Scientific Director Mesynthes Limited barnaby.may@ mesynthes · 55. T otal Cell Count A verage T...
1
Total Cell Count Average Total Blood Vessel Count per Frame Maximum Compression Load (N) Normalized Maximum Compression Load (N/mm) Yeild Stress (MPa) Maximum Load (N) Normalized Suture Retention Strength (N/mm) Normalized Suture Retention Strength (N/mm) Maximum Load (N) Modulus of Elasticity (Young's) (GPa) Concentration (mg/g) Concentration ( g/g) Barnaby May Ph.D Scientific Director Mesynthes Limited 69 Gracefield Road, Lower Hutt 5040, New Zealand [email protected] +64 21 056 9995 Endoform TM Dermal Template - Authentic Complexity Barnaby C. H. May, Stanley Lun, Sharleen M. Irvine, James N. Fisher, Evan Floden, Brian R. Ward Abstact: We have generated a new biologically-derived extracellular matrix (ECM) termed ‘Endoform’, to meet existing needs within wound care and emerging needs for tissue regeneration sca olds. A novel tissue processing procedure has been developed to render the ECM of ovine forestomachs acellular, while conserv- ing its inherent biological and physical properties. As such, Endoform retains a biologically rich collagen matrix, and importantly, a host of co-factors that are critical for cellular growth and tissue regeneration. An analysis of the microstructure of Endoform con rmed removal of the epithelial and muscle layers, partial re- tention of the basement membrane and the native 3-dimensional collagenaceous matrix. A biochemical comparison of Endoform and the commercial ECM-based product OaSIS™revealed that the two products were grossly similar. However, an important distinction was Endoform’s elevated concentrations of bronection , an important cell adhesion molecule. Endoform was shown to be non-toxic towards mammalian cell lines in vitro, and was well tolerated in a porcine full thickness excisional wound model. In a por- cine model cells penetrated into the Endoform matrix via open micro-architecture, and the matrix underwent a natural process of remodel- ling. Importantly, Endoform induced signi cantly higher levels of cell di erientation and vasculatization relative to OaSIS™ . A biophysical study demonstrated that Endoform had excellent performance characteristics for clinical applications. In some instances, Endoform out- performed leading commercial products. Summary: Endoform is a promising new material for clinical applications in wound care and tissue regeneration. Studies in vitro and in vivo have demonstrated its applicability and out- standing performance characteristics. Endoform’s properties can be attributed to its compli- mentary biochemical, architectural and biophysical properties, termed ‘authentic complexity’, which make this a unique and robust matrix material for immediate clinical applications. Macroscopic Characterization Biochemical Characterization The major and minor biochemical components Ovine Forestomach Endoform OaSIS™ In order to evaluate the in vivotolerance of Endoform and determine its e ectiveness in stimulating tissue regeneration and to undergo remodelling, we conducted a comprehensive wound healing study in pigs. The porcine wound healing model is generally considered to be the best of the animal models with which to study the wound healing process and to evaluate the e ectiveness of clinical or therapeutic interven- tions. Numerous studies have been published describing the use of exogenous extracellular matrix (ECM) products in wound healing studies in rodent and pig models, so this current study was well bench-marked in the literature (Hodde, 2002). female pig (approx. 18-20 kg) using a dermal punch. The wounds were created in four columns of 5 rows, spaced 3 cm apart, according to Figure 16. A total of ve animals were used in the study (100 wound sites). Each of the wounds is either untreated, or treated with sterile Endoform template (termed ‘FM’), 2-ply laminated Endoform (termed ‘LFM’) or the established wound matrix product, OaSIS™ . In each case, a circular piece of wound matrix (20 mm diameter) was applied to the wound and rehydrated in situ by the application of sterile saline. In order to average any positional bias, the location of the four treatments was changed be- tween each animal, such that the no two animals received the same treatment layout. Treated and untreated wounds were dressed identically. On days 0, 3, 7, 14, 28 and 42 all wounds were digitally imaged and the wound area and depth (height, if appropriate) were recorded. Addi- tionally, a single row of wounds from each animal was biopsied. While under anaesthesia the wound, as well as, a portion of normal tissue from the wound margins was surgically excised. All biopsies were formalin xed, mounted, sectioned and stained for analysis. Figure 16: Representativeimageof theporcinewound map used in thestudy. and Remodeling An examination of xed tissue biopsies taken during the course of the study provided evidence that Endoform was in ltrated by cells during the healing process (Figure 18). ECM’s (FM, LFM or OaSIS™ ) ap- peared as green ribbons in Gomori’s trichrome stained sections, and ECM’s were especially prominent the commercial product OaSIS™were visible for ap- proximately 14-28 days, after which time the matrices were fully degraded and mature collagen laid-down in