PMID 17147644 reports a randomized, evaluator-blinded study of a named copper-peptide gel added to standardized care for diabetic neuropathic ulcers. The immediate-treatment plantar-ulcer comparison favored Lamin Gel on percentage area closure, but not every endpoint or subgroup was convincingly positive. Variable follow-up, carried-forward observations, small groups and inconsistent reporting constrain the result. The study does not validate generic GHK-Cu, a cosmetic serum, an injectable preparation or a catalog lot.[1–4]
Study at a glance
| Field | Study record |
|---|---|
| Question | Did the historical Lamin 2% Gel formulation improve plantar-ulcer closure compared with its vehicle within standardized wound care? |
| Design | Multicenter, randomized, evaluator-blinded, vehicle-controlled; immediate and delayed treatment groups |
| Overall enrollment | 181 participants reported in the paper; the immediate plantar comparison contained 28 active and 32 vehicle participants |
| Material | PC1020, called prezatide copper acetate; the paper describes a 2:1 peptide-to-copper molar complex in a hydroxypropylmethylcellulose gel |
| Comparator | The corresponding hydroxypropylmethylcellulose vehicle with the same care program |
| Setting | Adults with diabetes and full-thickness lower-extremity ulcers meeting specified perfusion and other eligibility criteria |
| Timing | Immediate treatment for up to eight weeks followed by six weeks of observation; delayed groups had a different treatment schedule |
| Outcomes | Area closure, an approximately 98%-closure category, linear closure rate, ulcer-size analyses and adverse events including infection |
| Publication | Mulder et al., Wound Repair and Regeneration, October 1994; 2(4):259–269; DOI 10.1046/j.1524-475X.1994.20406.x |
| Registration and plans | No trial-registration identifier, separate original protocol or statistical analysis plan was obtained for this review |
| Commercial involvement | Three ProCyte-affiliated authors; the acknowledgments also identify involvement of its wound-healing development team |
The formulation and timing identify the historical experiment. They are not preparation, application or wound-care instructions.[1–3]
What material and care package were actually compared?
The paper identifies its copper-peptide material as PC1020, or prezatide copper acetate, and describes two peptide molecules per copper in the molar complex. Its immediate active formulation was named Lamin 2% Gel in a 3% hydroxypropylmethylcellulose vehicle. The study materials were supplied in coded sterile vials.[3]
This is more specific than the shorthand “GHK-Cu.” Copper coordination, stoichiometry, counterions, excipients and the finished formulation matter. The salt-form and counterion guide explains why a shared peptide name is not sufficient evidence of material equivalence.
All participants also received a standardized clinical care program, including debridement, dressing care, pressure-relieving footwear, diabetes and activity education, and management of infection when present. Thus, the randomized contrast concerns active gel versus vehicle inside that program. It does not isolate an effect without wound care or support extrapolation to intact skin, hair, scars or other wound settings.[1–3]
Enrollment, analysis groups and an unresolved discrepancy
The paper reports 181 participants overall and says all enrolled subjects were included in its safety and outcome analyses. Its plantar-ulcer Table 3 lists immediate vehicle n=32, immediate Lamin 2% n=28, delayed Lamin 2% n=39 and delayed Lamin 4% n=42: 141 in total. The headline immediate comparison therefore concerns 60 participants, not all 181.[3]
The February 2011 Australian evidence report instead describes 145 plantar and 36 other lower-extremity ulcers, while reproducing the same four plantar groups totaling 141. That difference is unresolved here. The accessible primary-table text has imperfect column ordering, and this review did not obtain a complete, visually verified primary participant-disposition table.[3,4]
The paper separately mentions four participants who left with baseline observations only and states they were included in the analysis. It does not establish that these four explain the technical report's 145-versus-141 discrepancy. Treating them as an invented exclusion group would contradict the stated inclusion rule.[3]
Eligibility required a full-thickness ulcer below the knee, adequately controlled diabetes and specified perfusion criteria. Conditions including osteomyelitis, gangrene and certain non-neuropathic ulcer causes were excluded. The findings cannot simply be transferred to all people with diabetes or all diabetic foot wounds.[3]
“End of study” was not a common week-eight measurement
Immediate treatment could last up to eight weeks, followed by six weeks of observation. For the end-of-study percentage-closure analysis, the paper used the wound measurement when a participant left the study, whether because of healing, an adverse event or another reason. Participants therefore did not necessarily contribute measurements at the same elapsed time.[3]
The weekly percentage-closure plot explicitly carried the last available observation forward after departure. Keeping a value in a graph is not evidence that the wound was repeatedly observed or that its condition remained unchanged. Moreover, infection that failed to respond after a week prompted discontinuation under the protocol, creating a potential connection between outcome, treatment changes and missing follow-up.[3]
These methods are described in the paper; they were not independently checked against a dated original analysis plan or participant-level data. The study-reading guide explains why the outcome, observation time and analyzed population must be considered together.
