The clinical effectiveness and safety of using epidermal growth factor, fibroblast growth factor and granulocyte-macrophage colony stimulating factor as therapeutics in acute skin wound healing: a systematic review and meta-analysis

Abstract Background Promoting wound healing is crucial to restore the vital barrier function of injured skin. Growth factor products including epidermal growth factor (EGF), fibroblast growth factor (FGF) and granulocyte-macrophage colony stimulating factor (GM-CSF) have been used for decades although no systematic evaluation exists regarding their effectiveness and safety issues in treating acute skin wounds. This has resulted in a lack of guidelines and standards for proper application regimes. Therefore, this systematic review and meta-analysis was performed to critically evaluate the effectiveness and safety of these growth factors on skin acute wounds and provide guidelines for application regimes. Methods We searched PubMed/Medline (1980–2020), Cochrane Library (1980–2020), Cochrane CENTRAL (from establishment to 2020), ClinicalTrials.gov (from establishment to 2020), Chinese Journal Full-text Database (CNKI, 1994–2020), China Biology Medicine disc (CBM, 1978–2019), Chinese Scientific Journal Database (VIP, 1989–2020) and Wanfang Database (WFDATA, 1980–2019). Randomized controlled trials (RCTs), quasi-RCTs and controlled clinical trials treating patients with acute skin wounds from various causes and with those available growth factors were included. Results A total of 7573 papers were identified through database searching; 229 papers including 281 studies were kept after final screening. Administering growth factors significantly shortened the healing time of acute skin wounds, including superficial burn injuries [mean difference (MD) = −3.02; 95% confidence interval (CI):−3.31 ~ −2.74; p < 0.00001], deep burn injuries (MD = −5.63; 95% CI:−7.10 ~ −4.17; p < 0.00001), traumata and surgical wounds (MD = −4.50; 95% CI:−5.55 ~ −3.44; p < 0.00001). Growth factors increased the healing rate of acute skin wounds and decreased scar scores. The incidence of adverse reactions was lower in the growth factor treatment group than in the non-growth factor group. Conclusions The studied growth factors not only are effective and safe for managing acute skin wounds, but also accelerate their healing with no severe adverse reactions.


Background
Skin maintains internal homeostasis and provides a barrier between our body and the outside environment [1]. Acute skin wounds break the barrier and expose the body to the risk of pathogen infections and fluid losses. Therefore, restoring skin integrity as soon as possible after wounding is the body's most effective way to restore the environment's balance, fight infections and prevent fluid and electrolyte disturbances from occurring. The speed of wound healing is of essential importance and can impact on the patient's prognosis [2].
In 1971, Frati and Scarpa reported the treatment of mouse burns with EGF [7]. The first human recombinant FGF-2 was reported in 1988 [8]. In 1989, Brown et al. reported in the New England Journal of Medicine that epidermal growth factor significantly accelerated the rate of healing of partial thickness skin wounds in a randomized clinical trial [9]. The development of growth factor products targeted at promoting wound healing has been thriving ever since and the clinical application of growth factors has become popular. In 1998, Fu et al. reported the result of a randomized placebo-controlled trial investigating the effect of recombinant bovine basic fibroblast growth factor (rbFGF) on burns healing. The study showed that rbFGF effectively decreased the time and improved the quality of healing. These favorable results started a wider trend of using growth factors in wound management [10]. In 2007, Ma et al. reported the use of recombinant human acidic FGF (rh-aFGF) for treating deep partial-thickness burns and skin graft donor site through a randomized, multicenter, doubleblind and placebo-controlled trial. The study demonstrated that rh-aFGF can promote the healing of both burn wounds and skin graft donor sites [11], which further strengthened the evidence of applying growth factor products to promote acute wound healing, including both burns and surgical wounds.
Currently, EGF, bFGF, aFGF and GM-CSF are approved growth factor products for use on acute skin wounds. During the past decades, the therapeutic use of these growth factors in acute wounds management has gradually become a customary practice in China, however, controversies have raged about the benefits and safety of the clinical implementation of distinct kinds of growth factor products. It is known that acute wounds naturally hold plenty of growth factors, which can stimulate cell proliferation and matrix production at the wound bed. Whether the growth factor receptors are saturated prior to the application of more growth factors to acute wounds is unknown. Secondly, deep acute wounds usually heal with hypertrophic scars. It is still unclear whether deep acute wounds heal with more (or less) severe scars under the use of growth factors. Moreover, in light of the economic costs and possible side-effects (such as carcinogenesis) of high local/systemic growth factor levels, it is unclear whether the practice of using exogenous growth factors for the therapy of acute wounds is a real necessity. In addition, whether growth factor treatments provide true benefits remains uncertain given their instability and short in vivo half-life [4,12,13].
Notably, a systematic evaluation of the effectiveness and safety of the available growth factor products used for acute skin wound therapy is missing. There is still the need to investigate whether the routine administration strategies used in clinical treatments suffice to guarantee the growth factor products' benefits. To address these issues, we performed the present systematic review and meta-analysis to assess the clinical effectiveness and safety of all currently clinically available growth factor products in treating acute skin wounds as compared to non-growth factor treatments. The results of

