Home / Research / KPV teardown

Paper-trail teardown · FDA review, patents, company records, trial registries and original papers read · Sources checked October 17, 2026

Does KPV calm an inflamed gut? In mice, again and again. In people, nobody has published a test.

KPV has a better animal record than most peptides on X. Labs in Texas, Italy, Germany, England and Atlanta have reported anti-inflammatory or healing effects in animals and cells, and oral KPV reduced colitis in mice in more than one lab.[7][13][14][17][19][26] But in 42 years, nobody has published a trial that gave KPV to people.[1][23][24] The only human tests we found were company-reported studies of a related two-part molecule used as a vaginal gel, none published in a journal, plus two anecdotes in a patent.[9][10][11] FDA reviewed KPV as a 0.1% skin cream, not the gut capsules people are taking.[1] The doses posted on X are about 100 times below the mouse colitis dose scaled to a person.[15][21][22][25]

Why this matters this week

On October 14, 2026 FDA sent an “Interim Access Policy for Compounding Using Certain Peptides” to the White House for review. Its text isn’t public, and the review hadn’t finished when we checked.[4] KPV is one of six peptides FDA’s compounding advisory committee backed in July, by 8 votes to 6 with 1 abstaining, after FDA’s own reviewers recommended against it.[5][1] On X, a post on October 13 announced “FDA advisers backed its 503A listing!”[25] We counted 172 posts about KPV as a peptide from October 10 to 16. They promise a lot: “will fix 90% of your gut issues,” “what it did to my colitis,” “healed 8 of 8 scraped corneas in 60 hours,” and a new obesity angle. People are swapping doses too: “200-500 mcg daily,” “500mcg of each per day.”[25] So we traced KPV from its first paper to FDA’s review.

What KPV is

KPV is three amino acids, lysine, proline and valine. They are the last three of alpha-MSH, a 13-amino-acid hormone your body makes that tans skin and damps inflammation.[1][8] It is small enough to make cheaply, which is part of why it sells. On its own, KPV doesn’t seem to work the way alpha-MSH does. In FDA’s summary of the research, it failed to displace alpha-MSH from its receptors and didn’t trigger their usual signal.[1] A London team found it still worked in mice whose main melanocortin receptor was broken.[17] FDA’s conclusion is that “the molecular targets underlying the pharmacological effects of KPV-related BDSs remain unknown.”[1]

Where it came from: a fever drug in rabbits

KPV started as a fever reducer. In 1984 James Lipton’s lab at UT Southwestern in Dallas gave it to rabbits with fevers, either into the brain (0.5 to 2 mg) or into a vein (2 to 200 mg). It lowered fever, though less potently than the full hormone.[6][8] In 1989 the same lab showed it reduced ear swelling in mice, its first anti-inflammatory paper.[7]

The University of Texas System got a patent on it in 1991.[8] Its anti-inflammatory evidence was a rat paw-swelling test. Six rats got 100 mg/kg of a modified form, with chemical caps on both ends (Ac-KPV-NH2). Six got the same weight of hydrocortisone, a steroid. The two worked about equally well. From that, the patent set human doses by borrowing hydrocortisone’s: about 0.2 to 3 mg/kg a day, and 0.2 to 3.5 mg/kg a day by mouth. It concluded KPV “can be used to reduce inflammation in man, in dosage not markedly different from that of hydrocortisone.”[8] For a 70 kg adult that is 14 to 245 mg a day. No person had taken it. The capped form in the patent is also not the plain KPV sold today, and FDA left capped KPV out of its review.[8][1]

The company that tried people, and stopped

In 1999 a company Lipton ran merged with a drug-delivery firm to form Zengen, in Woodland Hills, California, with Lipton as chief scientific officer.[10] Its lead drug, CZEN-002, was not KPV. It was two KPV-like units joined by a cysteine bridge. Zengen filed to test it in people in October 2002, as a vaginal gel for yeast infections.[10] In May 2004 it announced results: 20 women enrolled and 17 finished. There was no placebo group and no randomization. About 88% tested clear of yeast, with no severe reactions, and the drug wasn’t detectable in blood.[11] Lipton also cited earlier trials in China. Zengen said a larger trial would follow in Europe and Canada.[10][11] We found no results from it, no journal paper on any CZEN-002 trial in people, and a drug database lists the program as discontinued at Phase 2.[12][23][24] In 2003, while the trial was running, Lipton had told a trade paper, “Quite frankly, Zengen is not going to be in a financial position to take that very far.”[10]

Zengen also patented KPV for skin disease in 2005.[9] The patent’s only human evidence is two people. One had psoriasis and rubbed 1 mg of KPV in mineral oil on one forearm and hydrocortisone on the other for 11 days. The patent reports “marked improvement within minutes,” with relief lasting longer on the KPV arm. The other had a rash from a watchband.[9] No controlled study followed. In 2007 the patent was assigned back to Lipton and then to a small company, MSH Pharma. It lapsed in 2017 because nobody paid the maintenance fees.[9]

The gut story: real mouse results

The gut claims rest on solid-looking mouse work. In 2008 a team at Emory showed that KPV is carried into gut lining cells by PepT1, a transporter that normally absorbs small protein fragments from food. In cell tests, tiny amounts of KPV turned down NF-κB, a master switch for inflammation. Added to drinking water, KPV reduced chemically induced colitis in mice.[13] The same year a group in Münster, Germany, found KPV reduced colitis in two mouse models, and every treated mouse with a broken melanocortin receptor survived a colitis challenge.[14] In 2016 the Emory group found that the PepT1 transporter itself promoted colitis-linked tumors in mice, and that KPV in drinking water prevented them, but only in mice that had PepT1.[15] The group later packed KPV into nanoparticles to get more of it to the colon.[16]

This is a real strength. Unlike BPC-157, where about 77% of PubMed papers share one Zagreb co-author, KPV’s anti-inflammatory effect has turned up in independent groups on two continents, and PepT1 gives a plausible way for a swallowed tripeptide to reach gut cells.[7][13][14][17][19] It is still mice. Lab colitis is caused by chemicals or immune transfers, not by the mix of causes behind Crohn’s disease or ulcerative colitis in people.

