Background
The skin is the largest organ of the human body with numerous complex functions essential for our survival. Its primary function is to act as a protective barrier against the external environment. It can protect against harmful chemicals, ultraviolet radiation and pathogenic organisms, while at the same time producing vitamin D and regulating body temperature and moisture loss. Loss of the integrity of large portions of the skin as a result of injury, illness or burns can lead to major disability or even death; worldwide, burn injuries affect more than 11 million people annually [
1]. Wound repair is one of the most complex biological processes of human life. Skin grafting is one of the therapeutic strategies for covering the wound and limiting morbidity.
Currently, the use of an autologous, split-thickness skin graft (STSG) is commonplace as a reconstructive technique for the permanent closure of excised burns or following the excision of benign or malignant tumors. An STSG involves excision, using a dermatome, of the epidermis and part of the dermis, but leaves behind sufficient reticular dermis in the wound bed to enable the skin to regenerate itself. The greatest disadvantage of this technique is that it creates a donor site that is painful during healing, as well as often unsightly scars. The aim of donor-site management is thus to maintain an environment that promotes optimal healing and prevents morbidity, which can include pain and ultimately delayed healing (for review, see [
2]).
Theoretically, an appropriate STSG donor-site dressing should reduce patient discomfort, promote rapid healing, decrease pain, prevent hypertrophic scarring and make it possible to “re-crop” from existing donor sites (which can be a high priority in patients with extensive injuries such as thermal injuries). This is the reason for our interest in this clinical application.
It is both most cost-efficient and in the patient’s best interest for one dressing to be applied and remain in situ until healing is achieved. Unfortunately, if an alginate or hydrofiber dressing is left in situ throughout healing, the dressing is likely to dry out and possibly adhere to the wound bed [
3]. Furthermore, two systematic reviews of donor-site dressing found no clear evidence to support the choice of any particular dressing [
4,
5] except that they must be moist; occlusive or semi-occlusive dressings both fall into this category.
Currently, cell-based, engineered skin substitutes are showing promise for the treatment of acute and chronic wounds such as deep and partial burns, ulcers resistant to conventional therapies and surgical wounds [
6‐
10]. These various engineered tissue formats include allogeneic fibroblasts and keratinocytes in a bovine collagen matrix (OrCel®, Ortec International, Atlanta, GA, USA) [
11] or a bilayered living skin equivalent, Apligraf® (Organogenesis, Inc., Canton, MA, USA). Cultured neonatal fibroblasts derived from the human foreskin are combined with bovine type-I collagen to form a neodermis; this is then seeded with cultured neonatal keratinocytes which proliferate and differentiate [
12]. Approved in several countries, Apligraf®, as an allograft, has been used in acute wounds such as surgical excision sites and partial-thickness donor sites [
12]. Today, the major disadvantages of this biological dressing are its short shelf-life, its cost, its temporary nature [
13] and its susceptibility to immune rejection [
14,
15]. Previously, it has not been possible to provide both “seed and soil” in the same therapeutic agent, i.e., a complete epidermal layer has never been obtained (for review, see [
16]).
Cutaneous wound healing in the early gestation fetus is remarkably different from healing in adults. The most striking features of the fetal wound response are the speed and the absence of obvious scarring, an observation that was first reported more than 30 years ago [
17]. Studies in the marsupial embryo,
Monodelphis domestica, have shown that this fetal regeneration is not due to the moist, sterile environment of the uterus [
18]; rather, the nature of this regeneration mechanism could be the result of the difference between the fetal and adult immune responses [
19]. Indeed, in utero, the fetal environment demands that the immune system remains tolerant to maternal alloantigens. After birth, the sudden enormous exposure to environmental antigens, many of them derived from intestinal commensal bacteria, calls for rapid change to adapt distinct immune responses so that they are appropriate for early life. However, a study of adult skin grafted onto a fetus, infused with fetal blood and bathed in amniotic fluid found that the wound healed with scar formation [
20]. The conclusion of this study is that scarless healing properties are intrinsic to fetal skin and are not primarily the result of the fetal environment, i.e
., immune response and/or sterile fluid. In fact, after 24 weeks of gestation, fetal skin repair is histologically indistinguishable from adult skin. This fetal skin-specific phenotype appears to be dictated by quantitative and qualitative alterations in both the inflammatory and regenerative phases compared to normal adult wound healing, due to molecules secreted by fetal skin cells [
21]. Consequently, our hypothesis is that these fetal-skin-cell molecules will have a regenerative effect on adult wound healing.