a process of remodelling. No matrix was visible in wounds on day 42. Day 0 Day 3 Day 7 Day14 Day 28 Day 42 Figure 17: Representativeimagesillustrating wound healing over the42day timecourse OaSIS™- Day 7 OaSIS™- Day 14 Biaxial Strength - Ball Burst Strength Ball burst strength provides a measure of a biomaterials resistance to load when biaxial force is applied. As expected there was a dra- matic increase in the strength of the Endoform productsasadditional sheets were laminated to generate the series of multi-ply devices (2-, 3-, and 4-ply) (Figure 21). Uniaxial Strength Uniaxial strength measures the one- dimensional force tolerance of a biomaterial whereby a strip of material is clamped at either end and opposing forces are applied. The maxi- mum load of a sample is dependent on the inher- ent strength of the test material, as well as, the size and thickness of the test sample. We tested single and multi-ply Endoform products and de- termined the maximum load (Figure 23) for each Endoform product, as well as, the maximum tan- gential sti ness (N/mm) (data not shown), elon- gation at failure (mm) (data not shown), the modulus of elasticity (GPa) (Figure 24) and yield stress (MPa) (Figure 25). The yield stress is a term normalized to the dimensions of the test sample and can be used to compare similar products of di erent thicknesses. As expected, increasing the lamination state of the product signi cantly im- proved the uniaxial strength . However, lamina- tion did not statistically alter the maximum elon- gation, or modulus of elasticity. This suggests that the process of lamination increases the strength of the product, but does not alter its pliability, suggesting the handling of 1- and 4-ply products would be similar. 450 400 350 300 250 200 150 100 50 0 1-Ply 2-Ply 3-Ply 4-Ply Endoform Product Figure 21.Ball burst strength of multi-ply Endoform products. Thebiaxial strength of singleor multi-ply Endoform productswasdetermined using a Ball Burst Test according to ASTMD3797-89“Standard Test Method for Bursting Strength of Knitted Goods, Constant-Rate-of-Traverse(CRT) Ball-burst Test”, using an Instron 5800serieselectromechanical tester, tted with a ball-burst compression cage, whereby a 25.4mm stainlesssteel ball waspushed against thetest material at a feed-rateof 305+/-13mm/min, until failure. A1kNload cell wasused to record maximum compression load at failure(N). Error bars represent thestandard error of at least vesamples. 40 35 30 25 20 15 10 5 0 1-ply 2-ply 3-ply 4-ply Endoform Product Figure 23. Comparison of theuniaxial strength of Endoform singleand multi-ply products. A. Maximum load at failure(N); . Modulusof elasticity (Young’s) (GPa). Maximum load at failureof singleand multi-ply productswasdetermined using an Instron 5800serieselectromechanical tester. Variousmaterialswerecut to dog-boneshaped sampleswith a 0.6cm width. Sampleswereclamped with a gaugelength of 7.5cm. and elongated at a rateof 25.4mm/min until failure. Load (N) wasrecorded using a 500Nload cell. Sti nesswascalculated from the slopeof theload (N) versuselongation (mm) curve. Theload versuselongation curvewastransformed to a stress(N/m2) versusstrain curve, using a cross- sectional area of calculated from thethicknessof theproduct. Theslopeof this latter curvewasused to calculatethemodulusof elasticity, or Young’smodulus (GPa) (Figure 24). Error barsrepresent thestandard error of at least vesamples. Figure 24. Comparison of theModulusof elasticity (Young’s) (GPa) of Endoform singleand multi-ply products, callculated according from Figure 9. Error bars represent thestandard error of at least vesamples. To evaluate the suitability of the Endoform product line for implantation we compared the biophysical properties, as derived from uniaxial strength testing, with similar published data relating to the commercially available product Alloderm ™(Table 4). It is apparent from the spread of reported data that the consensus about the true strength, and elastic po- tential of Alloderm™has not been reached in the literature. Reported yield stress of Alloderm™range from 7.00±1.00 to 16.79±2.10 MPa. In order to make a meaningful comparison we calculated the mid-point of the data spread as 11.90 MPa, which is comparable to the yield stress of the 4-ply laminate (11.97±1.16 MPa). The two reported modulus of elasticity for Alloderm™are an order of magnitude di erent from each other making a comparison with the 4-ply product di cult. Table 4. Comparison of uniaxial strength propertiesof 4-ply Endoform and Alloderm™. N.D= no data. Errorsrepresent standard errorsfrom at least ve samples, or published values. Muscle Figure 1: RepresentativeH&E staining of theovineforestomach (left), and Endoform template(right). 10x Magni cation. Themuscle:ECMjunction and epithelium:ECMjunction areindicated by black arrowsin theleft panel. membrane (Figure 2), which would be ex- pected to have important implications to cell adhesion to Endoform. Nuclear staining (Figure 3) demon- strated Endoform was essentially acellular, and retained the complex collagenaceous matrix (Figure 4) Collagen IV Soluble Elastin Insoluble Elastin Total GAG’s Hyaluronic Acid Heparin Sulphate 46.6 ± 1.5 -9.7± 2.1 -6.6± 1.1 -3 112.0 ± 15.0 32.0 36.6 ± 5.7 4.2 54.0 ± 6.1 6.9 30.9 ± 6.4 8.8 27.5 ± 4.5 3.2 41.4 ± 5.0 5.3 3.9± 0.1 1.1 1.5± 0.3 0.2 7.3± 0.4 0.9 1.95 ± 0.02 0.6 0.4± 0.1 0.1 1.58 ± 0.11 0.2 N.D.