Twenty molecules are in late-stage development. You’ve heard of four. Here’s what the rest actually show.
Peptide Briefs · · Troy Snyder, Ed.D., MBA
If you follow this space on X, you could be forgiven for thinking there are four GLP-class molecules in the world: semaglutide, tirzepatide, retatrutide, and whatever a vendor is calling retatrutide this week. There are closer to twenty in late-stage development, several are already approved somewhere, and a few of them are starting to show up on research vendor menus under names most buyers have never seen in a paper.
This is a field guide to the ones that don’t get talked about. For each I’ll tell you what it is, who makes it, what the human evidence actually shows, and how far it is from a regulator’s desk. I grade the same way the compound pages do: A needs peer-reviewed pivotal data plus a completed regulatory safety review; B is peer-reviewed phase 3 or strong phase 2 without a regulator’s sign-off; C is phase 2 or less; D is preclinical or sponsor press releases only. Where a drug is approved in China but not reviewed by the FDA or EMA, I say so, because a Chinese approval is a real regulatory review and also not the same thing as a US one.
One framing note before the list. “GLP-1” has become shorthand for a whole family of gut-hormone drugs. Several of the molecules below don’t touch the GLP-1 receptor at all. They are here because they are being developed to do the same job and because they get lumped together on vendor sites.
The Chinese Pipeline
China is now the second center of gravity for this drug class, and two of its molecules are approved there. If you have seen a “Chinese retatrutide” referenced online, it is almost certainly one of the first two below.
UBT251 — the triple agonist Novo bought
What it is: A once-weekly synthetic peptide that activates GLP-1, GIP and glucagon receptors, the same three targets as retatrutide. Developed by United Bio-Technology (Hengqin), a subsidiary of The United Laboratories (TUL), Zhuhai. In March 2025 Novo Nordisk licensed worldwide rights outside Greater China for $200 million upfront.
Evidence: Two Chinese phase 2 trials, both reported by press release in 2026 and not yet peer-reviewed. In adults with overweight or obesity (baseline 92 kg, BMI 33), the 6 mg dose produced a mean weight loss of 19.7% at 24 weeks versus 2.0% on placebo. In type 2 diabetes (211 participants), HbA1c fell up to 2.16% versus 1.77% on semaglutide 1 mg, with weight loss up to 9.8% versus 4.8%. Novo is running a global phase 1b/2a in about 330 people, reading out in 2027, and TUL plans two Chinese phase 3s.
Why people call it “the Chinese retatrutide”: Same mechanism, similar early numbers (retatrutide’s phase 2 showed 24% at 48 weeks; 19.7% at 24 weeks is on a comparable trajectory), and it is Novo’s answer to Lilly’s molecule. On half-life, I can find no published pharmacokinetic data for UBT251 that would support a claim that it lasts longer than retatrutide. Both are dosed weekly. If a vendor is telling you otherwise, ask them for the paper.
Grade: C. Phase 2, press-release only, no peer review, no regulator.
Mazdutide — approved in China, dual GLP-1/glucagon
What it is: A once-weekly GLP-1 and glucagon receptor dual agonist, originally Lilly’s LY3305677, licensed to Innovent Biologics for China. Brand name Xinermei.
Evidence: The GLORY-2 phase 3 in Chinese adults with obesity reported 16.7% mean weight loss at 9 mg over 60 weeks; the earlier GLORY-1 phase 3 was published in NEJM in 2024 and the program has since appeared in JAMA and Lancet Diabetes & Endocrinology. A US phase 2 reported up to 18.1% at 16 mg over 32 weeks.
Regulatory: Approved by China’s NMPA for chronic weight management on June 27, 2025 and for type 2 diabetes on September 19, 2025. No FDA or EMA filing.
Grade: B. Peer-reviewed phase 3 and a completed regulatory review, but not by the FDA or EMA. If you want to call this an A for China, I won’t argue hard. For a US reader it is a B.
