Peptides for Gut Health: What Has Actually Been Tested in Humans?

You searched "best peptide for gut health" or "peptide to heal my gut." The honest question isn't which peptide is best. It's which ones have actually been tested in people, for what, and what happened. So that's what we checked.

The short answer

The two peptides pushed hardest for "gut healing," BPC-157 and KPV, have been studied mostly in rats, mice and cells. We found no full published human trial of either one for any gut condition; for BPC-157 there is only a 2005 conference abstract, which the FDA found inadequate. Larazotide is different: it went through several real human trials in celiac disease. The results were mixed, and its Phase 3 trial was terminated with no results posted. The one peptide here with strong human gut evidence, teduglutide, is a prescription drug for a rare, serious condition. It says nothing about everyday gut health.

First: "gut health" isn't something a trial can measure

A trial has to pick a condition and a thing to measure. "Gut health" is neither. Here's what the research on these peptides actually looked at:

• Celiac disease. An autoimmune reaction triggered by gluten. The standard management is a strict gluten-free diet. Larazotide trials measured symptom scores, a gut permeability test and blood antibody levels in people who already had a confirmed diagnosis.
• Inflammatory bowel disease (IBD). Chronic inflammation of the digestive tract that can cause real tissue damage. In animal studies, "colitis" is usually triggered on purpose with chemicals, which is a model, not the human disease.
• Short bowel syndrome with intestinal failure. So much bowel is missing or not working that a person needs fluids or nutrition through a vein (parenteral support). Trials measure how much of that IV support people can cut.
• "Leaky gut." A popular marketing phrase, not a clinical endpoint. Researchers can measure gut permeability with a sugar test (the lactulose to mannitol ratio). In two larazotide trials, that test didn't show a difference from placebo, and in one it was too variable to assess properly.

So when something promises to "support gut health," the first question is simple: which condition, measured how?

The evidence table

Here's what we found, laid out the way a trial is built. Where the source didn't give a detail, we say so instead of guessing. We left doses out on purpose.

BPC-157: a lot of rats, one human conference abstract

BPC-157 is the internet's favorite gut peptide. The animal work is real: in rats, for example, it sped up healing of surgically created fistulas between the colon and skin (PMID 18818478). A 2021 review from the same University of Zagreb research group describes healing in many rat tissues (PMID 34267654). Several of the BPC-157 papers we checked came from that one group, which is why independent replication matters.

Here's the odd part. Papers from that group describe BPC-157 as having been in clinical trials for inflammatory bowel disease under the code PL 14736, and call it safe in those trials (PMID 18818478). We searched PubMed and ClinicalTrials.gov for those trials and found no full published results and no IBD registry record. The FDA's July 2026 review of BPC-157 found one: a 2005 conference abstract of a placebo-controlled study in 53 people with mild to moderate ulcerative colitis, using a BPC-157 enema for two weeks. The FDA said the abstract left out how the main score was defined, who could join, the statistics and follow-up, so it can't support effectiveness or safety. A short abstract is not the same as trial results you can read.

The rest of the human data isn't about the gut. A 2025 systematic review of BPC-157 in sports medicine found 36 eligible studies: 35 preclinical and 1 clinical, a small retrospective knee pain report. It found no clinical safety data, and notes BPC-157 isn't FDA approved and is banned in professional sports (PMID 40756949). Other human reports are a 12 woman bladder pain pilot with no placebo group (PMID 39325560) and a 2 person IV safety pilot (PMID 40131143). A registered Phase 1 trial of oral tablets in healthy volunteers lists no results (NCT02637284).

KPV: cells and mice only

KPV is a three amino acid fragment of the hormone alpha-MSH. In 2007, researchers showed it calmed inflammation in human gut and immune cell lines, and that adding it to drinking water reduced chemically induced colitis in mice (PMID 18061177). A 2008 study found faster recovery in two mouse colitis models (PMID 18092346). A 2010 review called it a candidate worth developing for bowel inflammation, based on this kind of work (PMID 21222263).

Since then, a lot of KPV gut research has been about delivery technology: KPV packed into nanoparticles inside a gel for mice (PMID 28143741), or into a rectal gel for rats (PMID 34547895). That's a sign researchers think plain KPV has delivery problems. It also means those results belong to those engineered formulations. On ClinicalTrials.gov, a search for KPV returned zero trials. On PubMed, we found no human trial of KPV for any gut condition.

Larazotide: real human trials, mixed results

Larazotide is the one gut peptide in this group that was tested properly in people. It targets tight junctions, the seals between gut lining cells, and it was developed for celiac disease. ClinicalTrials.gov lists seven larazotide trials in celiac disease, from early safety studies to Phase 3.

The results are genuinely mixed. In two gluten challenge trials, larazotide did not change the gut permeability test compared with placebo, which is awkward for the "it fixes leaky gut" story (PMIDs 22825365 and 23163616). It did seem to limit symptoms at some lower doses. In the biggest published trial, 342 adults with ongoing symptoms on a gluten-free diet, only the lowest dose beat placebo on the main symptom score. The higher doses didn't (PMID 25683116). Side effects were reported as similar to placebo, or as generally well tolerated, in these trials.

