Advances in the management of leg ulcers – the potential role of growth factors

Abstract

Venous ulcers are a major health problem because of the increased costs of the treatment and the refractory nature of the ulcers. The treatment cost is estimated to be around 1 billion dollars per year in the United States (US), and the average cost for one patient over a lifetime exceeds $400 000. There has been an increasing trend in the use of growth factors in their management. Genetic engineering has revolutionised the research of wound healing, as the majority of recombinant growth factors are now available for in vitro and in vivo studies. Online searches of Medline, Pub Medical and Embase were carried out using the terms venous ulcers, leg ulcers, growth factors and growth hormone. The literature regarding the potential role of growth factors in the management of leg ulcers is reviewed. The important clinical studies are critically analysed with a view to appreciate the emerging therapies and the further research possibilities in the management of venous leg ulcers. Clinical results with the use of growth factors in non‐healing wounds are encouraging. However, small sample sizes and inconsistent end points in different clinical studies have been the main hurdle in reaching a definite conclusion. Further research is needed to provide the definite evidence. Future developments may include different delivery methods for the growth factors, use of different combinations of growth factors administered simultaneously or, sequentially, bioengineered skin grafts and chemical induction of angiogenesis with the use of gene transfer techniques.

Introduction

Venous disease is the commonest cause of leg ulceration, which is a major health problem because of the increased costs of the treatment and the refractory nature of the ulcers. The treatment cost is estimated to be around 1 billion dollars per year in the United States (US), and the average cost for one patient over a lifetime exceeds $400 000 (1). About 25% of the adult females and 15% of males in the United Kingdom (UK) are affected by varicose veins (2). Venous ulcers affect about 70 000–90 000 people at a single time in UK (3). The cost factor associated with their management is somewhere around £100–£600 millions per year, with community nursing time as the biggest element (4).

The standard treatment for venous ulceration includes leg elevation, compression therapy and absorbent dressings. Table 1 summarises the treatments currently employed for venous leg ulcers. Growth factors have recently gained importance in the management. Genetic engineering has revolutionised the research of wound healing, as the majority of recombinant growth factors are now available for in vitro and in vivo studies. Online searches of Medline, Pub Medical and Embase were carried out using the terms venous ulcers, leg ulcers, growth factors and growth hormone. The literature regarding the potential role of growth factors in the management of leg ulcers is reviewed. The important clinical studies are critically analysed with a view to appreciate the emerging therapies and the further research possibilities in the management of venous leg ulcers. Table 2 presents the mechanism of action, evidence and clinical trials for different growth factors used in treatment of venous ulcers.

Table 1.

Currently employed treatments for venous ulcers

Treatment of the underlying cause	e.g. diabetes, hypertension and sickle cell disease
Local therapy	Debridement (enzymatic and surgical)
	Antibiotics and antiseptics
	Skin grafts – split skin grafts, mesh grafts, pinch grafts and homologous keratinocytes
	Growth factors
	Pharmacological therapy – stanozolol, oxpentifylline, aspirin and thrombolytics
Compression therapy	High compression elastic bandages
Grade I = 20 mm of Hg	High compression inelastic bandages
Grade II = 21–30 mm of Hg	Multi‐layer compression bandages
Grade III = 31–40 mm of Hg	Short‐stretch bandages
Grade IV = 41–60 mm of Hg	Unna's boot
Surgery	Ulcer debridement, excision or grafting
	Varicose veins surgery, perforator ligation stripping and avulsions
	Venous valve reconstruction/replacement, venous by‐pass surgery

Table 2.

