Peptide Science Desk

Do peptides really work? The honest answer

Some do. Most wellness peptides are unproven.

Approved peptide drugs passed big trials in people. Most others have only animal or tiny studies.

By the Peptide Science Desk Editorial TeamFact-checked against FDA and official sources · Not yet reviewed by a licensed clinician. How we check facts

Short answer: Some do. Approved peptide medicines like semaglutide (Wegovy) and tirzepatide (Zepbound) worked in big trials with thousands of people. Most wellness peptides sold online, like BPC-157 or TB-500, have mostly animal studies or tiny human studies. So nobody can honestly promise they work in people. New to peptides? Start here: what are peptides?
Four scientists in lab coats look at a cell culture flask next to a microscope, black and white
Scientists look at cells growing in a flask (1987 photo). Many peptide studies so far were done on cells like these, not in people.Photo: National Cancer Institute (NIH), public domain

Do peptides really work?

  • "Peptides" is a huge group. Insulin is a peptide. So are Wegovy and Zepbound. So is BPC-157 sold online. They are not all the same.
  • The real question is: which peptide, for what, and was it tested in people?
  • We track 50 peptides. Of the 27 that are not FDA approved and not in late testing, none has strong proof in people for its main use. 9 have only animal or lab studies (or none), 12 have only small early studies, 4 have mixed results, and 2 were tested and did not clearly help.

Do peptides really work for weight loss?

  • Some approved ones do. In a big 68-week trial of 1,961 adults, people on semaglutide (Wegovy) lost about 14.9% of their weight on average. People on fake shots (placebo) lost about 2.4%.
  • In a 72-week trial of 2,539 adults, people on tirzepatide (Zepbound) lost about 15% to 21% on average, depending on the dose. The placebo group lost about 3.1%.
  • These are averages. Some people lost more and some lost less.
  • "Research" versions sold online are not the same products. The FDA says unapproved versions are not checked for safety, how well they work, or quality.

Do peptides really work for muscle growth?

  • Ipamorelin: no study in people tested it for muscle, sleep, or anti-aging.
  • CJC-1295: one small study showed growth hormone levels went up after a shot. It did not measure muscle or fat.
  • MK-677 (often sold as a peptide): one trial in older adults found a little more lean mass, but no gain in strength.
  • A higher hormone number on a blood test is not the same as bigger or stronger muscles.

Do peptides really work for skin, wrinkles, and hair?

  • GHK-Cu (copper peptide): one small study of 13 people found no clear change in wrinkles or redness when experts measured skin. Users said their skin looked better.
  • Hair: no peptide is FDA approved for hair loss. A few small studies of peptide lotions looked hopeful. See our guide on peptides for hair growth.

How long do peptides take to work?

  • For approved drugs, trials checked results at set times. The semaglutide trial measured weight at 68 weeks. The tirzepatide trial measured it at 72 weeks. Your doctor and the drug label are the right guide.
  • For most wellness peptides, there are no good studies in people. So any "results in 2 weeks" promise online is not based on real trials.

Why do people say peptides worked for them?

  • Placebo: in the semaglutide trial, people on fake shots still lost weight on average, because they also got a diet and exercise program.
  • Many changes at once: people often start a peptide with a new diet, more sleep, more workouts, or other peptides (a "stack"). Then nobody can tell which change helped.
  • Time: some aches and injuries get better on their own.
  • That is why good trials compare with a fake treatment, and neither the patients nor the doctors know who got what ("double-blind").

Are peptides worth it?

  • For approved peptide medicines used the approved way, big trials show real benefits and known side effects. Ask your doctor if one fits you.
  • For unapproved peptides, you would pay for something not proven in people. The FDA says products sold "for research purposes" are of unknown quality and may be harmful.
  • Whether a peptide works, whether it is safe, and whether it is FDA approved are 3 different questions. We rate each one on its own. Little safety data does not mean a peptide does not work. It means the risks are unknown.

