Peptide Therapy for Weight Loss: The Drug Is Tested, the Program Isn't

“Peptide therapy for weight loss” names a clinic product. It is not the name of a drug. It is a package: a GLP-1 or GIP receptor agonist, a dosing schedule, some monitoring, and often an add-on peptide the clinic markets as a booster.

The randomized trials that justify the whole category attach to exactly one piece of that package, the drug itself, at the specific dose a trial tested. Nobody has run a trial on the clinic’s version of the program, and most of what differentiates a branded “peptide therapy” protocol from the trial that got it approved rests on evidence far thinner than the headline number implies.

That is worth saying plainly before the numbers, because the numbers are genuinely strong. The problem is where clinics stop citing them.

What two randomized trials measured

The best human evidence for peptide weight loss comes from two placebo-controlled trials, each testing a single drug on its own.

In the STEP 1 trial, 1961 adults with overweight or obesity, none of them with diabetes, were randomized to once-weekly subcutaneous semaglutide 2.4 mg or placebo for 68 weeks, alongside a lifestyle intervention both groups received.^[1] Average weight change was -14.9% with semaglutide against -2.4% with placebo, a gap of 12.4 percentage points.^[1]

In the SURMOUNT-1 trial, 2539 adults with obesity were randomized to tirzepatide at 5 mg, 10 mg, or 15 mg weekly, or placebo, for 72 weeks.^[2] At the top 15 mg dose, average weight change was -20.9% against -3.1% with placebo.^[2]

Average body weight change from baseline

• Semaglutide 2.4 mg: -14.9
• Semaglutide placebo: -2.4
• Tirzepatide 15 mg: -20.9
• Tirzepatide placebo: -3.1

Highest tested dose of each drug versus its own trial's placebo arm.

Both were randomized, double-blind, and placebo-controlled, run by the STEP 1 Study Group and the SURMOUNT-1 Investigators respectively.^[1][2] That design is what lets anyone say the drug caused the weight loss rather than merely preceding it, whatever brand name sits on the vial. A clinic selling “peptide therapy” is selling access to that drug. It is not thereby selling a second trial proving its own version of the program works the same way.

What happens when you stop

An extension of the STEP 1 trial followed a subset of 327 participants after treatment and lifestyle support were withdrawn at week 68.^[3] By then, the semaglutide group had lost 17.3% of body weight. A year after stopping, they had regained two-thirds of it, for a net loss of 5.6% from where they started.^[3]

2/3 of the lost weight regained within a year of stopping semaglutide PMID 35441470

Bottom line: the trial evidence supports ongoing treatment. It does not support a short course that ends after a few months. A clinic that frames twelve weeks of peptide therapy as a fix misreads its own source material.

The mechanism the trials support

GLP-1 and GIP receptor agonism works mainly on the brain rather than directly on fat tissue. A 2025 review by Samms and Sloop describes GIP and GLP-1 receptor agonism both acting on the brain to suppress appetite, a mechanism with direct support from the physiology underlying the drugs tested in the weight-loss trials above.^[4] That is a real, physiology-grounded mechanism, and it is the one the weight-loss trials above measured the output of.

It is worth being specific about what this mechanism does not say. It does not say that any peptide which touches a related hormonal axis will produce the same weight change. That distinction is where the “therapy” part of peptide therapy tends to get loose.

What clinics add, and what evidence that addition has

Clinics rarely sell semaglutide or tirzepatide alone. The add-ons are where the evidence base changes shape entirely.

• CJC-1295. A placebo-controlled, double-blind trial by Teichman and colleagues tested this growth-hormone-releasing-hormone analog in healthy adults aged 21 to 61 and found it raised peak and cumulative growth hormone and IGF-1 concentrations.^[5] Body weight was not a measured outcome in that trial. Raising a hormone linked to metabolism is not the same finding as lowering a number on a scale, and the paper does not claim otherwise.
• AOD-9604. No human trial of this compound for weight loss turned up anywhere in a search of the published record, successful or otherwise. A claim this specific needs a citation this specific, and none exists to supply it.
• Compounded semaglutide. A pharmacovigilance study using the FDA’s Adverse Event Reporting System analyzed 707 reports involving compounded GLP-1 receptor agonists out of 81,078 total.^[6] Compounded products carried far higher reporting odds than FDA-approved versions for contamination (odds ratio 19.00) and for compounding or manufacturing errors (odds ratio 8.51).^[6] That is not a quality assurance a buyer can see on the vial label.

