What do peptides do? What published studies found

What do peptides do in the body?

Peptides are short chains of amino acids, the building blocks of proteins, and many of them work as messengers. The TGA describes peptides as chains found naturally in the body that take part in the signalling processes of growth, metabolism and tissue repair (TGA, peptide products page, 23 September 2026). Insulin is the best-known case: it was isolated in 1921 and remains a peptide medicine today (Sport Integrity Australia, Peptides explained, page read 14 October 2026). Drug makers copy or reshape some natural peptides to build medicines, and a 2021 review counted more than 80 that have reached the market (Muttenthaler et al., Nature Reviews Drug Discovery, 2021).

Your gut already makes GLP-1 (glucagon-like peptide-1, a hormone released after you eat). Semaglutide is a GLP-1 receptor agonist, a molecule built to fit the same receptor, the docking site on a cell where the hormone acts (Ashraf et al., Journal of Medical Internet Research, 2024). Tirzepatide fits both the GIP and GLP-1 receptors (Jastreboff et al., NEJM, 2022). Retatrutide, still an investigational peptide, fits GIP, GLP-1 and glucagon receptors (Jastreboff et al., NEJM, 2023). The retatrutide phase 2 trial page sets that trial out in full.

What a peptide does depends on its sequence, the order of its amino acids. Changing one amino acid can change which receptor a peptide fits, so peptides as a group do not share one effect. A result for one named peptide says nothing about another.

What are peptides used for?

More than 80 peptide drugs have reached the market for diseases including diabetes, cancer, osteoporosis, multiple sclerosis, HIV infection and chronic pain (Muttenthaler et al., Nature Reviews Drug Discovery, 2021). The TGA lists peptide-based medicines used to manage conditions such as diabetes, obesity and cardiovascular disease (TGA, peptide products page, 23 September 2026). Peptide products promoted online for weight loss, muscle growth, anti-ageing or recovery are a different group. Many are unapproved, and the TGA states that they have not been assessed by the regulator for safety, quality or effectiveness (TGA, 23 September 2026).

What have studies of therapeutic peptides found? The most direct records are randomised trials, where one group receives the peptide and another a placebo. The sections below report them one compound at a time, with the size of each trial and its limits. A 2022 review of the field notes that advances in production, modification and delivery have helped overcome the inherent drawbacks of peptides as drugs (Wang et al., Signal Transduction and Targeted Therapy, 2022).

Whether a given peptide may be imported, supplied or possessed in Australia is a separate legal question. The peptides and Australian law page records the Poisons Standard entries and the TGA position with dates.

Evidence limits

What do studies of research peptides not show?

In its briefing document for its July 2026 advisory committee meeting, the US Food and Drug Administration (FDA) found five published clinical studies of BPC-157, all short and small, and wrote that the evidence was insufficient to conclude that BPC-157 is effective for ulcerative colitis, the nominated use it reviewed (FDA briefing document, Pharmacy Compounding Advisory Committee, 23–24 July 2026).

Registries add to the picture. A search of ClinicalTrials.gov on 14 October 2026, by compound name only, returned no registered study for KPV, dihexa, IGF-1 LR3 or AHK-Cu, and four for BPC-157. A registry search by name can miss studies filed under other names, so the count is a floor, not proof that no study exists.

How do researchers tell whether a peptide works?

Researchers compare a measured outcome in people who received the peptide with the same outcome in people who received a placebo, usually in a trial where neither group knows which it got. The trials in this article measured body weight on a scale and skin hydration with instruments (Wilding et al., NEJM, 2021; Nukaly et al., Frontiers in Medicine, 2026). A personal sense that it worked is not a measure. A 2021 chart review of 17 people with knee pain who had received BPC-157, followed up by phone, used no tool to measure function or stiffness, which limits what it shows (Lee and Padgett, Alternative Therapies in Health and Medicine, 2021).