Ecnoglutide — the “biased” GLP-1
What it is: A long-acting GLP-1 receptor agonist from Sciwind Biosciences (Hangzhou), engineered to preferentially signal through the cAMP pathway rather than β-arrestin, which in theory means less receptor desensitization. Brand names Xianyida (diabetes) and Xianweiying (weight). Pfizer holds China commercialization rights.
Evidence: Phase 3 in China reported about 15.1% placebo-adjusted weight loss. The biased-agonism claim is mechanistically interesting and clinically unproven; nobody has shown in humans that it translates to better durability or tolerability than semaglutide.
Regulatory: NMPA approved for type 2 diabetes (January 30, 2026) and for weight management (March 6, 2026). No US filing announced.
Grade: B, same logic as mazdutide.
Bofanglutide (GZR18) — the every-two-weeks GLP-1
What it is: A long-acting GLP-1 receptor agonist from Gan & Lee Pharmaceuticals (Beijing), dosed once every two weeks. This is the one Chinese molecule where “longer half-life” is actually the selling point.
Evidence: Phase 3 in China with an extension ongoing; Gan & Lee has signed licensing deals with JW Pharmaceutical (Korea, $81 million, April 2026) and Lupin (India). Published data are thinner than for mazdutide or ecnoglutide.
Grade: C. It is a single GLP-1 agonist, not a triple, so anyone describing it as a longer-acting retatrutide is wrong on mechanism.
HRS9531 — Hengrui’s dual agonist
What it is: A GLP-1/GIP dual agonist (tirzepatide’s mechanism) from Jiangsu Hengrui, in phase 3 in China. Kailera Therapeutics licensed it for the rest of the world in 2024.
Grade: C pending phase 3 publication.
The Western pipeline you’ve probably skipped
Survodutide — the one that read out and underwhelmed
What it is: Boehringer Ingelheim and Zealand Pharma’s once-weekly GLP-1/glucagon dual agonist (BI 456906).
Evidence: SYNCHRONIZE-1 (NEJM, June 2026; 76 weeks, 1,481 people without diabetes) reported 12.2% and 13.0% at 3.6 and 6.0 mg versus 5.4% placebo on the primary estimand, or up to 16.6% if everyone adhered. SYNCHRONIZE-2 (NEJM, October 2026, type 2 diabetes) reported 8.2% and 9.8% versus 3.9%. A pre-specified substudy showed 63% liver fat reduction and 34% visceral fat reduction, and the earlier phase 2b in MASH showed histological improvement in up to 83% of patients. The cardiovascular outcomes trial completed in June 2026 with results expected before year end.
The honest read: This is semaglutide-range weight loss with a potentially better liver story. It is not a tirzepatide competitor on the scale.
Grade: B. Peer-reviewed phase 3, no regulatory review yet.
MariTide (maridebart cafraglutide) — the monthly one
What it is: Amgen’s AMG 133, a GLP-1 receptor agonist peptide fused to an antibody that blocks the GIP receptor. Yes, blocks. Tirzepatide activates GIP; MariTide antagonizes it; both lose weight. Nobody has fully explained this. The antibody scaffold gives it a half-life around three weeks, so it is dosed monthly, and Amgen is testing every-eight-weeks maintenance.
Evidence: Phase 2 (NEJM 2025) showed up to about 20% weight loss at 52 weeks in people without diabetes and about 17% with type 2 diabetes, with the striking feature that weight stayed off for months after the last dose. Phase 3 (MARITIME-1 and -2) is ongoing with an estimated primary completion in January 2027; as of October 2026 no phase 3 results have been reported. The phase 3 titration was slowed after high rates of nausea and vomiting in phase 2.
Grade: C, and the most interesting C on this list. If anything here deserves the label “longer half-life than retatrutide,” it is this, and it is a peptide-antibody conjugate, not something a peptide synthesizer can make.
Amycretin — GLP-1 and amylin in one molecule
What it is: Novo’s single peptide that activates both the GLP-1 receptor and the amylin receptor, available as a weekly injection and a daily pill. Think CagriSema’s two mechanisms fused into one chain.
Evidence: Phase 1b/2a injectable showed about 22% weight loss at 36 weeks; the oral form showed about 13% at 12 weeks. Phase 3 began in 2026.
