Scientific deep-dive
Zenagamtide Pill vs Shot: What the Phase 2 Lancet Reports Found
Two Lancet reports tested daily oral and weekly injected zenagamtide (amycretin) in type 2 diabetes. Here is how HbA1c, company-reported weight loss and side effects compare, and why the comparison is still indirect.
In the 36-week phase 2 program, the weekly zenagamtide shot had numerically larger blood sugar and weight results than the daily pill at the top doses, though the pill worked too. HbA1c fell 1.7% with the 40 mg injection and 1.4% with the 50 mg tablet in adults with type 2 diabetes,[1][2] and Novo Nordisk reported weight loss of up to 14.5% and 10.1%.[3] The abstracts do not report a randomized pill-versus-shot test, and both studied adults with type 2 diabetes rather than obesity alone.
About this article
The two Lancet abstracts report HbA1c and safety but no weight figures, so every weight number below for these reports comes from Novo Nordisk announcements, not peer-reviewed tables. For the drug’s earlier studies and background, see our amycretin pipeline profile.
One registered trial, two papers
The papers read like separate studies, but they share a single registration, NCT06542874, whose official title covers both once-weekly subcutaneous and once-daily oral dosing, with 448 people enrolled.[8] The oral report counts 186 participants and the injection report 262, and both describe the same 915 people screened between August 7 and December 27, 2024.[1][2]
The entry criteria matched: adults aged 18 to 75 with type 2 diabetes, HbA1c of 7.0% to 10.0%, a BMI of 23.0 to under 50.0, and stable metformin with or without an SGLT2 inhibitor. Each route had its own matched placebo, and the abstracts test each dose against that placebo.[1][2] So the side-by-side below is informative, but it is our comparison and carries no statistical test. Novo Nordisk funded both reports, and several authors are company employees.[1][2]
| Measure | Once-daily pill | Once-weekly injection |
|---|---|---|
| Maintenance doses | 6, 25, 50 mg | 0.4, 1.5, 5, 10, 20, 40 mg |
| Start and escalation | 1.5 mg, raised every 4 weeks | 0.2 mg, raised every 4 weeks |
| Randomized (drug / placebo) | 156 / 30 | 225 / 37 |
| Baseline HbA1c | Mean range 7.9–8.1% | Mean 7.8% |
| HbA1c change, lowest dose | −0.9% (6 mg) | −0.9% (0.4 mg) |
| HbA1c change, top dose | −1.4% (50 mg) | −1.7% (40 mg) |
| Difference vs placebo, top dose | −1.09% (95% CI −1.59 to −0.59) | −1.56% (95% CI −2.05 to −1.07) |
| Baseline weight | 101.1 kg | 99.2 kg |
| Weight change, best result vs placebo | Up to −10.1% vs −2.5% | Up to −14.5% vs −2.6% |
| Serious adverse events (drug / placebo) | 7 / 0 | 18 / 3 |
Blood sugar: close at the low end, wider at the top
The lowest doses of each route produced the same 0.9% drop in HbA1c, but placebo did more in the oral group (Novo reported −0.4% versus −0.2%),[3] so the placebo-adjusted effect was −0.5% for 6 mg by mouth[1] and −0.77% for 0.4 mg by injection.[2] At the top, oral 50 mg reached −1.09% against placebo with a confidence interval of −1.59 to −0.59,[1] while injected 40 mg reached −1.56%, with an interval of −2.05 to −1.07.[2] Those intervals overlap, and each active arm held about 50 people or fewer, so the size of the gap between routes is uncertain.
