Peptide side effects: what studies and regulators have recorded

Do peptides have side effects?

Yes, in the peptide medicine trials described below. In the 1,961-adult STEP 1 trial of semaglutide in adults without diabetes, nausea and diarrhoea were the most frequent adverse events and 4.5% of semaglutide participants stopped treatment because of gastrointestinal events against 0.8% with placebo, a finding limited to that maker-funded trial of that medicine (Wilding et al., NEJM, 2021). In the 2,539-adult SURMOUNT-1 trial of tirzepatide in adults without diabetes, adverse events led 4.3%, 7.1% and 6.2% of the three tirzepatide arms to stop, against 2.6% with placebo, a finding that holds for that tirzepatide trial and not for other peptides (Jastreboff et al., NEJM, 2022). Which side effects occur depends on the compound, so a list for one peptide does not carry over to another.

In the maker-funded phase 2 trial of 338 adults, gastrointestinal events with retatrutide were mostly mild to moderate and more frequent in the arms assigned higher amounts, and heart rate rose more in those arms, peaked at 24 weeks and declined afterwards; the findings apply to that investigational medicine in that trial population only (Jastreboff et al., NEJM, 2023).

Other peptide-axis medicines show different patterns. In a registry of 306 children who had an approved recombinant IGF-1 medicine (mecasermin), 80 had at least one hypoglycaemia adverse event, a finding that applies to that medicine and population only (Bang et al., Journal of Clinical Endocrinology and Metabolism, 2023). In a one-year trial of 65 healthy older adults, MK-677 (an oral small molecule, not a peptide) raised fasting blood glucose and cortisol, and the most frequent side effects were increased appetite, mild leg swelling and muscle pain; the trial was small and does not predict results in other groups (Nass et al., Annals of Internal Medicine, 2008).

Are peptides safe, and what is the risk?

The evidence differs by compound and product. The TGA states that unapproved peptide products have not been assessed for safety, quality or effectiveness, so there is no assurance they are safe, effective, correctly labelled or sterile (TGA, peptide products page, 23 September 2026). FDA concluded in July 2026 that there is insufficient clinical safety information to characterise the safety profile of BPC-157 (FDA briefing document, Pharmacy Compounding Advisory Committee, 23–24 July 2026). Whether peptides work is a separate question, covered on the page about what peptides do.

The risks named by regulators are about the product as well as the molecule. The TGA's Chief Medical Advisor, Professor Robyn Langham, said that the biggest risks with unapproved peptide products are not knowing what is in the vial, how much peptide it holds, whether it is sterile or whether it carries other contaminants (TGA, 7 May 2026). FDA names immunogenicity, the chance of an unwanted immune response, linked to peptide aggregation and peptide-related impurities (FDA, bulk drug substances page, content current as of 22 April 2026).

A 2026 scoping review of the published literature on six popular research peptides reported that they are often perceived as low risk despite the absence of efficacy or safety data for many of them, and the review summarises other studies and ran no trial of its own (Tewari et al., American Journal of Sports Medicine, 2026).

Evidence limits

What has not been shown about the safety of research peptides?

An absence of data is not evidence of safety. FDA wrote that its review of BPC-157 found no human pharmacokinetic studies for the routes it reviewed, and that ulcerative colitis can need long-term repeated treatment, so the information is insufficient to support patient safety over long-term use (FDA briefing document, 23–24 July 2026).

What side effects have been recorded for growth hormone peptides?

In a one-year trial of 65 healthy older adults, MK-677 (an oral small molecule, not a peptide) raised fasting blood glucose, lowered insulin sensitivity and raised cortisol, and the most frequent side effects were increased appetite, mild leg swelling and muscle pain, in a small trial that does not predict results in other groups (Nass et al., Annals of Internal Medicine, 2008). In a randomised trial of 123 older patients recovering from hip fracture, an ibutamoren (MK-677) safety signal led investigators to end the trial early; this finding concerns that postoperative population and does not establish the frequency of heart failure in other groups (Adunsky et al., Archives of Gerontology and Geriatrics, 2011; FDA, bulk drug substances page, content current as of 22 April 2026). Separately, FDA lists increased heart rate and a vasodilatory reaction, a widening of blood vessels, for CJC-1295, in a summary of limited clinical data that is not a trial and does not give how often either occurs (FDA, bulk drug substances page, content current as of 22 April 2026).

For GHRP-2 and GHRP-6, two growth hormone-releasing peptides, FDA states that it is aware of reports of serious adverse events with GHRP-2, including a higher insulin requirement, infection, pancreatitis and the death of critically ill study subjects, though causality has not been established. For GHRP-6 it cites a possible effect on cortisol and a rise in blood glucose through lower insulin sensitivity, as a possibility in its summary and not as a measured rate (FDA, content current as of 22 April 2026).