Growth factor therapies for venous ulcers

Growth factor	Source	Mechanism of action	Evidence/clinical trials/case reports
EGF	Platelets	Mitogenic effect on keratinocytes, fibroblasts	Clinical trials had shown improved healing in acute wounds (13). Venous ulcers treated with EGF applied topically shown ↑ healing rates and greater reduction in ulcer size (15)
	Macrophages
	Keratinocytes
FGF	Fibroblasts	Cellular proliferation, migration, angiogenesis and morphogenesis	Experimental studies in rats have shown enhanced healing 19, 20, 21
	Endothelial cells		Clinical trial showed no difference in healing between placebo and treatment arms (22)
KGF	Fibroblasts	Selective action on epithelial cells	Enhance granulation tissue formation. Double blind placebo controlled trial has shown improved healing and reduction in ulcer size in the group receiving 60 µg of repifermin. Higher recurrence rate (25)
PDGF	Platelets	Proliferation of smooth muscle cells and fibroblasts, increased granulation tissue formation	In vivo studies shown encouraging results (28, 29). Case series by Knighton et al. (30)–improved healing in 49 patients with chronic ulcers
	Macrophages		Randomised controlled trial by Robson et al. (31)–enhanced healing in pressure sores. Daily application of r‐PDGF has shown ↑ healing and ulcer size reduction in neuropathic lower extremity ulcers (33)
	Endothelial cells
GM‐CSF	Monocytes	Neutrophils proliferation and enhanced activity	Case series of patients with chronic venous ulcers receiving perilesional injection of rh‐GM‐CSF–80% healing in 4 weeks (40)
	Lymphocytes		Randomised controlled trial by Da Costa et al. (41, 42)–improved healing in treatment arm
	Fibroblasts		Case series with topical application of r‐HGM‐CSF–90% overall healing rate (43)
TGF β	Platelets	Stimulates fibroblasts and endothelial cells	No reported clinical trials
	Macrophages
	Fibroblasts

EGF, epidermal growth factor; FGF, fibroblast growth factor; GM‐CSF, granulocyte macrophage colony‐stimulating factor; KGF, keratinocyte growth factor; PDGF, platelet‐derived growth factor; TGF β, transforming growth factor β.

Growth substances are soluble signalling proteins affecting the process of normal wound healing, and they include growth factors and cytokines (5). Growth factors are chemical signalling peptides that act through specific cell surface receptors (6). They are the potential candidates for therapeutic interventions as they can stimulate the cells involved in wound healing (7), they are deficient or biologically ineffective in chronic non‐healing wounds (8, 9) and their efficacy has been established by in vitro studies.

Epidermal Growth Factor

Epidermal growth factor (EGF), first discovered by Cohen (10), is one of the first growth factors to be studied in wound healing. EGF is a polypeptide composed of 53 amino acids. It has a mitogenic effect on the cultured fibroblasts and keratinocytes (11). Franklin et al. (12) used it in experimental wounds in rabbit ears and found it to promote healing. In vivo results were not encouraging. Brown et al. used EGF in conjunction with silver sulphadiazine, on paired skin graft donor sites in a double blind clinical trial (13). Each patient served his own control. Each patient was treated with 10 µg/ml, the treatment group showed enhanced epidermal regeneration and 75–100% healing was improved by an average of 1·5 days. However, it must be remembered that this acute wound environment is quite different from chronic leg ulcers. Another similar study failed to show any statistically significant difference (14).

Falanga et al. (15) used recombinant human EGF in 35 patients with venous ulcers. This was a double blind randomised study, and the end point was complete ulcer re‐epithelialization at the end of 10 weeks. There were 17 patients in the treatment arm and 18 in the placebo. The former group received 10 µg/ml of h‐EGF applied topically daily. Patients were assessed weekly. They demonstrated a greater reduction in the ulcer size and improved healing rates in the group receiving r‐EGF. The mean percentage reduction in the ulcer size at the end of the study was 48% for the EGF group and 13% for the placebo group. However, the results were not statistically significant (P = 0·32). At the end of 10 weeks, 35% (6/17) in the treatment group and 11% (2/18) in the placebo group showed complete healing (P = 0·1). The sample size was small, and the results may have achieved statistical significance with a larger sample size or if the observation period was longer than 10 weeks. It is also questionable whether enough dose of EGF was used. It is shown that up to 40 µg of h‐EGF per day can be safely used (15).