Methods
This systematic review was conducted according to the guidelines for Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) [14]. It was based on the planned Participants, Intervention, Control, Outcome and Study design (PICOS) elements.

Search strategy
The searched databases included: PubMed/Medline (1980-2020); Cochrane Library (1980-2020); Cochrane CENTRAL (from establishment to 2020); ClinicalTrials.gov (from establishment to 2020); Chinese Journal Full-text Database (CNKI, 1994-2020); China Biology Medicine disc (CBM, 1978-2019); Chinese Scientific Journal Database (VIP, 1989-2020); and Wanfang Database (WFDATA, 1980(WFDATA, -2019. With the combination of subject words and free words, the search terms included two categories: (1) 'epidermal growth factor', 'basic fibroblast growth factor', 'acid fibroblast growth factor', and 'granulocyte-macrophage colony stimulating factor'; and (2) 'trauma', 'wound', 'burn', and 'surgery'. The logical relationship was created with 'OR' and 'AND'; and the search formula was thereafter developed according to the characteristics of the different databases. For example, the search strategy for PubMed was: ((epidermal growth factor OR EGF) OR (basic fibroblast growth factor OR bFGF) OR (acid fibroblast growth factor OR aFGF) OR (granulocytemacrophage colony stimulating factor OR GM-CSF)) AND ((superficial OR surgical OR burn) AND wounds)). A preretrieval process improved the searches strategy. In addition, we conducted a manual search of unpublished studies and conference materials, tracking also the references of the included literature. For the analysis we included studies reported in both Chinese and English.

Inclusion and exclusion criteria
The inclusion and exclusion criteria are listed in Table 1.

Study selection
Two researchers independently read the titles and abstracts to exclude the literature that did not meet the inclusion criteria. As a further safeguard, the full texts of the literature that might have met the inclusion criteria were read and evaluated. At the same time, the following information was extracted: author, publication date, research type, characteristics of research objects, sample number, loss of or withdrawal from interview, intervention measures and measurement indicators, and more. For multiple studies published in the same literature, the required data were acquired according to their research contents. In the case of repetitive reports, the study included only the latest or the most comprehensive ones.
Quality evaluation The quality of the included research method was evaluated via Jadad's scale, which is an internationally recognized clinical trial scoring standard, as it includes data about random method, allocation concealment, blind use, loss of follow-up, withdrawal and outcome. The score range was 1-5 points, including 1-2 points for lower quality and 3-5 points for higher quality.

Study selection and characteristics
In total, our preliminary screening selected 7573 papers. After screening titles, abstracts and full-texts ( Figure 1) we kept 229 papers including 281 studies, which consisted of 207 randomized controlled trials (RCTs) and 74 clinical controlled trials (CCTs) with a total of 30 562 patients. The basic characteristics of the included studies and the results of the methodological quality evaluations are shown in Table 2 [10,11,. All the growth factors in these studies were applied topically. In all studies, the patients' basic characteristics were comparable (p > 0.05) between intervention groups and control groups.
Infection rate of second-degree burn wounds Seven studies [16,33,58,76,79,80,82] including 395 cases with superficial second-degree burn wounds compared the infection rate of growth factor and other non-growth factor treatment methods. There turned out to be no statistical heterogeneity between the results (p = 0.24; I 2 = 25%). Therefore, the fixed effect model was used for meta-analysis ( Figure S3, see online supplementary material). The results showed that the infection rate was lower in the growth factor treatment group than in the non-growth factor group, and the difference was statistically significant (RR = 0.52; 95% CI: 0.39-0.69; p < 0.00001). Seventeen studies [16,58,76,80,82,91,94,108,118,119,122,124,128,131,132,135,136] enrolling a total of 1389 patients were conducted to compare the infection rate of deep second-degree burn wounds between growth factor and other non-growth factor treatments. The results showed no statistical heterogeneity (p = 0.54; I 2 = 0%). Hence, the fixed effect model was used for meta-analysis ( Figure S4, see online supplementary material). The results showed that the infection rate was lower in the growth factor group than in the non-growth factor treatment group (RR = 0.52: 95% CI: 0.42 ∼ 0.64; p < 0.00001).
Vancouver scar scale score of deep second-degree burn wounds Five studies [101,104,108,122,123] including 413 patients compared growth factor with other non-growth factor treatments concerning the deep second-degree burn scar score. The follow-up time was between 6 and 12 months. The results showed the presence of statistical heterogeneity (p = 0.004; I 2 = 74%). Therefore, the random effect model was used for meta-analysis ( Figure 4). The results showed that the Vancouver scar scale score of the growth factor treatment group was improved as compared with that of the non-growth factor group (5.23 ∼ 5.67 vs 6.51 ∼ 8.4, i.e. 2.45 lower than that of the non-growth factor treatment group) (MD = −2.45; 95% CI: −3.29 ∼ −1.6; p = 0.004).

