BPC-157 vs GHK-Cu: Two Evidence Gaps, Compared with Lab Results

The most important difference is where each compound's evidence gap sits. BPC-157 has a large body of rodent research and only three tiny, uncontrolled human pilots, so the gap is between animals and people. GHK-Cu has small human studies of skin creams, so its gap is between skin and injection, which is how it is sold as a vial. A DailyMed search on September 30, 2026 found no drug label for BPC-157, and GHK-Cu turned up only as an ingredient in topical skin products. No trial has compared the two.

At a glance

How they work differently

BPC-157 is a designed fragment. The name stands for "body protection compound," and the 15-amino-acid version is a partial sequence of a compound reported to be isolated from human gastric juice (Chang et al., Journal of Applied Physiology, 2011). Reviews describe it as acting through several overlapping pathways in animal and cell work: VEGFR2 and nitric-oxide signaling through Akt-eNOS (which drives new blood vessel growth), and ERK1/2 signaling, with effects on fibroblasts and inflammation (McGuire et al., Current Reviews in Musculoskeletal Medicine, 2025). In one rat-cell study, it sped the outgrowth and migration of tendon cells, likely through the FAK-paxillin pathway (Chang et al., 2011).

GHK-Cu is a small, natural peptide that carries a copper ion. A 2018 review says it is present in human plasma, can be released from tissue after injury, and has been known since its discovery in 1973. The same review reports that, in a public gene-expression database, GHK changed the activity of about 31% of human genes by 50% or more, which its authors treat as a possible mechanism alongside copper delivery (Pickart and Margolina, International Journal of Molecular Sciences, 2018). Worth knowing: one author's 1973 doctoral thesis on the peptide is in that review's reference list, so it is a review by the field's originator, and gene-database results are a hypothesis-generator, not proof of an effect in a person.

Both mechanisms are described mostly from cells and animals, and neither has been shown to explain an effect in people.

What the evidence shows

Head to head. We searched PubMed on September 30, 2026 for papers naming both compounds. The results were narrative reviews, not trials, and we found no study that compared them or tested them together. Numbers from separate studies of different tissues and endpoints cannot be lined up, so we don't.

BPC-157. The bulk of the evidence is in rodents, and a systematic review that searched through June 2024 found 36 eligible studies, of which 35 were preclinical and one was clinical (Vasireddi et al., HSS Journal, 2025). A 2026 review of its drug development says there is no approved formulation and no completed Phase II clinical trial (Mateescu et al., Pharmaceutics, 2026). Human data since then is three small pilots, all with the same first author and all run at clinics:

• A retrospective chart review of knee injections in 17 patients, 16 of whom were reached by phone. Of those 16, 14 reported relief (four had received BPC-157 together with another peptide). There was no control group and no standardized outcome tool (Lee and Padgett, Alternative Therapies in Health and Medicine, 2021).
• An uncontrolled pilot in 12 women with interstitial cystitis who had not responded to the standard drug. All 12 rated the result 80% or 100% on a self-report scale, and no adverse events were reported (Lee et al., 2024).
• A safety pilot of intravenous infusion in two healthy adults, with no measurable changes on heart, liver, kidney, thyroid or glucose markers and no side effects reported (Lee and Burgess, 2025).

A 2026 study also showed BPC-157 relaxing rings of human artery tissue, taken from 12 bypass surgeries, in a lab dish, an early mechanistic finding that its authors say needs in vivo work (Yildirim et al., Journal of Clinical Medicine, 2026). No controlled human trial shows that it heals a tendon, a gut or a joint. See our BPC-157 guide.

GHK-Cu. This is where the topical-versus-injectable gap matters. For skin creams, the 2018 review cites a facial-cream study in 71 women and an eye-cream study in 41 women, each lasting about three months, and a comparison in which collagen production improved in 70% of women using GHK-Cu, versus 50% with a vitamin C cream and 40% with retinoic acid. Look at how the review sources them, though: the first two are cited to 2002 conference proceedings, and the collagen comparison to a 1998 pilot (Pickart and Margolina, 2018). Independent reviews are cooler. A 2025 review found a "surprising absence of clinical studies" of GHK-Cu as an anti-wrinkle ingredient (Mortazavi et al., BioImpacts, 2025). A 2026 systematic review says most randomized trials are small and conducted on facial cosmetic creams (Najafi et al., 2026). A third 2026 review, searching through August 12, 2026, calls the clinical evidence sparse, notes that a 13-participant study after laser treatment was negative on objective endpoints, and that a small 2026 eyebrow study reported positive hair outcomes without defining the GHK-Cu form or local exposure (Mateescu et al., 2026).