What we found when we did the math on doses

The mouse colitis studies used 100 micromoles of KPV per liter of drinking water.[15] That’s about 34 mg per liter. Mice drink roughly 6 to 11 mL a day per 30 g of body weight, depending on strain.[21] That works out to about 9 mg/kg a day. FDA’s standard method for converting an animal dose to a human starting dose divides a mouse dose by 12.3.[22] For a 70 kg adult, the mouse colitis dose comes to about 50 mg a day, with a range of 37 to 74 mg depending on how much the mice drank.[15][21][22] The 1991 patent’s oral range is 14 to 245 mg a day.[8] The doses people post on X are 0.2 to 0.5 mg a day, sometimes 1 to 2 mg.[25] That’s 100 to 250 times below the scaled mouse dose.

This is our arithmetic, and it has limits. Body-surface scaling is a convention for picking a safe first dose in a trial, not proof that a dose works. A drug acting inside the gut may depend more on how much reaches the colon than on body weight, and we assumed the mice drank a typical amount. No human study shows what any oral dose of KPV does.[22][1] The point is narrower: the doses on X were never tested, and the mouse results came from far higher ones.

What FDA actually reviewed

KPV reached FDA because Wells Pharmacy Network nominated it in March 2021 for a 0.1% topical cream or gel, for “wound healing and inflammatory conditions” such as psoriasis and eczema.[1][3] Wells withdrew the nomination in April 2026. FDA reviewed KPV anyway, on its own initiative.[3][1] The nomination itself was muddled: its certificate of analysis named one form of KPV in its title and gave the chemical formula of another.[1]

The skin cream has its own problem. In a 2017 lab test, plain KPV applied to donated human skin didn’t get through in any amount the test could detect. It only passed after the skin was punctured with microneedles.[18] FDA cited that study.[1]

Meanwhile FDA found KPV sold online as injections, oral capsules, creams and nasal sprays, and in a “regenerative combo” with BPC-157, TB-500, AOD9604 and follistatin. Claimed uses run from inflammatory bowel disease to mast cell disorders, histamine intolerance, COVID-19 recovery, Lyme disease and mold.[1] Yet FDA-registered outsourcing pharmacies reported making no KPV products at all from January 2017 to June 2025.[1] If a pharmacy listing does come, FDA notes it “may not be limited to a specific use,” so a review of a skin cream could end up covering oral gut capsules.[1]

What’s known in people

Almost nothing. FDA wrote that “No published studies were found in which non-compounded or compounded drug products containing KPV … were used in humans.”[1] Its public list says it found no human exposure data “via any route of administration.”[2] We searched PubMed ourselves and got 140 records on KPV; none is a clinical trial.[23] ClinicalTrials.gov lists no study of KPV, Lys-Pro-Val or CZEN-002.[24] The closest thing to human data is a 1998 test-tube study in which blood drawn from people with HIV made fewer inflammatory signals when KPV was added in the lab.[26]

Checking the claims going around this week

“Healed 8 of 8 scraped corneas in 60 hours.” That’s from a 2006 study in rabbits. Their corneas were scraped and treated with KPV eye drops at 1 to 10 mg/mL, four times a day. At 60 hours, eight of eight treated corneas had healed and none on placebo had.[19] It says nothing about swallowing or injecting KPV.

“KPV for obesity.” The new study behind it, published in 2026 by a team in South Korea, treated mouse fat cells in a dish and gave oral KPV to mice on a fattening diet. The mice gained less weight.[20] No person was involved.

“It fixed my colitis.” One person’s report can’t separate KPV from the other things people take with it, the natural ups and downs of colitis, or placebo. Many of these posts describe KPV taken with BPC-157 or in blends.[25] No trial has tested KPV in colitis in people.[24]

Safety: what isn’t known

FDA found no studies of how KPV is absorbed or cleared in any animal, and no standard toxicity tests: none of a single large dose, repeated dosing, damage to genes, effects on pregnancy or fertility, or cancer.[1] FDA’s adverse-event database had no reports through December 3, 2025.[1] FDA also noted that because plain KPV barely crosses skin, a cream might cause little harm elsewhere in the body; that says nothing about capsules or injections.[1][18] That isn’t reassurance, since almost no one has taken it under medical supervision. FDA also flagged that peptide products can clump into aggregates that may provoke immune reactions, and that there is no information to assess that risk for KPV.[1] KPV has no official quality standard, and there is no FDA-approved product to compare against.[1] Gray-market vials and blends are not checked by anyone.

Bottom line

KPV is one of the better-supported peptides in animals: several independent labs, a plausible route into gut cells, and colitis results in mice from more than one lab.[13][14][15][17] That makes it worth a real trial. It has never had one. Forty-two years after the first rabbit study, the human evidence is an uncontrolled gel study of a different molecule, two anecdotes in a lapsed patent, and a test tube of blood.[11][9][26] FDA reviewed it as a skin cream that plain KPV can’t get through, while people take it by mouth at doses far below what helped mice.[1][18][15][25] If the White House clears FDA’s policy and pharmacies can make KPV, that will settle whether it can be sold. It won’t settle whether it works.

This page analyzes regulatory documents, patents, company records, trial registries and published research. It is not medical advice and does not recommend using, sourcing, or dosing KPV. The PubMed count, dose scaling and X counts are Jabbit’s own; inputs are cited above.