Growth Hormone

Growth hormone, also known as somatotropin, is a protein hormone of about 190 amino acids, which is synthesized and secreted by cells called somatotrophs in the anterior pituitary. It is a major participant in control of several complex physiologic processes, including growth and metabolism. Growth hormone acts on the liver and other tissues to stimulate the production of insulin‐like growth factor I (IGF‐I), which is responsible for the growth‐promoting effects. A randomised double blind placebo controlled trial by Rasmussen et al. (16) evaluated the effects of topical application of growth hormone in patients with chronic leg ulcers. The study was conducted for 12 weeks. Nineteen patients received 1 IU/cm² of the ulcer area dose of growth hormone 5 days per week and 18 received a placebo. All patients received compression therapy. The healing rates were 16% per week in treatment group and 3% per week in placebo group (P = 0·02). Fifty per cent reduction in the ulcer size was observed in 58% (11/19) and 18% (2/11) of the patients in treatment and placebo group, respectively. This was a small sample size further reduced because eight patients from this study were withdrawn as their ulcers healed before 2 weeks. The results are not superior to the other published series. The majority of patients in the study were elderly with a mean age of 80 years. They used single layer compression, which is different from the four‐layer compression bandage system and can explain the low healing rates. The dressings are to be changed more frequently (5 days per week), and it is a very expensive treatment. A single tube of growth hormone costs around $1962, hence cannot be recommended unless a significant difference in outcomes is proved.

Fibroblast Growth Factor

Fibroblast growth factor (FGF) was described about 40 years ago (17), has angiogenic properties and results in its action through specific receptors resulting in cellular proliferation, migration, morphogenesis and angiogenesis (18). It has nine homologous polypeptides, and the most extensively studied of these is basic FGF (bFGF). FGF has shown enhanced wound healing in animal studies. Experimental wounds were created in rats and then FGF was injected around the incisions and an increased tensile strength of the wounds was observed (19). Similarly other researchers have shown that FGF can reverse the healing defect in infected and diabetic wounds in experimental rats (20, 21).

The effects of FGF have been evaluated in a clinical trial on patients with diabetic ulcers, but the results were not convincing (22). This study involved 17 patients who received 0·25–0·75 µg/cm² of FGF daily for 6 weeks and then twice weekly for 12 weeks. Three of nine patients healed in the treatment group as compared with five out of eight in placebo group. Similarly, the weekly reduction in the ulcer size and the percentage area healed at the end of the study did not differ between the two groups. This was very small sample size to derive any conclusions, and it is possible that the dose used was low or the frequency of administration was insufficient to show any significant response. King et al. (23) found encouraging results with the use of FGF in treating leg ulcers in a similar study.

Keratinocyte Growth Factor

Keratinocyte growth factor (KGF) was discovered in 1989. It belongs to FGF family and is also known as FGF‐10. It has a highly selective action on epithelial cells and also enhances granulation tissue formation by recruiting fibroblasts (24). Recently, a double blind placebo controlled multi‐centre study has been carried out using recombinant KGF‐2 (repifermin) in patients with venous leg ulcers (25). This trial included 90 patients. Two different doses (20 µg and 60 µg) of topical application of repifermin were used in the treatment arm. They were applied twice weekly for 12 weeks. All the standard wound care for venous ulcers including compression therapy was provided. At 12 weeks, 29% of placebo‐treated wounds healed, and this increased to 40% when the results were stratified to include the wounds of <15 cm² size and <18 months duration. The reduction in the wound size for all wounds was 56% in placebo, 75% for repifermin 20 µg and 73% for repifermin 60 µg. These results were statistically significant. Results were also statistically significant if both 20 µg and 60 µg dose groups were pooled together and tested for 75% and 90% healing at the end of 12 weeks. Although the results were lower in patients receiving 60 µg as compared with those receiving 20 µg, it should be noted that more than half of the ulcers in the former group were more than 24 months old, and it is likely that they would have taken longer time. In the placebo group, which also received the compression therapy, healing rates at 12 weeks were around 29%, which is well below the standard healing rates achieved by the compression therapy alone (26). The authors have not provided enough information about the follow up of these patients. It appears as they were followed only for one month. The recurrence rates shown are 13% for the placebo and 16% for the group receiving 60 µg of repifermin. Hence, the quality of the epithelial cover provided by the use of repifermin is questionable.

