A Prospective, Postmarket, Compassionate Clinical Evaluation of a Novel Acellular Fish-skin Graft Which Contains Omega-3 Fatty Acids for the Closure of Hard-to-heal Lower Extremity Chronic Ulcers

Introduction. A novel piscine acellular fish-skin graft product has 510k clearance on the US market. This product (Omega3, Kerecis, Isafjordur, Iceland) is to be used similarly to extracellular matrices (ECMs) on the market (eg, bovine and porcine) except that it contains fats, including omega-3 polyunsaturated fatty acids that have been associated with anti-inflammatory properties in many studies. While many current ECMs are effective on open wounds, studies have largely excluded application to hard-to-heal ulcers. To test this product in a real-world environment, the authors chose to look specifically at hard-to-heal ulcers based on previously defined wound and patient factors..Methods. The primary objective was to assess the percentage of wound closure area from baseline after 5 weekly fish-skin graft applications in 18 patients with at least 1 “hard-to-heal” criteria. Patients underwent application of the fish skin for 5 sequential weeks, followed by 3 weeks of standard of care. Wound area, skin assessments, and pain were assessed weekly. Results. A 40% decrease in wound surface area (P < 0.05) and a 48% decrease in wound depth was seen with 5 weekly applications of the fish-skin graft and secondary dressing (P < 0.05). Complete closure was seen in 3 of 18 patients by the end of the study phase. Conclusion. This fish-skin product appears to provide promise as an effective wound closing adjunctive ECM. This is true when used in this compassionate setting, where many other products fail. This study lacks a control arm and an aggressive application schedule, but the investigators believe it represents real-world practice.

A cellular fish skin is increasingly being used clinically, as a readily safe and effective alternative tissue source for wound repair in chronic nonhealing ulcers of many etiologies.¹ Current mammalian acellular dermal matrices (ADM) raise concerns of the potential for autoimmune response, risk of prion diseases, and potential cultural or religious issues that may prohibit the use of porcine or bovine products in many countries.¹ The fish-skin material has at least 2 fundamental differences from other biologic materials on the market: 1) no disease transmission risk exists from fish to humans; and 2) the product still contains fats, which are removed during processing of  mammalian products. When grafted, the acellular fish skin provides a natural structure to the wound bed that contains natural skin elements and bioactive lipids, which not only acts as a scaffold for revascularization and repopulation of the patients’ cells, but provides anti-inflammatory and antimicrobial properties as well. An in vitro study comparing fish skin to a human amnion/chorion-derived product shows the fish-skin graft is an ideal platform for a 3-dimensional (3D) ingrowth of cells. The fish skin was able to support 3D ingrowth and proliferation of fibroblasts while no cell ingrowth was seen in the amnion/chorion membrane-derived product.²

In its natural state, fish skin is metabolically active and serves as a protective barrier for piscine species from their harsh aquatic environment. Basic features of fish and human skin are evolutionary conserved. The main differences between human and teleost fish skin are the presence of scales instead of hairs and a lack of keratinized layer in the fish skin.³ All scales are removed from the acellular fish-skin graft, while it still consists of 3 basic layers: epidermis, dermis, and hypodermis. In the natural state, the epidermal layer consists of an outer layer rich in microfilaments and collagen; an intermediate layer, with contents such as unicellular mucous secreting glands; and a third anchoring the basal epithelial cells layer

Following injury, mucous cells from the intermediate epithelial layer margins help close wounds by secreting lysosomes, immunoglobulin, C-reactive peptides, and lymphocytes which are transported to the damaged area.^3-5 The fish dermis is a highly vascularized collagenous matrix composed of fibroblasts, pigment cells, and scales. Its primary function is to strengthen and protect skin against tensile force. The hypodermis separates the inner face of the dermis from the subjacent muscle cells, which is composed of loosely organized collagen, chromatophores, vasculature, and adipose cells.⁵ Neutral and acidic glycoconjugates and antimicrobial peptides (AMP) are secreted and expressed from fish skin; examples of AMPs include hepcidin, defense-like peptides, apolipoproteins, and piscidin. This secretion aids in defense against invading pathogens. Antimicrobial peptides found in fish skin have shown to repair wounds and act effectively against pathogenic bacteria, fungi, viruses, or parasites. ⁴