The numbers in charts

Weight loss in big obesity drug trials

% of starting body weight lost, drug vs placebo (fake shot)

Different trials, different people: not a head-to-head comparison.

Show the numbers as a table
Weight loss in big obesity drug trials: % of starting body weight lost, drug vs placebo (fake shot)
TrialDrugPlacebo
Tirzepatide 15 mg (Zepbound)
SURMOUNT-1 · 2,539 adults · 72 weeks		−20.9%−3.1%
Semaglutide 2.4 mg (Wegovy)
STEP 1 · 1,961 adults · 68 weeks		−14.9%−2.4%
Liraglutide 3.0 mg (Saxenda)
SCALE · 3,731 adults · 56 weeks		−7.4%−3.0%
Mean % change in body weight at each trial's main endpoint. The liraglutide percentages come from the Saxenda FDA label (Table 4, Study 1); the trial paper gives the number of adults and the length. Sources: Jastreboff et al., NEJM 2022 (SURMOUNT-1) (PMID 35658024); Wilding et al., NEJM 2021 (STEP 1) (PMID 33567185); Pi-Sunyer et al., NEJM 2015 (SCALE) (PMID 26132939); Saxenda FDA label, Table 4, Study 1 (DailyMed).

How many people each peptide was tested in

Biggest study in people we list for each one (log scale, so small numbers still show)

The popular online peptides were tested in tens of people, or none. The approved drugs were tested in thousands.

Show the numbers as a table
How many people each peptide was tested in: Biggest study in people we list for each one (log scale, so small numbers still show)
PeptidePeople in the study
GLP-1: Wegovy / Ozempic (semaglutide) (FDA approved)
SELECT heart trial		17,604 people
GLP-1: Trulicity (dulaglutide) (FDA approved)
REWIND heart trial		9,901 people
GLP-1: Saxenda / Victoza (liraglutide) (FDA approved)
SCALE		3,731 people
Cagrilintide (part of CagriSema) (In testing)
REDEFINE 1 (with semaglutide)		3,417 people
GLP-1 pill: Foundayo (orforglipron) (FDA approved)
ATTAIN-1 · not a peptide		3,127 people
GLP-1: Zepbound / Mounjaro (tirzepatide) (FDA approved)
SURMOUNT-1		2,539 people
Tymlos (abaloparatide) (FDA approved)
ACTIVE		2,463 people
GLP-1: Retatrutide (In testing)
TRIUMPH-1		2,339 people
Forteo (teriparatide) (FDA approved)
Fracture trial		1,637 people
Vyleesi (bremelanotide, PT-141) (FDA approved)
RECONNECT (2 trials)		1,267 people
Thymosin alpha-1 (Not FDA approved)
TESTS sepsis trial · no benefit		1,106 people
Egrifta (tesamorelin) (FDA approved)
Belly fat trial		412 people
Ipamorelin (Not FDA approved)
Phase 2 (gut, after surgery)		114 people
Semax (Not FDA approved)
Not randomized		110 people
Selank (Not FDA approved)
Small randomized study		70 people
BPC-157 (Not FDA approved)
All 3 small studies added up		30 people
GHK-Cu (Not FDA approved)
Skin product trial		13 people
TB-500 (Not FDA approved)
No study of TB-500 itself		0 people
KPV (Not FDA approved)
Mice only		0 people
MOTS-c (Not FDA approved)
Mice only		0 people
AOD-9604 (Not FDA approved)
Rats only		0 people
Bar length uses a log scale: each step is 10 times more people. For TB-500 we found 2 small trials of the full thymosin beta-4 protein (18 and 73 people), not TB-500 itself. Each number is on that peptide's page with a link to the study. Sources: Our 50 peptide pages, "Studies in people" tables; Lincoff et al., NEJM 2023 (SELECT) (PMID 37952131); Jastreboff et al., NEJM 2026 (TRIUMPH-1) (PMID 42814954).

