Eur J Dermatol 2014; 24(2): 210-6
Therapy Anneke BROWN1 Matthias AUGUSTIN2 3 ¨ Michael JUNGER 4 Markus ZUTT Joachim DISSEMOND5 Eberhard RABE6 Roland KAUFMANN7 Margrit SIMON8 9 ¨ Markus STUCKER Sigrid KARRER10 Wolfgang KOENEN11 Wolfgang VANSCHEIDT12 Karin SCHARFETTER-KOCHANEK13 Uwe WOLLINA14 Thomas KRIEG1,15,16 Sabine A. EMING1,15,16 1
Department of Dermatology, University of Cologne Kerpener Str. 62, 50937 Cologne, Germany 2 Department of health service research in dermatology and health care profession, University of Hamburg-Eppendorf, Germany 3 Department of Dermatology, Ernst-Moriz-Arndt-University, Greifswald, Germany 4 Department of Dermatology, University of Göttingen, Germany 5 Department of Dermatology, University of Essen, Germany 6 Department of Dermatology, University of Bonn, Germany 7 Department of Dermatology, Johann-Wolfgang-Goethe-University of Frankfurt, Germany 8 Dermatological Practice, Hauptstr. 131, Berlin, Germany 9 Departement of Dermatology, Ruhr-University of Bochum, Germany 10 Department of Dermatology, University of Regensburg, Germany 11 Department of Dermatology, University of Mannheim, Germany 12 Dermatological Practice, Paula-Modersohn-Platz 3, Freiburg, Germany 13 Department of Dermatology, University of Ulm, Germany 14 Department of Dermatology, Hospital Dresden-Friedrichstadt, Germany 15 Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Germany 16 Center for Molecular Medicine (CMMC), University of Cologne, Germany
Randomized standard-of-care-controlled trial of a silica gel fibre matrix in the treatment of chronic venous leg ulcers* Background: Chronic venous leg ulcers (CVU) are a common, unresolved medical problem. Silica gel fibre (SGF) is a novel biodegradable inorganic material developed to serve as a carrier substrate for the local release of pharmaceutical agents facilitating tissue repair. Objectives: To assess the performance and safety of SGF in subjects with CVU. Methods: Open, randomized, standard-of-care-controlled, multi-centre trial. Subjects (ITT 120 patients) received either SGF in addition to standard treatment or standard-of-care treatment (S-o-C) alone. The primary performance variable was the time to healing of the target ulcer until the end of a 12 week treatment period. Results: SGF was well tolerated. Mean time to healing up to week 12 was 85.62 days for the SGF group (SE ± 1.5) and 79.66 days for the S-o-C group (SE ± 1.77) (p-value = 0.217). There was no statistically relevant difference regarding the incidence of complete healing of the target ulcers by weeks 12 and 24 between the SGF and the S-o-C groups (p-value >0.05). Conclusion: SGF is well tolerated and offers a promising perspective as a carrier substrate for the local release of active pharmaceutical agents into the wound site to promote tissue repair. Key words: biodegradable matrix, silica gel fibre matrix, venous ulcer
Reprints: S. A. Eming doi:10.1684/ejd.2014.2344
Article accepted on 12/12/2013 ∗ Registration number and name of trial registry: Study number SGF27101, ClinicalTrials.gov identifier: NCT00998673. Study title “Randomized standard-of-care-controlled trial of a SGF wound dressing in the treatment of chronic venous leg ulcers”. Information of trial protocol: http://clinicaltrials.gov/ct2/show/NCT00998673.
210
EJD, vol. 24, n◦ 2, March-April 2014
To cite this article: Brown A, Augustin M, Jünger M, Zutt M, Dissemond J, Rabe E, Kaufmann R, Simon M, Stücker M, Karrer S, Koenen W, Vanscheidt W, ScharfetterKochanek K, Wollina U, Krieg T, Eming SA. Randomized standard-of-care-controlled trial of a silica gel fibre matrix in the treatment of chronic venous leg ulcers. Eur J Dermatol 2014; 24(2): 210-6 doi:10.1684/ejd.2014.2344
C
hronic leg ulcers are a common clinical problem in Western countries. Approximately 60-70% of chronic leg ulcers are due to chronic venous insufficiency [1]. In Germany, 0.1% of the population suffers from chronic venous leg ulcers [2] and it is estimated that this disease creates costs of about 2 to 2.5 billion euros per year [3]. Current conservative venous leg ulcer management consists of compression therapy, debridement and moist dressings. A large variety of diverse wound dressings exist, ranging from a simple wound gauze to complex, biologically active products containing, for example, growth factors or hyaluronic acid [4]. However, randomized controlled prospective studies testing and demonstrating the efficiency of current modern wound products are largely missing. The development of biodegradable matrix scaffolds that promote tissue repair and regeneration represents a perspective for advanced therapy of acute and chronic skin wounds [5]. Currently, there are no products of bioengineered scaffolds/matrices that completely recapitulate the complex physiology, biological texture and function of uninjured skin. Thus, the development of a suitable material that meets the critical physical, biological and mechanical properties of healing skin remains a major challenge. In recent experimental studies, biomaterials have been successfully combined with growth factors or other wound healing promoting factors stimulating the repair response [6, 7], and some of these approaches are currently under investigation in clinical trials. SGF is a biodegradable inorganic material developed for tissue regeneration. The SGF matrix fibres are composed of a non-crystalline poly hydroxy silicic acid ethyl ester. The individual fibres have a mean diameter of 50m. SGF does not contain viable or non-viable, animal or human compounds or a pharmaceutically active substance. It is hypothesized that SGF promotes tissue growth by providing a 3-dimensional structure that functions as a scaffold for ingrowing cells and vascular sprouts. Furthermore, SGF offers important biomaterial properties that make it a favourable material to deliver and release pharmaceutically active substrates into the wound site. Based on pharmacokinetic and histological analysis in pre-clinical skin wound healing studies in mini pigs, SGF is progressively degraded by hydrolysis during the healing response and replaced by host granulation tissue within 3-13 weeks. The terminal decomposition product is orthosilicic acid [8]. The SGF material degrades slower compared to known biodegradable wound matrices which are based on collagen, cellulosis or polyglycolic acid gels. The SGF matrix was developed by Bayer Innovation GmbH (Düsseldorf, Germany; Bayer Code KGF271) based on a technology discovered 10 years ago by the Fraunhofer Institute for Silicate Research (Wurzburg, Germany). SGF has achieved CE approval in Europe and the groundwork has been laid for FDA 510(k) approval. Bayer Innovation GmbH holds an ISO 13485 certificate. This is the first study investigating the use of SGF in the clinic. Other devices with similar characteristics regarding its use, application and claim are for example Alloderm® , Graftjacket® , Oasis® and Primatrix® . All those products provide extracellular matrix scaffolds, which are biodegradable and acellular, although of different biological origin (human, porcine or bovine). The aim of this study was to test the tolerability and safety profile of SGF for the local treatment of chronic skin wounds.