Acute wounds in healthy individuals heal through a relatively orderly, linear sequence of physiological events that include hemostasis, inflammation, epithelialization, fibroplasia, and maturation.^6,7 In chronic wounds, there is an increase in metalloproteinases (MMP) activity and a decrease of their counteractive MMP inhibitors activity. There is also a decreased proliferation and responsiveness to growth factor from fibroblasts, thus leading to impaired migration of keratinocytes and impaired gap junctions, which ultimately hinder the healing process. Chronic, nonhealing ulcers are more likely to occur in patients with underlying disorders, such as peripheral artery disease, diabetes, and venous insufficiency.^8-10 There is no single primary factor that contributes to impaired wound healing, but it is well known that chronic wounds usually fail to progress through the stages of wound healing and are arrested in the inflammatory stage. An excessive amount of protein mediators, such as proinflammatory cytokines, play a prominent role in the molecular and cellular processes during the inflammatory stage of skin healing and are known to delay wound healing.¹¹ Research has shown the effects of bioactive lipid mediators—omega-3 polyunsaturated fatty acids, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)—reduce inflammatory responses and the transmigration of proinflammatory cytokines across the endothelium.^11-13 Omega-3 polyunsaturated fatty acids, EPA, and DHA are found predominately in fish skin, oils, plasma, and cellular tissues.¹

The fish-skin graft is a decellularized fish skin harvested from codfish in the North Atlantic. It is a skin substitute containing collagen, fibrin, proteoglycans, and glycosaminoglycans, with the potentially added benefits of bioactive lipid mediators. It furnishes a complex scaffold that provides an optimal environment for a favorable host tissue response, a response characterized by restoration of tissue structure and function, while delivering anti-inflammatory EPA and DHA type omega-3 fatty acids. The primary objective of this study is to assess the percentage of wound closure for hard-to-heal ulcers.

All patients signed an institutional review board informed consent consistent with the Helsinki Accord for Human Subjects. In this prospective, nonblinded, noncomparative, postmarket, clinical evaluation of the fish-skin graft, a total of 20 patients were enrolled in the study but only 18 completed the trial. The patients were older than 18 years and recruited from an urban tertiary wound care program. The subjects had ulcers that were of full-thickness and had either > 20 cm² or had been present for at least 52 weeks, both factors defined as markers of hard-to-heal wounds. If patients had ulcers in bilateral extremities, the larger ulcer was undertaken for treatment enrollment. Study participants suffered from either a combination of venous insufficiency with an ankle-brachial index (ABI) of 0.7 to 1.3, diabetes mellitus with an ABI ≥ 1.3 or a toe pressure > 40 mm Hg, or peripheral artery disease with an ABI of 0.5 to 0.7. 

Patients had to be willing to use appropriate offloading and suitable compression therapy. Exclusion criteria consisted of patients undergoing chemotherapy, being treated with immunosuppressive drugs or corticosteroids, or had been previously evaluated for the treatment. Comorbidities included hypertension, tobacco use, diabetes mellitus type 2, and hepatitis C virus; most notably, renal failure was not an exclusion criteria. There was no blinding in this study, as the sample size was small and the product definitively placed. This was an open study with both patients and clinicians aware of the products being used. 

The fish-skin graft is available in sizes 3 cm x 3.5 cm, 3 cm x 7 cm, and 7 cm x 10 cm, and it is indicated for partial-thickness and full-thickness wounds and skin loss injuries as well as superficial and second-degree burns. The fish-skin graft was moistened with normal saline and applied directly to the wound. It was held in place with surgical adhesive and surgical strips. A secondary dressing that delivers ongoing moisture or is moisture retentive was necessary. The fish-skin graft can be reapplied weekly and does not require the removal of the previously applied product since it is gradually resorbed and remodeled in the wound.

