Compounds with a graded row

Ordered by rung, strongest first. The rung rates the study design; the chip beside it says what the study measured. Both are needed.

1. Collagen Peptides

   human RCT Direct outcome Not approved for this condition

   Who was studied

   adults with stage II or III pressure ulcers, n=120 randomised and 112 completed, oral 10 g/day for 16 weeks

   A randomised double-blind placebo-controlled study measured PUSH score, PSST score and wound area in pressure ulcers over 16 weeks; the high-specific-peptide collagen hydrolysate arm differed from placebo. The population is pressure ulcers in long-term care, not diabetic foot ulcers, surgical wounds or burns, and no trial in those populations is on file for this record.

   Source PubMed 1 primary source read for this row

2. GHK-Cu

   human RCT Direct outcome Not approved for this condition

   Who was studied

   adults with diabetic neuropathic plantar foot ulcers, multicentre, randomised, evaluator-blinded, vehicle-controlled; enrolment number not reported in the abstract

   Graded fresh after the merge of copper-tripeptide-1 and ghk-cu: of the combined citation set, the only source that measured wound closure in humans is Mulder 1994 in Wound Repair and Regeneration, which reported 98.5% median area closure of plantar ulcers under topical GHK-Cu gel against 60.8% for vehicle. The other merged citations are skin-ageing, alopecia and in vitro keratinocyte work and measure nothing about wound closure. A phase 2 split-wound trial of topical GHK-Cu gel in standardised punch-biopsy wounds in healthy adults (NCT07437586, n=60 estimated) is recruiting and has no results on file.

   Source PubMed 2 primary sources read for this row

3. LL-37 (Cathelicidin)

   human RCT Direct outcome Not approved for this condition

   Who was studied

   adults with hard-to-heal venous leg ulcers, n=34, 3-week open-label placebo run-in then a 4-week randomised double-blind phase, topical 0.5, 1.6 or 3.2 mg/mL twice weekly

   Wound area and healing rate constants were measured directly; the two lower doses healed faster than placebo and the highest dose did not. A separate randomised double-blind trial in a distinct population, diabetic foot ulcers with mild infection (Arch Dermatol Res 2023), measured granulation index and reported faster healing but no change in IL-1alpha, TNF-alpha or aerobic bacterial counts; its enrolment number is not stated in the abstract.

   Source PubMed 2 primary sources read for this row

4. Thymosin Beta-4

   human RCT Direct outcome Not approved for this condition

   Who was studied

   adults with a full-thickness pressure ulcer 5-70 cm2 present at least 1 month, n=72, topical, up to 84 days; a parallel trial enrolled n=72 with venous stasis ulcers

   Two completed randomised, double-blind, placebo-controlled phase 2 dose-response trials of topical thymosin beta-4 in chronic ulcers, one in pressure ulcers and one in venous stasis ulcers, each n=72. The registered primary outcome in both was safety and tolerability over 84 days rather than closure, and no results are posted on either registry entry. These are chronic ulcer populations; no trial in acute surgical wounds or burns is on file.

   Source ClinicalTrials.gov 2 primary sources read for this row

5. BPC-157

   animal in vivo Direct outcome Not approved for this condition

   Who was studied

   rats, alkali-burn skin wound model, topical application

   Wound closure, re-epithelialisation, granulation tissue and collagen deposition were measured in rats, not humans; the same paper's proliferation, migration and angiogenesis findings are in vitro. No human study measuring wound closure with BPC-157 is on file: the only registered human study (NCT02637284) is a phase 1 safety and pharmacokinetics study in healthy volunteers with no wound endpoint. Every wound-closure figure for this compound is a rodent figure.

   Source PubMed 2 primary sources read for this row

6. Carnosic Acid

   animal in vivo Direct outcome Not approved for this condition

   Who was studied

   adult mice, cutaneous wound model, including mice lacking TRPA1 in sensory neurons; topical rosemary extract

   Wound healing speed and fibrosis were measured in mice, and the effect was absent in mice lacking sensory-neuron TRPA1. The compound is a rosemary diterpene, not a peptide, despite the record name, and no human wound study is on file.

   Source PubMed 1 primary source read for this row

7. GHK (Glycyl-L-Histidyl-L-Lysine)

   animal in vivo Direct outcome Not approved for this condition

   Who was studied

   streptozotocin-induced diabetic rats; excision wounds treated with biotinylated GHK incorporated into collagen matrices

   Wound contraction rate, granulation-tissue collagen and skin antioxidant status were measured in diabetic rats, not humans. The copper-free GHK record has no human wound trial on file; the human diabetic-ulcer trial in the corpus used the copper complex, which is a different record.

   Source PubMed 1 primary source read for this row

8. TB-500

   animal in vivo Direct outcome Not approved for this condition

   Who was studied

   db/db diabetic mice and 26-month-old aged mice; full-length thymosin beta-4 and a seven-amino-acid actin-binding-domain peptide, the TB-500 sequence, applied to dermal wounds

   Wound contraction, keratinocyte migration and collagen deposition were measured in diabetic and aged mice, and the seven-residue fragment performed comparably to full-length thymosin beta-4. The human ulcer trials on file used full-length thymosin beta-4, not this fragment, and the only registered human TB-500 study (NCT07487363) is a cardiovascular safety and pharmacokinetics study that measures no wound outcome.

   Source PubMed 2 primary sources read for this row

Checked, and nothing recorded

1 compound was checked against wound healing and produced no gradeable row.

• KPV — checked, no row on file

A compound is here because someone looked and the sources did not support a row — not because nobody looked. Under the rubric, uncertainty resolves to no row, and a row resting on a vendor page or a clinic blog is worse than no row at all.