Platelet‐Derived Growth Factor

Platelet‐derived growth factor (PDGF) was first described by Ross (27). Three different isomers have been described (PDGF‐AA, PDGF‐BB and PDGF‐AB). The response of different cell types to the application of PDGF depends upon the expression of different receptors. PDGF has a potent effect on the proliferation of fibroblasts and smooth muscle cells, which express β receptors. That is why most of the research in PDGF has been focused on PDGF‐BB. Animal studies with PDGF have been very encouraging with an increase in the granulation tissue formation and increased tensile strength (28, 29). This has led to the development of a formulation with the recombinant PDGF (rh‐PDGF‐BB) and carboxymethyl cellulose gel (becaplermin), which has been used topically in different trials with promising results. Most of the work with becaplermin has been in patients with diabetic foot ulcers, and no big trail in patients with leg ulcers has been reported.

Knighton et al. (30) studied PDGF in the treatment of 49 patients with chronic cutaneous ulcers. Patients were classified on the basis of wound severity index. After treatment with autologous PDGF, they observed improved healing with a mean 100% healing time of 10·5 weeks. Robson et al. (31) conducted a double blind randomised placebo controlled trial involving 20 patients with pressure ulcers. Each patient in the treatment arm received 10 µg of r‐PDGF. After 28 days, the mean reduction in the wound size was 21% for the treatment group and 6·4% for the placebo group. The results were statistically significant. Similar findings were observed in another trial by Mustoe et al. (32).

Steed (33) in a randomised controlled trial studied the use of r‐PDGF in the treatment of lower extremity diabetic neuropathic ulcers. They had 118 patients each receiving a daily topical application of r‐PDGF (2·2 µg/cm² of ulcer area). Forty‐eight per cent of the patients in the treatment arm healed as compared with the 25% in the placebo arm (P < 0·01). The median reduction in the wound area was better in the treatment group but was not statistically significant. Although this was a randomised study, the mean ulcer area in the placebo group was much larger than the treatment group (9·0 cm² vs. 5·5 cm²), and this could have skewed the results in terms of complete healing rates. The reported healing rates for uncomplicated diabetic neuropathic ulcers with contact casting are around 70%, and if taken as benchmark, the results of this study are fairly low (34).

Becaplermin is the first recombinant growth factor approved by the Food and Drugs Administration (FDA) for the treatment of diabetic chronic neuropathic ulcers (35). Several clinical studies have shown its efficacy. The treatment with becaplermin is expensive ($400/100 g) and requires frequent dressing changes. It has not been studied in trials for other wound types including leg ulcers. A combined analysis of four recent randomised controlled trials of rh‐PDGF in the treatment of diabetic foot ulcers has shown that 100 µg/ml becaplermin used once daily is effective in improving healing, but the number needed to treat to achieve healing in one extra patient could vary from 7 to 25 (36). It seems a reasonable choice in those wounds, which have not healed after optimum conservative treatment.

Platelet‐Derived Wound Healing Formula

Platelet‐derived wound healing formula (PDWHF) is a patient‐specific mixture of growth substances. It has shown efficacy in preclinical studies. But a randomised controlled trial failed to show significant results (37). Another similar trial of topical autologous platelet lysate in the treatment of venous ulcers did not show any difference between the treatment and the control arms (38). As this formula contains a combination of growth factors released from the patients' own platelets, there is a possibility that some of the released growth factors may have an inhibitory effect on healing.

Transforming Growth Factor β

Transforming growth factor β (TGF β) is a family of naturally occurring peptides rather than a single factor. TGF β1 has been shown to accelerate wound healing in animal models. A very few studies of TGF β in wound healing are published. Robson et al. (39) studied the effects of recombinant TGF β2 on diabetic foot ulcers in a double blind randomised controlled trial. They had three groups of patients. One group received standard wound care, other received a placebo sponge and the next group received sponge containing TGF β2 in varying concentrations. Higher wound healing rates were seen in patients treated with TGF β2 as compared with placebo but surprisingly enough the results of the group receiving standard wound care alone were the best (71% vs. 58% and 61%). However, these results reported for standard wound care group are much higher as compared with the other published literature (24). A contributing factor may be the small sample size, and in fact this study needs to be replicated with a large sample to detect statistically significant effect.