Grade: C. Early, but this is where Novo is putting its chips after the CagriSema head-to-head miss.
Eloralintide — Lilly’s amylin-only answer
What it is: A selective amylin receptor agonist (no calcitonin receptor activity, unlike cagrilintide), weekly injection.
Evidence: Phase 2 showed up to 20% weight loss at 48 weeks as monotherapy, which is remarkable for an amylin drug on its own. The eloralintide-plus-tirzepatide combination (presented at EASD, September 30, 2026) showed up to 23.3% at 48 weeks in obesity with type 2 diabetes versus 14.8% for tirzepatide 15 mg alone. Phase 3 started late 2025.
Grade: C, likely B within a year.
Pemvidutide, VK2735, petrelintide, berobenatide
Briefly: pemvidutide (Altimmune) is a GLP-1/glucagon dual in phase 2 with a MASH focus; VK2735 (Viking) is a GLP-1/GIP dual in phase 3 with both injectable and oral forms; petrelintide (Zealand, partnered with Roche) is another long-acting amylin analogue in phase 2; berobenatide (Pfizer, formerly Metsera’s MET-097i) is an ultra-long-acting GLP-1 dosed monthly, in phase 3, with 12.3% placebo-adjusted weight loss at 28 weeks in phase 2b. All C or D today.
What this means if you are reading vendor listings
Three patterns to watch for.
Names that aren’t names. “GLP-3,” “Triple G,” “Reta-L,” “long-acting GLP” are catalog labels, not molecules. Every compound on this page has a development code and, for most, an INN. A vendor who cannot tell you which one they are selling cannot have tested its identity.
Half-life claims without pharmacokinetics. A peptide’s half-life comes from its lipid tail or antibody scaffold and is measured in a phase 1 trial. Retatrutide’s is about six days; semaglutide’s about seven; cagrilintide’s about seven to eight; MariTide’s about three weeks. Any claim of a “longer half-life” version of a weekly peptide should come with a published number. If the vendor’s answer is a different fatty acid chain, that is a different molecule with no human data at all.
Approval that isn’t your approval. Mazdutide and ecnoglutide are approved drugs in China. That is a meaningful review. It is not an FDA review, and it does not change the legal status of research-grade material in the US, where the FDA stated in June 2026 that sales of unapproved GLP-class products to consumers are illegal and Lilly filed six lawsuits in August 2026 against sellers of retatrutide, including research-use-only vendors.
The field is moving fast enough that half of this page will need updating within six months: CagriSema’s FDA decision is due this quarter, retatrutide’s filing is planned for early 2027, MariTide phase 3 reads out around the same time, and SYNCHRONIZE-CVOT is due before year end. I’ll post grade changes as they land.
Sources
- United Laboratories / Novo Nordisk, UBT251 phase 2 topline releases, February 24 and March 25, 2026
- Innovent Biologics, mazdutide NMPA approvals (June 27 and September 19, 2025); Ji L et al., Once-weekly mazdutide in Chinese adults with obesity or overweight, NEJM 2024, doi:10.1056/NEJMoa2411528
- Sciwind Biosciences, ecnoglutide NMPA approval announcements, January 30 and March 6, 2026
- Boehringer Ingelheim, SYNCHRONIZE-1 topline (April 28, 2026) and ADA 2026 presentation; SYNCHRONIZE-1 and SYNCHRONIZE-2 in NEJM, June and October 2026
- Amgen, MariTide phase 2 (NEJM 2025); ClinicalTrials.gov MARITIME-1 and MARITIME-2 records
- Eli Lilly, eloralintide phase 2 and EASD 2026 presentation (September 30, 2026)
- Novo Nordisk, REDEFINE 4 company announcement, February 23, 2026
- FDA statement on unapproved incretin products, June 2026; Eli Lilly press release on retatrutide lawsuits, August 12, 2026
Peptide Briefs is an independent reference and does not sell anything. Its owners co-own a research-peptide vendor; that is disclosed on every page. Nothing here is medical advice.
Leave a Reply