- N.D.- 2.1± 0.1 -5 whole Endoform had comparable biochemistry to OaSIS™apart from elevated bronectin concentra- tions. Importantly, the basement membrane pro- teins laminin and collagen IVwere present, a nd- ing that parallels IHCstudies (Figure 2). Further studies are being conducted to understand the biological signi cance of these ndings, as well as identify additional matrix components. (m g/g ± SE) Fibronectin 15.30 ± 1.17 13.67 ± 1.64 5.00 ± 0.50 10x 10x Figure 18:Histological examination of Endoform (LFM) and OaSIS™- treated wounds. Woundswerebiopsied on days7, 14, and 42, formalin xed, sectioned and stained with Gomori’strichromestain (red = cytoplasm, green=collagen, blu/black=nuclei). Slides wereimaged at 4x, 10x (days42) and 40x (days7and 14). Higher magni cationsaretaken in thearea of the4x imageoutlined in red. ExogenousECM’sareindicated by black arrows. At days7and 14, exogenousECM’sappear asgreen-stained strati ed sheets. $" $! #" #! " ! Figure 25.Comparison of theyield stressof 1- and 2-ply Endoform productswith commercial dural repair products. Errorsrepresent standard errorsfrom at least vesamples, or published data. Data taken from Table 4. Table 5.Comparison of uniaxial strength propertiesof 1-ply and 2-ply Endoform productswith commercial dural repair products. Errorsrepresent standard errors from at least vesamples, or published values. and 2-ply Endoform products, were compared with published data for the dural repair products DuraGuard™and Durarepair™ (Table 5and Figure 25). The 1- and 2-ply Endoform products had a similar yield stress to Dura-Guard™(13.5±3.34 MPa) and all three products out performed Dura-Derm™(6.27±4.20 MPa). The stron- gest of the products, Durarepair had a yield stress of 22.7±2.83 MPa. The 1- and 2-ply Endoform products (Young’s modulus of 0.04±0.01 and 0.05±0.01 GPa, respectively) had better elastic properties than Dura-Guard™and Durarepair™(0.08±0.02 and 0.07±0.01 MPa, respectively). DuraDerm™demonstrated the best elastic potential (modulus 0.002±0.009 GPa), but as noted above, was the weakest of the products surveyed. DNA 9.0± 0.5 2.6 2.2± 0.5 0.3 2.7± 0.2 0.4 Lipid 50.2 ± 2.9 14.4 59.0 ± 5.2 5.7 44.7 ± 4.0 5.7 Table 1:Major biochemical componentsof theovineforestomach, and thematrices, Endoform and OaSIS™. 1Errorsrepresent standard error from triplicateexperiments. 2Errorsrepresent standard errors from threeindependent production lots, tested in triplicate. 3Percentagecomposition based on total collagen from thehydroxyprolineanalysisonly, encludescollagen III, collagen IV and solublecolla- gen. 4N.D. = Not detected. 5Percentagecomposition excludesheparin sulphateasthisisincluded in total GAG’s. 60 40 20 0 Total Soluble Collagen Collagen III Collagen IV Soluble Elastin Insoluble Elastin Total Sulphated GAGs' Figure 4:Low (left) and high (right) resolution SEMimagesof theEndoform templateshow- ing therelatively smooth luminal surface(left) and thecomplex collagenaceousmatrix under- lying thebasement membrane(right). Figure 3:Representativeimagesof Hoescht 33258stained sectionsof ovineforestomach and Endoform template. Cellsappear asintense blue uorescencewhilethematrix and connectivetissuesappear more di use. 20x Magni cation. Hodde, J. (2002). "Naturally occurring sca olds for soft tissue repair and regeneration." Tissue Eng 8(2): 295-308. Lemer, M. L., D. C. Chaikin, et al. (1999). "Tissue strength analysis of autologous and cadaveric allografts for the pubovaginal sling." Neurourol Urodyn 18(5): 497-503. Morgan, A. S., T. McI , et al. (2004). "Biomechanical properties of materials used in static facial suspension." Arch Facial Plast Surg 6(5): 308-10. Sclafani, A. P., S. A. McCormick, et al. (2002). "Biophysical and microscopic analysis of homologous dermal and fascial materials for facial aesthetic and reconstructive uses." Arch Facial Plast Surg 4(3): 164-71. Vural, E., N. McLaughlin, et al. (2006). "Comparison of biomechanical properties of alloderm and enduragen as static facial sling biomaterials." Laryngoscope 116(3): 394-6. Zerris, V. A., K. S. James, et al. (2007). "Repair of the dura mater with processed collagen devices." J Biomed Mater Res BAppl Biomater 83(2): 580-8. FGF2 1.70 ± 1.38 1.47 ± 0.32 4.85 ± 0.84 Table 2: Minor biochemical componentsof theovine forestomach, Endoform and OaSIS™. 1Errorsrepresent stand- arderror from triplicateexperiments. 2Errorsrepresent stand- ard errorsfrom threeindependent production lots, tested in triplicate. Endoform Increases Cell Di eren- tiation and Vascularization Cell Di erientation: In order to quantify cell proliferation, immunohistochemistry of biopsied tissue was conducted using the cell marker Ki67. Ki67 is a protein that is ex- pressed during all active phases of the cell cycle, and is therefore a useful marker of cell proliferation and cellular activity. Ki67 is not expressed in resting cells. Formalin xed biopsy tissues were mounted, sectioned and active cells were then detected using a Ki67 primary antibody and staining developed using HRP- conjugated secondary antibody and DABvia an automated Bond Max™Automated IHC/ISH Staining System (Leica Microsystems Instruments). Using ImageJ(National