Granulocyte Macrophage Colony‐Stimulating Factor

Several clinical trials have shown that granulocyte macrophage colony‐stimulating factor (GM‐CSF) can speed up the healing when compared with placebo. Barbolla‐Escaboza et al. (40) used a single perilesional injection of 300 µg of rh‐GM‐CSF in 10 patients with chronic leg ulcers and observed 80% complete healing response within 4 weeks. This is a very small sample again; however, it supports the hypothesis that GM‐CSF may help to heal chronic wounds.

Da Costa et al. (41) conducted a randomised controlled trial, with patients in the treatment arm receiving a single perilesional injection of rh‐GM‐CSF and the control arm receiving a placebo. They found 50% (8/16) patients have their ulcer healed by week 4, with only one ulcer healing in the control group. This study was terminated earlier as they detected significant difference between the two groups. However, the sample size was small. Each patient received only one injection regardless of the size of the ulcer. They had ischaemic, venous and diabetic ulcers in their study. It is surprising that they just covered the ulcer with an ordinary dressing and did not use compression bandages. They have also not provided any follow up data regarding recurrence rates.

Two years later, they repeated the study with a larger sample size (n = 60) and tried to improve the study design (42). Only venous ulcers were included in the trial, and the patients received standard four‐layer compression bandages, and the patients were followed up for 6–12 months for re‐ulceration. This time they also studied two different doses of GM‐CSF applied for 4 weeks. The results were statistically significant (P < 0·005). They used 200 µg and 400 µg of GM‐CSF perilesionally. After 13 weeks, complete healing was observed in 61% (11/18) receiving 400 µg and in 57% (12/21) receiving 200 µg. However, in the control arm, only 19% (4/21) healed completely. This led to the recommendation that GM‐CSF 400 µg/day is the optimal dose. It must be noted that the healing rate in the control arm is much lower than the benchmark of 55% healing using compression therapy (24). They used plain gauze dressings, and the outcomes may have been different if moist wound environment with adequate compression therapy was used.

Attempts at using topical rh‐GM‐CSF in the treatment of venous ulcers have met some success. In a trial with topical application of GM‐CSF, 47 of 52 ulcers healed completely with an average healing period of 19 weeks (range 3–46) and a 90% overall healing rate (43). Although in this study they did not compare this with a control, they did study nine chronic ulcers, which have been refractory to standard treatment for the last 46 weeks. and their healing rate was 89% with the mean healing time of 19 weeks. The re‐ulceration rate over a year was 6%. We compare these results with a study by Mayer et al. (44), who used standard compression therapy for venous ulcers and achieved over all healing rate of 73% at a year. However, their re‐ulceration rate over a year was 30%. A controlled trial may be required to prove that in clinical setting GM‐CSF improves healing rates than the standard care.

Conclusion

Growth factors have shown encouraging results in the management of leg ulcers in experimental and controlled clinical studies. These promising results in the preclinical studies have not been translated into clinical treatments (45). A point to remember is that most types of wounds made in animals are not very representative of human venous leg ulcers. Small sample sizes and inconsistent end points in different clinical studies have been the main hurdle in reaching a definite conclusion. The results of growth factor trials in diabetic and pressure ulcer patients should be interpreted with caution in context with venous ulcers as the aetiology of venous ulcers involving the underlying venous hypertension is different from that of pressure necrosis seen in diabetic foot and neuropathic ulcers, with a different tissue microenvironment. Future developments may include different delivery methods for the growth factors, use of different combinations of growth factors administered simultaneously (46) or, sequentially, bioengineered skin grafts and chemical induction of angiogenesis with the use of gene transfer techniques.

Acknowledgements

This article is based on an assignment, submitted to the Wound Healing Research Unit, University of Wales Cardiff, as a requirement for MSc in wound healing and tissue repair.