If you've spent any time reading about GLP-1 receptor agonists, you've internalized a tradeoff: better efficacy comes with more nausea. Semaglutide gets you to ~15% weight loss; about a third of patients pay for it with weeks of GI misery. Tirzepatide pushes higher; the nausea curve climbs with it. Retatrutide hits 24%+ in Phase 3, and the discontinuation-due-to-AE rate climbs too.
Ecnoglutide is the first GLP-1 to try a different strategy entirely. Not a different receptor target. Not a new amino-acid sequence trick. A different signal through the same receptor.
On March 6, 2026, China's National Medical Products Administration approved ecnoglutide injection for chronic weight management — making it the first cAMP-biased GLP-1 receptor agonist to reach market anywhere in the world. The Phase 3 data backing the approval, published in Nature Communications on January 7, 2026 and in The Lancet Diabetes & Endocrinology across 2025, is worth a careful look — because the nausea numbers don't look like a normal GLP-1.
What "biased agonism" actually means
When GLP-1 binds its receptor, the receptor does two things in parallel:
- Activates Gαs / cAMP signaling — the pathway that drives insulin secretion, satiety, and most of the therapeutic benefits.
- Recruits β-arrestin — a regulatory pathway that pulls the receptor off the cell surface (internalization), causes desensitization, and contributes to nausea, vomiting, and other GI signaling cascades.
Native GLP-1 and the marketed analogs — semaglutide, liraglutide, tirzepatide's GLP-1 arm, dulaglutide — activate both pathways. They're "balanced" agonists. The cAMP pathway gives you the benefits; the β-arrestin arm hands you the bill.
Ecnoglutide was engineered to pull those signals apart. It's a 38-residue peptide modified from native GLP-1 with a fatty-acid side chain for albumin binding (long half-life, once-weekly dosing). The structural tweaks shift its signaling profile so that it preferentially activates cAMP while minimizing β-arrestin recruitment.
In cell-based assays, the receptor stays at the surface longer with ecnoglutide than with semaglutide. Less internalization → less desensitization → sustained activity per dose, and — critically — the β-arrestin arm that gets blamed for GI side effects is largely sidestepped.
That's the theory. Phase 3 is where theory meets bodies.
SLIMMER: the weight loss trial
The SLIMMER trial (n=664) ran at Peking University People's Hospital and 33 other Chinese centers, enrolling adults with overweight or obesity (BMI ≥24 kg/m² with comorbidity, or ≥28 kg/m²) and without type 2 diabetes. Randomization was 1:1:1:1 to once-weekly ecnoglutide 1.2 mg, 1.8 mg, 2.4 mg, or placebo, for 48 weeks.
Weight loss at week 40
| Arm | Mean % weight change | ≥5% loss | ≥10% loss |
|---|---|---|---|
| Placebo | +0.1% | — | — |
| Ecnoglutide 1.2 mg | −9.1% | 77% | — |
| Ecnoglutide 1.8 mg | −10.9% | 84% | — |
| Ecnoglutide 2.4 mg | −13.2% | 87% | — |
At 48 weeks the top dose pushed to a 15.4% mean reduction, with 92.8% of patients hitting clinically meaningful (≥5%) weight loss.
For context: that puts ecnoglutide roughly on par with semaglutide 2.4 mg (Wegovy), which delivered ~14.9% at 68 weeks in STEP-1. It does not beat tirzepatide 15 mg (~20.9% in SURMOUNT-1) and is well below retatrutide 12 mg (~24.2% at 48 weeks in Phase 2). On efficacy alone, ecnoglutide is competitive, not category-leading.
The interesting column isn't in that table.
The tolerability story
In SLIMMER, the most common adverse events were the usual GLP-1 suspects — nausea, vomiting, diarrhea, constipation — but the rates were notably below the historical curve for unbiased agonists. Discontinuation due to adverse events was low across all dose arms.
The EECOH-1 diabetes trial gives us the cleaner number. In a 211-patient placebo-controlled study published in Nature Communications (January 7, 2026):
| Adverse event (24 weeks) | Placebo | Ecno 0.6 mg | Ecno 1.2 mg |
|---|---|---|---|
| Nausea | 9.9% | 7.2% | 12.7% |
| Diarrhea | 5.6% | 11.3% | 24.6% |
| Vomiting | low | low | modest |
Read that nausea row twice. A GLP-1 at its highest tested diabetes dose produced nausea at 12.7%, against a placebo rate of 9.9%. The delta is ~3 percentage points.
For comparison, in the SUSTAIN trials semaglutide 1.0 mg produced nausea in roughly 20–22% of patients (vs ~6–8% placebo) — a delta of ~14 percentage points. Liraglutide's nausea rate in LEAD trials ran 28–40%. The ecnoglutide signal is in a different zip code.
The catch: diarrhea was higher on ecnoglutide than on placebo (24.6% vs 5.6% at 1.2 mg). The side-effect profile didn't disappear — it shifted. Biased signaling appears to dampen the central nausea/vomiting pathway more effectively than the lower-GI motility effects. Whether patients prefer diarrhea over nausea is a real question; for most people the answer is yes.
EECOH-2: head-to-head against dulaglutide
The EECOH-2 trial (n=621) is the harder test — a 52-week, open-label, active-controlled, non-inferiority design comparing ecnoglutide 0.6 mg and 1.2 mg against dulaglutide 1.5 mg in T2D patients on metformin.