Institute of Health) images were processed to quantify the number of Ki67-positive cells per frame (Figure 19). Generally, cell proliferation spiked on days 7 and 14 for all four treatment groups. On both days 7 and 14, Endoform-based treatments had signi cantly greater cell pro- liferation than OaSIS™ -treated and untreated groups (P<0.01 one-way ANOVA using GraphPad Prism). This proliferative phase resolved over time with Ki67-positive cells returning to ‘baseline’ on day 42 in all treatment groups. Vascularization: In order to quantify the extent of endotheliazation and the develop- ment of vasculature in the healing wounds we used immunohistochemistry coupled with digital quanti cation methods. Fixed biopsy tissues were sectioned, mounted and stained with an anti-CD34 antibody at 1:100 dilution and visualised using an HRP-conjugated secondary antibody and DABstaining, before light counterstaining with Mayer’s Haemotoxylin (Figure 20). Tissue biopsies taken at day 3 were excluded from this analysis given the absence of a clearly de ned dermal layer. Stained slides were digitally imaged at 40x magni cation and the number of vessels quanti ed using ImageJ. The average total number of vessels count per frame is given in Figure 20, for each of the four treatment groups over the course of the experiment (day 3 ex- cluded). There was a statistically signi cant increase in the number of blood vessels in wounds treated with either of the Endoform templates, relative to untreated wounds. The increase in total blood vessels relative to untreated wounds was evident on days 14 (P<0.01 FM and LFM), 28 (P<0.01 FM and LFM) and 42 (P<0.01 FM and P<0.05 LFM). In comparison, OaSIS™treatment did not increase the total number of blood vessels relative to the untreated control. Suture Retention Strength Sutute retention testing is a practical clinical consideration that determines the resistance of a biomaterial to suture‘pull-out’. The test protocol is similar to uniaxial strength, however in this instance one edge of the test material is clamped, while the other is secured to the opposing clamp via a suture placed through the test material. Lamination of the Endoform increased the resis- tant of the product to suture pull-out (Figure 26). A comparative analysis of Endoform prducts and the dural repair products Durarepair™and Dura- Guard™is shown in Table 6and Figure 27. Taking into account the relative thicknesses of the four products via normalized suture reten- tion the four products had equivalent potential to resist suture pull-out. 20 18 16 14 12 10 8 6 4 2 0 1-ply 2-ply 3-ply 4-ply Endoform Product Figure 26. Comparison of thesutureretention strength of multi-ply Endoform products. Samplesof multi-ply Endoform productsweretested for sutureretention according to ANSI/AAMI VP20–1994Guidelinesfor Cardiovascular ImplantsVascu- lar ProsthesesMeasured in Newton’s. Suturesweremadein 4cm x 2.5cm samples, using suturewith a bite-depth of 2mm. Load at failurewasrecorded using a Instron 5800serieselectromechanical tester, tted with a 100Nload cell using an advancerateof 100mm/min. Load at failurewasde ned asa 90% reduction in the observed load. Thefreeend of thesamplewasheld in a 25mm vicegrip, whilethe suturewasattached to theopposing clamp via a stainlesshook. Error barsrepre- sent thestandard error of at least six samples. 90 80 70 60 50 40 30 20 10 0 4-Ply OFM Alloderm Strattice Surgisis Material Figure 28.Comparison of thenormalized sutureretention strength of 4-ply Endo- form product and commercially availableimplant matrices. Errorsrepresent standard errorsfrom at least vesamples, or from published data. No error reported for Surgisis™. Alloderm™and Strattice™tested with a bitedepth of 10 mm, 4-ply Endoform and Surgisis™tested with a bitedepth of 2mm. 30 25 20 15 10 5 0 1-Ply Endoform 2-Ply Endoform Dura-Guard Durepair Material Figure27.Comparison of thenormalized sutureretention strength of Endoform productsand dural repair products. Errorsrepresent standard errorsfrom ve independant samples, or from published data. Dura-Guard™ (Zerris, James et al. 2007) Durarepair™ (Zerris, James et al. 2007) 10.02±1.35 0.40±0.01 25.1±3.4 12.38±2.10 0.50±0.02 24.76±5.19 Mesynthes Limited, 69 Grace eld Road, Lower Hutt 5040, New Zealand. Ovine forestomach Endoform Endoform is prepared using a propri- etory processing method to decellularize and delaminate ovine forestomach tissue. Total (m g/g ± SE)1 % composition (m g/g ± SE)2 % composition (m g/g ± SE) 1 % composition 3 of Endoform and OaSIS™were quanti ed using established techniques. Endoform was shown to be a ‘rich’ biological Epithelium Histology was used to demonstrate e ective removal of the epithelial and muscle layers (Figure 1). Immunohistochemistry staining for collagen IVand laminin demonstrated Collagen Soluble Collagen Collagen 142.7 ± 9.7 40.5 821.0. ± 9.0 86.4 629.7 ± 39.7 80.6 55.3 ± 3.21 -71.2 ± 16.3 -48.7 ± 5.3 -3 matrix containing connective, adhesion and signal- ing proteins, all of which play critical roles in cell adhesion, di erientation and proliferation. The matrix was mainly composed of collagen, including types III and IV. Elastin was present in high concen- ECM the existence of a partially intact basement III 61.3 ± 0.7 -196.8 ± 6.9 -171.5 ± 11.1 -3 trations as well as GAG’s and bronectin. On the In Vivo E cacy - Porcine Model of Wound Healing On day 0 of the study, a total of 20 full thickness 20 mm diameter wounds were surgically created on the back of a 6 week old anaesthetised Endoform Persistence Endoform - Day 7 40x 40x at day 7 (indicated by black arrows, Figure 10). Cells Endoform - Day 14 were clearly visible within the exogenous ECM colla- gen sca old at day 7. Both the Endoform matrices and Biophysical Characterization %! 