Adverse reactions of deep second-degree burn wounds
Three studies [95,96,124], including 522 patients with deep second-degree burn wounds, compared the incidence of adverse reactions after the treatment with growth factor vs. other non-growth factor treatments. The results showed that no statistical heterogeneity occurred (p = 0.29; I 2 = 20%), so the fixed effect model was used for meta-analysis ( Figure S5, see online supplementary material). The results showed that the incidence of adverse reactions was lower in the growth factor treatment group than in the non-growth factor group (RR = 0.35; 95% CI: 0.19-0.67; p = 0.001).

Adverse reaction of traumata and surgical wounds
Six studies [157,171,197,215,219,221] including 622 patients with traumata and surgical wounds compared the incidence of adverse reactions after growth factor treatment or other non-growth factor treatment methods. The results were statistically heterogeneous (p < 0.0001; I 2 = 84%). Hence, the random effect model was used for meta-analysis ( Figure S7, see online supplementary material). The results showed that the incidence of adverse reactions was lower in the growth factor group than in the control group (RR = 0.55; 95% CI: 0.46 ∼ 0.65; p < 0.00001).

Discussion
Growth factors are important biologically active molecules which can markedly impact on the wound environment, leading to rapid increases in cell migration, proliferation and differentiation, while regulating the cellular responses inherent to the wound healing process [14]. Recombinant growth factors have been used as adjunctive treatments for acute wounds to accelerate healing, however, the effectiveness and safety of administering these growth factor products under such conditions had not been systematically analyzed. In 2016, Zhang et al. [242] performed a meta-analysis concerning growth factor therapy in cases of partial thickness burns. Thirteen studies with a total of 1924 participants were included and the results showed that the topical application of growth factors including FGF, EGF and GM-CSF significantly reduced wound healing time as compared with standard wound care alone. Although these preliminary results seemed to be encouraging, the authors pointed out that high-quality and adequately powered trials were still needed to further confirm their conclusions. Another meta-analysis performed by Abdelhakim et al. included 9 clinical studies and has shown that local bFGF treatment accelerated wound healing and prevented pathological scarring. In a similar fashion, the author pointed out that further research was needed to indicate more clinical advantages [243].
In this systematic review, we performed a comprehensive search of relevant clinical studies published in either Chinese or English. We included many studies published in Chinese which had not been considered for evaluation before. Our data show that as compared to non-growth factor treatments, the therapeutic use of growth factor products including FGF, EGF and GM-CSF for acute wounds significantly changed the healing outcome in terms of lessening healing time, heightening healing rate and reducing incidence of infections and adverse reactions. Therefore, our study results positively support the therapeutic use of the current clinically available growth factor products for acute wounds, especially in the case of wounds that tend to have longer healing time.
However, one must point out that out of the 229 studies considered, only 3 were conducted outside China (i.e. in Japan) and reported in English, while the remaining 226 articles, including 7 reported in English and 219 in Chinese, were all carried out within China and reported by Chinese researchers. During the screening period, one randomized clinical trial conducted in the USA showed that epidermal growth factor accelerated skin-graft-donor sites wound healing significantly [9]. However, the types of outcome measurements in this study could not be combined with those from other included studies to conduct meta-analysis. Thus although it was eventually excluded, the results of this study did support our general conclusions. We have to admit that the lack of clinical data from other countries and areas has reduced the evidence's power level. This is especially true considering that most of the included studies are rated as low-quality ones (Jadad score: 1-2 for 202 papers, 4-5 for 6 papers only). The lack of sufficient clinical data from other countries and areas outside Asia is likely caused by the lack of available growth factor products for treating acute wounds in these places. Becaplermin in Regranex ® is the only U.S. Food and Drug Administration (FDA) approved recombinant PDGF product and is only indicated for the treatment of neuropathic ulcers in diabetics. This product carried a boxed warning from the FDA and due to safety issues has been withdrawn in Europe [244]. We were only able to find one study using PDGF gel to treat acute fullthickness punch biopsy wounds on 7 healthy subjects [245]. The results of the study showed PDGF gel was effective in promoting wound healing, which was in accord with the general results of this meta-analysis. Since PDGF has not been officially approved for use on acute wounds, we did not include PDGF in this meta-analysis. However, we believe that when PDGF becomes more widely used for treating acute wounds in the future, it will be meaningful to conduct a more comprehensive evaluation regarding the efficacy and safety issues of all the important growth factor products that are still lacking evidence for clinical use today.