Percentage area closure favored immediate treatment
In the immediate plantar comparison, median percentage area closure was 98.5% with Lamin Gel and 60.8% with vehicle, reported p<0.05. The paper describes rank-based testing for percentage-closure comparisons. Those p values do not, by themselves, establish a pure difference in population medians or provide a confidence interval for that difference.[1–3]
The abstract also reports a roughly threefold faster closure rate. In the methods, this is a linear wound-edge advance measure derived from area and perimeter over observed follow-up, not percentage closure per day. This review does not convert the percentage-closure values into a linear rate.[1,3]
A later technical-report table labels numerical values resembling the primary paper's percentage-closure means as a rate in mm/day. Because those units and the primary table do not agree, the disputed numerical rate values are not reproduced here.[3,4]
Near-complete closure was less certain
The later Australian evidence table defines its categorical outcome as at least 98% wound closure and reports the following risk ratios. These are the technical report's estimates, not confidence intervals presented in the primary article.[4]
| Reported analysis group | Immediate Lamin Gel | Immediate vehicle | Risk ratio, 95% confidence interval |
|---|---|---|---|
| All plantar ulcers | 15/28 | 10/32 | 1.71, 0.94–3.14 |
| Smaller-ulcer subgroup | 9/14 | 9/16 | 1.14, 0.64–1.93 |
| Larger-ulcer subgroup | 6/14 | 1/16 | 6.86, 1.31–42.26 |
The overall interval includes 1.0, so the favorable point estimate does not establish a conventionally significant overall categorical-closure difference. The larger-ulcer interval is extremely wide and derives from only 30 participants.[4]
There is also a threshold-reporting limitation: the primary methods use greater-than-98% wording with rounding, while other primary prose and the technical report use at-least-98% wording. These descriptions should not silently be treated as an identical executable rule. The source data needed to reclassify individual wounds were not available.[3,4]
Ulcer-size results do not prove a treatment interaction
The reported smaller-ulcer medians were 100% with active gel and 99.6% with vehicle, p>0.05. For the larger-ulcer group, the abstract reports 89.2% versus −10.3%, p<0.01; a negative closure percentage indicates an increase in measured area. The main article's corresponding vehicle value is −10.4%, another small reporting difference rather than a basis for reconstructing data.[1–3]
The exact boundary is inconsistent: the primary Results section defines smaller ulcers as less than 100 mm² and larger ulcers as at least 100 mm², whereas the abstract describes the larger group as greater than 100 mm². This review retains the source labels without deciding where a wound exactly at 100 mm² belonged.[1,3]
The nonsignificant smaller-ulcer result is not proof of equivalence or no effect. Conversely, a significant result in one subgroup and a nonsignificant result in another does not establish that the treatment effects differ. No formal interaction result or independent dated plan establishing the cutoff and analysis was obtained. See the subgroup and interaction guide.
Delayed treatment and multiplicity
The delayed active groups did not show a statistically significant difference from vehicle on the reported closure comparisons. These are not evidence of equivalence, and they do not prove that a particular treatment timing is optimal.[3]
The article describes several tests for different outcomes and group comparisons, including rank-based tests, a test of mean linear closure rate and categorical comparisons. It does not report an overall familywise multiplicity-control scheme covering all endpoints, timing groups and subgroups. This review therefore treats the reported p values and later confidence intervals as nominal, not as independently verified confirmatory evidence.[3,4]
The confidence-interval and multiplicity guide explains why favorable estimates, precision and control of repeated testing are separate questions.
Infection and safety monitoring
The immediate plantar comparison reported infection in about 7% with Lamin Gel versus 34% with vehicle, p<0.05. The Australian evidence table gives the corresponding counts as 2/28 and 11/32.[1–4]
The full article does describe safety monitoring: weekly questions about adverse events, clinical examinations and scheduled laboratory testing, including blood counts, liver and kidney measures, glycemic measures and serum copper. It reports no statistically significant difference in overall adverse-event rates between active and vehicle groups. That is not a statement that no adverse events occurred.[3]
However, a comprehensive participant-level event listing, severity and attribution breakdown, complete arm-specific withdrawal accounting and detailed laboratory-result tables were not available in the reviewed material. Infection-related discontinuation and variable observation times further limit a simple safety comparison. A lower observed infection proportion cannot establish broad safety or an antimicrobial treatment claim.[3]
The adverse-event guide distinguishes recorded events, attribution, treatment discontinuation and evidence of tolerability.