Then came Phase 3 (NCT03569007). It enrolled 307 people and is listed as terminated, with a completion date of July 2022 and no results posted as of 15 September 2026. The record doesn't give us results, so we can't tell you exactly how it went. What we can say: there's no posted Phase 3 result, and we found no approval. "It was in Phase 3" is not evidence that it works.

Teduglutide: what a proven gut peptide looks like

For contrast, here's a peptide with strong human gut evidence. Teduglutide is a lab-made version of GLP-2, a gut hormone. In a 24 week Phase 3 trial of 86 people with short bowel syndrome who depended on IV nutrition, 63% on teduglutide cut their IV support by more than 20%, vs 30% on placebo (PMID 22982184). An earlier 83 person trial was less tidy: the lower dose worked on the main endpoint, the higher dose didn't reach significance (PMID 21317170). A 2013 review reports it was approved by the FDA and the European Medicines Agency in 2012 (PMID 24191254).

Real drugs come with real safety tracking too. A 2025 analysis found 4,533 adverse event reports for teduglutide in the FDA's reporting database, though that kind of data can't prove the drug caused them (PMID 40219687). The lesson: this is a specialist drug for a rare, serious condition, with a hard endpoint and ongoing monitoring. It is not evidence that any peptide improves gut health in people without that condition.

Evidence gaps and negative results

• BPC-157: no controlled human trial for any gut condition found. The IBD trials mentioned in its own research papers have no results we could locate.
• BPC-157 safety: a 2025 systematic review found no clinical safety data (PMID 40756949).
• KPV: zero human trials found in either database. All gut evidence is cell or rodent work.
• Larazotide: no permeability benefit shown in two trials. Higher doses were no better than placebo in the largest published trial. Phase 3 terminated with no posted results.
• Teduglutide: even here, one trial's higher dose missed significance, and post-marketing adverse event reports are substantial.
• Healthy people: we found no trial of any of these peptides for "gut health" in people without a diagnosed gut condition.

Why rat studies and one-off reports don't answer your question

• Animals aren't patients. A rat with a surgically made fistula, or a mouse given a chemical to inflame its colon, is a model. Plenty of things that work in models don't work in people.
• Form and route matter. KPV results came from drinking water, nanoparticle gels and rectal gels. BPC-157 results came from rat abdominal injections or a bladder injection in patients. A result from one form doesn't carry over to a different product taken a different way.
• No comparison group, no conclusion. Symptoms go up and down on their own, and expectation is powerful. The 12 person BPC-157 bladder pilot had no placebo group, so it can't separate the peptide from everything else.
• Tiny numbers hide side effects. A 2 person safety pilot can't pick up anything but the most common problems.
• Even good human trials can be confusing. In larazotide trials, lower doses sometimes did better than higher ones. That's exactly why bigger confirmatory trials exist, and why a terminated Phase 3 leaves the question open.
• A registry record isn't a result. Registering a trial shows it was planned. It doesn't tell you what happened.

Questions to ask before believing a gut claim

Bring these to a gastroenterologist or your doctor, or just run them in your head when you see a claim online:

1. What exact condition is this supposed to help, and do I actually have that diagnosis?
2. Was it tested in people, or only in cells and animals?
3. Was there a placebo or comparison group?
4. How many people, for how long, and what was measured?
5. Was it the same form and route as the product being promoted?
6. Were results published in a journal, or is there only a registry entry, a press release or a testimonial?
7. What side effects were reported, and was the study big enough to spot uncommon ones?
8. Is there an approved treatment for my condition with stronger evidence?
9. Which of my symptoms mean I should get tested first rather than experiment?

What does have evidence

For diagnosed gut conditions, the path runs through a GI doctor and treatments tested for that condition. For celiac disease, a strict gluten-free diet is the standard management, which is exactly why larazotide was tested as an add-on to it rather than a replacement.

Fiber is worth a mention, with a caveat. A 2019 Lancet series of meta-analyses (185 prospective studies and 58 clinical trials) linked higher fiber intake with 15 to 30% lower rates of death, heart disease, stroke, type 2 diabetes and colorectal cancer, with the biggest risk reductions at 25 to 29 g a day (PMID 30638909). That's about long-term disease risk in the general population. It excluded people with chronic disease, and it doesn't measure gut symptoms. It's good general health evidence, not a gut repair treatment.

The honest verdict

The gut peptide hype is running way ahead of the human evidence. BPC-157 and KPV are interesting lab and animal stories; BPC-157's only human gut report is a 2005 abstract the FDA called inadequate. Larazotide got a serious human test in celiac disease and came back with mixed results and an unfinished Phase 3. Teduglutide shows what real proof looks like, and it's for a rare condition, not for your digestion. If your gut is genuinely off, get it diagnosed. Curious where the law lands? See are peptides legal.

What this does not mean