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. Magainins

   human RCT Direct outcome Not approved for this condition

   Who was studied

   adults with mildly infected diabetic foot ulcers, n=835 across two double-blind randomised trials (studies 303 and 304), topical cream twice daily versus oral ofloxacin, 14 days

   The molecule tested was pexiganan (MSI-78), a synthetic 22-residue analogue of magainin 2, not a native frog magainin. Clinical improvement was 85-90% and microbiological eradication 42-47%, statistically equivalent to oral ofloxacin; study 303 failed its equivalence test. The FDA declined approval in 1999 on the grounds of no advantage over existing treatment, and a later placebo-controlled phase 3 programme (NCT01590758 / NCT01594762, OneStep-1 and -2) separated pexiganan cream from placebo on neither clinical nor microbiological response. No regulator has approved it.

   Source PubMed 2 primary sources read for this row

2. LL-37 (Cathelicidin)

   human RCT Surrogate marker Not approved for this condition

   Who was studied

   adults with diabetic foot ulcers, n=25 (13 LL-37, 12 placebo), Jakarta; 8 of the 25 ulcers were mildly infected at baseline, the rest uninfected; topical LL-37 cream 0.5 mg/g, 0.025 mL/cm2 twice weekly for 4 weeks

   Synthetic LL-37 was actually administered, which separates it from the rest of the endogenous skin peptides here. The primary endpoint was granulation index, which improved; aerobic bacterial colonisation counts from wound swabs, the infection-related measure, did not fall, and no infection-resolution endpoint such as cure of cellulitis was assessed. The trial was not designed as an infection trial and most enrolled ulcers were uninfected.

   Source PubMed Central 1 primary source read for this row

3. Psoriasin (S100A7)

   human observational Mechanistic only Not approved for this condition

   Who was studied

   healthy human skin, in vivo and explant, challenged with Escherichia coli; endogenous psoriasin was neutralised with a blocking antibody rather than administered; n not reported in the abstract

   Recorded, and not evidence for this condition

   This measured what the body already makes: psoriasin secreted by keratinocytes was identified as the principal E. coli-killing activity of skin, and blocking it with an antibody removed that activity. Psoriasin was never given to anyone, and no study measured whether administering it treats a skin infection. Presence at a site, or loss of it in diseased skin, is not evidence of treatment effect.

   Source PubMed 1 primary source read for this row

4. Dermcidin

   animal in vivo Direct outcome Not approved for this condition

   Who was studied

   mice with 5 mm dorsal burn wounds infected with Acinetobacter baumannii ATCC 19606; DCD-1L applied topically at 64 ug/mL (8x MIC) once daily for 10 days; group sizes and sex not reported in the abstract

   DCD-1L, the proteolytic fragment of dermcidin, was administered rather than merely measured, and bacterial burden was the outcome: reductions of 1.15, 2.80 and 5.32 log10 CFU/mL on days 1, 5 and 10, with faster wound closure. A companion catheter model reduced biofilm formation by 33-67%. All of this is mouse work; no human study administering dermcidin or DCD-1L for a skin infection is on file.

   Source PubMed Central 1 primary source read for this row

5. RNase 7

   in vitro Mechanistic only Not approved for this condition

   Who was studied

   human primary keratinocytes, 3D organotypic skin equivalents and stratum corneum extracts, challenged with Corynebacterium amycolatum and C. xerosis; no human or animal subjects

   Recorded, and not evidence for this condition

   Recombinant RNase 7 inhibited bacterial growth in culture dose-dependently, and neutralising endogenous RNase 7 in stratum corneum extracts allowed bacterial outgrowth. Both are bench measurements of skin-surface biology; no study administered RNase 7 to a person or an animal, and no clinical skin infection outcome such as impetigo, infected ulcer or surgical site infection has been measured.

   Source PubMed Central 1 primary source read for this row

Checked, and nothing recorded

2 compounds were checked against skin infection and produced no gradeable row.

• Defensins — checked, no row on file
• Histatins — 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.