SUSTAIN-7 deep-dive: semaglutide vs dulaglutide once-weekly in T2D (Pratley 2018)
Last verified 2026-05-28 · Phase 3b · Completed (results published Feb 1, 2018) · NCT02648204 ↗
SUSTAIN-7 (NCT02648204) is the head-to-head phase-3b trial that directly pitted two of the most-prescribed once-weekly GLP-1 receptor agonists against each other in adults with type 2 diabetes. Novo Nordisk randomized 1,201 patients on stable metformin monotherapy across 194 sites in 16 countries 1:1:1:1 to once-weekly subcutaneous semaglutide 0.5 mg, dulaglutide 0.75 mg, semaglutide 1.0 mg, or dulaglutide 1.5 mg, in an open-label design with 40 weeks of treatment plus a 5-week safety follow-up. The primary endpoint was change in HbA1c from baseline to week 40, with body weight as the confirmatory secondary endpoint. Results published by Pratley and colleagues in The Lancet Diabetes & Endocrinology on February 1, 2018 showed that at both matched-tier comparisons (low-dose and high-dose), semaglutide produced larger HbA1c reductions and meaningfully greater weight loss than dulaglutide, with comparable gastrointestinal-event rates.
- Enrollment
- 1,201
- Duration
- 40 weeks of randomized treatment plus a 5-week safety follow-up (final visit at week 45)
- Drug
- Semaglutide (vs dulaglutide once-weekly, head-to-head)
- Population
- Adults aged 18 and older with type 2 diabetes and HbA1c 7.0%-10.5% (53-91 mmol/mol) on stable metformin monotherapy (≥1,500 mg/day or maximum tolerated dose) for at least 90 days before screening. Patients with a history of pancreatitis, a personal or family history of medullary thyroid carcinoma or MEN2, eGFR <60 mL/min/1.73 m², NYHA class IV heart failure, or proliferative retinopathy were excluded. Across the four arms (n = 301/299/300/299), mean baseline HbA1c was 8.2%, mean body weight was 95.2 kg, mean BMI was about 33 kg/m², and mean diabetes duration was 7.4 years.
Primary endpoint
Change from baseline in HbA1c at week 40 — semaglutide vs dulaglutide at matched dose tiers (on-treatment without rescue medication, primary analysis)
Treatment arm
Semaglutide 0.5 mg: −1.5 percentage points (SE 0.06); semaglutide 1.0 mg: −1.8 percentage points (SE 0.06)
Comparator
Dulaglutide 0.75 mg: −1.1 percentage points (SE 0.05); dulaglutide 1.5 mg: −1.4 percentage points (SE 0.06)
Treatment difference: Low-dose comparison (sema 0.5 vs dula 0.75): ETD −0.40 pp (95% CI −0.55 to −0.25; P<0.0001). High-dose comparison (sema 1.0 vs dula 1.5): ETD −0.41 pp (95% CI −0.57 to −0.25; P<0.0001).
Mean baseline HbA1c was 8.2% across arms. Semaglutide was superior to dulaglutide at both dose tiers — the high-dose contrast (1.0 mg sema vs 1.5 mg dula) is the comparison most often quoted in clinical guidelines because both are the highest then-approved T2D doses for each drug.