The oral curve flattened quickly in this dose range: 25 mg gave −1.3% and 50 mg gave −1.4%.[1] Company headline figures for HbA1c run slightly higher than the published ones, at up to −1.8% by injection and −1.5% by mouth,[3] and a June 2026 Novo release ties the −1.8% to a dose-response model.[4] We use the Lancet numbers for HbA1c. They share one limit with the company figures: both abstracts report an efficacy estimand, using on-treatment data without rescue medication, so they estimate the effect had people stayed on treatment without rescue, not what happened across everyone assigned.[1][2]
Weight: what the company reported
Weight is the question most readers care about, and it is the weakest part of this record for now. Novo’s November 2025 announcement gave up to −14.5% with injections against −2.6% on placebo, and up to −10.1% with tablets against −2.5%.[3] By our subtraction, that is about 11.9 and 7.6 percentage points beyond placebo. A June 2026 release for the American Diabetes Association meeting put the 40 mg injection at 14.6%, against 2.1% with placebo, from a baseline of about 219 pounds. The same release’s dose-response model gave up to −14.5% at 40 mg, or −11.81 points against placebo (95% CI −15.37 to −8.25).[4] Both announcements describe results as if people stayed on treatment without rescue medication, not across everyone assigned.[3][4] These are sponsor summaries rather than the peer-reviewed tables, and neither Lancet abstract gives a weight figure to set beside them.
Two details suggest week 36 understates where the higher doses might have ended up. Novo reported no weight loss plateau at week 36 for the higher doses of either route, and people on the highest injected dose had been at their final maintenance dose for only 4 weeks.[3] The announcement gives no equivalent figure for the pill. Whether a longer course would widen or narrow the gap between routes is unknown.
Dose by mass is very different
A person on oral 50 mg takes 350 mg a week, about nine times the mass of the 40 mg weekly injection.[1][2] The abstracts do not report how much of the oral dose reaches the bloodstream, so that arithmetic describes the doses tested and nothing more. Pill-taking conditions are still being worked out: a phase 1 study of 320 planned participants is looking at how meal timing and water volume change absorption.[11]
Side effects by route
The oral abstract gives usable rates. Gastrointestinal events occurred in 26% of the 6 mg group, 41% at 25 mg and 47% at 50 mg, against 23% on placebo.[1] Seven participants on the pill had serious adverse events and none on placebo did.[1] The injection abstract says only that most adverse events were gastrointestinal and mild to moderate, with no rates by dose. It lists 18 participants with serious adverse events across the zenagamtide groups and three on placebo.[2] No deaths occurred with either route.[1][2]
Without injection-arm GI rates, these abstracts cannot tell you which route is easier on the stomach. Trials this size also cannot detect rare harms. A small company-run phase 1 study found that renal impairment had no clinically relevant effect on drug exposure, but it gave 42 people a single 0.3 mg injection, so it says little about long-term tablet use.[12]
Why these results differ from the 24% headline
Earlier studies in people with overweight or obesity, which did not require diabetes, produced larger figures. A phase 1b/2a injection study reported −24.3% at 60 mg and −22.0% at 20 mg at week 36, though its authors noted that a large number of participants withdrew.[5] A review of the oral phase 1 work cites weight loss of up to 13.1% over 85 days.[6] That oral study was built mainly to assess safety, with weight as an exploratory endpoint, and its multiple-dose 12-week part randomized 60 people at one research unit.[7] Those numbers come from different populations and designs, and they should not be read as what the phase 2 doses would do in people without diabetes.
What phase 3 will and will not answer
Phase 3 will add larger, longer data for each form. AMAZE 9 is testing five oral dose levels against placebo in about 950 people with obesity and no diabetes, with percent weight change at week 76 as the main outcome.[9] AMAZE 7 pits weekly injected zenagamtide against two semaglutide doses in about 650 people, with the main outcome at week 84.[10] Neither is designed as a pill-versus-shot trial, so any comparison across them will still be indirect.
JAMA described zenagamtide as unapproved in October 2026 and reported on a black market for it.[13] Products sold online under either name have not been shown to be the drug tested in these trials; our guide to unapproved GLP-1 sourcing covers the risks of that market. If a daily pill is the main draw, the phase 3 record for orforglipron in obesity is further along than this one.