A trial also fixes its primary endpoint, the main outcome it set out to measure, before the results arrive. STEP 1 named two: the percentage change in body weight and a weight reduction of at least 5% (Wilding et al., NEJM, 2021). Choice of measure changes the answer. A 2026 meta-analysis of 19 randomised skin trials with 1,341 participants found that peptides improved hydration and brightness while effects on elasticity and density were inconsistent, a pooled result across different peptides and products (Nukaly et al., Frontiers in Medicine, 2026).

How quickly did peptides show effects in published studies?

The time to a measured effect in the trials cited here ran from weeks to more than a year, depending on the peptide and the outcome. In the maker-funded STEP 1 trial of 1,961 adults without diabetes, body-weight results were reported at 68 weeks, for one medicine in that population (Wilding et al., NEJM, 2021). In the maker-funded SURMOUNT-1 trial of 2,539 adults without diabetes, they were reported at 72 weeks, for a second medicine in that population (Jastreboff et al., NEJM, 2022). In the maker-funded phase 2 retatrutide trial of 338 adults, weight was reported at 24 and 48 weeks, for an investigational medicine in that trial population (Jastreboff et al., NEJM, 2023).

In the maker-funded phase 2 trial in 338 adults, the highest-dose arm was down 17.5% at 24 weeks and 24.2% at 48 weeks, results limited to that investigational medicine and trial population (Jastreboff et al., NEJM, 2023).

The skin trials were shorter. A split-face trial of 93 women ran 12 weeks (Robinson et al., International Journal of Cosmetic Science, 2005), and a three-arm trial of 21 women ran eight weeks (Aruan et al., Journal of Clinical and Aesthetic Dermatology, 2023). The BPC-157 knee report relied on recall of events 6 to 12 months earlier (Lee and Padgett, 2021).

What do studies show about peptides and weight loss?

Three peptides have large randomised trials for body weight: semaglutide, tirzepatide and retatrutide, all of which act on gut-hormone receptors. In the 68-week STEP 1 trial of 1,961 adults without diabetes, average body weight fell 14.9% with semaglutide against 2.4% with placebo, in a maker-funded trial of an approved medicine that applies to that population (Wilding et al., NEJM, 2021). In the 72-week SURMOUNT-1 trial of 2,539 adults without diabetes, the three tirzepatide arms averaged weight loss of 15.0% to 20.9% against 3.1% with placebo, in a second maker-funded trial whose findings do not carry over to other medicines (Jastreboff et al., NEJM, 2022). In the maker-funded phase 2 retatrutide trial of 338 adults, the highest-dose arm lost 24.2% at 48 weeks against 2.1% with placebo, a result limited to that investigational medicine and trial population (Jastreboff et al., NEJM, 2023).

These numbers describe trial medicines studied under medical supervision in selected adults, and the trials were funded by the makers of those medicines (Wilding et al., 2021; Jastreboff et al., 2022, 2023). Semaglutide and tirzepatide are approved medicines. Retatrutide is an investigational medicine that the TGA states has not been approved by it or any comparable international regulator (TGA safety advisory, 14 September 2026). Research material is a laboratory compound, not the trial medicine and not a substitute for any medicine.

The same trials recorded harms. In the maker-funded STEP 1 trial of 1,961 adults, nausea and diarrhoea were the most common adverse events with semaglutide (Wilding et al., NEJM, 2021). In the maker-funded SURMOUNT-1 trial of 2,539 adults and the maker-funded phase 2 retatrutide trial of 338 adults, gastrointestinal events were the most common with tirzepatide and retatrutide, and the findings describe those medicines and populations only (Jastreboff et al., NEJM, 2022; Jastreboff et al., NEJM, 2023). The peptide side effects page sets out what trials and regulators recorded.

Do growth hormone peptides work, and which is strongest?

No head-to-head trial of growth hormone-releasing compounds was identified in the sources checked for this page, so none is ranked here. In a 26-week trial of 412 adults with HIV and abdominal fat, tesamorelin cut visceral fat by 15.2% while it rose 5.0% with placebo, in a trial of one condition in which 86% of participants were men (Falutz et al., NEJM, 2007). In a one-year trial of 65 healthy older adults, MK-677 (an oral small molecule, not a peptide) raised fat-free mass by 1.1 kg against a 0.5 kg fall with placebo, with no change in strength or function, in a small trial of one group (Nass et al., Annals of Internal Medicine, 2008).