EJD, vol. 24, n◦ 2, March-April 2014
Material and methods The study was performed as an open, 1:1-randomized, parallel group, standard-of-care (S-o-C) controlled, multicentre trial with 14 study centres in Germany (12 university hospitals and 2 dermatological practices) and one in the USA (centre for advanced wound care). The study was approved by the local ethic committees.
Study subjects Inclusion criteria were men or women aged ≥ 18 years with a chronic venous leg ulcer (proven by duplex or Doppler sonography and by ankle/brachial arterial Doppler pressure index ≥0.8 and ≤1.3), ulcer duration minimum 3 months up to 5 years, ulcer size 5 - 40 cm2 , viable wound bed with granulation tissue and no exposed muscle, tendon or bone. Treatment with active wound care agents had to be paused for 14 days before start of study treatment and no topical antibiotics, antiseptics, enzymatic debridement agents were allowed within 7 days before the start of the study treatment. Exclusion criteria were pregnant or nursing women, ulcers of a non-venous aetiology and pre-existing conditions, like signs of infection of the ulcer, medications and therapies inhibiting wound healing or unstable diseases like uncontrolled diabetes with HbA1c >8%, uncontrolled organ failure or active malignancies. The subjects had to give their signed informed consent.
Randomization The randomization list was generated by the Biometrical Department of Winicker Norimed GmbH (Nürnberg, Germany). Randomization numbers were assigned to eligible subjects in ascending order at each centre. Enrolment of subjects was competitive across all participating centres. The subjects were randomized at the baseline visit (Day 0) in a 1:1 ratio to 1 of 2 treatment groups (SGF or S-o-C). Since time to healing was expected to increase with longer disease duration and ulcer size [9-11], stratification by disease duration (ulcer present ≤1 year versus >1 year) and ulcer size (20 cm2 ) was applied in order to avoid bias.
Determination of sample size Based on the current literature [12, 13], it was assumed that no healing would occur during the first 12 days after randomization, i.e. that the study period with possible healing events would have a duration of 72 days. The assumption of a healing rate of 80% under 84-day SGF treatment and a healing rate of 64% under S-o-C treatment (e.g. SGF treatment 20% superior to S-o-C treatment) corresponded – under the assumption of exponentially distributed healing events – to a median time to healing of 61 days in the S-o-C group and 43 days in the SGF group. Choosing a significance level of ␣ = 0.05 and a power of  = 80%, a two-sided log-rank test was determined to require n = 110 subjects in each group. Assuming further a drop-out rate of 10%, a total of 246 randomized subjects were determined to be necessary.
211
Study plan The study plan is illustrated in figure 1. The 2 treatment groups received either SGF in addition to S-o-C treatment of their target ulcer (SGF group) or S-o-C treatment alone (S-o-C group). The treatment consisted of complete sharp or ultrasonic debridement of all non-viable tissue after local anaesthesia (e.g Emla® Cream, AstraZeneca GmbH, Wedel, Germany) was applied at least 60 minutes before, followed by application of SGF and/ or S-o-C as defined below. The SGF patches used in this study were 0.2 cm thick and had a size of 4.7 cm × 4.7 cm. SGF wound dressing was cut to the exact size of the target ulcer and was applied once at baseline visit by a specially trained study nurse in the study centre. SGF was not removed and remained on the ulcer. Reapplication of SGF during the 12 week treatment phase was only performed if the material had been completely absorbed. S-o-C consisted of non-adhesive wound dressing (Mepilex® for exudative wounds or Mepitel® for non-exudative wounds, Mölnlycke Healthcare, ErkrathUnterfeldhaus, Germany) and compression therapy (a 4-layer bandaging system (Profore® , Smith & Nephew, Hamburg, Germany)) from the base of the toes to just below the knee. Change of the covering non-adhesive dressing and compression therapy were performed at least twice weekly by a specially trained wound nurse, an ambulant nursing service or at the study centre. Debridement or cleansing of the wounds with isotonic sodium chloride solution was repeated at the biweekly visits as necessary. The treatment period lasted either until complete ulcer healing or for a maximum of 12 weeks. Treatment could be prolonged to 24 weeks if the target ulcer had decreased but not healed by week 12. Subjects in the S-o-C group whose target ulcers had not healed could switch to SGF treatment after week 12 for 4 weeks, and then for an additional 8 weeks if the target ulcer was responsive to the SGF
treatment. The subjects visited the study centres biweekly during the treatment period and monthly during the follow-up period. Ulcers were photographed and assessed at each visit based on standard wound healing parameters (ulcer size, exudate amount, wound edge, amount of epithelialisation, amount of remaining SGF wound matrix (SGF group), recurrence of ulcer (healed ulcers)). Physical examinations were carried out at baseline, weeks 12 and 24. Adverse events (AE), concomitant therapies and compliance were assessed at each visit, vital signs at each visit during the treatment period and at week 24. The following treatment-related rules to stop the study were applied: repeated necrosectomy (>2×), increase in ulcer size ≥20%, no decrease in ulcer size after 6 weeks of treatment.
