EVIDENCE MATRIX / SIDE BY SIDE
Four Compounds, Unequal Evidence
A model-first comparison of what each compound is, what researchers measured, and where inference stops.
In plain English
These four entries should not be ranked on one “best peptide” scale. They are different kinds of molecules, studied in different systems for different outcomes. KPV has anti-inflammatory findings in cells and animal models but no published human clinical trial. NAD+ is a cellular coenzyme, and the clearest human trials test precursors that raise a blood biomarker. BPC-157 has extensive animal repair research but only tiny human pilot evidence. GHK-Cu has small topical human studies, plus broader cell and gene-expression work.
The useful comparison is therefore methodological. Was the experiment done in cells, animals, ex vivo human tissue, or people? Was the endpoint a pathway, a biomarker, a visible tissue change, or a clinical outcome? Was the material the exact compound, a protected formulation, a precursor, or a combination product? Each answer changes how far a finding can travel. The matrix below keeps measured benefit, adverse or cautionary signals, and unresolved questions in separate columns.
The evidence matrix
| Dimension | KPV | NAD+ | BPC-157 | GHK-Cu |
|---|---|---|---|---|
| Molecule | Three-amino-acid alpha-MSH fragment | Cellular dinucleotide coenzyme; not a peptide | Synthetic fifteen-amino-acid peptide | Three-amino-acid copper complex |
| Main research area | Inflammation and tissue repair | Redox metabolism and NAD+-consuming enzymes | Angiogenesis, gastric protection, repair | Skin matrix, copper delivery, hair and wound biology |
| Strongest evidence layer | Cells and animal models [1][3][6] | Short human precursor trials and reviews [8][9][10][12] | Animal studies; tiny human safety pilot [13][14][16][17] | Small topical or combination human studies, reviews, ex vivo skin [18][20][22] |
| Measured signal | Reduced inflammatory activity; model-specific healing | Increased blood NAD+; selected metabolic endpoints | VEGFR2-linked angiogenesis and animal repair | Collagen-related signals, hair-count result in a combination study, skin penetration |
| Key caution | No published human trial | Biomarker gains do not prove longevity or prevention | Human efficacy and long-term safety unknown | Topical evidence does not support systemic claims |
| Regulatory frame | Unapproved research compound | Form-dependent supplement and compounded-product context | Unapproved investigational compound | Cosmetic topical ingredient context; systemic use unapproved |
Mechanism is a starting line
KPV’s mechanism links PepT1 uptake to reduced NF-kB and MAP-kinase inflammatory signaling [3]. BPC-157’s best-defined pathway links VEGFR2 internalization to Akt and endothelial nitric-oxide signaling [16]. GHK-Cu connects copper handling with fibroblast matrix production and broad gene-expression patterns [19][21]. NAD+ sits beneath many cellular processes as both a redox carrier and a substrate for enzymes such as sirtuins, PARPs, and CD38 [11].
All four mechanisms are biologically meaningful. None, by itself, gives a clinical effect size. A pathway can explain why a result occurred in a model and guide the next experiment. It cannot replace randomized outcomes, pharmacokinetics, adverse-event tracking, or replication. This is why the desk records mechanism in one row and evidence maturity in another.
What counts as human evidence
NAD+ precursor studies provide the most conventional human intervention data in this group: randomized trials measure blood NAD+ and selected functional or metabolic outcomes [9][10][12]. Even there, the 2025 review finds limited clinical efficacy and sparse tissue-level understanding [8].
GHK-Cu has human-facing evidence, but attribution is complicated. A hair trial used a combination product containing GHK [20], while penetration work used excised human skin [22]. BPC-157’s two-person study was a safety pilot without an efficacy endpoint [13]. KPV has no published human clinical trial. “Human evidence” is therefore not one category: randomized intervention, combination trial, ex vivo tissue, and tiny uncontrolled exposure answer different questions.
Measured benefit, adverse signal, open question
For KPV, measured benefits are reduced inflammatory endpoints and faster repair in models; the open question is essentially the full human translation. For NAD+, measured benefits include precursor-driven blood NAD+ increases and a muscle-insulin-sensitivity result, while unchanged body composition and HbA1c in that study limit a broader claim [10]. For BPC-157, animal repair endpoints coexist with theoretical cautions tied to angiogenesis and a clinical sample too small to characterize safety [13][16]. For GHK-Cu, topical and combination-study signals coexist with skin-delivery limits and no basis for systemic anti-aging claims [18][20][22].
No single score captures those tradeoffs honestly. The calm reading is a set of bounded statements: what changed, what did not, how the study was designed, and what still requires a better experiment.
How to use the comparison
The matrix works as a claim-checking tool. A statement about KPV should trace back to a preclinical inflammation or repair model. A claim about NAD+ should distinguish the intact molecule from NMN or NR and identify whether the endpoint is blood concentration or health outcome. A BPC-157 claim should disclose the animal-heavy evidence base. A GHK-Cu claim should name the topical, ex vivo, or combination context.
From there, the individual files add study detail: KPV, NAD+, BPC-157, and GHK-Cu. The reference ledger provides the full source trail.