MK-677 (ibutamoren) is an orally active small molecule, not a peptide, although it is grouped with growth hormone secretagogues (Sigalos and Pastuszak, Sexual Medicine Reviews, 2018). The two trials used different people (adults with HIV, healthy older adults), different lengths (26 weeks, one year) and different outcomes (Falutz et al., 2007; Nass et al., 2008). No trial in the sources checked compared one compound with another, so the figures cannot be lined up. A 2018 review concluded that few long-term, rigorously controlled studies of these compounds exist (Sigalos and Pastuszak, 2018).

Growth hormone itself has been tested in young, fit people. A 2008 systematic review of 27 study samples with 303 treated participants found that growth hormone in young, fit people raised lean body mass by 2.1 kg while strength and exercise capacity did not seem to improve, and that soft-tissue swelling and fatigue were more common, a review of growth hormone itself and not of a secretagogue (Liu et al., Annals of Internal Medicine, 2008). The tesamorelin research guide covers that compound's trial record in more detail.

Are peptides anti-aging? What do skin studies show?

Skin studies measured appearance, not ageing itself. A 2026 meta-analysis of 19 randomised trials with 1,341 participants found that oral and topical peptides improved skin hydration and brightness, with a modest effect on wrinkles (a mean difference of 0.27) and inconsistent effects on elasticity and density, a pooled result across different peptides and products (Nukaly et al., Frontiers in Medicine, 2026). A 2023 meta-analysis of 26 trials with 1,721 participants found oral hydrolysed collagen improved skin hydration compared with placebo and noted several biases in the included trials (Pu et al., Nutrients, 2023). Both reviews measured skin outcomes only.

Which peptide is the strongest anti-ageing one? No trial ranking them was identified in the sources checked. The one head-to-head record reviewed here is an eight-week trial of 21 Indonesian women comparing acetylhexapeptide-3 cream, palmitoyl pentapeptide-4 cream and placebo, in which the pentapeptide cream appeared to give better results, and the authors called it an initial study needing more participants and a longer follow-up (Aruan et al., Journal of Clinical and Aesthetic Dermatology, 2023).

Sagging skin is usually measured as elasticity. In the 2023 meta-analysis of 26 oral collagen trials with 1,721 participants, which noted several biases in the included trials, elasticity showed no significant difference by collagen source (p = 0.21) or by measurement method (p = 0.06), while hydration effects varied with source and with how long the trial ran (Pu et al., Nutrients, 2023). The review split trials by collagen source, so a result for one source is not a result for all.

What peptides are in skincare, and do peptide moisturisers work?

A 2009 review sorted skincare peptides into four groups: signal peptides, enzyme-inhibitor peptides, neurotransmitter-inhibitor peptides and carrier peptides (Gorouhi and Maibach, International Journal of Cosmetic Science, 2009). In a 12-week placebo-controlled, split-face trial of 93 women aged 35 to 55, a palmitoyl pentapeptide moisturiser, built to stimulate collagen production, improved wrinkles and fine lines compared with the same moisturiser without it, in one trial of one formulation that does not establish results for other peptides (Robinson et al., International Journal of Cosmetic Science, 2005). Copper-binding GHK-Cu has a larger lab record than a human one: a 2026 systematic review that kept 64 of 1,247 screened records found most randomised human trials of it small and run on facial creams (Najafi et al., Archives of Internal Medicine Research, 2026).

The 2026 meta-analysis of 19 randomised trials with 1,341 participants reported that its wrinkle result was largely driven by oral polypeptides, a pooled result across different peptides and products (Nukaly et al., Frontiers in Medicine, 2026). Topical trials are also short and small: 21 women for eight weeks (Aruan et al., 2023) and 93 women for twelve weeks (Robinson et al., 2005). A 2007 review of cosmeceutical peptides noted that their effects may not be as rigorously tested as those of most FDA-regulated drugs, while the supporting evidence was growing (Lupo and Cole, Dermatologic Therapy, 2007).

