percentage points
Population
- Adults with type 2 diabetes inadequately controlled on metformin
- Mean age 56.6 years
- Mean diabetes duration 8.6 years
- Mean baseline A1c 8.28%
- Mean baseline weight 93.7 kg
- Type 1 diabetes
- eGFR <45 mL/min/1.73 m²
- History of pancreatitis
- Proliferative diabetic retinopathy
- Diabetic maculopathy
- Retinopathy requiring urgent treatment
Interventions
- 5 mg, 10 mg, or 15 mg subcutaneous once weekly
- Started at 2.5 mg weekly
- Increased by 2.5 mg every 4 weeks until assigned dose
- 1 mg subcutaneous once weekly
- Started at 0.25 mg weekly
- Dose doubled every 4 weeks to 1 mg
Primary outcome
A1c reduction
Mean change from baseline at 40 weeks
Tirzepatide was superior to semaglutide 1 mg for A1c reduction at every tested dose
Weight loss
−7.6 kg
1.9 kg greater loss than semaglutide
−9.3 kg
3.6 kg greater loss than semaglutide
−11.2 kg
5.5 kg greater loss than semaglutide
−5.7 kg
Additional outcomes
82–86% with tirzepatide
79% with semaglutide
69–80% with tirzepatide
64% with semaglutide
27–46% with tirzepatide
19% with semaglutide
15–36% with tirzepatide
8% with semaglutide
Safety
17–22% with tirzepatide
18% with semaglutide
13–16% with tirzepatide
12% with semaglutide
6–10% with tirzepatide
8% with semaglutide
0.6%, 0.2%, and 1.7% with tirzepatide
0.4% with semaglutide
5–7% with tirzepatide
3% with semaglutide
GI adverse effects were most common and usually mild to moderate
Clinical interpretation
- Tirzepatide lowered A1c more than semaglutide 1 mg at all tested doses
- Weight loss increased substantially with higher tirzepatide doses
- Hypoglycemia was uncommon with both therapies in patients receiving metformin without insulin or sulfonylureas
- GI adverse effects were common with both incretin-based therapies
- SURPASS-2 established strong glycemic and weight-loss efficacy for tirzepatide in T2DM
Limitations
- Open-label trial
- Only 40 weeks of treatment
- Semaglutide was tested at 1 mg rather than the subsequently approved 2 mg diabetes dose
- Trial was designed for glycemic efficacy rather than cardiovascular outcomes
- Patients were receiving background metformin, limiting extrapolation to other treatment settings
- Funded by Eli Lilly, the manufacturer of tirzepatide