Injury recovery.
The regenerative peptide class is the most-discussed off-label category and the one with the largest evidence gap. Mechanism is suggestive; human RCT data is sparse.
How this class moves the outcome.
Angiogenesis & granulation
Preclinical models show increased capillary formation and granulation tissue at injury sites — the leading hypothesis for accelerated soft-tissue repair.
Actin / cytoskeletal remodeling
Thymosin β-4 sequesters G-actin and promotes cell migration in wound-healing assays.
Extracellular matrix signalling
GHK-Cu modulates collagen and glycosaminoglycan expression in fibroblast culture — relevant to skin and connective-tissue repair.
Ranked by human-evidence strength.
- 01
BPC-157
Body Protection Compound 157Synthetic 15-aa sequence derived from a gastric juice protein. Reported anecdotally for soft-tissue and tendon recovery.
Gastric pentadecapeptidet½ ~30 min (subcutaneous, rodent)Preclinical only - 02
TB-500
Thymosin Beta-4 fragmentSynthetic fragment of Thymosin β4. Reported for recovery; small clinical trials in cardiac and ophthalmic contexts.
Actin-sequestering peptidet½ ~2–3 hPreclinical only - 03
Thymosin β4 (full)
TB4Full-length 43-aa Tβ4. Distinct from the TB-500 fragment; clinical trials in dry-eye and cardiac repair.
Actin-sequestering peptidet½ ~2 hHuman observational - 04
GHK-Cu
Copper Tripeptide-1Naturally occurring copper-binding tripeptide. Topical use studied for skin remodelling; injectable use is anecdotal.
Copper-binding tripeptidet½ Short (minutes, plasma)Human observational
What the human data actually shows.
Each bar is the proportion of candidates on this brief at that evidence tier. RCT first, observational second, preclinical and anecdotal labeled and never inflated.
See full methodology →Indexed human trials and reviews driving the candidate rankings above. Sorted by study type, then recency.
- moderateSikiric P, Seiwerth S, Rucman R, et al.Preclinical2018unreviewed
- moderatePickart L, Vasquez-Soltero JM, Margolina APreclinical2015unreviewed
- moderateChang CH, Tsai WC, Hsu YH, Pang JHPreclinical2014unreviewed
- moderateGoldstein AL, Kleinman HKPreclinical2012unreviewed
- moderateCrockford D, Turjman N, Allan C, Angel JPreclinical2010unreviewed
- moderateKrivic A, Anic T, Seiwerth S, Huljev D, Sikiric PPreclinical2008unreviewed
Where this class can hurt you.
- Most products are research-use-only — quality, sterility, and identity are not regulator-verified.
- Route of administration (oral vs injected) materially changes exposure and is often miscommunicated.
- Long-term systemic exposure data in humans does not exist for most of this class.
- Human RCT data is sparse — most evidence is animal model or self-report.
- Route of administration affects local vs systemic exposure; oral vs injected is not equivalent.
- Regulatory status varies — many of these are research-use-only in most jurisdictions.
Decide between candidates.
Put the top 4 candidates side by side.
Mechanism · half-life · routes · evidence tier · safety signals. A decision matrix, not a leaderboard.
Current scientific consensus.
- Preclinical wound-healing models
- Topical skin-repair signal (GHK-Cu)
- Reliable human tendon-repair benefit
- Joint-pain resolution
- Effective human dose
- Long-term systemic safety
- Whether oral routes deliver active compound
"Mechanistically plausible, clinically unproven. Treat human use as early-stage experimentation, not therapy."
Where this is heading.
Mechanistically plausible, clinically unproven. Treat human use as early-stage experimentation, not therapy.
We re-grade this brief quarterly. Confidence and outcome grades shift as new RCTs publish, safety data accumulates, and replication efforts read out. Changes are logged publicly on the evidence changelog.
Matched protocol templates.
Matching templates…