1200 1000 800 600 400 200 0 4-Ply Endoform Alloderm Strattice Surgisis Material Figure 22.Ball burst strength of ECM-based products, normalized to thickness. Data taken from Table3. Errorsrepresent standard errorsfrom at least ve samples, or from published data. No thicknesserror reported for Surgisis™. 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 1-ply 2-ply 3-ply 4-ply Endoform Product We compared the biaxial strength of the 4-ply Endoform laminate with published ball burst data for the commercial products, Alloderm™(LifeCell Corporation), Strattice™(LifeCell Corporation), and Surgisis™(Cook Biotechnology) (Table 3and Figure 22). Using this analysis, we demonstrated that the relative biaxial strengths of the four products were statistically similar. Maximum Compression Load (N±SE) Thickness (mm) Normalized Maximum Compression Load (N/mm±SE) 4-ply Endoform™ 361.5±24.9 0.47±0.01 773.7±68.6 Alloderm™ (Boguszewski, Dyment et al. 2008) Strattice™ (Boguszewski, Dyment et al. 2008) Surgisis™ (package insert) 1781.5±80.2 1.9±0.1 937±106.0 1059.7±181.8 1.49±0.07 711.2±155.4 440±81 0.76 578.9 Modulusof Elasticity (Youngs’) (GPa±SE) Yield Stress (MPa) Thickness (mm±SE) Alloderm™(Lemer, Chaikin et al. 1999) N.D Alloderm™ (Morgan, McI et 16.79±2.10 N.D al. 2004) Alloderm™(Choe, Kothandapani et al. 2001) Alloderm™(Vural, N.D 10.55±2.37 0.9±0.1 N.D 7.20±2.56 1.00±0.05 Table 3. Comparison of ball-burst propertiesthe4-ply Endiform and commer- cial implant products. Errorsrepresent standard errorsfrom at least vesam- ples, or from published data. No thicknesserror reported for Surgisis™. 4-ply Endoform™ 0.06±0.01 11.97±1.16 0.47±0.01 McLaughlin et al. 2006) Alloderm™(Gouk, Lim et al. 2008) Alloderm™ (Sclafani, McCormick et al. 2002) N.D 15.25±7.13 1.34±0.05 0.014±0.0015 7±1 0.45±0.05 0.001±0.002 8.64±3.31 1.89±0.30 The uniaxial strength properties and thicknesses of the 1- Collagen IV Ovine forestomach Endoform 1Ovine Forestomach 2Endoform (m g/g ± SE) 1OaSIS™ (m g/g ± SE) Laminin 6.30 ± 0.24 9.65 ± 4.54 6.00 ± 0.30 Endoform - Day 42 40x OaSIS™- Day 42 40x 1-Ply Endoform 2-Ply Endoform DuraDerm Dura-Guard Durarepair Product Modulusof Elasticity (Youngs’) (GPa±SE) Yield Stress (MPa) Thickness (mm±SE) 1-ply Endoform™ 0.04±0.01 10.15±1.81 0.25±0.01 2-ply Endoform™ 0.05±0.01 9.77±1.68 0.31±0.01 DuraDerm™(Sclafani, McCormick et al. 2002) Dura-Guard™(Zerris, Jameset al. 2007) Durarepair™(Zerris, Jameset al. 2007) 0.002±0.009 6.27±4.20 1.4±0.2 0.08±0.02 13.5±3.3 0.400±0.001 0.07±0.01 22.7±2.8 0.50±0.02 Laminin 220 200 180 160 20 15 10 140 120 100 80 5 0 Fibronectin Laminin FGF2 Endoform OaSIS™ Figure 2:Representativeimmunohistochemistry imagesof collagen IV and laminin stained ovineforestomach and Endoform template. Immunostaining appeared asred-brown. Basement membraneisindicated with black arrows. 10x Magni cation. Ovine forestomach Endoform Matrix Cell nucelii Endoform OaSIS™ Figure 5:Comparison of themajor biochemical componentsof thematrices, Endoform and OaSIS™. Errorsfor Endoform represent the standard error from threeindependent production lots, tested in triplicate. Errorsfor ovineforestomach and OaSIS™represent stand- ard errorsfrom triplicateexperiments. Figure 6: (insert) Comparison of theminor biochemical componentsof thematrices, Endoform and OaSIS™. Errorsfor Endoform represent thestandard error from threeindependent production lots, tested in triplicate. Errorsfor ovineforestomach and OaSIS™represent standard errorsfrom triplicateexperiments. Boguszewski, D. V., N. A. Dyment, et al. (2008). Biomechanical Comparison of Abdominal Wall Hernia Repair Materials. ASME2008 Summer Bioengineering Conference Marriott Resort, Marco Island, Florida, ASME Choe, J. M., R. Kothandapani, et al. (2001). "Autologous, cadaveric, and synthetic materials used in sling surgery: comparative biomechanical analysis." Urology 58(3): 482-6. Gouk, S. S., T. M. Lim, et al. (2008). "Alterations of human acellular tissue matrix by gamma irradiation: histology, biomechanical property, stability, in vitro cell repopulation, and remodeling." J Biomed Mater Res BAppl Biomater 84(1): 205-17. Hyaluronic Acid DNA Total Lipid 1200 1100 1000 900 ** ** ** 70 60 50 ** ** ** ** ** * 800 700 600 500 400 300 ** 40 30 20 10 200 100 0 0 7 14 28 42 3 7 14 28 42 Day FM LFM OaSIS Untreated Figure 19: Quanti cation of cell proliferation during wound healing. (top left). Representativeimageof Ki67-stained tissuesection (40x). (top right). Arepresentative’mask’ of thesameKi67-stained imagegener- ated using ImageJand used to quantify thenumber of Ki67-positvecells. (bottom). Thetotal number of Day FM LFM OaSIS Untreated Figure 20: Quanti cation of blood vesselsin treated woundsduring thetimecourse. (top left). Representa- tiveimageof a CD34-stained tissuesection. (top right). Thesameimagesplit