Masking, commercial involvement and source access
Randomization, coded preparations and blinded outcome evaluation are reported strengths. They do not establish that every participant or care provider was blinded, or that masking was maintained. A complete allocation-sequence and concealment procedure, masking-success assessment and independent original protocol/SAP were not obtained. The randomization and masking guide explains these separate safeguards.[3]
Wiley identifies Leonard Patt, Marie Hanley and Gordon Duncan as ProCyte-affiliated authors; Patt was the corresponding author. The primary acknowledgments name the company's wound-healing development team and other contributors. A separate, exhaustive account of funding, data ownership, analysis control and publication decisions was not available. These disclosures are relevant without proving that commercial involvement caused a biased result.[2,3]
For this review, the complete publicly displayed author-uploaded article text was read on ResearchGate. The page identifies an upload by Leonard Patt on June 29, 2023; its platform date differs from the 1994 publication date verified with Wiley and PubMed. A full publisher PDF and a visually checked complete primary-table set were not obtained. The Australian evidence-table pages and all five FDA document pages were visually inspected.[1–6]
A fresh official PubMed metadata response contained no listed correction relationships on September 3, 2026. This is a limited database observation, not proof that no notice exists anywhere. No registration identifier was reported in the reviewed sources; absence of an obtained record is not proof that no private protocol existed.[1–3]
The later FDA device clearance is a different question
FDA record K970153 describes a June 5, 1997 substantial-equivalence decision for Lamin-2 Hydrating Gel, a hydrogel dressing. It is a device-clearance record, not drug approval of generic GHK-Cu and not independent validation of the 1994 trial's efficacy conclusions.[5,6]
The submitted device description lists water, glycerin, carbomer, prezatide copper acetate and preservatives. That differs from the paper's sterile hydroxypropylmethylcellulose trial formulation. A related name or shared ingredient does not establish an identical finished product.[3,6]
The clearance letter expressly restricted labeling claims about accelerated wound healing or epithelialization and about treatment or cure of a wound. The permitted dressing context must not be transformed into those stronger claims. This page does not establish current commercial availability or the status of any other product.[6]
The approved-drug versus research-material guide provides the broader identity boundary: a paper, ingredient name and product-specific regulatory record answer different questions.
What the study can and cannot support
The study offers a historically controlled comparison of one topical formulation plus standardized care, with favorable percentage-closure and infection findings in its immediate plantar analysis. It also contains null comparisons, broad uncertainty around categorical subgroup estimates, variable endpoint timing, carried-forward measurements and reporting discrepancies.[1–4]
It does not establish universal wound healing, a durable benefit after complete follow-up, an optimal regimen, effectiveness without clinical care, comprehensive safety, or equivalence of another copper-peptide material. No result authenticates the identity, amount, purity, sterility, stability or performance of a catalog lot.
Frequently asked questions
Was this randomized?
The paper describes a multicenter randomized, evaluator-blinded, vehicle-controlled study. That does not verify allocation concealment or complete masking of everyone involved.[1–3]
How many participants supplied the headline result?
The immediate plantar analysis contained 28 active and 32 vehicle participants. Enrollment was 181 across ulcer sites and treatment schedules; the later report's 145 plantar total does not reconcile with its displayed groups totaling 141.[3,4]
Was the result simply measured at week eight?
No. Immediate treatment could last up to eight weeks with subsequent observation, but the end-of-study analysis used the measurement at departure. The weekly graph carried last observations forward.[3]
Did every wound endpoint favor Lamin Gel conclusively?
No. The overall categorical-closure confidence interval included no difference, smaller-ulcer and delayed-treatment comparisons were nonsignificant, and the larger-ulcer estimate was imprecise. These null results do not establish equivalence.[3,4]
Was safety unmonitored?
No. Weekly adverse-event questioning and scheduled laboratory monitoring were reported. Detailed outcome reporting remains incomplete, so the study should not be described as proving broad safety.[3]
Did FDA approve the trial's healing claims?
No such conclusion follows. The later K970153 device clearance involved a different described hydrogel and explicitly restricted accelerated-healing and wound-treatment/cure labeling claims.[5,6]
Does this validate another GHK-Cu product?
No. The material, finished formulation, study population, clinical-care program and regulatory record are specific. They cannot establish the quality or equivalence of another preparation.
Primary and official sources
- PubMed: PMID 17147644
- Wiley version-of-record abstract and article metadata
- Primary article: publicly displayed author-uploaded full text
- February 2011 Australian diabetes-foot technical report, Table 124
- FDA device-clearance record K970153
- FDA K970153 summary, clearance letter and indications
Review and corrections
Sources checked September 3, 2026 (America/Edmonton and UTC). Serum Research takes editorial responsibility for this source-based summary. No independent clinical, statistical, analytical or regulatory specialist review is claimed. It is not medical advice, a product-quality assessment or a material-equivalence conclusion. See how we review evidence, the study-records library and the research library. Report a source or transcription issue through Contact Us.