Efficacy assessments The primary and secondary performance variables are shown in textbox 1. An ulcer was regarded as healed when the skin was completely re-epithelised without drainage or dressing requirements. The primary performance analysis was based on the intention-to treat (ITT) dataset and performed by two-sided log rank test (significance level ␣ = 5%). It was repeated with a Cox regression analysis checking for centre effects and with the per protocol (PP) analysis set. Kaplan-Meier estimates for the time to healing were calculated for both treatment groups. Performance analyses were in parts repeated by strata to compare corrected vs. uncorrected data (disease duration ≤1 year vs. >1 year). The second performance analyses were based on ITT dataset as well as on PP dataset. They were performed by sample statistics, frequency tables and exploratory tests as appropriate. Pain due to venous leg ulcer between and during dressing changes were assessed by 2 visual analogue scales (VAS). The subjects marked the degree of their pain
Study treatment
S-o-C up to 12 weeks*
SGF + S-o-C up to 12 weeks
Complete healing
Generally responsive but not completely healed
12 week follow-up
Complete healing
SGF + S-o-C up to 12 weeks*
S-o-C or other treatment regimen*
Generally responsive but not completely healed
12 week follow-up
Switch to SGF for 4 weeks
SGF + S-o-C up to 8 weeks if wound size reduced by 50%
S-o-C or other treatment regimen* if not responsive*
*4-week
safety follow-up after last treatment
S-o-C or other treatment regimen*
Figure 1. Overview of the different treatment groups.
212
EJD, vol. 24, n◦ 2, March-April 2014
Box 1. Primary and secondary performance variables Primary performance variable: time to healing of the target ulcer up to the end of the 12 week treatment period (calculated as: date of confirmed healing minus date of baseline visit plus 1 (days)) Secondary performance variables: time to healing of the target ulcer up to week 24 (for subjects who continued treatment after week 12) incidence of complete wound healing of the target ulcer at week 4, 8 and 12 rate of recurrence of the target ulcer at the end of the 12 week follow-up period size of target ulcer at different time points (absolute size and change) ulcer related pain between visits and during dressing change patient benefit index physician’s and subject’s assessment of dressing performance
on a horizontal scale of 100 mm labelled “no pain” at the extreme left and “unbearable pain” at the extreme right. Patient benefit index (PBI) measures patient-defined treatment objectives and benefits, particularly in the course of a treatment. In this study a validated version developed for chronic wounds, consisting of 22 items was used (Version PBI-W22).
subject received 1 SGF patch at baseline and 4 reapplications of smaller patches in total during the entire study (medians). In terms of material size, approximately 11cm2 SGF wound dressing was applied on average to each subject at baseline and 24cm2 in total during the entire study (medians).
Demographic and baseline data The description of the demographic and baseline characteristics of the study population focuses on the ITT data and is shown in table 1. All subjects were Caucasians. The target ulcers of most subjects were located at the inner ankles and affected the epidermis/dermis or subcutis. For the majority, the ulcer’s wound edge was distinct/attached to wound base and the amount of wound exudate was moderate or light. 26 of 60 subjects in the SGF group, 29 of 60 subjects in the S-o-C group and 7 of 19 subjects in the switched to SGF group had a disease duration 0.05).
Time to healing up to week 24 Mean time to healing up to week 24 was 128.73 days for the SGF group (SE ± 4.79) and 114.2 days for the S-o-C group (SE ± 5.44) (p-value = 0.142).
Incidence of complete healing Complete healing of the target ulcers by week 12 was documented for 10 subjects in the SGF group (16.7%) and for 16 subjects in the S-o-C group (26.7%) (p = 0.268), by week 24 it was documented for 15 subjects in the SGF group (25%) and for 21 subjects in the S-o-C group (35%) (p = 0.319) (figure 2). Among the latter, 2 subjects had been switched to SGF after week 12. Stratification by disease duration for different periods showed p-values ≥0.089 (Fisher’s exact test). However, the healing rates in the stratum with an ulcer duration ≤1 year were higher than in subjects with longer disease duration, in particular in the S-o-C group.