In a phase 2 trial of 117 enrolled patients after bowel surgery, 114 of whom formed the safety analysis, any adverse event occurred in 87.5% of the ipamorelin group and 94.8% of the placebo group, and the authors called the peptide well tolerated, though it did not beat placebo on the main outcome; the findings concern postoperative patients in a short trial and not other groups (Beck et al., International Journal of Colorectal Disease, 2014). For the same bowel-surgery trial of 117 enrolled adults, 114 in the safety analysis, FDA records serious adverse events including death; its summary gives no rate or determination of causation, and it refers to the same trial, not a second one (Beck et al., 2014; FDA, bulk drug substances page, 22 April 2026). In a 2008 systematic review of 27 study samples with 303 treated participants, growth hormone in young, fit people was linked to more soft-tissue swelling and fatigue, and that review concerned growth hormone itself and not a secretagogue (Liu et al., Annals of Internal Medicine, 2008).

Which peptide is the safest, according to the studies?

No study ranking peptides by safety was identified in the sources checked for this page on 14 October 2026, so this page names no safest peptide. The most fully documented safety records on this page come from large maker-funded trials of two approved medicines, semaglutide in 1,961 adults and tirzepatide in 2,539 adults, and neither trial compares the two medicines (Wilding et al., NEJM, 2021; Jastreboff et al., NEJM, 2022). A 2026 literature review of growth hormone-axis research compounds sorted them into evidence tiers, from compounds with regulatory-grade randomised trial data to compounds for which that review found no human studies, and it is a review and not a safety trial (Dominikowski et al., Frontiers in Endocrinology, 2026). The safety of a compound and the quality of the material sold are separate questions.

No trial comparing the safety of one growth hormone secretagogue against another was identified in the sources checked for this page. A 2018 review called for long-term safety work, including cancer incidence and mortality, because few long-term, rigorously controlled studies exist (Sigalos and Pastuszak, Sexual Medicine Reviews, 2018).

Dominikowski et al. (2026) is a literature review; the individual study accounts on this page give the populations and limits for the adverse-event results.

Are peptides safer than steroids or testosterone?

No study comparing the safety of peptides with that of steroids or testosterone was identified in the sources checked for this page, so this page cannot say. Each has its own documented effects. In a cross-sectional study of 140 weightlifters, men who had used anabolic-androgenic steroids for two or more years had lower heart pumping function (ejection fraction 52% against 63%) and more coronary plaque than non-users, and the observational design shows an association, not a cause, and involved no peptide (Baggish et al., Circulation, 2017). In the 5,246-man TRAVERSE trial of men with hypogonadism and heart-disease risk, testosterone replacement was noninferior to placebo for major cardiovascular events (7.0% against 7.3%), with more atrial fibrillation, acute kidney injury and pulmonary embolism, and the findings apply to that medicine in that population and involved no peptide (Lincoff et al., NEJM, 2023). A 2018 review described growth hormone-axis peptides as having a thinner trial record (Sigalos and Pastuszak, Sexual Medicine Reviews, 2018).

The two studies differ in design. TRAVERSE was a randomised trial in men with hypogonadism and heart-disease risk, with testosterone studied for a medical reason. The weightlifter study was observational and looked at illicit steroid use. Neither involved a peptide, so neither can stand in for a peptide result.

The word peptide also covers very different molecules, from insulin to growth hormone-releasing compounds. Each has its own recorded adverse events, set out in the table below.

What side effects have been recorded for women?

The records cited on this page do not report side effects by sex for most compounds, so the female-specific data on this page are thin. The 412-adult tesamorelin trial in people with HIV enrolled 86% men, which limits what it shows for women (Falutz et al., NEJM, 2007), and the maker-funded phase 2 retatrutide trial of 338 adults enrolled 48.2% women but reported gastrointestinal events and heart-rate rises that were larger in higher-amount arms for all participants together, a finding limited to that trial population (Jastreboff et al., NEJM, 2023). FDA's July 2026 briefing records one case report of a 40-year-old woman with widespread skin darkening and gum darkening after exposure to a product labelled for research purposes only that contained BPC-157 and TB-500, with the same reactions again on re-exposure two weeks later, and because the product held two peptides FDA could not assign the reaction to either one, and a single case report gives no rate (FDA briefing document, 23–24 July 2026).

The same FDA review lists one report of a 55-year-old woman with nine days of redness and swelling around the site where a compounded BPC-157 product was given, who was also exposed to a compounded thymosin product, and FDA noted that the second product limits what that single report can show (FDA briefing document, 2026).

On pregnancy, FDA found one published study in pregnant rats, which did not suggest malformations, but no study covering a full reproductive cycle and no carcinogenicity studies of BPC-157; an animal result does not establish a human result, the passage this page draws on gives no size for that rat study, and the document reports no human pregnancy data (FDA briefing document, 2026).

Is a synthetic peptide safe, and what can go wrong with the product?

Whether a peptide is made in a laboratory does not decide its safety; identity, purity, sterility and impurities do. In one case, TGA laboratory testing of a single sample of a product sold as retatrutide found undeclared semaglutide and no retatrutide, at approximately eight times the semaglutide level found in semaglutide products on the Australian Register of Therapeutic Goods, after a consumer reported uncontrollable vomiting and a torn oesophagus that needed hospital treatment, and one report does not give a rate (TGA safety advisory, 14 September 2026). In a 2024 test of three semaglutide vials bought online, the authors reported endotoxin in all three and a measured purity of 7.7% to 14.37% against 99% stated on the labels, and the findings describe those three vials only (Ashraf et al., Journal of Medical Internet Research, 2024).