into a HDAB mask using ImageJ. Blood vesselsappear black after deconvolution of theimageand arecounted accordingly. Ki67-positivecellsin three40x framestaken from theepithelial layer and three40x framestaken from the regenerating dermal layer werecounted using IHC and digital methods. Each of thefour treatment groups weresampled at thedaysindicated. **P<0.01signi cancerelativeto untreated control using one-way (bottom). Averagetotal number of blood vesselscounted per frame, analyzed for each tissuebiopsy. Error barsrepresent standard errorsfrom the20biopsiesanalyzed for each treatment group at thetimepoints indicated. **P<0.01and *P<0.05signi cancerelativeto untreated control using one-way ANOVA. ANOVA. Suture Retention - Maximum Load (N±SE) Thickness (mm±SE) Normalized Suture Retention (N/mm) 4-ply Endoform™ 16.0±1.3 0.47±0.01 34.2±2.8 2 Alloderm™ (Boguszewski, Dyment et al. 2008) Strattice™ (Boguszewski, Dyment 135.2±11.1 1.9±0.13 71.2±10.71 10 59.9±2.3 1.49±0.07 40.2±3.43 10 et al. 2008) Surgisis™(Oasis Product Insert) 13.7±3.2 0.76 18.0 2 Table 7.Comparison of thesutureretention strength of Endoform productsand commercially avail- ableimplant matrices. Errorsrepresent standard errorsfrom at least vesamples, or from published data. No error reported for Surgisis™. 1-ply Endoform™ 4.7±0.4 0.25±0.01 18.9±1.5 2-ply Endoform™ 7.1±0.5 0.31±0.01 23.2±1.6 Table 6.Comparison of thesutureretention strength of Endoform productsand commercially availabledural repair matrices. Errorsrepresent thestandard error of vesamples, or published data. The comparative study was extended to determine the performance of Endoform products relative to com- mercially available implant products, Alloderm™ , Strat- tice™and Surgisis™ . The analysis is summarized in Table 7and Figure 28. The 4-ply Endoform laminate product was signi cantly more resistant to suture pull-out than the Surgisis™product (normalized suture retention strengths, 34.2±2.8 and 18.0 N/mm, respectively). The reported suture retention strengths of both Alloderm™ and Strattice™have employed a modi cation to the Bite Depth (mm) ASTM standard, using a bite-depth of 10 mm (Table 5). The normalized suture retention strengths of these products was 71.2±10.71 and 40.2±3.43 N/mm, respec- tively. In comparison, the normalized suture retention strength of the 4-ply Endoform product was 34.2±2.8 N/mm, using a bite-depth of 2 mm. We suspect that the suture retention strength of the 4-ply product would equivalent to both Alloderm™and Strattice™if all three products were tested in side-by-side experiments using equivalent bite-depths. This notion is supported by the observation that the 4-ply Endoform has similar yield stress under uniaxial testing (see Table 5), and that the three products have similar normalized ball-burst strengths (Table 3). Suture Retention - Maximum Load (N±SE) Thickness (mm±SE) Normalized Suture Retention (N/mm)
Transcript of Scientific Director Mesynthes Limited barnaby.may@ mesynthes · 55. T otal Cell Count A verage T...
Tota
l C
ell
Cou
nt
Ave
rage
Tot
al B
lood
Ves
sel C
ount
per
Fra
me
Max
imum
Com
pres
sion
Loa
d (N
)
Nor
mal
ized
Max
imum
Com
pres
sion
Lo
ad (N
/mm
)
Yeild
Stre
ss (M
Pa)
Max
imum
Loa
d (N
)
Nor
mal
ized
Sut
ure
Ret
entio
n S
treng
th (N
/mm
)
Nor
mal
ized
Sut
ure
Ret
entio
n S
treng
th (N
/mm
)
Max
imum
Loa
d (N
)
Mod
ulus
of E
last
icity
(You
ng's
) (G
Pa)
Con
cent
ratio
n (m
g/g)
Con
cent
ratio
n (
g/g)
Barnaby May Ph.DScientific DirectorMesynthes Limited69 Gracefield Road, Lower Hutt 5040, New Zealand
[email protected]+64 21 056 9995
EndoformTM Dermal Template -Authentic Complexity
Barnaby C. H. May, Stanley Lun, Sharleen M. Irvine, James N. Fisher, Evan Floden, Brian R. Ward
Abstact: We have generated a new biologically-derived extracellular matrix (ECM) termed ‘Endoform’, tomeet existing needs within wound care and emerging needs for tissue regeneration sca�olds. A novel tissueprocessing procedure has been developed to render the ECM of ovine forestomachs acellular, while conserv-ing its inherent biological and physical properties. As such, Endoform retains a biologically rich collagenmatrix, and importantly, a host of co-factors that are critical for cellular growth and tissue regeneration. Ananalysis of the microstructure of Endoform con�rmed removal of the epithelial and muscle layers, partial re-tention of the basement membrane and the native 3-dimensional collagenaceous matrix. A biochemical
comparison of Endoform and the commercial ECM-based product OaSIS™ revealed that the two products were grossly similar. However, animportant distinction was Endoform’s elevated concentrations of �bronection , an important cell adhesion molecule. Endoform was shownto be non-toxic towards mammalian cell lines in vitro, and was well tolerated in a porcine full thickness excisional wound model. In a por-cine model cells penetrated into the Endoform matrix via open micro-architecture, and the matrix underwent a natural process of remodel-ling. Importantly, Endoform induced signi�cantly higher levels of cell di�erientation and vasculatization relative to OaSIS™. A biophysicalstudy demonstrated that Endoform had excellent performance characteristics for clinical applications. In some instances, Endoform out-performed leading commercial products.