What to raise with a prescriber
- Ask whether a clinical trial is open to you, and what it would require. The AMAZE 9 and AMAZE 7 entries exclude people with diabetes and those with prior GLP-1-based treatment, which AMAZE 7 limits to the year before screening.[9][10]
- If you have type 2 diabetes, ask how these 36-week HbA1c changes compare with the options available to you now, since these reports used placebo rather than an approved drug as the comparator.[1][2]
- If needle avoidance is your main concern, say so. Approved oral options exist, and this phase 2 program does not establish how the two zenagamtide forms compare over the long term.
Frequently Asked Questions
References
- 1.Mora P, Aroda VR, Asong M, Blüher M, Carlander AF, Kaltoft M, Vilsen LN, Rosenstock J. Efficacy and safety of once-daily oral zenagamtide, a novel unimolecular GLP-1 and amylin receptor agonist, in adults with type 2 diabetes: a multicentre, randomised, parallel, double-blind, placebo-controlled, dose-finding, phase 2 trial. Lancet. 2026. PMID: 42532079.
- 2.Mora P, Aroda VR, Asong M, Blüher M, Heftdal LD, Carlander AF, Vilsen LN, Rosenstock J. Efficacy and safety of once-weekly subcutaneous zenagamtide, a novel unimolecular GLP-1 and amylin receptor agonist, in type 2 diabetes: a multicentre, randomised, parallel, double-blind, placebo-controlled, dose-finding, phase 2 trial. Lancet. 2026. PMID: 42532080.
- 3.Novo Nordisk. Novo Nordisk phase 2 trial with amycretin reports significant weight loss and HbA1c reduction in type 2 diabetes. Company announcement No 38 / 2025. Novo Nordisk company announcement. 2025. https://www.novonordisk.com/content/nncorp/global/en/news-and-media/news-and-ir-materials/news-details.html?id=916463
- 4.Novo Nordisk. Zenagamtide Phase 2 T2D Data at ADA 2026. Novo Nordisk press release. 2026. https://www.novonordisk-us.com/media/press-releases/zenagamtide-phase2-t2d-ada-2026.html
- 5.Dahl K, Toubro S, Dey S, Duque do Vale R, Flint A, Gasiorek A, Heydorn A, Jastreboff AM, Key C, Petersen SB, Vegge A, Adelborg K. Amycretin, a novel, unimolecular GLP-1 and amylin receptor agonist administered subcutaneously: results from a phase 1b/2a randomised controlled study. Lancet. 2025. PMID: 40550231.
- 6.Panou T, Gouveri E, Popovic DS, Papanas N. New Amylin-Based Agonists for the Treatment of Obesity and Type 2 Diabetes Mellitus. Diabetes Ther. 2026. PMID: 42803913.
- 7.Gasiorek A, Heydorn A, Gabery S, Hjerpsted JB, Kirkeby K, Kruse T, Petersen SB, Toubro S, Vegge A, Key C. Safety, tolerability, pharmacokinetics, and pharmacodynamics of the first-in-class GLP-1 and amylin receptor agonist, amycretin: a first-in-human, phase 1, double-blind, randomised, placebo-controlled trial. Lancet. 2025. PMID: 40550229.