Size of target ulcer In both treatment groups the mean size of the target ulcer decreased from baseline to week 12. Stratification by dis-
ease duration and ulcer size by means of Wilcoxon test did not show any significant differences between the treatment groups (figure 3). The ulcer size reduction of the mean target ulcer size observed in subjects switched to SGF was similar to that in the SGF group. There were no marked differences regarding the incidence of complete healing up to week 12 or up to week 24 (Fisher’s exact test p-value ≥0.083 or ≥0.151) stratified by target ulcer size at baseline (20 cm2 ) between the treatment groups. Healing rate and ulcer size appeared inversely related, in particular in the SGF group.
Recurrence of ulcer Recurrence of an ulcer already healed was seen in 1 subject in the SGF group and in 1 subject in the S-o-C group.
Ulcer related pain between visits and during dressing changes In both treatment groups the mean pain between and during dressing changes decreased from baseline to week 12, without indicating any marked differences between the 2 treatment groups when analysed by ANCOVA (p-value ≥0.105). The analysis of the data stratified by disease duration showed that the baseline pain level during dressing changes was lower in subjects with disease duration >1 year than in subjects with disease duration ≤1 year.
Patient benefit index (PBI) PBI data might give a slightly distorted picture due to a relatively high number of missing values (e.g. PBI scores week 12: 10%). However the available data do not indicate any marked differences between treatment groups
60,00%
50,00%
40,00%
30,00%
20,00%
10,00%
0,00% SGF (N = 26)
S-o-C (N = 29)
switched to SGF (N = 7)
SGF (N = 34)
Ulcer present ≤ 1 year
switched to SGF (N = 12)
S-o-C (N = 31)
Ulcer present > 1 year week 12
week 24
Figure 2. Incidence of complete healing – stratified by disease duration (ITT data). From week 14 onwards, subjects who switched from S-o-C to SGF were analyzed in the S-o-C group as well as in the switchedto-SGF.
214
EJD, vol. 24, n◦ 2, March-April 2014
60,00%
50,00%
40,00%
30,00%
20,00%
10,00%
0,00% SGF (N = 25)
S-o-C (N = 29)
switched to SGF (N = 8)
Ulcer size < 10cm2
SGF (N = 24)
S-o-C (N = 21)
switched to SGF (N = 6)
Ulcer size 10-20cm2 week 12
S-o-C (N = 11)
S-o-C (N = 10)
switched to SGF (N = 5)
Ulcer size > 20cm2
week 24
Figure 3. Incidence of complete healing – stratified by ulcer size at baseline (ITT data). From week 14 onwards, subjects who switched from S-o-C to SGF were analyzed in the S-o-C group as well as in the switchedto-SGF.
(p-values ≥0.175). At week 12 the majority of subjects (who answered the respective question) stated that the treatment had helped them “very much” or at least “quite a lot” to reach their baseline goals (see also Demographic and baseline data).
Physician’s and subject’s assessment of dressing performance Both the SGF matrix and the wound dressing used in the S-o-C group were described as easy to apply by physicians and subjects. Both dressings generally met the physicians’ expectations and were considered to be used or recommended again for the treatment of chronic venous leg ulcers. The majority of the subjects in both treatment groups described the progress of healing as good or very good (SGF-group: 88.9%, S-o-C-group: 79.2%) and most subjects were “satisfied” or “very satisfied” with the overall performance of the dressings (SGF group: 95.5%; S-o-C group: 91.7%).
Adverse events (AEs) Device related AEs were observed in 14 subjects in the SGF group and in 18 subjects in the S-o-C group. None of the reported device-related AEs were serious.
Discussion In this study we tested the performance and safety of a novel, inorganic, biodegradable SGF scaffold for the local treatment of subjects with chronic venous leg ulcers. The EJD, vol. 24, n◦ 2, March-April 2014
findings show that the SGF matrix is degraded and absorbed by the wound environment in a biocompatible fashion. No side effects related specifically to the biomaterial were observed. Furthermore, the rate of wound closure in the SGF group versus the S-o-C group was comparable. Thus, the novel developed SGF matrix is well tolerated and biocompatible. The findings in this study also indicate that the SGF matrix itself did not exert specific pro-healing activities. A possible reason for this may be that, particularly in the SGF group, sharp debridement of the ulcers was consequently not performed at each visit. The number of subjects who received sharp debridement in the SGF group compared to the S-o-C group decreased from baseline (SGF group: 34 subjects vs. S-o-C group: 24 subjects) towards the end of the study (SGF group: 7 subjects vs. S-o-C group: 17 subjects). Based on the current findings it is not clear if debridement was less often indicated in the SGF group or if it was not performed for other reasons, e.g. hesitation to remove SGF patches for the procedure. Frequent surgical debridement leads to a better healing rate [14, 15]. Thus, for future studies a consequent and systematic surgical debridement should be part of the study plan. Another explanation for not finding significant differences in the healing rate between the SGF and the S-o-C group might be that the portion of subjects with “hard to heal” ulcers was high compared to other published studies [10, 11, 16]. In fact, the incidence of ulcer healing in the S-o-C group was with 26.7% after 12 weeks and 35% after 24 weeks, lower than results which had been reported in previous studies and which provided the basis for our study design. Gelfand et al., for example, observed a healing rate of chronic venous ulcers of 45.2% after 12 weeks of standard wound therapy with limb compression [11]. Wong et al. observed even higher
215
healing rates of approximately 60% after 12 weeks and approximately 70% after 24 weeks [17]. The subjects included in both of these studies had disease durations between 4 weeks and 1 year, respectively, indicating a mean of 3 months. In the presented study about 50% of the subjects suffered from their ulcers for longer than 1 year. The observed healing rate was higher in subjects with disease duration
Therapy Anneke BROWN1 Matthias AUGUSTIN2 3 ¨ Michael JUNGER 4 Markus ZUTT Joachim DISSEMOND5 Eberhard RABE6 Roland KAUFMANN7 Margrit SIMON8 9 ¨ Markus STUCKER Sigrid KARRER10 Wolfgang KOENEN11 Wolfgang VANSCHEIDT12 Karin SCHARFETTER-KOCHANEK13 Uwe WOLLINA14 Thomas KRIEG1,15,16 Sabine A. EMING1,15,16 1