Summary: Endoform is a promising new material for clinical applications in wound care andtissue regeneration. Studies in vitro and in vivo have demonstrated its applicability and out-standing performance characteristics. Endoform’s properties can be attributed to its compli-mentary biochemical, architectural and biophysical properties, termed ‘authentic complexity’,which make this a unique and robust matrix material for immediate clinical applications.
Macroscopic Characterization Biochemical CharacterizationThe major and minor biochemical components
Ovine Forestomach Endoform OaSIS™
In order to evaluate the in vivo tolerance of Endoform and determine its e�ectiveness in stimulating tissue regeneration and to undergo remodelling, we conducted a comprehensive wound healing study in pigs. The porcine wound healing model is generally considered to be the best of the animal models with which to study the wound healing process and to evaluate the e�ectiveness of clinical or therapeutic interven- tions. Numerous studies have been published describing the use of exogenous extracellular matrix (ECM) products in wound healing studies in rodent and pig models, so this current study was well bench-marked in the literature (Hodde, 2002).
female pig (approx. 18-20 kg) using a dermal punch. The wounds were created in four columns of 5 rows, spaced 3 cm apart, according to Figure16. A total of �ve animals were used in the study (100 wound sites).
Each of the wounds is either untreated, or treated with sterile Endoform template (termed ‘FM’), 2-ply laminated Endoform (termed ‘LFM’) orthe established wound matrix product, OaSIS™. In each case, a circular piece of wound matrix (20 mm diameter) was applied to the wound andrehydrated in situ by the application of sterile saline. In order to average any positional bias, the location of the four treatments was changed be-tween each animal, such that the no two animals received the same treatment layout. Treated and untreated wounds were dressed identically.
On days 0, 3, 7, 14, 28 and 42 all wounds were digitally imaged and the wound area and depth (height, if appropriate) were recorded. Addi-tionally, a single row of wounds from each animal was biopsied. While under anaesthesia the wound, as well as, a portion of normal tissue fromthe wound margins was surgically excised. All biopsies were formalin �xed, mounted, sectioned and stained for analysis.
Figure 16: Representative image of the porcine wound map used in the study.
and RemodelingAn examination of �xed tissue biopsies taken
during the course of the study provided evidence thatEndoform was in�ltrated by cells during the healingprocess (Figure 18). ECM’s (FM, LFM or OaSIS™) ap-peared as green ribbons in Gomori’s trichromestained sections, and ECM’s were especially prominent
the commercial product OaSIS™ were visible for ap-proximately 14-28 days, after which time the matriceswere fully degraded and mature collagen laid-downin a process of remodelling. No matrix was visible inwounds on day 42.
Day 0 Day 3 Day 7
Day 14 Day 28 Day 42
Figure 17: Representative images illustrating wound healing over the 42 day time course
OaSIS™- Day 7
OaSIS™- Day 14
Biaxial Strength - Ball Burst Strength
Ball burst strength provides a measure of a biomaterials resistance to load when biaxial force is applied. As expected there was a dra- matic increase in the strength of the Endoform products as additional sheets were laminated to generate the series of multi-ply devices (2-, 3-, and 4-ply) (Figure 21).
Uniaxial StrengthUniaxial strength measures the one-
dimensional force tolerance of a biomaterial whereby a strip of material is clamped at either end and opposing forces are applied. The maxi- mum load of a sample is dependent on the inher- ent strength of the test material, as well as, the size and thickness of the test sample. We tested single and multi-ply Endoform products and de- termined the maximum load (Figure 23) for each Endoform product, as well as, the maximum tan- gential sti�ness (N/mm) (data not shown), elon- gation at failure (mm) (data not shown), the modulus of elasticity (GPa) (Figure 24) and yield stress (MPa) (Figure 25). The yield stress is a term normalized to the dimensions of the test sample and can be used to compare similar products of di�erent thicknesses. As expected, increasing the lamination state of the product signi�cantly im- proved the uniaxial strength . However, lamina- tion did not statistically alter the maximum elon- gation, or modulus of elasticity. This suggests that the process of lamination increases the strength of the product, but does not alter its pliability, suggesting the handling of 1- and 4-ply products would be similar.
450
400
350
300
250
200
150
100
50
01-Ply 2-Ply 3-Ply 4-Ply
Endoform Product
Figure 21. Ball burst strength of multi-ply Endoform products. The biaxial strength of single or multi-ply Endoform products was determined using a Ball Burst Test according to ASTM D 3797-89 “Standard Test Method for Bursting Strength of Knitted Goods, Constant-Rate-of-Traverse (CRT) Ball-burst Test”,using an Instron 5800 series electromechanical tester, �tted with a ball-burstcompression cage, whereby a 25.4 mm stainless steel ball was pushed againstthe test material at a feed-rate of 305 +/-13 mm/min, until failure. A 1 kN loadcell was used to record maximum compression load at failure (N). Error barsrepresent the standard error of at least �ve samples.