A 2018 review of laboratory work reports that GHK, the tripeptide behind GHK-Cu, stimulated collagen, elastin and glycosaminoglycan synthesis in skin cells, a cell result that does not establish the same effect in a person's skin (Pickart and Margolina, International Journal of Molecular Sciences, 2018). The GHK-Cu research guide lists the compound's published record.

Can peptides heal a torn meniscus?

A PubMed search on 14 October 2026 did not identify a human study showing that a peptide healed a torn meniscus, and a single search can miss studies filed under other terms. The search 'BPC-157 meniscus' returned two records. One was a 2026 scoping review that searched six peptides against musculoskeletal tissues including meniscus and found human data limited to a handful of weakly controlled studies, with no study count in its abstract (Tewari et al., American Journal of Sports Medicine, 2026). The other was a 2021 chart review of 17 people with knee pain who had received BPC-157, which lists meniscus tears among many causes of knee pain and notes that follow-up MRI scans were still needed, so it does not show healing of a tear (Lee and Padgett, Alternative Therapies in Health and Medicine, 2021).

The wider record is mostly animal work. A 2025 systematic review of 36 BPC-157 studies, 35 of them preclinical, found improved outcomes in animal muscle, tendon, ligament and bone injury models, and one small retrospective clinical report of relief in 7 of 12 patients with chronic knee pain of unspecified cause, so the animal findings do not establish the same effects in humans (Vasireddi et al., HSS Journal, 2025).

FDA's July 2026 briefing noted that BPC-157 had been nominated for tendonitis, that the nomination lacked enough information for FDA to evaluate that use, and that FDA identified no clinical studies in that population (FDA briefing document, 23–24 July 2026). The BPC-157 research guide sets out the compound's study record.

What do studies show about KPV, AHK-Cu and dihexa?

Studies of KPV, AHK-Cu and dihexa were done in cells, tissue samples or animals, and a ClinicalTrials.gov search by compound name on 14 October 2026 found no registered human study for any of them, although a name search can miss studies filed under other names. In a 2008 mouse study of two colitis models, with no mouse count in its abstract, the three-amino-acid peptide KPV reduced the incidence of colitis, a result in mice that does not establish an effect in humans (Dalmasso et al., Gastroenterology, 2008). In a second 2008 mouse study of two colitis models, with no mouse count in its abstract, KPV was linked to earlier recovery and less colonic inflammation, again in mice only (Kannengiesser et al., Inflammatory Bowel Diseases, 2008). In a 2007 laboratory study, with no sample count in its abstract, AHK-Cu lengthened human hair follicles kept in culture and increased the growth of cultured hair-papilla cells, which does not establish an effect on a person's scalp (Pyo et al., Archives of Pharmacal Research, 2007). In rats, with no animal count in the abstract, dihexa reversed scopolamine-induced memory deficits and showed activity in aged rats, a result that does not establish an effect in humans and that comes from a paper carrying a 2021 notice of concern on PubMed, so it needs cautious reading (McCoy et al., Journal of Pharmacology and Experimental Therapeutics, 2013).

FDA states that it has not identified any human exposure data for KPV or dihexa acetate, and that it lacks important information on whether either would cause harm to humans (FDA, bulk drug substances page, content current as of 22 April 2026). FDA's advisory committee met on KPV for wound healing and inflammatory conditions on 23 July 2026 (FDA meeting page, 23–24 July 2026).

AHK-Cu is the copper complex of the three-amino-acid peptide alanyl-histidyl-lysine. Its hair study used follicles in a dish, so it cannot say what happens on a person's scalp (Pyo et al., 2007). Dihexa's rat work used scopolamine, a drug that impairs memory, and aged rats, and it does not establish an effect in people (McCoy et al., 2013).