Statistics. This was a pilot/feasibility study addressing real-world effectiveness, not an efficacy study. As such it was not powered to do anything other than provide evidence of a 20% or greater wound area reduction over the first 5 weeks of therapy. The SPSS software (SPSS, Inc, Chicago, IL) statistical package was used, and the student t test and Chi-square test were applied.

Data collection. At the initial visit (week 1), investigators verified completion of informed consent status and eligibility checklist. A complete history and physical exam were performed, and concomitant medications were recorded. Having met the inclusion criteria, the study participants were treated with a weekly placement of appropriately sized fish-skin graft, Allevyn (Smith & Nephew, Hull, UK) foam, secondary dressing, and compression for 5 weeks. At each study visit, the wound was photographed and total surface area of the reference ulcers (cm²) and the area of nonviable tissue (cm²) were measured using planimetry. The ulcer and surrounding skin assessments noted were: level of odor (none, slight, moderate, or strong); level of exudate (none, slight, moderate, or heavy); condition of the surrounding skin (healthy, inflamed, macerated, dry and flaky, or other); and any signs of erythema (no reddening, pink, or red). Adverse events, changes in medication, and assessment of visual pain scale of 1-10 were also documented. For weeks 6-8, dressings were removed and secondary dressings with the foam were placed. The ulcer and surrounding skin assessments were performed and photographed, pain assessment and any adverse events or concomitant medications were recorded. Patients interrupting treatment for more than 7 consecutive days were withdrawn from the study. All adverse events reported by the patient, or in response to questioning or observation by the investigator, were recorded and addressed properly.

The trial consisted of 20 participants representing 20 wounds. Two patients were withdrawn from the study due to loss of follow-up. Therefore, 18 participants representing 18 ulcers successfully completed the study. The study population of recruited individuals (n = 18) comprised 14 (78%) males and 4 (22%) females. The mean age was 55 years (range 31-84 years). The mean initial wound size was 8.2 cm² (range 1.5-25.5 cm²), and the mean ulcer age was 35 months (range 3-126 months) (Table 1). 

Over a 5-week period, 5 weekly applications of the extracellular matrix (ECM) and secondary dressing (P < 0.05) resulted in a 40% decrease in wound surface area (Figures 1, 2, and 3). A 48% decrease in wound depth was seen with 5 weekly applications of the ECM and secondary dressing (P < 0.05) (Figure 2). No significant change was seen in surface area or depth from weeks 5  to 8 with secondary dressing alone. A nonstatistically significant reduction in reported pain and drainage was seen. Three out of the 18 patients received complete wound closure by the end of the study phase.

It is widely accepted that healing chronic wounds requires moist environments, an effective antimicrobial barrier, and protection against MMP proliferation.^8,14 An ideal dressing should address all the above issues at a significantly reduced cost of care. Biomedical tissue engineering is a promising approach that has received great attention to restore wound-bed structure in chronic nonhealing wounds. In Armstrong et al,¹⁵ punch biopsy wounds in 20 subjects were treated with a collagen matrix or Monsel’s solution. The collagen matrix produced less inflammation, had a lower incidence of wound infection, was associated with a faster re-epithelialization rate, and healed with a better appearance at 4 weeks than did Monsel’s solution.¹⁵ Many mammalian and porcine collagen tissues have been widely used as ECM structural protein applications for skin substitutes. Important components of these ECM include glycosaminoglycans, proteoglycans, fibronectin, and growth factors, which promote granulation and epithelialization of dermal wounds.^15-19 The fish-skin graft is composed of previously mentioned components plus the added benefits of omega-3 lipids. Using a natural, piscine 3D scaffold product has early data supporting efficacy compared to porcine ECMs and represents a novel approach to treat these hard-to-heal lower extremity wounds.^16,20 Baldursson et al¹⁶ conducted a noninferiority study that compared the effect of fish skin ADM against porcine small intestine submucosa ECM in the healing process of 162 full-thickness, 4-mm wounds on the forearm of 81 volunteers. The fish-skin product was noninferior at the primary end point, healing at 28 days compared to porcine ADM. Furthermore, the wounds treated with fish-skin grafts healed significantly faster and exhibited no autoimmune reactions.¹⁶

