GHK-Cu vs BPC-157 vs TB-500: What's the Difference?

GHK-Cu vs BPC-157 vs TB-500: what is the difference?

GHK-Cu is a copper-bound tripeptide with human wound-healing trials; BPC-157 is a synthetic 15-amino-acid peptide with cell, rat, and three small uncontrolled human reports; TB-500 is the seven-amino-acid Ac-LKKTETQ fragment, not full-length thymosin beta-4. Their evidence tiers are therefore different, and none establishes a proven treatment claim or long-term safety. The molecules should not be ranked by borrowed results.¹⁸¹³

1. Chemical identity: three names, three substances

A comparison begins with the molecule actually tested. GHK-Cu is glycyl-L-histidyl-L-lysine complexed with copper, commonly called a copper tripeptide. Its short sequence and metal complex are relevant: evidence about unbound GHK, copper ions, or an unspecified “copper peptide” should not automatically be treated as evidence about every GHK-Cu formulation. The skin-focused background article What Is GHK-Cu? explains that distinction in more detail.

BPC-157 is a synthetic pentadecapeptide, meaning a chain of 15 amino acids. The tendon literature describes it as a partial sequence associated with body protection compound and gastric juice, but that description does not demonstrate that a manufactured peptide is a naturally circulating human repair factor. Its published history is concentrated in laboratory and animal models, with three later, uncontrolled human reports. The companion explainer What Is BPC-157? covers the identity question separately.⁸⁹

TB-500 is a synthetic Ac-LKKTETQ heptapeptide, corresponding to residues 17–23 of thymosin beta-4 (Tβ4). Full-length Tβ4 is a separate 43-amino-acid protein. The 1991 biochemical paper concerns full-length Tβ4 and actin sequestration; it does not establish that the shorter fragment has the same behavior.¹⁰ This is why full-length Tβ4 trials cannot be relabeled as TB-500 trials. See What Is TB-500? and the related BPC-157 vs TB-500 comparison.

2. GHK-Cu: from matrix signals to human wound trials

Cell and laboratory evidence

GHK-Cu has a coherent cell-biology literature, but its findings remain model-level evidence. Fibroblast experiments reported stimulation of collagen synthesis, while another study examined MMP-2 and TIMP expression in matrix turnover. These results identify pathways worth testing; they do not prove that intact human skin receives the same exposure, builds a correctly organized matrix, or improves in appearance. A review describes broader collagen, elastin, and antioxidant signals, but a pathway marker is not a clinical endpoint.¹²

Animal evidence

In a rat experimental-wound chamber, the copper tripeptide increased measures including collagen, total protein, and glycosaminoglycans. That adds tissue context to the culture findings, but chamber contents are not equivalent to closure, tensile strength, scar quality, or restored function in a person.³ A separate irradiated-rat flap study did not provide a uniformly positive result: its prespecified vascular and ischemic outcomes did not establish improved repair.⁷ GHK-Cu therefore reaches a supportive but mixed animal tier, not a universal repair conclusion.

Human evidence

GHK-Cu reaches the human randomized-trial tier, which is more than a cell or animal signal. A multicenter, randomized, evaluator-blinded, placebo-controlled study tested a specific GHK-Cu gel in diabetic neuropathic plantar ulcers alongside standardized wound care and reported greater percentage area closure than vehicle in that defined setting.⁴ This is the phase 2/3 wound-healing evidence associated with GHK-Cu: the molecule, gel, disease population, care pathway, and endpoint matter. It cannot be generalized to an unspecified serum, intact facial skin, or every wound.

Other participant studies were narrower. A biopsy pilot measured procollagen staining in normal skin, and a small randomized post-laser study found a subjective preference but no significant objective erythema advantage.⁵⁶ For a fuller appraisal of those endpoints, read GHK-Cu Research: Skin, Collagen, and Repair. Human RCT evidence is a meaningful tier, but it is not proof of a general treatment claim.

3. BPC-157: promising models, limited human observation

Cell and laboratory evidence

Chang and colleagues studied rat Achilles tendon explants and cultured rat fibroblasts. The reported observations included tendon outgrowth, cell migration, cell spreading, and survival under oxidative stress, with proposed involvement of focal-adhesion signaling. These are laboratory observations in rat-derived systems. They can generate a hypothesis about tendon biology, but they do not measure a human injury outcome or justify an expected result.⁸

Animal evidence

In a rat Achilles-tendon transection model, BPC-157 was associated with better reported functional index, failure load, and histology than controls. The intact-animal setting is a higher evidence tier than a dish, while remaining limited by species, injury model, exposure, follow-up, and replication. The study does not establish that human tendon injuries heal faster or stronger with BPC-157.⁹

Human evidence

BPC-157 has human reports, but only at the small uncontrolled-observation tier. A retrospective chart review included 17 patients; mixed exposure in the report makes a BPC-only conclusion difficult. A 12-patient interstitial-cystitis pilot reported symptom changes without a control group, and a two-adult report described a brief safety observation. These three papers do not provide a randomized comparison, establish effectiveness, or settle long-term safety.¹⁰¹¹¹² Thus, BPC-157 reaches human observational evidence, not human RCT evidence.