Department of Dermatology, University of Cologne Kerpener Str. 62, 50937 Cologne, Germany 2 Department of health service research in dermatology and health care profession, University of Hamburg-Eppendorf, Germany 3 Department of Dermatology, Ernst-Moriz-Arndt-University, Greifswald, Germany 4 Department of Dermatology, University of Göttingen, Germany 5 Department of Dermatology, University of Essen, Germany 6 Department of Dermatology, University of Bonn, Germany 7 Department of Dermatology, Johann-Wolfgang-Goethe-University of Frankfurt, Germany 8 Dermatological Practice, Hauptstr. 131, Berlin, Germany 9 Departement of Dermatology, Ruhr-University of Bochum, Germany 10 Department of Dermatology, University of Regensburg, Germany 11 Department of Dermatology, University of Mannheim, Germany 12 Dermatological Practice, Paula-Modersohn-Platz 3, Freiburg, Germany 13 Department of Dermatology, University of Ulm, Germany 14 Department of Dermatology, Hospital Dresden-Friedrichstadt, Germany 15 Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Germany 16 Center for Molecular Medicine (CMMC), University of Cologne, Germany
Randomized standard-of-care-controlled trial of a silica gel fibre matrix in the treatment of chronic venous leg ulcers* Background: Chronic venous leg ulcers (CVU) are a common, unresolved medical problem. Silica gel fibre (SGF) is a novel biodegradable inorganic material developed to serve as a carrier substrate for the local release of pharmaceutical agents facilitating tissue repair. Objectives: To assess the performance and safety of SGF in subjects with CVU. Methods: Open, randomized, standard-of-care-controlled, multi-centre trial. Subjects (ITT 120 patients) received either SGF in addition to standard treatment or standard-of-care treatment (S-o-C) alone. The primary performance variable was the time to healing of the target ulcer until the end of a 12 week treatment period. Results: SGF was well tolerated. Mean time to healing up to week 12 was 85.62 days for the SGF group (SE ± 1.5) and 79.66 days for the S-o-C group (SE ± 1.77) (p-value = 0.217). There was no statistically relevant difference regarding the incidence of complete healing of the target ulcers by weeks 12 and 24 between the SGF and the S-o-C groups (p-value >0.05). Conclusion: SGF is well tolerated and offers a promising perspective as a carrier substrate for the local release of active pharmaceutical agents into the wound site to promote tissue repair. Key words: biodegradable matrix, silica gel fibre matrix, venous ulcer
Reprints: S. A. Eming doi:10.1684/ejd.2014.2344
Article accepted on 12/12/2013 ∗ Registration number and name of trial registry: Study number SGF27101, ClinicalTrials.gov identifier: NCT00998673. Study title “Randomized standard-of-care-controlled trial of a SGF wound dressing in the treatment of chronic venous leg ulcers”. Information of trial protocol: http://clinicaltrials.gov/ct2/show/NCT00998673.
210
EJD, vol. 24, n◦ 2, March-April 2014
To cite this article: Brown A, Augustin M, Jünger M, Zutt M, Dissemond J, Rabe E, Kaufmann R, Simon M, Stücker M, Karrer S, Koenen W, Vanscheidt W, ScharfetterKochanek K, Wollina U, Krieg T, Eming SA. Randomized standard-of-care-controlled trial of a silica gel fibre matrix in the treatment of chronic venous leg ulcers. Eur J Dermatol 2014; 24(2): 210-6 doi:10.1684/ejd.2014.2344
C
hronic leg ulcers are a common clinical problem in Western countries. Approximately 60-70% of chronic leg ulcers are due to chronic venous insufficiency [1]. In Germany, 0.1% of the population suffers from chronic venous leg ulcers [2] and it is estimated that this disease creates costs of about 2 to 2.5 billion euros per year [3]. Current conservative venous leg ulcer management consists of compression therapy, debridement and moist dressings. A large variety of diverse wound dressings exist, ranging from a simple wound gauze to complex, biologically active products containing, for example, growth factors or hyaluronic acid [4]. However, randomized controlled prospective studies testing and demonstrating the efficiency of current modern wound products are largely missing. The development of biodegradable matrix scaffolds that promote tissue repair and regeneration represents a perspective for advanced therapy of acute and chronic skin wounds [5]. Currently, there are no products of bioengineered scaffolds/matrices that completely recapitulate the complex physiology, biological texture and function of uninjured skin. Thus, the development of a suitable material that meets the critical physical, biological and mechanical properties of healing skin remains a major challenge. In recent experimental studies, biomaterials have been successfully combined with growth factors or other wound healing promoting factors stimulating the repair response [6, 7], and some of these approaches are currently under investigation in clinical trials. SGF is a biodegradable inorganic material developed for tissue regeneration. The SGF matrix fibres are composed of a non-crystalline poly hydroxy silicic acid ethyl ester. The individual fibres have a mean diameter of 50m. SGF does not contain viable or non-viable, animal or human compounds or a pharmaceutically active substance. It is hypothesized that SGF promotes tissue growth by providing a 3-dimensional structure that functions as a scaffold for ingrowing cells and vascular sprouts. Furthermore, SGF offers important biomaterial properties that make it a favourable material to deliver and release pharmaceutically active substrates into the wound site. Based on pharmacokinetic and histological analysis in pre-clinical skin wound healing studies in mini pigs, SGF is progressively degraded by hydrolysis during the healing response and replaced by host granulation tissue within 3-13 weeks. The terminal decomposition product is orthosilicic acid [8]. The SGF material degrades slower compared to known biodegradable wound matrices which are based on collagen, cellulosis or polyglycolic acid gels. The SGF matrix was developed by Bayer Innovation GmbH (Düsseldorf, Germany; Bayer Code KGF271) based on a technology discovered 10 years ago by the Fraunhofer Institute for Silicate Research (Wurzburg, Germany). SGF has achieved CE approval in Europe and the groundwork has been laid for FDA 510(k) approval. Bayer Innovation GmbH holds an ISO 13485 certificate. This is the first study investigating the use of SGF in the clinic. Other devices with similar characteristics regarding its use, application and claim are for example Alloderm® , Graftjacket® , Oasis® and Primatrix® . All those products provide extracellular matrix scaffolds, which are biodegradable and acellular, although of different biological origin (human, porcine or bovine). The aim of this study was to test the tolerability and safety profile of SGF for the local treatment of chronic skin wounds.
