SURPASS-2: TZ2 vs. semaglutide head-to-head

SURPASS-2 is the Phase 3 trial published in the New England Journal of Medicine in 2021 that directly compared TZ2 and semaglutide in adults with type-2 diabetes. The trial matters for research design because it's one of the few head-to-head incretin comparisons in the literature — and it cleanly answers the question "does dual GIP/GLP-1 agonism actually outperform single GLP-1 agonism, or is it just a marketing claim?"

Trial design

1,879 adults with T2D, randomized 1:1:1:1 to:

• TZ2 5 mg weekly SC
• TZ2 10 mg weekly SC
• TZ2 15 mg weekly SC
• Semaglutide 1 mg weekly SC (the maximum approved diabetes dose at trial start)

Primary outcome: change in HbA1c at 40 weeks. Secondary: body weight change, proportion reaching HbA1c < 7%, glycemic excursions, lipid panel.

Headline findings

Arm	HbA1c change	Weight change	HbA1c < 7%
Semaglutide 1 mg	−1.86%	−5.7 kg	79%
TZ2 5 mg	−2.01%	−7.6 kg	82%
TZ2 10 mg	−2.24%	−9.3 kg	86%
TZ2 15 mg	−2.30%	−11.2 kg	86%

All three TZ2 doses showed greater HbA1c reduction than semaglutide 1 mg, with statistical significance. The weight-loss gap was larger: TZ2 15 mg achieved nearly double the weight reduction of semaglutide at the maximum diabetes dose.

What it tells us about dual agonism

The interesting science is in the dose-response. At matched doses (5 mg TZ2 vs. 1 mg semaglutide — both at the lower end of their respective ranges), TZ2 produced 33% greater weight reduction. This implies that activating GIP receptors in addition to GLP-1 receptors adds a measurable, separable effect — not just "more drug" but a qualitatively different signal.

For research designs that want to isolate GLP-1 effects specifically, semaglutide is the appropriate tool. For research that needs the broader incretin signal (or wants to study the GIP contribution), TZ2 gives you both receptors. SURPASS-2 quantifies the practical difference between the two.

Adverse events

GI events were the dominant tolerability signal in both compounds. Nausea, diarrhea, vomiting were each in the 17–25% range for TZ2 arms, and 18–24% for semaglutide — broadly comparable. Hypoglycemia was rare in both (these are glucose-dependent secretagogues), which is methodologically important: research models comparing incretin compounds against insulin-based interventions should expect very different hypoglycemia profiles.

What the trial doesn't tell you

SURPASS-2 used semaglutide 1 mg, which was the maximum diabetes-indicated dose at the time. Higher semaglutide doses (2.4 mg, the Wegovy obesity dose) were not in this comparison. So the trial cleanly answers "TZ2 vs. semaglutide-at-T2D-dose" but does not answer "TZ2 vs. semaglutide-at-obesity-dose." A later head-to-head (SURMOUNT-5) addresses that question for obesity research; check the published literature for findings.

Notes

This is a research summary of published clinical data. It is not a recommendation for any specific research protocol or design. For research use only.

Reference

1. Frías JP, Davies MJ, Rosenstock J, et al. NEJM 2021;385:503–515. PMID: 34170647