Studies on the impact of omega-3 fatty acids on healing and inflammation have found them to be mostly beneficial.^11-13,21 The most interesting quality is the anti-inflammatory effect of omega-3, EPA, and DHA. McDaniel et al,¹¹ gathered 18 individuals randomized to 28 days of either EPA + DHA supplementation (active) or placebo. After 28 days, the active group had significantly higher plasma levels of EPA + DHA and lower 15-lipoxygenase. On day 28, eight 8-mm blisters were created on the forearms of the patients to initiate inflammation and produce wound fluid for lipid mediator and proinflammatory cytokines quantifications. The active group had lower mean levels of myeloperoxidase and more re-epithelialization on day 5 postwounding.¹² While this study did not assess these properties as they may apply to topically applied omega-3 fatty acids, future in vivo studies now underway will assess such inflammatory responses.

The purpose of this study was to evaluate the safety and effectiveness of the fish-skin graft in 20 eligible patients with at least 1 ulcer meeting the “hard-to-heal” criteria. The etiologies of these wounds were either/or a combination of diabetic, venous, or arterial disease. The primary objective of this study is to assess the percentage of wound closure area from baseline after 5 applications of fish skin. The authors chose to apply the fish-skin graft for 5 weeks at weekly intervals, which in many similar studies represents a near ideal application cycle.^17,22 They also chose, as a primary endpoint, a well-recognized surrogate marker of eventual wound closure.²³ The uses of a surrogate endpoint other than complete closure has been well established; a surrogate endpoint usually occurs early in the course of therapy, this time frame is intended to assess real-world outcomes of a patient’s response to treatment. Clinicians can use valid surrogate markers for rapid screening of potential therapies, therefore aiding in the discovery of novel treatments that correlate with true clinical outcomes.^23-25

Gelfand et al²³ assessed the surrogate endpoints for the treatment of venous leg ulcers in which 56,488 wounds from 29,189 patients were analyzed. The study demonstrated in a large, diverse patient population that the percent change in area, log healing rate, and log area ratio at the fourth week of care can serve as important surrogate markers of complete wound healing at 12 or 24 weeks of care. The authors recommend using the time frame of approximately 4 weeks of treatment for studies being conducted to screen initial wound healing agents in diseases in which the true outcome is delayed.²³ Sheehan et al²⁶ showed that in diabetic foot ulcers there was an absolute wound area reduction at 4 weeks (82% reduction vs. 25% reduction; P < 0.001), which correlated with the absolute closure rate at 12 weeks (58% reduction vs. 9% reduction; P < 0.01).²⁶

The authors’ data indicates that the fish-skin graft exceeded their expectations, with > 20% surface area of the patients’ wounds closing between weeks 0 to 5 with 5 weekly fish-skin graft applications and secondary dressings. Given the increased rate of closure, no threat of known disease transmission or immune reactivity, benefits of bioactive lipid mediators, and ease of use, the fish-skin graft appears to be at least noninferior to ADMs derived from mammalian tissue. This product can be used as a suitable, safe, alternative for skin substitute. Possibly, the most fundamental difference between fish skin and porcine or mammalian ECM are the beneficial effects of bioactive lipid mediators.

The limitations of this pilot study include a small sample size, lack of a control arm, and an aggressive application schedule. Investigational studies are needed with larger sample sizes to assess further efficacy of the acellular fish-skin graft and to compare its effectiveness to other ECM-derived scaffolds. Studies designed in a similar fashion to other studies using mammalian ECM would be interesting.^17,22

This product appears to provide promise as an effective wound closing adjunctive ECM. This is especially true when used in this compassionate setting, where many other products fail. However, clearly a better body of evidence is necessary in regards to its best application.

From the Mount Sinai St. Luke’s Hospital and Mount Sinai Roosevelt, New York, NY 

Address correspondence to:
John C. Lantis II, MD, FACS
Mount Sinai St. Luke’s Hospital and Mount Sinai Roosevelt
New York, NY
JLantis@chpnet.org

Disclosure: The research product was supplied free of charge by Kerecis (Isafjordur, Iceland).