EJD, vol. 24, n◦ 2, March-April 2014
Material and methods The study was performed as an open, 1:1-randomized, parallel group, standard-of-care (S-o-C) controlled, multicentre trial with 14 study centres in Germany (12 university hospitals and 2 dermatological practices) and one in the USA (centre for advanced wound care). The study was approved by the local ethic committees.
Study subjects Inclusion criteria were men or women aged ≥ 18 years with a chronic venous leg ulcer (proven by duplex or Doppler sonography and by ankle/brachial arterial Doppler pressure index ≥0.8 and ≤1.3), ulcer duration minimum 3 months up to 5 years, ulcer size 5 - 40 cm2 , viable wound bed with granulation tissue and no exposed muscle, tendon or bone. Treatment with active wound care agents had to be paused for 14 days before start of study treatment and no topical antibiotics, antiseptics, enzymatic debridement agents were allowed within 7 days before the start of the study treatment. Exclusion criteria were pregnant or nursing women, ulcers of a non-venous aetiology and pre-existing conditions, like signs of infection of the ulcer, medications and therapies inhibiting wound healing or unstable diseases like uncontrolled diabetes with HbA1c >8%, uncontrolled organ failure or active malignancies. The subjects had to give their signed informed consent.
Randomization The randomization list was generated by the Biometrical Department of Winicker Norimed GmbH (Nürnberg, Germany). Randomization numbers were assigned to eligible subjects in ascending order at each centre. Enrolment of subjects was competitive across all participating centres. The subjects were randomized at the baseline visit (Day 0) in a 1:1 ratio to 1 of 2 treatment groups (SGF or S-o-C). Since time to healing was expected to increase with longer disease duration and ulcer size [9-11], stratification by disease duration (ulcer present ≤1 year versus >1 year) and ulcer size (20 cm2 ) was applied in order to avoid bias.
Determination of sample size Based on the current literature [12, 13], it was assumed that no healing would occur during the first 12 days after randomization, i.e. that the study period with possible healing events would have a duration of 72 days. The assumption of a healing rate of 80% under 84-day SGF treatment and a healing rate of 64% under S-o-C treatment (e.g. SGF treatment 20% superior to S-o-C treatment) corresponded – under the assumption of exponentially distributed healing events – to a median time to healing of 61 days in the S-o-C group and 43 days in the SGF group. Choosing a significance level of ␣ = 0.05 and a power of  = 80%, a two-sided log-rank test was determined to require n = 110 subjects in each group. Assuming further a drop-out rate of 10%, a total of 246 randomized subjects were determined to be necessary.
211
Study plan The study plan is illustrated in figure 1. The 2 treatment groups received either SGF in addition to S-o-C treatment of their target ulcer (SGF group) or S-o-C treatment alone (S-o-C group). The treatment consisted of complete sharp or ultrasonic debridement of all non-viable tissue after local anaesthesia (e.g Emla® Cream, AstraZeneca GmbH, Wedel, Germany) was applied at least 60 minutes before, followed by application of SGF and/ or S-o-C as defined below. The SGF patches used in this study were 0.2 cm thick and had a size of 4.7 cm × 4.7 cm. SGF wound dressing was cut to the exact size of the target ulcer and was applied once at baseline visit by a specially trained study nurse in the study centre. SGF was not removed and remained on the ulcer. Reapplication of SGF during the 12 week treatment phase was only performed if the material had been completely absorbed. S-o-C consisted of non-adhesive wound dressing (Mepilex® for exudative wounds or Mepitel® for non-exudative wounds, Mölnlycke Healthcare, ErkrathUnterfeldhaus, Germany) and compression therapy (a 4-layer bandaging system (Profore® , Smith & Nephew, Hamburg, Germany)) from the base of the toes to just below the knee. Change of the covering non-adhesive dressing and compression therapy were performed at least twice weekly by a specially trained wound nurse, an ambulant nursing service or at the study centre. Debridement or cleansing of the wounds with isotonic sodium chloride solution was repeated at the biweekly visits as necessary. The treatment period lasted either until complete ulcer healing or for a maximum of 12 weeks. Treatment could be prolonged to 24 weeks if the target ulcer had decreased but not healed by week 12. Subjects in the S-o-C group whose target ulcers had not healed could switch to SGF treatment after week 12 for 4 weeks, and then for an additional 8 weeks if the target ulcer was responsive to the SGF
treatment. The subjects visited the study centres biweekly during the treatment period and monthly during the follow-up period. Ulcers were photographed and assessed at each visit based on standard wound healing parameters (ulcer size, exudate amount, wound edge, amount of epithelialisation, amount of remaining SGF wound matrix (SGF group), recurrence of ulcer (healed ulcers)). Physical examinations were carried out at baseline, weeks 12 and 24. Adverse events (AE), concomitant therapies and compliance were assessed at each visit, vital signs at each visit during the treatment period and at week 24. The following treatment-related rules to stop the study were applied: repeated necrosectomy (>2×), increase in ulcer size ≥20%, no decrease in ulcer size after 6 weeks of treatment.