Although this meta-analysis has brought to light encouraging results, the collection of the latter from limited countries and areas (mainly in China) increases the bias of the study. From this standpoint, the evidence supporting the routine therapeutic use of growth factor products for acute wounds is still weak. More high-quality clinical studies and clinical studies from outside of China are needed to further confirm the efficacy, necessity and safety of their clinical application. Despite the possible bias of the conclusions drawn from clinical studies, the current data do show some potential merits of using growth factors to promote acute wound healing. It is interesting to note that several of the included studies focused on the healing of surgical wounds entailing high risks of contamination and infection, such as in the case of perianal surgery [154,214,218,219,223,224,226]. Growth factors were beneficial as they decreased the healing time of such wounds, and therefore decreased the chances of infection and of the development into chronic wounds. Thus, the therapeutic use of growth factors in cases with surgical wounds susceptible to contamination and infection could be a beneficial practice. Again, the need remains for more evidence reported by higher-quality studies.
Moreover, we noted that therapeutically using growth factors for acute wounds not only increased the speed of healing, but also improved the quality of healing in the case of deep wounds. It is well worth pointing out that with growth factors treatments, deep second-degree burn wounds healed with lower scar scores [101,104,108,122,123], which is an important indicator for routine clinical use. It is well known that an increased wound healing time is an important risk factor for hypertrophic scarring in second-degree burns [246]. The current data showed that, instead of causing 'an overgrowth', growth factor treatments safely reduced wound healing time by 5.63 days while concurrently decreasing the degree of hypertrophic scarring. Similarly, in their study Abdelhakim et al. [243] also pointed out that bFGF might prevent pathological scarring through several cellular mechanisms, such as interfering with myofibroblasts formation and inducing apoptosis. However, longer follow-up times and large-scale clinical trials are still needed to confirm this scar-reducing effect and the causal relationship with reduced wound healing times.
Notably, most of the studies included in this systematic review used only a single growth factor either by itself or combined with other non-growth factor treatments and proved their effectiveness. However, it is yet to be proven that combining different growth factors achieves better clinical results, or whether the contrary is true. Since applying supraphysiological doses of growth factor(s) correlates with an increased risk of cancer, the importance of controlling the spatial-temporal release of growth factors at the wound site and of overcoming this challenge is probably crucial for any successful growth factor-based therapy [244]. Also, as different growth factors partake in the various stages of the wound healing process, using a single growth factor may not suffice for best wound healing. A sophisticated growth factor delivery system enabling a controlled spatialtemporal delivery [13], mimicking the synergistic wound healing activity of the combined release profiles of growth factors in real physiological situations, could be a promising direction for future research. Currently, the use of platelet rich plasma (PRP) to promote refractory wound healing has already supplied a hint for applying growth factor compounds in a more effective fashion. However, PRP has not been routinely used on acute wounds due to economic considerations. More in-depth study of the PRP's spatial-temporal working mechanism might provide stronger evidence to develop recombinant growth factor combination products for promoting acute wound healing in the future.

Conclusions
With the systematic review and evaluation of the currently available evidence, we conclude that the therapeutic use of growth factors including EGF, FGF and GM-CSF is effective and safe in the treatment of acute skin wounds, especially in the case of wounds entailing higher risks of infection. However, the need still remains for more higher-quality studies to further strengthen our conclusion.

Supplementary data
Supplementary data is available at Burns & Trauma Journal online.