The question everyone on a GLP-1 eventually asks is: what happens when I stop?

The clearest randomized answer so far comes from SURMOUNT-MAINTAIN — a trial designed to test whether continuing tirzepatide was better than stopping it. It was. Dramatically. In the placebo arm, 67% of participants (60 of 90) received rescue tirzepatide after regaining at least half of the weight they had lost. With continued full-dose treatment: 8%.

That's the best-case scenario: pre-selected responders, supervised trial setting, rescue therapy available.

Now there is a sham-controlled pivotal dataset designed specifically for the off-ramp.

What REMAIN-1 Actually Tested

Fractyl Health's REMAIN-1 pivotal study took a different approach. Instead of comparing "keep taking the drug" vs "stop," it asked: if someone has already stopped, does a single duodenal procedure help them maintain their results?

Study design:

308 participants in the pivotal cohort
All participants reached at least 15% weight loss on tirzepatide
Drug was stopped
Participants were randomized 2:1 to either the Revita procedure (endoscopic duodenal ablation) or a genuine sham procedure
Double-blind: neither participants nor investigators knew which group they were in

This is a sham-controlled design — the gold standard for device trials. The sham group underwent an identical-looking procedure with no active treatment.

The Six-Month Results

Announced October 16, 2026, in the modified intention-to-treat (mITT) population of 308 participants (least-squares mean weight regain):

Weight regain:

Revita: 5.1%
Sham: 8.2%
p=0.0023

Weight retention:

Revita retained approximately 80% of GLP-1-induced weight loss
Sham retained approximately 67%

Responders:

32% of Revita participants were weight-stable or continued losing weight at six months
18% in the sham group

What This Is — and What It Isn't

This is:

The first randomized, double-blind, sham-controlled pivotal trial specifically designed for the moment after stopping a GLP-1
Statistically significant (p=0.0023)
The strongest design standard available for evaluating a procedure

This is not:

A comparison between Revita and continuing on tirzepatide
Evidence that Revita is better than drug continuation (it almost certainly is not)
A study without meaningful limits

The limits you need to know:

These are company-reported topline results. They come from Fractyl's press release, not a peer-reviewed paper. Arm sizes, dropouts and full safety data have not been published.

The release reports a p-value but no confidence interval for the difference. A p-value tells you whether a result is unlikely to be chance. A confidence interval tells you the range of plausible effect sizes. Without it, you can't judge how large the true difference might be.

The most striking number is post hoc. Fractyl also reported 3.8% regain in patients with a complete ablation of at least 20 cm. Subgroups chosen after the fact generate hypotheses; they don't confirm them.

This is six-month data. The endpoint that determines whether the result holds — the share of participants still maintaining at least 5% total weight loss at 12 months — is the second co-primary endpoint, and it won't read out until Q1 2027.

The Context That Makes This Important

Three datasets now exist on the post-GLP-1 off-ramp, each using a different approach:

SURMOUNT-MAINTAIN (Lancet, 2026): Tirzepatide vs placebo for people who had already lost weight on the drug. In the placebo arm, 67% (60 of 90) received rescue tirzepatide after regaining at least half of their lost weight during the 52-week maintenance period. Source: Horn et al., The Lancet 2026;407:2305-2318. PMID 42119587.

Tufts Food is Medicine Institute pilot (Obesity Science & Practice, October 2026): 39 adults randomized at the moment of stopping to usual care, a digital wellness app, or medically tailored meals. Four months. Relative to usual care, both intervention groups gained about 7 percentage points less weight (app −7.16%, 95% CI −12.27 to −2.04; meals −7.17%, 95% CI −12.20 to −2.14). Weight was an exploratory outcome; pilot only. Source: Hollis-Hansen et al., DOI 10.1002/osp4.70194. PMID 42846130.

REMAIN-1 (Fractyl Health, October 2026): 308 participants, randomized, sham-controlled, six months. 5.1% vs 8.2% regain. The only one of the three with a sham control matched to the active procedure.

These three studies aren't directly comparable — different populations, interventions, and time frames. But they converge on one finding: the moment you stop is where results get made or lost, and structured interventions perform better than nothing.

What This Means If You're Currently on a GLP-1

Revita is not approved. Fractyl plans a De Novo FDA filing in late Q4 2026 and expects the review clock to start only after the 12-month endpoint reads out. A CPT code (1122T) for the procedure takes effect January 1, 2027 — a billing code, not an approval.

You cannot access the Revita procedure today outside of a clinical trial.

But what you can do today is build the infrastructure that REMAIN-1 proves matters:

Before you stop the drug (whether by choice or because coverage ends):

Get a body composition measurement as your baseline
Know your daily protein target (set it with your clinician or dietitian) and log your meals so you can see whether you actually hit it. PeptidesGPT members can log meals, including branded foods, and track protein against their goals.
Build resistance training habits now, while the drug is reducing friction
Talk to your prescriber about what a maintenance plan looks like before you need one

The drug produces the weight loss. What you build during the treatment window determines what you keep after it.

Sources

Fractyl Health, Inc. "Fractyl Health's Revita Meets Six-Month Co-Primary Endpoint in Ongoing REMAIN-1 Pivotal Study." GlobeNewsWire, October 16, 2026.
Horn DB, et al. "Tirzepatide for maintenance of bodyweight reduction in people with obesity in the USA (SURMOUNT-MAINTAIN)." The Lancet 2026;407(10545):2305-2318. DOI 10.1016/S0140-6736(26)00656-2. PMID 42119587.
Hollis-Hansen K, et al. "Preventing Weight Regain After GLP-1 Therapy Discontinuation: A Pilot Randomized Controlled Trial." Obesity Science & Practice, October 7, 2026. DOI 10.1002/osp4.70194. PMID 42846130.

PeptidesGPT is an educational platform. This content does not constitute medical advice. The Revita procedure is investigational and not FDA-approved. Decisions about GLP-1 therapy, discontinuation, or any other treatment should be made with a licensed healthcare provider.

What Happens to Your Weight After You Stop a GLP-1? New Sham-Controlled Data on the Off-Ramp