Efficacy assessments The primary and secondary performance variables are shown in textbox 1. An ulcer was regarded as healed when the skin was completely re-epithelised without drainage or dressing requirements. The primary performance analysis was based on the intention-to treat (ITT) dataset and performed by two-sided log rank test (significance level ␣ = 5%). It was repeated with a Cox regression analysis checking for centre effects and with the per protocol (PP) analysis set. Kaplan-Meier estimates for the time to healing were calculated for both treatment groups. Performance analyses were in parts repeated by strata to compare corrected vs. uncorrected data (disease duration ≤1 year vs. >1 year). The second performance analyses were based on ITT dataset as well as on PP dataset. They were performed by sample statistics, frequency tables and exploratory tests as appropriate. Pain due to venous leg ulcer between and during dressing changes were assessed by 2 visual analogue scales (VAS). The subjects marked the degree of their pain
Study treatment
S-o-C up to 12 weeks*
SGF + S-o-C up to 12 weeks
Complete healing
Generally responsive but not completely healed
12 week follow-up
Complete healing
SGF + S-o-C up to 12 weeks*
S-o-C or other treatment regimen*
Generally responsive but not completely healed
12 week follow-up
Switch to SGF for 4 weeks
SGF + S-o-C up to 8 weeks if wound size reduced by 50%
S-o-C or other treatment regimen* if not responsive*
*4-week
safety follow-up after last treatment
S-o-C or other treatment regimen*
Figure 1. Overview of the different treatment groups.
212
EJD, vol. 24, n◦ 2, March-April 2014
Box 1. Primary and secondary performance variables Primary performance variable: time to healing of the target ulcer up to the end of the 12 week treatment period (calculated as: date of confirmed healing minus date of baseline visit plus 1 (days)) Secondary performance variables: time to healing of the target ulcer up to week 24 (for subjects who continued treatment after week 12) incidence of complete wound healing of the target ulcer at week 4, 8 and 12 rate of recurrence of the target ulcer at the end of the 12 week follow-up period size of target ulcer at different time points (absolute size and change) ulcer related pain between visits and during dressing change patient benefit index physician’s and subject’s assessment of dressing performance
on a horizontal scale of 100 mm labelled “no pain” at the extreme left and “unbearable pain” at the extreme right. Patient benefit index (PBI) measures patient-defined treatment objectives and benefits, particularly in the course of a treatment. In this study a validated version developed for chronic wounds, consisting of 22 items was used (Version PBI-W22).
subject received 1 SGF patch at baseline and 4 reapplications of smaller patches in total during the entire study (medians). In terms of material size, approximately 11cm2 SGF wound dressing was applied on average to each subject at baseline and 24cm2 in total during the entire study (medians).
Demographic and baseline data The description of the demographic and baseline characteristics of the study population focuses on the ITT data and is shown in table 1. All subjects were Caucasians. The target ulcers of most subjects were located at the inner ankles and affected the epidermis/dermis or subcutis. For the majority, the ulcer’s wound edge was distinct/attached to wound base and the amount of wound exudate was moderate or light. 26 of 60 subjects in the SGF group, 29 of 60 subjects in the S-o-C group and 7 of 19 subjects in the switched to SGF group had a disease duration 0.05).
Time to healing up to week 24 Mean time to healing up to week 24 was 128.73 days for the SGF group (SE ± 4.79) and 114.2 days for the S-o-C group (SE ± 5.44) (p-value = 0.142).
Incidence of complete healing Complete healing of the target ulcers by week 12 was documented for 10 subjects in the SGF group (16.7%) and for 16 subjects in the S-o-C group (26.7%) (p = 0.268), by week 24 it was documented for 15 subjects in the SGF group (25%) and for 21 subjects in the S-o-C group (35%) (p = 0.319) (figure 2). Among the latter, 2 subjects had been switched to SGF after week 12. Stratification by disease duration for different periods showed p-values ≥0.089 (Fisher’s exact test). However, the healing rates in the stratum with an ulcer duration ≤1 year were higher than in subjects with longer disease duration, in particular in the S-o-C group.