Secondary endpoints
| Endpoint | Treatment | Comparator | Difference |
|---|---|---|---|
| Change from baseline in body weight (kg) at week 40 — confirmatory secondary, low-dose comparison From a mean baseline of about 96 kg, semaglutide 0.5 mg roughly doubled the weight reduction seen with dulaglutide 0.75 mg. | Semaglutide 0.5 mg: −4.6 kg (SE 0.28) | Dulaglutide 0.75 mg: −2.3 kg (SE 0.27) | ETD −2.26 kg (95% CI −3.02 to −1.51; P<0.0001 for superiority) |
| Change from baseline in body weight (kg) at week 40 — confirmatory secondary, high-dose comparison The 3.5-kg gap at the high-dose contrast is the headline weight-loss finding of SUSTAIN-7 and the basis for clinical preference of semaglutide when weight reduction is a co-target with glycemic control. | Semaglutide 1.0 mg: −6.5 kg (SE 0.28) | Dulaglutide 1.5 mg: −3.0 kg (SE 0.27) | ETD −3.55 kg (95% CI −4.32 to −2.78; P<0.0001 for superiority) |
| HbA1c <7.0% (53 mmol/mol; ADA target) at week 40 Nearly four out of five patients on semaglutide 1.0 mg reached the standard ADA glycemic target by week 40. | Semaglutide 0.5 mg: 68%; semaglutide 1.0 mg: 79% | Dulaglutide 0.75 mg: 52%; dulaglutide 1.5 mg: 67% | Odds ratio favored semaglutide at both tiers; P<0.0001 vs dulaglutide for both comparisons |
| HbA1c ≤6.5% (48 mmol/mol; AACE target) at week 40 The stricter AACE target is reached by two-thirds of high-dose semaglutide patients and fewer than half of high-dose dulaglutide patients. | Semaglutide 0.5 mg: 47%; semaglutide 1.0 mg: 67% | Dulaglutide 0.75 mg: 30%; dulaglutide 1.5 mg: 45% | Odds ratio favored semaglutide at both tiers; P<0.0001 |
| Weight loss ≥5% at week 40 | Semaglutide 0.5 mg: 44%; semaglutide 1.0 mg: 63% | Dulaglutide 0.75 mg: 23%; dulaglutide 1.5 mg: 30% | Odds ratio favored semaglutide at both tiers; P<0.0001 |
| Weight loss ≥10% at week 40 More than one in five patients on semaglutide 1.0 mg achieved ≥10% weight loss — a threshold rarely reached by dulaglutide 1.5 mg in this trial. | Semaglutide 0.5 mg: 14%; semaglutide 1.0 mg: 22% | Dulaglutide 0.75 mg: 3%; dulaglutide 1.5 mg: 8% | Odds ratio favored semaglutide at both tiers; P<0.0001 |
| Change from baseline in fasting plasma glucose at week 40 (mmol/L) | Semaglutide 0.5 mg: −2.2 mmol/L; semaglutide 1.0 mg: −2.8 mmol/L | Dulaglutide 0.75 mg: −1.9 mmol/L; dulaglutide 1.5 mg: −2.2 mmol/L | Reductions favored semaglutide at both tiers |
| Change from baseline in systolic blood pressure at week 40 (mmHg) Blood-pressure reduction was modest in every arm and broadly comparable between drugs. | Semaglutide 0.5 mg: −2.7 mmHg; semaglutide 1.0 mg: −5.0 mmHg | Dulaglutide 0.75 mg: −2.3 mmHg; dulaglutide 1.5 mg: −4.3 mmHg | Similar small reductions across arms; not a primary differentiator |
| Change from baseline in fasting lipid profile at week 40 (ratio to baseline) | Total cholesterol, LDL, and triglycerides decreased modestly with semaglutide at both doses; HDL essentially unchanged | Comparable directional changes with dulaglutide at both doses; smaller magnitudes | Numerical differences favored semaglutide on triglycerides; lipid changes were not prespecified primary differentiators |
| Composite endpoint: HbA1c <7.0% without severe or BG-confirmed symptomatic hypoglycemia and no weight gain at week 40 This composite — sometimes called the "triple goal" — captures the clinical ideal of reaching glycemic target without trading off hypoglycemia or weight, and is the endpoint most often cited when SUSTAIN-7 is summarized in formulary reviews. | Semaglutide 0.5 mg: 60%; semaglutide 1.0 mg: 74% | Dulaglutide 0.75 mg: 41%; dulaglutide 1.5 mg: 53% | Odds ratio favored semaglutide at both tiers; P<0.0001 |
Adverse events
| Event | Treatment rate | Comparator rate |
|---|---|---|
| Any gastrointestinal disorder GI events were the most common AE category in every arm. At matched high doses, the GI burden was numerically similar between semaglutide 1.0 mg and dulaglutide 1.5 mg. | Semaglutide 0.5 mg: 43% (129/301); semaglutide 1.0 mg: 44% (133/300) | Dulaglutide 0.75 mg: 33% (100/299); dulaglutide 1.5 mg: 48% (143/299) |