Fewer heart attacks, strokes and heart deaths

% of people who had one of these, drug vs placebo, in 2 big heart trials

Different trials and different people. Lower is better.

Show the numbers as a table
Fewer heart attacks, strokes and heart deaths: % of people who had one of these, drug vs placebo, in 2 big heart trials
TrialDrugPlacebo
Semaglutide 2.4 mg (Wegovy)
SELECT · 17,604 adults with heart disease and extra weight, no diabetes · about 40 months		6.5%8.0%
Dulaglutide (Trulicity)
REWIND · 9,901 adults with type 2 diabetes		12.0%13.4%
Main result of each trial: heart attack, stroke, or death from heart causes. In SELECT, 20% fewer people had one (hazard ratio 0.80). Sources: Lincoff et al., NEJM 2023 (SELECT) (PMID 37952131); Gerstein et al., Lancet 2019 (REWIND) (PMID 31189511).

Two big trials where the peptide did not help

% of very sick hospital patients who died, peptide vs placebo

Almost the same in both groups: no clear benefit.

Show the numbers as a table
Two big trials where the peptide did not help: % of very sick hospital patients who died, peptide vs placebo
TrialPeptidePlacebo
Thymosin alpha-1 for sepsis
TESTS · 1,106 adults · deaths by day 28		23.4%24.1%
VIP (aviptadil) for severe COVID-19
TESICO · 461 adults · deaths by day 90		38%36%
Two different trials and diseases. Different trials, different people: not a head-to-head comparison. Sources: Wu et al., BMJ 2025 (TESTS) (PMID 39814420); Brown et al., Lancet Respir Med 2023 (TESICO) (PMID 37348524).

More charts: Peptides by the numbers

What studies in people show

Abstracts checked on PubMed on October 19, 2026. Results are averages from each study. No doses on purpose.

Study typePeopleMain finding
Big placebo-controlled trial, 68 weeks: semaglutide (STEP 1)1,961 adultsWeight fell 14.9% on average vs 2.4% on placebo. Nausea and diarrhea were the most common side effects.
Big placebo-controlled trial, 72 weeks: tirzepatide (SURMOUNT-1)2,539 adultsWeight fell about 15% to 21% on average vs 3.1% on placebo. Stomach side effects were most common.
Placebo-controlled trial, 48 weeks: retatrutide (not approved)338 adultsWeight fell up to 24.2% on the highest dose vs 2.1% on placebo. Bigger trials are still running.
Review of all human studies (2026): BPC-15730 people in 3 studiesOnly 3 small studies, none with a placebo group. Proof rated "very low."
Small randomized study: GHK-Cu skin products13Experts saw no clear change in wrinkles or redness. Very small.
Randomized trial: MK-677 in older adults65A little more lean mass, but no gain in strength. Blood sugar went up.
TB-500 itself in people0We found no published study of TB-500 in people.
Check any peptide in 1 secondFree FDA Status Checker: approved, being tested, or not proven.

FAQ

What counts as real proof that a peptide works?

A trial in people that compares the peptide with a fake treatment (placebo), picks who gets what by chance, and is published so others can check it.

Is a Reddit success story proof?

No. One person’s story cannot rule out placebo, other changes they made, or time. It can be a reason to ask good questions.

Does "clinically studied" on a label mean it works?

Not by itself. Ask what was studied: the same product, in people, against placebo, and how many people.

Which peptides are proven to work?

FDA-approved peptide medicines for their approved uses, like semaglutide and tirzepatide for weight. Use our FDA status checker to look up any peptide.

If a peptide has little safety data, does that mean it does not work?

No. Safety and proof are separate. Little safety data means the risks are unknown, not that it fails.

Related pages

We never share doses or tell you where to buy. Always ask your doctor. Print our doctor questions sheet.

Sources

Not medical advice. Medical disclaimer.