Fetal wound healing features an absence of acute inflammation in association with a low level of pro-inflammatory chemokines, such as interleukin (IL)-8 [
22], and a high level of anti-inflammatory cytokines, such as IL-10. This is the opposite of the situation in adult wound healing [
23]. In mice lacking IL-10, fetal wounds display substantial inflammatory cell infiltrates and develop scars.
Furthermore, the fetal fibroblasts proliferate at a faster rate [
24], with growth and migration rates decreasing with age [
25].
In addition, fetal fibroblasts differ from adult in collagen synthesis, with a huge ratio of collagen type III to collagen type I: 3:1, in contrast to 1:3 in adults [
26].
In this regard, a novel acellular collagen matrix derived from fetal bovine dermis has recently been designed for dressing partial- and full-thickness wounds (PriMatrix®, from TEI Biosciences Inc., Boston, MA, USA). However, although this scaffold is particularly rich in type-III collagen [
27], it contains no cytokines or growth factor crucial for wound healing.
The novel nature of our regenerative dressing project is based on four properties of fetal skin cells, which provide an attractive alternative to the keratinocytes and fibroblasts from neonatal foreskin commonly used as skin substitutes:
-
Thanks to their low immunogenicity and their immunosuppressive properties they can be used in an allogeneic context without risk of immune rejection
-
As a result of the factors they secrete, fetal cells could improve the quality and speed of healing, something which is a constant problem for surgeons. Interestingly, scarless fetal skin healing appears to be largely dependent on the fetal tissue itself and is not reliant on the specific in utero environment [
20,
28], giving fetal skin cells great intrinsic potential for wound-healing management
-
Pain is decreased by the anti-inflammatory properties of the secreted factors
-
Finally, their high proliferation capacity means that from a single skin sample, we can amplify and bank clinical-grade keratinocytes and fibroblasts that can be available to surgeons in 4 days
Furthermore, working on these fetal cells brought us to the conclusion [
29] that:
-
Keratinocytes and fibroblasts can dramatically inhibit allogeneic peripheral blood mononuclear cell (PBMC) proliferation in a dose-dependent manner, confirming that they induce an immunotolerant state
-
This immunosuppressive activity of fetal keratinocytes and fibroblasts mainly relies on IDO (indoleamine 2,3-dioxygenase) activity. Through its ability to locally decrease tryptophan availability, IDO is recognized to exert an immunosuppressive effect on T cells requiring tryptophan to proliferate
-
Combining fibroblasts and keratinocytes in co-culture experiments strongly enhances the production of wound-healing growth factors and cytokines (GM-CSF, IL-8, IL-1α, VEGF-A)
These interesting conclusions have led us to build a fetal bioconstruct (biological dressing) that mixes fetal fibroblasts and keratinocytes at a ratio of 1:1 on a type-I collagen matrix (European patent WO 2014 090961A1). This is referred to as CICAFAST, like the trial.
In this trial, the aim of the first application in humans is to test the efficacy of the biological dressing versus a conventional treatment by means of a randomized study. The patient will be their own control, thus avoiding the factors that can influence the healing of a graft donor site. The STSG will be made on the inner part of the thigh at two different sites on the same thigh (upper and lower). The treatment for each site will be chosen by randomization to avoid place bias.
We chose Jelonet® as the conventional dressing since both the Ear, Nose and Throat (ENT) Department and the Burns Center, which are the two investigation centers, use it daily in their practice. Jelonet® (a paraffin-gauze dressing) is the reference treatment for management of the graft donor site. It is essential to differentiate between the graft donor site where a Jelonet® dressing is left in place throughout the duration of healing and the recipient site of the graft or of the wound where the ADAPTIC® (primary dressing made of knitted cellulose acetate fabric and impregnated with a petrolatum emulsion) and UrgoTul® (lipidocolloid dressing) dressings are changed every 24 to 48 h.