| Nausea Most common single GI event; typically mild-to-moderate and peaks during dose-escalation. | Semaglutide 0.5 mg: ≈20%; semaglutide 1.0 mg: ≈22% | Dulaglutide 0.75 mg: ≈13%; dulaglutide 1.5 mg: ≈20% |
| Diarrhea Comparable across arms. | Semaglutide 0.5 mg: ≈13%; semaglutide 1.0 mg: ≈15% | Dulaglutide 0.75 mg: ≈12%; dulaglutide 1.5 mg: ≈17% |
| Vomiting | Semaglutide 0.5 mg: ≈7%; semaglutide 1.0 mg: ≈8% | Dulaglutide 0.75 mg: ≈5%; dulaglutide 1.5 mg: ≈11% |
| Severe or BG-confirmed symptomatic hypoglycemia Severe-or-BG-confirmed symptomatic hypoglycemia was uncommon in every arm and broadly comparable between drugs in this metformin-only background population. | Semaglutide 0.5 mg: 1.7%; semaglutide 1.0 mg: 0.7% | Dulaglutide 0.75 mg: 1.3%; dulaglutide 1.5 mg: 2.0% |
| Premature treatment discontinuation due to adverse events Gastrointestinal events were the most common reason for discontinuation in every arm. | Semaglutide 0.5 mg: 8%; semaglutide 1.0 mg: 10% | Dulaglutide 0.75 mg: 4%; dulaglutide 1.5 mg: 7% |
| Diabetic retinopathy complications A small numerical imbalance was observed, consistent with the broader SUSTAIN program (most prominently SUSTAIN-6) and is the basis for the retinopathy precaution in the semaglutide label. | Semaglutide 0.5 mg: 1.3%; semaglutide 1.0 mg: 2.3% | Dulaglutide 0.75 mg: 1.0%; dulaglutide 1.5 mg: 0.7% |
| Fatal adverse events Six total deaths in the trial; none were judged by investigators to be related to study drug. | Semaglutide 0.5 mg: 1 death; semaglutide 1.0 mg: 1 death | Dulaglutide 0.75 mg: 2 deaths; dulaglutide 1.5 mg: 2 deaths |
Clinical significance
SUSTAIN-7 is the trial that established semaglutide as the strongest A1C-reducing GLP-1 receptor agonist on the market at the time of publication. Until February 2018, comparisons between once-weekly GLP-1s relied on indirect cross-trial inference. SUSTAIN-7 settled the question head-to-head: at both matched dose tiers, semaglutide cut HbA1c by roughly an additional 0.4 percentage points and produced 2.3 to 3.5 kg of additional weight loss versus dulaglutide, with a similar tolerability profile. The findings reshaped diabetes-society guideline language on GLP-1 selection, pushed payers to differentiate between drugs that had previously been lumped together as a class, and laid the groundwork for the SUSTAIN-6 cardiovascular-outcomes data and the obesity-dose semaglutide program (STEP-1 onward) to position semaglutide as the reference GLP-1 against which next-generation incretins would be benchmarked.
Frequently asked questions
Why did semaglutide beat dulaglutide on HbA1c?
The two drugs share a mechanism — long-acting GLP-1 receptor agonism — but their pharmacology differs. Semaglutide has higher receptor affinity at the GLP-1 receptor than dulaglutide and reaches higher steady-state exposure at its approved T2D doses (0.5 mg and 1.0 mg). SUSTAIN-7 was designed to compare matched dose tiers (0.5 vs 0.75 mg and 1.0 vs 1.5 mg) and showed that at both tiers semaglutide cut HbA1c by an extra 0.4 percentage points (estimated treatment difference −0.40 pp at the low tier and −0.41 pp at the high tier, both P<0.0001). The trial was powered for HbA1c non-inferiority with a margin of 0.4%, and the result cleared not only non-inferiority but statistical superiority at both contrasts.
Should patients on Trulicity switch to Ozempic?
SUSTAIN-7 is a head-to-head trial, not a switch study, so it cannot directly tell us whether switching from dulaglutide to semaglutide adds incremental benefit on top of dulaglutide already-achieved A1C and weight reductions. What it does show is that in a randomized population, semaglutide produces about 0.4 percentage points more HbA1c reduction and 2.3 to 3.5 kg more weight loss at matched dose tiers. Whether that gap is worth a switch depends on whether the patient is at goal on dulaglutide, how much additional weight or glycemic benefit they need, payer coverage, and whether the higher-dose semaglutide 2.0 mg (approved later under SUSTAIN FORTE data) or the obesity-dose 2.4 mg (Wegovy) is in scope. The treatment decision belongs to the prescribing clinician.
How big was the weight-loss difference, and was it clinically meaningful?
