SUSTAIN 7 trial results favored semaglutide over dulaglutide for blood sugar and weight at the doses tested. Over 40 weeks, the main blood-sugar measure fell about 0.4 percentage points more with semaglutide in both comparisons. The trial enrolled people with type 2 diabetes. Its results don't rank every weight-loss treatment or predict one person's outcome.
What question did SUSTAIN 7 ask?
Could two weekly diabetes medicines produce different results in people whose blood sugar remained high on metformin, a common diabetes medicine?
Both drugs mimic GLP-1, a gut hormone involved in blood sugar and appetite. The trial compared semaglutide 0.5 mg with dulaglutide 0.75 mg, and semaglutide 1.0 mg with dulaglutide 1.5 mg. These were two planned comparisons, not equivalent doses for switching medicines. Everyone continued metformin; there was no placebo, or inactive treatment.
Richard Pratley and colleagues published the findings in The Lancet Diabetes & Endocrinology in 2018. Novo Nordisk funded, designed and analyzed the trial.
Who took part, and for how long?
Researchers randomly assigned 1,201 adults at 194 sites in 16 countries. Participants were at least 18, with group averages around 55–56 years. They had type 2 diabetes and were taking metformin alone.
Their HbA1c, a measure of average blood sugar over roughly three months, had to be 7.0%–10.5%. People with moderately reduced kidney function or worse were excluded, as were those with past pancreatitis, or inflammation of the pancreas. Certain advanced eye conditions requiring urgent treatment also ruled people out.
Treatment lasted 40 weeks, followed by five weeks of follow-up. The trial was open-label: participants and clinicians knew which medicine each person received. Of those assigned, 1,199 received medication; one person in each dulaglutide group did not.
What was measured, and what happened after extra treatment?
The primary endpoint, the main result chosen in advance, was the change in HbA1c at week 40. Weight change was a planned secondary measure. The registry record, NCT02648204, records both.
The headline analysis used readings while people remained on assigned treatment and before rescue medication, an extra diabetes drug for persistently high blood sugar. A statistical model accounted for missing readings. A check that also counted readings taken after people stopped treatment or added rescue medication still favored semaglutide, by a little less: about 0.3 HbA1c points in each comparison (study appendix).
SUSTAIN 7 trial results: blood sugar and weight
Semaglutide's estimated HbA1c advantage was 0.40 percentage points in the lower-dose pair and 0.41 in the higher-dose pair. These differences cleared the study's statistical tests.
| Weekly study treatment | People who received medication | Average HbA1c change at week 40 | Average weight change |
|---|---|---|---|
| Semaglutide 0.5 mg | 301 | −1.5 percentage points | −4.6 kg |
| Dulaglutide 0.75 mg | 299 | −1.1 percentage points | −2.3 kg |
| Semaglutide 1.0 mg | 300 | −1.8 percentage points | −6.5 kg |
| Dulaglutide 1.5 mg | 299 | −1.4 percentage points | −3.0 kg |
A percentage-point change is a subtraction between percentages. A fall from 8.2% to 6.7%, for example, is 1.5 points. That example explains the units; it isn't an individual forecast.
The modeled weight differences favored semaglutide by 2.26 kg and 3.55 kg, respectively. Those are differences between trial groups taking active medicines, not comparisons with no treatment.
Stomach symptoms and stopping treatment
About one in three to one in two participants reported gastrointestinal symptoms, meaning stomach or bowel complaints. Nausea, vomiting and diarrhea were common. The paper's Table 4 reports:
| Reported event | Semaglutide 0.5 mg, n=301 | Dulaglutide 0.75 mg, n=299 | Semaglutide 1.0 mg, n=300 | Dulaglutide 1.5 mg, n=299 |
|---|---|---|---|---|
| Any stomach or bowel complaint | 129 (43%) | 100 (33%) | 133 (44%) | 143 (48%) |
| Nausea | 23% | 13% | 21% | 20% |
| Diarrhea | 14% | 8% | 14% | 18% |
| Vomiting | 10% | 4% | 10% | 10% |
| Stopped treatment because of any side effect | 24 (8%) | 14 (5%) | 29 (10%) | 20 (7%) |
| Stopped because of stomach or bowel effects | 16 (5%) | 6 (2%) | 18 (6%) | 14 (5%) |
Serious events, such as those requiring hospital care, occurred in 6%–8% across groups. Six participants died: one in each semaglutide group and two in each dulaglutide group. Those small counts cannot settle differences in rare harms.
Stomach complaints weren't consistently more frequent with semaglutide across both pairs. Stopping because of side effects was more frequent with semaglutide in each comparison.
Stopping medication also differed from leaving the study. In the table's order, 47, 27, 51 and 36 people stopped treatment early for any reason. Altogether, 72 left the trial. Someone could stop medication but still attend follow-up.
The limits, and what later research adds
Knowing the assigned drug can influence symptom reporting. The narrow enrollment criteria also limit how widely these averages apply. A blood-sugar result cannot establish a heart-outcome comparison; SUSTAIN-6 covers that separate question.
A 2020 follow-up analysis in BMJ Open examined the same participants by age, sex, diabetes duration, starting blood sugar and body size. Findings generally favored semaglutide across those groups. This was a post hoc analysis, meaning the subgroup questions were examined after the main study. Small subgroups and reuse of the same trial limit what it adds.
As of November 5, 2026, the day this article was written, a Diabetes Care study published October 27, 2026 offers a newer comparison in people with diabetes receiving hemodialysis, a blood-filtering treatment for kidney failure. Its abstract reports that semaglutide, compared with dulaglutide, was associated with fewer deaths and hospitalizations; against liraglutide the differences were not statistically significant. It used national dialysis-registry and Medicare data rather than random assignment. That separate population extends the research question without changing whom SUSTAIN 7 studied.
How the doses fit today's prescribing conversation
The Ozempic injection label, revised 05/2026, lists a maximum weekly dose of 2 mg. The Trulicity label, revised 03/2026, lists 4.5 mg for adults. SUSTAIN 7 didn't compare those maximum doses.
For the wider diabetes picture, see semaglutide for type 2 diabetes. The tirzepatide versus semaglutide comparison concerns different trials and another medicine.
SUSTAIN 7 tested manufactured preparations, not compounded semaglutide. Semaglutide prescribed through Promise is dispensed by a licensed U.S. compounding pharmacy as a compounded medication, which is different from an FDA-approved product: the formulation offered here is not FDA-approved. FDA explains this distinction. A licensed provider may still prescribe — that decision is between you and your doctor.
What belongs in the visit
The useful question is how the evidence fits someone's blood-sugar goals, kidney function, eye history, current medicines and previous side effects. The prescriber sets the actual dose and any change of medicine. SUSTAIN 7's dose pairs aren't a conversion chart.
At Promise, a licensed provider reviews every request, and not everyone qualifies.