For injection, the support we found is animal work summarized in the 2018 review, and no published human trial. FDA's page for injectable GHK-Cu says there are "limited data in humans" to inform safety. See our GHK-Cu guide.

A plain reading. Neither compound has strong human evidence. GHK-Cu has more human data than BPC-157, but it concerns skin products and is mostly small and industry-linked, so it is the wrong evidence for a vial. BPC-157's human data is directly about injection or infusion, but it consists of three uncontrolled pilots. The evidence is not strong for either.

Side effects and risks compared

Neither has an established human safety profile for injection, and the risk information differs mostly in how much is missing.

For BPC-157, reviews describe preclinical safety studies with no adverse effects across several organ systems but no clinical safety data beyond the tiny pilots, and flag risks from unregulated manufacturing and contamination (Vasireddi et al., 2025). A 2026 review lists pathological angiogenesis and carcinogenesis among the potential risks of unapproved peptides such as BPC-157 (Moiz et al., Missouri Medicine, 2026); that is a theoretical concern about a compound that promotes blood vessel growth, not an observed human harm.

For GHK-Cu, the skin-cream record is long: the 2018 review says it has been used in cosmetic products for decades without adverse effects, which is the authors' summary, not a safety trial. That record does not carry over to injection. FDA says compounded injectable GHK-Cu may pose immunogenicity risk because of aggregation and peptide-related impurities, and that human data are limited. For BPC-157, FDA raises the same immunogenicity concern and says it found no or only limited safety information for the proposed routes (FDA, content current as of April 22, 2026). On that page, current as of April 22, 2026, both BPC-157 and injectable GHK-Cu sit in the table of nominations withdrawn by the nominators, which is not an approval or a safety clearance.

What the lab tests show for each

As of September 30, 2026, VendorVerify's database has 41 BPC-157 vials from 14 vendors (31 lab reports, November 2024 to August 2026). Median purity was 99.57%, with 33 of 38 vials at 99% or higher and 2 under 95%. The median measured amount was 17.8% over label, 2 of 34 vials landed within 5% of label, 13 measured more than 20% over and 1 more than 50% over. 3 of the 41 vials failed our definition (empty, under 95% pure, more than 20% under label, or more than 50% over), or 7.3%.

For GHK-Cu, the database has 29 vials from 9 vendors (16 reports, February 2025 to August 2026). Median purity was 99.90%, with 25 of 26 at 99% or higher. The median measured amount was 15.2% over label, 5 of 28 vials landed within 5%, 14 measured more than 20% over and 10 more than 50% over. As of September 30, 2026, ten vials had failed, or 34%, all from overfill rather than purity.

Both show excellent purity and heavy overfill, and GHK-Cu's overfill was more extreme. But 41 and 29 vials from 14 and 9 vendors are small samples, and one vendor with several vials can move them, so these numbers cannot rank the two compounds against each other. Purity and amount also say nothing about sterility, endotoxin, or whether the peptide is the right one at the molecular level.

Here is who has the strongest independent lab record for BPC-157 right now. The table shows verdicts and flags vendors with more than 5% failed vials; ranked isn't automatically good.

And the same for GHK-Cu, where the overfill pattern means more vendors may carry flags:

Buying smarter

Check a vendor's record for the specific compound: the BPC-157 results page and the GHK-Cu results page show how many reports sit behind each score. Treat a vendor-posted COA as a starting point (how to read a peptide COA, fake peptide COAs). With overfill this common for both, a quantity test on your own vial tells you more than a purity retest; how to get a peptide tested covers ordering identity, purity and quantity testing, and endotoxin testing is the add-on for anything injected. If your report comes back with a fail, stop using that batch, tell anyone who bought from the same run, bring the report to the vendor, and submit it to VendorVerify.

Database figures without a date update automatically from VendorVerify's lab results; dated figures are snapshots. This isn't medical advice; both compounds are sold for research use only, and the human evidence described above is limited.