- 8.Novo Nordisk A/S. Safety and Efficacy of Once-weekly Subcutaneous and Once-daily Oral NNC0487-0111 in Participants With Type 2 Diabetes: a Dose Finding Study. ClinicalTrials.gov NCT06542874. 2024. https://clinicaltrials.gov/study/NCT06542874
- 9.Novo Nordisk A/S. Efficacy and Safety of Once-daily Oral Zenagamtide in Participants With Obesity (AMAZE 9). ClinicalTrials.gov NCT07720271. 2026. https://clinicaltrials.gov/study/NCT07720271
- 10.Novo Nordisk A/S. Efficacy and Safety of Zenagamtide s.c. Once-weekly Compared to Semaglutide s.c. Once-weekly in Participants With Obesity (AMAZE 7). ClinicalTrials.gov NCT07668414. 2026. https://clinicaltrials.gov/study/NCT07668414
- 11.Novo Nordisk A/S. A Research Study Looking at the Effect of Timing of Meal After Dosing and Water Volume at Dosing on How the Medicine Zenagamtide is Taken up by the Body of Participants With Overweight or Obesity. ClinicalTrials.gov NCT07815067. 2026. https://clinicaltrials.gov/study/NCT07815067
- 12.Oldenburg LIK, Haugaard SP, Karlsson T, Kegel-Hübner V, Parwez M, Wiingaard C, Flint A. Renal Impairment Does Not Affect Pharmacokinetics, Safety or Tolerability of Zenagamtide. Diabetes Obes Metab. 2026. PMID: 42443140.
- 13.Rubin R. Unapproved GLP-1 Drugs Retatrutide and Zenagamtide Sold Online Are No Bargain. JAMA. 2026. PMID: 42853576.
Related research
Amycretin (Zenagamtide): Novo's Next-Gen Dual Amylin/GLP-1 Drug
Novo Nordisk's amycretin is an oral + subQ dual amylin/GLP-1 receptor co-agonist. Phase 1 oral showed up to -13.1% body weight at 12 weeks; phase 1b/2a subQ reached -24.3% at 36 weeks at 60 mg. We review the published data and the phase 3 program now underway.
11 min read
Cagrilintide: The Amylin Analogue Behind CagriSema — What the Evidence Shows
Cagrilintide is Novo Nordisk's investigational long-acting amylin analog and the amylin half of CagriSema. The phase 2 monotherapy trial showed about -10.8% weight loss at 26 weeks. What it is, the trial data, and what it means for patients.
8 min read
REDEFINE: The Full CagriSema Trial Program (2026)
The full CagriSema REDEFINE program: REDEFINE 1 (22.7% adherent weight loss), REDEFINE 2 in diabetes, and the REDEFINE 4 head-to-head miss versus tirzepatide.
13 min read
The Next GLP-1 Pills: Elecoglipron, Aleniglipron and Other Oral Drugs in Trials
A wave of daily small-molecule GLP-1 pills is moving through trials after Foundayo and the Wegovy pill. What 2026 trials of elecoglipron, aleniglipron, safiglipron (HRS-7535), VCT220 and TERN-601 showed, why TERN-601 was dropped, and why none is available yet.
8 min read
GLP-1 Pipeline Tracker: Every Next-Gen Weight-Loss Drug in Development (2026)
A tracker of next-generation GLP-1 and incretin drugs: retatrutide, CagriSema, MariTide, survodutide, amycretin, mazdutide, ecnoglutide and the approved oral pill orforglipron (Foundayo), with stage and peak reported weight loss.
7 min read
ZUPREME-1: Petrelintide Phase 2 Weight Loss Results
In the phase 2 ZUPREME-1 trial, weekly petrelintide lowered weight by up to 9.8% at 28 weeks and 10.7% at 42 weeks versus 1.7% on placebo, with nausea in 20%. What the doses, estimand and side effects show, and what the trial leaves open.
8 min read
Where to get GLP-1 safely: vetted online providers
Vetted telehealth providers that prescribe online. We compare pricing, form, and states served.
No insurance needed · vetted by our editors
WeightLossRankings.org is reader-supported. When you buy through links on our site, we may earn an affiliate commission. How we make money
CoreAge Rx
Lowest-price compounded semaglutide and tirzepatide with one flat price at every dose
Read full review →From $79/mo
“starting at” price
From $79/mo
sale price
Gala
Compounded semaglutide or tirzepatide at one all-doses price on a yearly plan, with free shipping
Read full review →From $129/mo
starting dose