GHK-Cu and the Ten Weeks Nobody Has Actually Tested

GHK-Cu shows evidence of collagen remodelling activity in preclinical models, influencing decorin expression, matrix metalloproteinase balance, and angiogenesis in cell and animal studies. Human evidence is limited to one null RCT of a topical cream that did not test the remodelling window. Regulatory status for injectable forms remains unresolved in most jurisdictions.

A wound closes in about two weeks. What it looks like six months later, a thin pale line or a thick ropey cord, gets decided in the ten weeks after that, while everyone’s stopped changing the dressing and moved on.

That’s the part of wound biology most peptide chatter skips straight past. GHK-Cu gets filed under wound-healing and left there, one compound, one job, case closed. But healing a wound and building a good scar are two different jobs, run by different cells on different timelines. The copper-peptide research base is almost entirely built around the first one. Whether any of it survives the trip into the second is the real question, and almost nobody’s asking it.

What is GHK-Cu and where does it come from?

GHK is three amino acids stitched together: glycine, histidine, lysine. On its own it’s a small peptide with an unusually strong grip on copper. Bind that copper and you get GHK-Cu, the form almost all of the research is actually about.

It was first identified in human blood in 1979, not invented in a lab, found in the body. That’s solid ground. It’s a real molecule the body already makes, not an invented compound with a marketing department standing behind it.

How does GHK-Cu influence collagen remodelling in preclinical models?

In cells grown in a dish and animal wound models, GHK-Cu does a lot at once. It pushes cells to make more decorin, a protein that helps organize collagen fibers into a tight weave instead of a loose tangle. It shifts the balance of enzymes that chew up old collagen to make room for new. It kicks off new blood vessel growth, acts as an antioxidant, and dials down inflammation while limiting scar overgrowth. That’s a consistent list, repeated across a lot of separate lab and animal studies.

It’s also entirely preclinical. Every one of those effects comes from cells in a plate or a wound on a rat, never from a scar on a human arm three months after surgery. Consistent across preclinical systems is a real claim. It is not the same claim as consistent in human tissue, and that distance doesn’t close itself.

Same compound, opposite jobs

Building more collagen and tearing collagen down sound like opposite instructions. GHK-Cu gets credit for doing both.

That’s not automatically a contradiction. A scar isn’t built by laying down collagen once and walking away. It’s built by laying down a rough first draft, then spending weeks knocking out the badly organized fibers and resetting better ones in their place. Think of a stonemason pulling loose bricks before resetting a wall straight. Some enzymes do the knocking-out. Decorin and fresh collagen do the resetting. A compound that nudges both in the right direction, at the right point in that process, isn’t confused. It’s doing remodeling.

The catch is timing. Lab dishes don’t run on a clock the way skin does. They get a fixed dose, not a signal delivered at week five versus week nine. Whether GHK-Cu’s real-tissue timing lines up with the weeks a scar is actually deciding its own shape is a separate question from whether the molecule can nudge those enzymes and decorin at all.

What parts of the GHK-Cu mechanism remain untested?

One more piece gets added to the GHK-Cu story a lot: that it strengthens collagen crosslinking by feeding copper to the enzyme that welds collagen strands into strong fiber. If true, that would matter for scar strength.

Nobody’s shown it. Not in a dish, not in an animal, not anywhere. It’s a reasonable guess. That enzyme does need copper, and GHK-Cu does deliver copper. But reasonable isn’t the same as tested. Until someone runs the actual experiment, this piece of the mechanism is speculation.

One lab, most of the papers

The most-cited summary of what GHK-Cu does was written by the same lab that’s been studying it for decades, a thorough document, but not an outside check on itself. It’s also, like nearly everything else in this literature, written by the lab that’s been running this compound since 1979.

Almost all of what’s known comes from one research group, and the human evidence for injection specifically is sparse compared to what exists for creams. That’s not an accusation. It’s a plain fact about the shape of the evidence: one research program generated most of what’s known, and independent replication hasn’t caught up yet.

The window this would actually need to work in

A wound closes fast. The remodeling phase that decides whether the resulting scar sits flat or thick runs for weeks after that, roughly week three through week twelve. That’s when collagen fibers get reorganized, crosslinked, and pared down by that same knock-out-and-reset process.

That’s the window where GHK-Cu’s mechanism, if it holds up outside a dish, would need to be doing something. Not at the moment of injury. Not during initial closure. During the slow months afterward, when the dressing’s long gone and the tissue is quietly rebuilding its own architecture. Almost none of the existing research is designed around that window.

Does GHK-Cu reach target tissue in sufficient concentrations?

Say the mechanism holds. Say GHK-Cu genuinely nudges collagen turnover the right direction during weeks three through twelve. None of that matters if the compound never reaches the deep layers of skin in usable amounts for that long.

The data that exists points the other way. A 1997 rat study (PMID 9187381) found GHK breaks down fast after injection, which raises an obvious follow-up nobody’s answered: does any dosing schedule hold a working concentration at a wound site for ten straight weeks? The literature doesn’t say.

Inject it and regulators treat it as a compound with unresolved safety questions. Put it in a cream and the same regulators treat it as routine. That split runs on how the peptide gets into the body, not on whether the molecule does anything different once it’s inside.

The one human trial that exists, and what it didn’t test

There’s exactly one published randomized controlled human trial of GHK-Cu. It tested a skincare cream against placebo and came back indistinguishable from placebo on the objective endpoints it measured.

That result gets cited a lot as proof GHK-Cu doesn’t work. It’s weaker evidence than that framing suggests. The trial measured a topical cream applied to skin, not a compound delivered to reach a wound site during the remodeling window. It didn’t test injection. It didn’t measure scar architecture in the weeks that actually build a scar. A null result on a different question isn’t a null result on this one. It just means this one has never been run.

What would actually answer the question

A trial built to answer the real question would need to deliver GHK-Cu to a wound site, hold a working concentration there through the twelve-week remodeling window, and measure scar architecture on biopsy rather than closure speed or a photograph. Nobody’s designed that trial yet, and running it means solving the delivery problem first.

The honest answer isn’t that GHK-Cu works or that it doesn’t. It’s that the lab-dish data and the one null skincare trial were never testing the same thing. What would matter for scars has never been directly tested.