The CJC-1295 pattern is not unique to weight loss. It shows up across the peptide market: a real, measurable biomarker change gets marketed as if it were the outcome a buyer wants. ARA-290 has more human trial data behind it than most peptides sold online, and even there the gap between what was measured and what gets implied in marketing copy keeps showing up rather than closing. Among peptides pitched specifically at stubborn fat rather than overall body weight, the trial record narrows even further, which is the subject of a separate look at peptides for visceral fat.

How the trials dosed it

In the SURMOUNT-1 trial, participants were escalated over a 20-week period before settling into their assigned 10 mg or 15 mg weekly maintenance dose.^[2] In STEP 1, participants on semaglutide reached and held a 2.4 mg weekly dose for the 68-week treatment period.^[1]

Neither trial tested a compounded version, a combination with an added peptide, or an accelerated schedule. None of that has a trial behind it with body weight as the measured outcome. This describes what the trials used; it is not dosing advice. Whatever a clinic protocol does differently from the published titration, it is doing on evidence that has not been collected. That absence is different from evidence that contradicts the trials.

Side effects, documented and undocumented

The trials themselves are candid about adverse events. In STEP 1, nausea and diarrhea were the most common side effects with semaglutide, typically transient and mild to moderate, and 4.5% of the semaglutide group discontinued for gastrointestinal reasons against 0.8% on placebo.^[1] A meta-analysis of ten tirzepatide trials covering 6836 participants found the same pattern for that drug: gastrointestinal events were the most commonly reported adverse events, occurring more often at higher doses, with nausea and diarrhea the most frequent at any dose.^[7]

A separate meta-analysis of nine tirzepatide trials totaling 9871 participants found a more specific signal: a significantly higher rate of gallbladder or biliary disease with tirzepatide compared with a pooled control group of placebo and basal insulin.^[8]

What none of this covers is combination use. No published trial has measured the safety of pairing a GLP-1 or GIP agonist with an additional peptide like CJC-1295 in the same protocol. That is not a reassuring unknown, it is simply unmeasured, and a clinic that reassures a patient on the strength of the drug’s trial safety data is answering a question about ingredient A while quietly adding ingredient B.

What's been tested

By component of a typical peptide therapy program.

What a clinic’s own results can’t prove

A clinic that reports its own patients lost weight on its peptide therapy program is describing an uncontrolled observation. There is no comparison group, no randomization, and usually no publication. Without those, the clinic’s number cannot separate the drug’s own effect from the diet counseling bundled into the visit, from the fact that people who pay for a monitored program tend to be more motivated than average, or from plain regression to the mean among patients who started at their heaviest.

This is the difference between interventional and observational evidence, and it is not a technicality. A randomized trial assigns the drug and compares it against something; that structure is what allows a causal sentence like “semaglutide caused this weight loss.” A clinic’s case series does not have that structure, and no amount of patient testimonials substitutes for it.

The causal claim the trials support is scoped tightly: this drug, at this dose, in this population, for this long. It does not extend automatically to a different vial, a different combination, or a different clinic’s protocol just because the brand name on the intake form matches the one in the trial.

What this means in practice: the evidence supports the specific drug at the specific dose in the published trials, taken on an ongoing basis. It does not extend to whatever a clinic adds alongside it, and it offers no safety data on combinations nobody has studied.

None of this is an argument against semaglutide or tirzepatide, whose trial data is about as strong as anything in the peptide and incretin space, closer in evidentiary weight to an established drug class than to the natural peptides sold with a wellness-brand gloss and no trial behind them at all. It is an argument against treating a clinic’s branded bundle as if the trial’s conclusions traveled with it intact. Those conclusions stop at the vial the trial used.

If a program adds a peptide, changes the schedule, or buys from a compounding pharmacy, ask what trial, if any, covers that specific change. For semaglutide and tirzepatide alone, the answer is a strong one. For almost everything clinics sell around them, the honest answer is still: no trial yet.

This article is for research and informational purposes only and is not intended to diagnose, treat, cure, or prevent any disease. The peptides discussed here are sold for research use only and are not for human consumption. Nothing in this article constitutes medical advice. Consult a qualified clinician before making changes to a health, training, or supplementation protocol.