From a mean baseline body weight of roughly 96 kg, semaglutide 0.5 mg cut weight by 4.6 kg vs 2.3 kg with dulaglutide 0.75 mg (estimated treatment difference −2.26 kg; P<0.0001) and semaglutide 1.0 mg cut weight by 6.5 kg vs 3.0 kg with dulaglutide 1.5 mg (estimated treatment difference −3.55 kg; P<0.0001). On categorical responder analyses, 22% of semaglutide 1.0 mg patients reached ≥10% weight loss versus 8% on dulaglutide 1.5 mg. That gap is clinically meaningful in the context of co-occurring obesity — it is the size of difference that often determines whether a patient drops a BMI category, qualifies for a different anti-obesity-medication step, or sees a measurable change in cardiometabolic risk markers.
How does SUSTAIN-7 compare to SURPASS-2?
Both are open-label, head-to-head, phase-3 trials in T2D patients on metformin. SUSTAIN-7 (n = 1,201, 40 weeks) compared semaglutide (0.5 and 1.0 mg) against dulaglutide (0.75 and 1.5 mg), the leading GLP-1 monotherapy. SURPASS-2 (n = 1,879, 40 weeks) compared tirzepatide (5, 10, 15 mg) against the highest then-approved T2D semaglutide dose (1.0 mg). The hierarchy that emerged across the two trials is: dulaglutide 1.5 mg cut A1C by −1.4 pp and weight by 3.0 kg; semaglutide 1.0 mg cut A1C by −1.8 pp and weight by 6.5 kg (SUSTAIN-7); tirzepatide 15 mg cut A1C by −2.30 pp and weight by 11.2 kg (SURPASS-2). Each step up the ladder adds roughly 0.4 percentage points of A1C reduction and several kilograms of weight loss — a useful framework for understanding why class-level statements about “GLP-1s” obscure real clinical differences.
Was the rate of hypoglycemia comparable between the two drugs?
Yes. Severe-or-BG-confirmed symptomatic hypoglycemia was uncommon and broadly comparable in every arm — 1.7% on semaglutide 0.5 mg, 0.7% on semaglutide 1.0 mg, 1.3% on dulaglutide 0.75 mg, and 2.0% on dulaglutide 1.5 mg. SUSTAIN-7 used a metformin-only background, which keeps hypoglycemia risk low for any GLP-1; in patients also taking sulfonylureas or insulin, hypoglycemia rates would be higher with both drugs and the trial does not directly inform that scenario. The composite endpoint of “HbA1c <7% without hypoglycemia and without weight gain” was reached by 74% of semaglutide-1.0-mg patients versus 53% of dulaglutide-1.5-mg patients, suggesting that the higher glycemic effect of semaglutide did not come at the cost of more hypoglycemia in this population.
Did the gastrointestinal-tolerability profile differ between semaglutide and dulaglutide?
The overall rate of gastrointestinal adverse events was similar at the matched high dose: 44% on semaglutide 1.0 mg and 48% on dulaglutide 1.5 mg. At the low-dose tier, semaglutide 0.5 mg had numerically more GI events than dulaglutide 0.75 mg (43% vs 33%). Premature discontinuation for adverse events ran 10% on semaglutide 1.0 mg and 7% on dulaglutide 1.5 mg, with GI events the most common reason in every arm. The picture is that semaglutide buys its incremental glycemic and weight effect at roughly the same GI cost as dulaglutide rather than substantially more, which is part of why the trial result moved guideline language and payer formularies.
References
- 1.Pratley RE, Aroda VR, Lingvay I, et al.; SUSTAIN 7 investigators. Semaglutide versus dulaglutide once weekly in patients with type 2 diabetes (SUSTAIN 7): a randomised, open-label, phase 3b trial. Lancet Diabetes Endocrinol. 2018. PMID: 29397376.
- 2.U.S. National Library of Medicine. Efficacy and Safety of Semaglutide Versus Dulaglutide as add-on to Metformin in Subjects With Type 2 Diabetes (SUSTAIN 7) — Study Record. ClinicalTrials.gov, NCT02648204. 2018. https://clinicaltrials.gov/study/NCT02648204
- 3.Frías JP, Davies MJ, Rosenstock J, et al.; SURPASS-2 Investigators. Tirzepatide versus Semaglutide Once Weekly in Patients with Type 2 Diabetes. N Engl J Med. 2021. PMID: 34170647.
- 4.U.S. Food and Drug Administration. Ozempic (semaglutide) injection — Original FDA Approval Letter and Prescribing Information (December 5, 2017). Drugs@FDA, NDA 209637. 2017. https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/209637lbl.pdf