The three vials in that 2024 test also held 28.56% to 38.69% more semaglutide than their labels stated, a finding that describes those vials only (Ashraf et al., 2024). Endotoxin, a toxin from bacterial cell walls, is one thing a lab report can screen for; the endotoxin, sterility and purity page explains the test and how to read a certificate of analysis covers what a report can and cannot show.

In a 2015 laboratory audit of purchased quorum-sensing research peptides requested at a purity of at least 95.0%, 44.0% met that specification, and the finding concerns that sampled peptide class and does not establish the quality of other products (Verbeke et al., Journal of Pharmaceutical Analysis, 2015). In the same audit, the main compound of one sample had a different structure from the one requested and the laboratory's results differed widely from the supplier certificates, again for that peptide class only (Verbeke et al., Journal of Pharmaceutical Analysis, 2015). Those were a different class of peptides from the ones on this page, but the supply chain problem is the same kind.

TGA laboratory testing of five seized Melanotan II nasal spray bottles with the same label found that the estimated amount varied by more than a factor of two between bottles, a finding that describes those five bottles only (TGA, 17 August 2026). The TGA's Chief Medical Adviser said these risks are particularly concerning for products given into the body by needle, because contaminated or non-sterile manufacturing can raise the risk of infection and harmful immune responses including anaphylaxis (TGA, 14 September 2026).

Are peptides in skincare safe?

Published safety records for skincare peptides come from trials of specific products, and no source on this page establishes safety for skincare peptides as a group. A 2026 meta-analysis of 19 randomised trials with 1,341 participants reported minimal adverse events across oral and topical peptide trials, and its findings describe the peptides and products those trials tested (Nukaly et al., Frontiers in Medicine, 2026). In a 12-week split-face trial in 93 women, the tested palmitoyl pentapeptide formulation was reported as well tolerated by the skin, and this does not establish the safety of other peptides or routes (Robinson et al., International Journal of Cosmetic Science, 2005). A 2026 systematic review that kept 64 of 1,247 screened records noted that most randomised human trials of GHK-Cu are small and run on facial creams (Najafi et al., Archives of Internal Medicine Research, 2026).

The TGA notes that a peptide product may be regulated as a cosmetic if its ingredients, claims and presentation are limited to cosmetic purposes, and as a therapeutic good if it is promoted for a therapeutic purpose (TGA, 23 September 2026). The peptides and Australian law page records the legal position by compound.

Is dihexa safe, and what is known about IGF-1 LR3 side effects?

No human safety data for dihexa were identified in the sources checked for this page. FDA states that it has not identified any human exposure data for products containing dihexa acetate by any route, and that it lacks important information on whether the substance would cause harm to humans (FDA, content current as of 22 April 2026). A ClinicalTrials.gov search by compound name on 14 October 2026 returned no dihexa study, and a name search can miss studies filed under other names. The published rat work (McCoy et al., Journal of Pharmacology and Experimental Therapeutics, 2013) carries a 2021 notice of concern recorded on PubMed, so its findings need cautious reading and do not establish an effect in humans. The same ClinicalTrials.gov name search returned no registered human study for IGF-1 LR3, and a 2026 review lists it among unregulated research compounds that act on the growth hormone-IGF-1 axis (Dominikowski et al., Frontiers in Endocrinology, 2026).

The approved IGF-1 medicine mecasermin is a different molecule from IGF-1 LR3. Its registry of 306 children recorded hypoglycaemia in 80 of them (Bang et al., Journal of Clinical Endocrinology and Metabolism, 2023), and that finding does not transfer to IGF-1 LR3. The 2026 review flags biologically plausible but unproven concerns about promoting cell growth for this axis (Dominikowski et al., 2026).

Why do online forum reports not settle whether research peptides are safe?

Forum posts cannot show cause, and regulators say the same about their own reports. FDA's drug adverse event database held three reports for BPC-157 through 4 December 2025, and its foods database held three more that were clouded by multi-ingredient supplements, and FDA states that it cannot draw definitive conclusions about safety from the drug database, because there is no certainty that an event was due to the product and reports often lack detail (FDA briefing document, 23–24 July 2026). A post also cannot show what was in the vial. The TGA calls unmarked vials, code-only labels and missing ingredient information major red flags (TGA, 7 May 2026).

The database has a second limit. Compounding pharmacies generally do not report adverse events to FDA, so unless someone files a report the agency may not know of an event (FDA briefing document, 2026). Few reports therefore does not mean few events.

Personal reports and chart reviews are also weak on measurement. A 2021 chart review of 17 people with knee pain who had received BPC-157 used no validated tool to measure function or stiffness, which limits what it shows (Lee and Padgett, Alternative Therapies in Health and Medicine, 2021). The page on what published studies found explains how trials measure effects.