Peptide Side Effects by Class: What Is Documented, What Is Anecdotal

What side effects do peptides have?

Start with legal status, because it decides what is known. Semaglutide, tirzepatide, bremelanotide and elamipretide are FDA-approved drugs. Their labels list side effects with percentages. Retatrutide and MK-677 are still being tested, and their side effects were recorded in registered studies. BPC-157, TB-500, injected GHK-Cu, Semax, Selank, DSIP and Epitalon are not approved for any use in the US. As our FDA status tracker covers, several went through an FDA panel vote in July 2026. But no pharmacy can legally make them today, and none has a safety record. This article sorts each group into two columns: what a study or label counted, and what only shows up on forums.

How to read the table

A counted side effect means a rate measured against a dummy treatment in a proper study, or listed on an FDA label. A user report means a symptom that shows up on forums, in clinic ads or in single-patient write-ups, with no count of how many people tried it. Stories can invent problems that don't exist. They can also miss ones that only show up when 2,000 people are followed for a year. Each row says how strong the evidence is.

GLP-1 drugs: what the big studies counted

The GLP-1 drugs are the only peptide group where you can talk about side effects with real numbers. Stomach problems go up with the dose, bunch up while the dose is being raised and usually fade. That is why the studies call them 'mild to moderate' even when more than 40% of people get them. In TRIUMPH-1, the 4, 9 and 12 mg retatrutide groups had nausea in 28.6%, 38.4% and 42.4% (14.8% on a dummy shot). Vomiting hit 10.6%, 22.8% and 25.3% (4.8% on a dummy shot). And 4.1%, 6.9% and 11.3% quit because of side effects (4.9% on a dummy shot).

Retatrutide also caused a side effect the other two drugs didn't: dysesthesia, a tingling or extra-sensitive feeling in the skin. In TRIUMPH-1 it hit 5.1%, 12.3% and 12.5% by dose, against 0.9% on a dummy shot. It was higher in the smaller study of people with knee arthritis, TRIUMPH-4 (445 people, 68 weeks): 8.8% on 9 mg and 20.9% on 12 mg, against 0.7%. Lilly said it was mostly mild and rarely made people stop. That is the 'about one in five' number that goes around, and it comes from the top dose in one study, not the drug as a whole. Our retatrutide versus tirzepatide comparison covers how much weight people lost in those studies.

Melanocortins: the Vyleesi label

The Vyleesi label also records a short rise in blood pressure, about 6 mmHg on the top number and 3 mmHg on the bottom, peaking 2 to 4 hours after a dose. Dark skin patches showed up in 1%, more often in darker skin. The nausea was bad enough that 8% quit because of it. Melanotan II, the unapproved cousin, works on the same receptor and brings the same nausea. It also has single-patient reports of new and darkening moles. Nobody has ever counted how often that happens, because nobody has run a study.

Growth hormone boosters: the MK-677 blood sugar and swelling data

That is what you would expect from pushing growth hormone and IGF-1 back up to young-adult levels. Growth hormone works against insulin, and holding water is a known effect. The study called the drug 'generally well tolerated'. That phrase gets quoted a lot. The blood sugar line gets quoted less. For the injected boosters (ipamorelin, CJC-1295, sermorelin, the GHRPs), there are short studies of blood levels and hormone response, but nothing that followed healthy people for a year. So the water retention and numb, tingly hands described on forums can't be given a rate. They make sense given how these drugs work, but nobody has counted them.

Mitochondrial peptides: the Forzinity label

Elamipretide is the one mitochondrial peptide with a label, because it was approved in September 2025 for Barth syndrome (see our Forzinity explainer). The prescribing information rests on TAZPOWER, a study of 12 patients who each tried the drug and a dummy shot, so the percentages are rough. Every patient had a reaction at the shot spot on the drug, against 67% on the dummy shot. Redness hit 100% against 25%, and hard lumps and itching 67% against 17%. There is an allergy warning covering rash, facial swelling and cough. A daily shot of a peptide under the skin causes skin reactions in most people. That is the counted reality behind the vaguer 'shot spot issues' stories about MOTS-c, which has no data like this at all.

Healing peptides: BPC-157 and TB-500 have no human side-effect data

The 'no known side effects' claim comes from animals. The animal studies of BPC-157 are many, and they show no sign of poisoning. That is not the same as being safe in people. The HSS Journal review said it plainly: side effects are possible from unregulated making, from contamination and because nobody knows how safe it is in people. The worry that keeps coming up is a theory. A peptide picked because it grows new blood vessels and tissue could do the same for a tumor. There is no evidence in people either way, and Harvard Health flags it as a reason for caution in anyone with cancer or at risk of it. What users report (soreness where they inject, headache, nausea, tiredness) reads like the side effects of injecting anything.

Copper peptides and brain peptides

For GHK-Cu, the counted use is in skin creams. Skin irritation is reported, but with no rate, and there is no data on injecting it. The nausea, low blood pressure and tiredness people describe after injecting it are stories. Semax and Selank have Russian studies in people that say they were well tolerated, without saying how many people had problems. There is no Western side-effect tracking. An irritated nose, headache and irritability are the common forum reports. Treat them as believable, not measured.

Why products bought online add a second layer of risk

That difference is the point of the Harvard Health and American Medical Association pieces on injected peptides published in 2026. The doctors quoted are not saying every peptide is dangerous. They are saying that with unapproved products, nobody has checked the strength, purity, sterility or storage. The real short-term risks are allergic reactions, including severe ones, and contamination. The long-term effects are unknown because the studies were never done. Our guide to reading a lab certificate explains what a certificate can and can't tell you.

The bottom line

A side-effect claim that comes with a study name, how many people were in it and a dummy-treatment rate is counted. One that comes with 'users report' is not. For GLP-1 drugs, bremelanotide, MK-677 and elamipretide, the records exist and are worth reading in full. For BPC-157, TB-500, copper peptides and brain peptides, the true statement in September 2026 is that nobody has counted.