Size of target ulcer In both treatment groups the mean size of the target ulcer decreased from baseline to week 12. Stratification by dis-
ease duration and ulcer size by means of Wilcoxon test did not show any significant differences between the treatment groups (figure 3). The ulcer size reduction of the mean target ulcer size observed in subjects switched to SGF was similar to that in the SGF group. There were no marked differences regarding the incidence of complete healing up to week 12 or up to week 24 (Fisher’s exact test p-value ≥0.083 or ≥0.151) stratified by target ulcer size at baseline (20 cm2 ) between the treatment groups. Healing rate and ulcer size appeared inversely related, in particular in the SGF group.
Recurrence of ulcer Recurrence of an ulcer already healed was seen in 1 subject in the SGF group and in 1 subject in the S-o-C group.
Ulcer related pain between visits and during dressing changes In both treatment groups the mean pain between and during dressing changes decreased from baseline to week 12, without indicating any marked differences between the 2 treatment groups when analysed by ANCOVA (p-value ≥0.105). The analysis of the data stratified by disease duration showed that the baseline pain level during dressing changes was lower in subjects with disease duration >1 year than in subjects with disease duration ≤1 year.
Patient benefit index (PBI) PBI data might give a slightly distorted picture due to a relatively high number of missing values (e.g. PBI scores week 12: 10%). However the available data do not indicate any marked differences between treatment groups
60,00%
50,00%
40,00%
30,00%
20,00%
10,00%
0,00% SGF (N = 26)
S-o-C (N = 29)
switched to SGF (N = 7)
SGF (N = 34)
Ulcer present ≤ 1 year
switched to SGF (N = 12)
S-o-C (N = 31)
Ulcer present > 1 year week 12
week 24
Figure 2. Incidence of complete healing – stratified by disease duration (ITT data). From week 14 onwards, subjects who switched from S-o-C to SGF were analyzed in the S-o-C group as well as in the switchedto-SGF.
214
EJD, vol. 24, n◦ 2, March-April 2014
60,00%
50,00%
40,00%
30,00%
20,00%
10,00%
0,00% SGF (N = 25)
S-o-C (N = 29)
switched to SGF (N = 8)
Ulcer size < 10cm2
SGF (N = 24)
S-o-C (N = 21)
switched to SGF (N = 6)
Ulcer size 10-20cm2 week 12
S-o-C (N = 11)
S-o-C (N = 10)
switched to SGF (N = 5)
Ulcer size > 20cm2
week 24
Figure 3. Incidence of complete healing – stratified by ulcer size at baseline (ITT data). From week 14 onwards, subjects who switched from S-o-C to SGF were analyzed in the S-o-C group as well as in the switchedto-SGF.
(p-values ≥0.175). At week 12 the majority of subjects (who answered the respective question) stated that the treatment had helped them “very much” or at least “quite a lot” to reach their baseline goals (see also Demographic and baseline data).
Physician’s and subject’s assessment of dressing performance Both the SGF matrix and the wound dressing used in the S-o-C group were described as easy to apply by physicians and subjects. Both dressings generally met the physicians’ expectations and were considered to be used or recommended again for the treatment of chronic venous leg ulcers. The majority of the subjects in both treatment groups described the progress of healing as good or very good (SGF-group: 88.9%, S-o-C-group: 79.2%) and most subjects were “satisfied” or “very satisfied” with the overall performance of the dressings (SGF group: 95.5%; S-o-C group: 91.7%).
Adverse events (AEs) Device related AEs were observed in 14 subjects in the SGF group and in 18 subjects in the S-o-C group. None of the reported device-related AEs were serious.
Discussion In this study we tested the performance and safety of a novel, inorganic, biodegradable SGF scaffold for the local treatment of subjects with chronic venous leg ulcers. The EJD, vol. 24, n◦ 2, March-April 2014
findings show that the SGF matrix is degraded and absorbed by the wound environment in a biocompatible fashion. No side effects related specifically to the biomaterial were observed. Furthermore, the rate of wound closure in the SGF group versus the S-o-C group was comparable. Thus, the novel developed SGF matrix is well tolerated and biocompatible. The findings in this study also indicate that the SGF matrix itself did not exert specific pro-healing activities. A possible reason for this may be that, particularly in the SGF group, sharp debridement of the ulcers was consequently not performed at each visit. The number of subjects who received sharp debridement in the SGF group compared to the S-o-C group decreased from baseline (SGF group: 34 subjects vs. S-o-C group: 24 subjects) towards the end of the study (SGF group: 7 subjects vs. S-o-C group: 17 subjects). Based on the current findings it is not clear if debridement was less often indicated in the SGF group or if it was not performed for other reasons, e.g. hesitation to remove SGF patches for the procedure. Frequent surgical debridement leads to a better healing rate [14, 15]. Thus, for future studies a consequent and systematic surgical debridement should be part of the study plan. Another explanation for not finding significant differences in the healing rate between the SGF and the S-o-C group might be that the portion of subjects with “hard to heal” ulcers was high compared to other published studies [10, 11, 16]. In fact, the incidence of ulcer healing in the S-o-C group was with 26.7% after 12 weeks and 35% after 24 weeks, lower than results which had been reported in previous studies and which provided the basis for our study design. Gelfand et al., for example, observed a healing rate of chronic venous ulcers of 45.2% after 12 weeks of standard wound therapy with limb compression [11]. Wong et al. observed even higher
215
healing rates of approximately 60% after 12 weeks and approximately 70% after 24 weeks [17]. The subjects included in both of these studies had disease durations between 4 weeks and 1 year, respectively, indicating a mean of 3 months. In the presented study about 50% of the subjects suffered from their ulcers for longer than 1 year. The observed healing rate was higher in subjects with disease duration