4. TB-500: separate the fragment from full-length Tβ4

Cell and animal evidence

Full-length Tβ4 is biologically studied as an actin-sequestering protein, and reviews describe preclinical wound findings for that full-length molecule.¹³¹⁴ Those studies do not establish equivalent cell or animal evidence for Ac-LKKTETQ. The most accurate fragment-specific tier is therefore an identity and safety-evidence gap, not a positive animal-efficacy tier borrowed from its parent protein.

Human evidence

Full-length Tβ4 has phase 2 wound studies, including a randomized venous-ulcer trial and reports discussing two phase 2 trials.¹⁵¹⁶ Those trials tested full-length Tβ4, not TB-500. The FDA compounding safety entry for TB-500 (thymosin beta-4 fragment LKKTETQ) states that it has not identified human exposure data for products containing the fragment and notes important safety gaps, including concerns about immunogenicity, aggregation, and impurities.¹⁷ TB-500 therefore reaches no identified human-exposure tier. Full-length Tβ4 results cannot fill that gap.

5. Side-by-side comparison

6. How to read the comparison responsibly

“Which is best?” is not answered by counting favorable headlines. GHK-Cu has the strongest human evidence of these three for a specific wound question, including randomized controlled data, but that does not make every GHK-Cu product a proven treatment. BPC-157 has a clearer tendon-focused preclinical story than clinical story: its animal results are hypotheses about translation, while its human literature consists of uncontrolled reports. TB-500 cannot inherit either full-length Tβ4’s cell biology or its human wound trials.

Keep molecule, model, comparator, outcome, and follow-up together. A collagen marker is not wound closure; wound-chamber material is not restored tissue function; a symptom report without a comparator is not an RCT; and a parent protein is not a fragment. “No identified human exposure data” is also not the same as proof of harm—it is a statement about an evidence gap that prevents a confident safety or efficacy conclusion.

None of these compounds has a proven general treatment claim. The FDA compounding notes for both BPC-157 and the TB-500 fragment should be read as safety context, not as a substitute for clinical trials.¹⁸¹⁷ This article is an evidence comparison, not a product recommendation. Readers wanting a broader framework can explore the Repair Protocol book or the Glow Stack book; neither book changes the molecule-specific evidence described above.

Frequently asked questions

Are GHK-Cu, BPC-157, and TB-500 versions of the same peptide?

No. GHK-Cu is a copper-bound tripeptide, BPC-157 is a synthetic 15-amino-acid pentadecapeptide, and TB-500 is the Ac-LKKTETQ seven-amino-acid fragment. Their chemical identities and evidence histories must be kept separate.

Which of the three has the strongest human evidence?

GHK-Cu reaches randomized human wound evidence in a defined diabetic-ulcer trial. BPC-157 has three small uncontrolled human reports, while TB-500 has no identified human exposure data according to the FDA entry for the fragment.

Does full-length thymosin beta-4 research prove that TB-500 works?

No. Full-length Tβ4 is a separate 43-amino-acid protein, whereas TB-500 is a seven-amino-acid fragment. Findings from the parent protein may motivate research, but they do not establish matching fragment activity or safety.

Does BPC-157 have human randomized-trial evidence?

No randomized BPC-157 trial is identified in the cited evidence. The three papers are a retrospective chart review, a 12-patient uncontrolled interstitial-cystitis pilot, and a two-adult brief safety report, so broad efficacy and long-term safety remain unsettled.

Does GHK-Cu human wound evidence prove that a cosmetic product heals skin?

No. The randomized evidence tested a specific GHK-Cu gel in diabetic neuropathic plantar ulcers with standardized wound care. That defined trial cannot establish that every cosmetic formulation, intact skin application, or unrelated wound has the same outcome.

The bottom line

GHK-Cu reaches the furthest human evidence tier, but only for specific tested formulations and wound settings. BPC-157 reaches animal and limited human observational tiers, with no human RCT in the cited literature. TB-500 is a fragment whose human evidence cannot be borrowed from full-length Tβ4; the FDA identifies no human exposure data for the fragment. For more evidence-first reading, visit all articles.