HbA1c reduction at week 32:
| Arm | HbA1c change |
|---|---|
| Dulaglutide 1.5 mg | −1.65% |
| Ecnoglutide 0.6 mg | −1.91% (Δ −0.26%, p=0.0002) |
| Ecnoglutide 1.2 mg | −1.89% (Δ −0.24%, p<0.001) |
Ecnoglutide met non-inferiority and crossed into statistical superiority, though the trial's authors note the absolute difference (~0.25%) is not clinically meaningful by typical thresholds. The reductions were sustained out to week 52. Discontinuation rates were comparable: 3–4% across all arms.
The honest read: ecnoglutide is as good as dulaglutide on glycemic control, with a comparable safety profile and once-weekly dosing. The dulaglutide audience now has a structurally differentiated option.
How does this stack up against the field?
| Drug | Mechanism | Best Phase 3 weight loss | Nausea (top dose) | Status |
|---|---|---|---|---|
| Semaglutide (Wegovy) | Balanced GLP-1 | ~14.9% (68 wk) | ~44% | FDA approved |
| Tirzepatide (Zepbound) | GLP-1 / GIP | ~20.9% (72 wk) | ~29% | FDA approved |
| Retatrutide | GLP-1/GIP/glucagon | ~24.2% (48 wk) | ~35–55% | Phase 3 |
| Orforglipron | Oral small-molecule GLP-1 | ~14.7% (72 wk) | ~25% | Phase 3 |
| Ecnoglutide | cAMP-biased GLP-1 | ~15.4% (48 wk) | ~12.7% | Approved (China) |
The 12.7% number is the headline. If it holds up in larger and more diverse populations, ecnoglutide could become the GLP-1 for the patient who wanted semaglutide-class results but quit semaglutide because they were nauseous for three months.
What we don't know yet
A few important caveats:
Population generalizability. Every pivotal trial was conducted in China, in a Chinese population. GLP-1 response and side-effect profiles can differ by ethnicity (East Asian populations have historically shown stronger glycemic responses and somewhat different GI tolerability patterns than Western populations). Real-world data outside China is essentially zero.
No head-to-head against semaglutide. EECOH-2 used dulaglutide as the comparator — a reasonable choice for the Chinese market where dulaglutide is widely prescribed, but not the most informative comparison for Western prescribers. A head-to-head against Wegovy or Zepbound would settle the question of whether biased agonism actually shifts the tolerability curve in like-for-like patients.
No cardiovascular outcomes data. Semaglutide has SELECT. Tirzepatide has SURMOUNT-MMO underway. Ecnoglutide has nothing yet. For a drug meant to be taken indefinitely by cardiometabolic patients, CV outcomes data will matter for regulators and payers — particularly the FDA and EMA if Sciwind pursues those markets.
Long-term receptor pharmacology. Biased agonism is a young field. We have a strong theoretical and short-term basis to believe that reduced β-arrestin recruitment changes desensitization kinetics. We don't yet know whether that translates into longer-lasting weight loss after discontinuation, or whether it changes the trajectory of beta-cell preservation in T2D. These are multi-year questions.
FDA pathway is unclear. Sciwind has run a Phase 1 trial in the US (NCT05299944) and signed regional licensing deals, but there's no public IND for a global pivotal program. China-only approval is meaningful clinically but commercially limited. A US filing — if it happens — would likely require either bridging studies in a Western population or full new pivotal trials.
What about the oral version?
Sciwind's XW004 is an oral formulation of ecnoglutide using a SNAC-style absorption enhancer (the same broad family of technology that made oral semaglutide / Rybelsus possible). Phase 1 data in 2024 showed proof-of-concept bioavailability and the expected PK profile. If a biased oral GLP-1 reaches Phase 3 with a similarly low-nausea signal, the implications get substantially more interesting — oral semaglutide's nausea rate is one of the limitations of Rybelsus and the recently launched oral Wegovy.
That program is years behind the injectable. Don't expect oral ecnoglutide data of meaningful size before 2027.
Why this matters beyond ecnoglutide
The interesting question isn't really will ecnoglutide become the next Wegovy. It probably won't, at least not in the West, where the regulatory road is long and the incumbents have a multi-year lead.
The interesting question is whether biased agonism is a generalizable design strategy for the next generation of metabolic peptides — and whether the cAMP / β-arrestin split is the right lever.
If the SLIMMER and EECOH-1 tolerability signals hold up in larger trials and across populations, expect to see biased agonism applied to:
- GIP / glucagon dual and triple agonists, where the nausea ceiling currently caps usable doses.
- Amylin analogs (cagrilintide, amycretin's amylin arm), where similar β-arrestin pathways are implicated in GI signaling.
- Oral peptide formulations, where bioavailability constraints already force lower doses — making any tolerability improvement disproportionately valuable.
This is also a notable validation of the biased agonism hypothesis broadly — a strategy that's been talked about for opioid receptors (oliceridine), angiotensin (TRV027), and others for fifteen years with mixed real-world results. Ecnoglutide is the first commercially approved drug where the biased signaling is the central marketing claim and the Phase 3 data appears to back it up.
Bottom line
Ecnoglutide isn't going to dethrone tirzepatide on raw efficacy. It probably won't reach US patients for years, if ever. And the side-effect profile shifts rather than disappears — more diarrhea, less nausea.
But the 12.7% nausea rate at the top diabetes dose is a genuinely interesting signal, and it's the first evidence in a real Phase 3 program that you can keep the cAMP-driven benefits of GLP-1 agonism while substantially dialing back the β-arrestin-driven side effects. That's a meaningful proof-of-concept, and the second-generation drugs built on it could matter more than ecnoglutide itself.
For now, file ecnoglutide under: important for what it proves, not for who it treats — yet.
This article is for research and educational purposes only. Ecnoglutide is approved in China for chronic weight management and is not approved by the FDA or other Western regulators for any indication. Nothing here is medical advice; weight-management therapies should be discussed with a qualified clinician.