This clinical trial should enable the development of a new strategy for STSG donor-site wound healing. The pain of STSG healing during the first days is well known and results from the exposure of sensory nerve endings [
30]. Reducing this pain will also reduce the use of analgesic drugs, and sick leave.
Our biological dressing address the essential need for surgeons to “re-crop” from existing donor sites, e.g., for thermal-burn patients. By accelerating healing, improving the appearance of the scar and decreasing pain, we will improve the treatment conditions for skin grafts.
Discussion
The aim of this first human-use trial is to prove both the absence of significant AEs and the efficacy of this new biological dressing.
This trial is the result of multidisciplinary work with the Burns, ENT and Dermatology Departments, as well as with a cell-therapy unit which produced clinical-grade cell banks. A GMP manufacturing process was developed and validated to produce the dressing and conduct regulatory preclinical studies. No sign of toxicity was observed in efficacy/toxicity studies in Wistar rats. In the biodistribution study, clinical monitoring of nude rats showed no signs of acute toxicity of the dressing. The histological conclusion of these studies shows that the use of CICAFAST dressing leads to an improvement in the quality of wound healing compared to the control group treated with the conventional Jelonet® dressing.
Furthermore, a tumorigenicity test was performed by inoculating significant amounts of fetal fibroblasts and keratinocytes in nude mice (10 × 106 cells/animal). Masses were palpable at the administration site (subcutaneous) in all animals following injection. In the light of the regressive character of these masses, two animals were sacrificed on D14 before the masses disappeared completely. At the autopsy, the observations revealed no abnormal organs. Histological analyses concluded that the masses were related to the injection of cells (well-differentiated epithelium) but were not considered as malignant. In addition, it is important to consider that the fibroblasts and keratinocytes (skin cells) were injected subcutaneously. In the absence of a competent local immune system, these cells found an environment conducive to remaining in place for a few weeks before disappearing completely in 24 days. At autopsy of the mice (D84–12 weeks), the observations revealed no abnormal organs or visible tumors. Histological analyses confirmed the absence of tumor. No tumorigenic fetal fibroblasts and keratinocytes were highlighted. No sign of human deoxyribonucleic acid (DNA) was quantified in the organs of the 29 nude mice treated.
As the preclinical studies were favorable, the first clinical trial was written and conducted.
At the time of writing this protocol, in addition to the fact that this is the first administration in humans and that we aimed to check for toxicity, efficacy was also important.
We would have liked a double-blind study in addition to the control and randomization, and that is how we wrote the first version of the protocol. However, the difference in structure, appearance and healing results between the two wounds for the first few patients made it impossible to blind the study either for the patient, the investigator or the expert evaluating the photos of the wound healing. However, to give the protocol maximum scientific level and avoid bias, the following three points have been added:
-
To avoid the problem of variable wound depth in ulcers or burns, we chose the graft donor-site model, where the depth is always the same
-
To avoid co-morbidities that could affect healing, or to avoid known allocation bias due to the non-blinded way for the patient, each patient acted as their own control in our trial. The two dressings – the reference dressing and our new biological dressing – will be applied at the graft donor site, and the healing of one will be evaluated against the other
-
The position of the dressing is randomized so as not to favor one position (and, therefore, one dressing) more than the other
Furthermore, to obtain the best perspective on healing with or without CICAFAST and because it is unethical to perform biopsies on a barely healed wound, in addition to the opinion of the investigator and the patient (P-OSAS), photos will be taken for expert review and the wound will be viewed using reflectance confocal microscopy.
Nevertheless, the lack of blinding is the weak point of our study. CICAFAST is a pilot study, phase 1/2, first-in-human clinical trial. However, if the results of this study are promising, we will then undertake a randomized, double-blind trial where only one dressing per patient would be applied (CICAFAST or Jelonet®) and the difference of the dressing’s release would be evaluated.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.