40
35
30
25
20
15
10
5
01-ply 2-ply 3-ply 4-ply
Endoform Product
Figure 23. Comparison of the uniaxial strength of Endoform single and multi-ply products. A. Maximum load at failure (N); . Modulus of elasticity (Young’s) (GPa). Maximum load at failure of single and multi-ply products was determined using an Instron 5800 series electromechanical tester. Various materials were cut to dog-bone shaped samples with a 0.6 cm width. Samples were clamped with a gauge length of 7.5 cm. and elongated at a rate of 25.4 mm/min until failure. Load (N) was recorded using a 500 N load cell. Sti�ness was calculated from the slope of the load (N) versus elongation (mm) curve. The load versus elongation curve was transformed to a stress (N/m2) versus strain curve, using a cross- sectional area of calculated from the thickness of the product. The slope of this latter curve was used to calculate the modulus of elasticity, or Young’s modulus (GPa) (Figure 24). Error bars represent the standard error of at least �ve samples.
Figure 24. Comparison of the Modulus of elasticity (Young’s) (GPa) of Endoform single and multi-ply products, callculated according from Figure 9. Error bars represent the standard error of at least �ve samples.
To evaluate the suitability of the Endoform product linefor implantation we compared the biophysical properties, asderived from uniaxial strength testing, with similar publisheddata relating to the commercially available product Alloderm™ (Table 4). It is apparent from the spread of reported datathat the consensus about the true strength, and elastic po-tential of Alloderm™ has not been reached in the literature.Reported yield stress of Alloderm™ range from 7.00±1.00 to16.79±2.10 MPa. In order to make a meaningful comparisonwe calculated the mid-point of the data spread as 11.90 MPa,which is comparable to the yield stress of the 4-ply laminate(11.97±1.16 MPa). The two reported modulus of elasticity forAlloderm™ are an order of magnitude di�erent from eachother making a comparison with the 4-ply product di�cult.
Table 4. Comparison of uniaxial strength properties of 4-ply Endoform and Alloderm™. N.D = no data. Errors represent standard errors from at least �ve samples, or published values.
MuscleFigure 1: Representative H&E staining of the ovine forestomach (left), and Endoform template (right). 10x Magni�cation. The muscle:ECM junction and epithelium:ECM junction are indicated by black arrows in the left panel.
membrane (Figure 2), which would be ex- pected to have important implications to cell adhesion to Endoform.
Nuclear staining (Figure 3) demon-strated Endoform was essentially acellular,and retained the complex collagenaceousmatrix (Figure 4)
CollagenIV
SolubleElastin
InsolubleElastin
TotalGAG’s
HyaluronicAcid
HeparinSulphate
46.6 ± 1.5 -3 9.7± 2.1 -3 6.6± 1.1 -3
112.0 ± 15.0 32.0 36.6 ± 5.7 4.2 54.0 ± 6.1 6.9
30.9 ± 6.4 8.8 27.5 ± 4.5 3.2 41.4 ± 5.0 5.3
3.9± 0.1 1.1 1.5± 0.3 0.2 7.3± 0.4 0.9
1.95 ± 0.02 0.6 0.4± 0.1 0.1 1.58 ± 0.11 0.2
N.D.4 - N.D.4 - 2.1± 0.1 -5
whole Endoform had comparable biochemistry to OaSIS™ apart from elevated �bronectin concentra- tions.
Importantly, the basement membrane pro-teins laminin and collagen IV were present, a �nd-ing that parallels IHC studies (Figure 2). Furtherstudies are being conducted to understand thebiological signi�cance of these �ndings, as well asidentify additional matrix components.
(m g/g ± SE)
Fibronectin 15.30 ± 1.17 13.67 ± 1.64 5.00 ± 0.50
10x 10xFigure 18: Histological examination of Endoform (LFM) and OaSIS™- treated wounds. Wounds were biopsied on days 7, 14, and 42, formalin �xed, sectioned and stained with Gomori’s trichrome stain (red = cytoplasm, green=collagen, blu/black=nuclei). Slides were imaged at 4x, 10x (days 42) and 40x (days 7 and 14). Higher magni�cations are taken in the area of the 4x image outlined in red. Exogenous ECM’s are indicated by black arrows. At days 7 and 14, exogenous ECM’s appear as green-stained strati�ed sheets.
$"
$!
#"
#!
"
!
Figure 25. Comparison of the yield stress of 1- and 2-ply Endoform products with commercial dural repair products. Errors represent standard errors from at least�ve samples, or published data. Data taken from Table 4.
Table 5. Comparison of uniaxial strength properties of 1-ply and 2-ply Endoformproducts with commercial dural repair products. Errors represent standard errorsfrom at least �ve samples, or published values.
and 2-ply Endoform products, were compared with published data for the dural repair products DuraGuard™ and Durarepair™ (Table 5 and Figure 25). The 1- and 2-ply Endoform products hada similar yield stress to Dura-Guard™ (13.5±3.34 MPa) and all threeproducts out performed Dura-Derm™ (6.27±4.20 MPa). The stron-gest of the products, Durarepair had a yield stress of 22.7±2.83MPa. The 1- and 2-ply Endoform products (Young’s modulus of0.04±0.01 and 0.05±0.01 GPa, respectively) had better elasticproperties than Dura-Guard™ and Durarepair™ (0.08±0.02 and0.07±0.01 MPa, respectively). DuraDerm™ demonstrated the bestelastic potential (modulus 0.002±0.009 GPa), but as noted above,was the weakest of the products surveyed.
DNA 9.0± 0.5 2.6 2.2± 0.5 0.3 2.7± 0.2 0.4
Lipid 50.2 ± 2.9 14.4 59.0 ± 5.2 5.7 44.7 ± 4.0 5.7
Table 1: Major biochemical components of the ovine forestomach, and the matrices, Endoform and OaSIS™. 1Errors represent standard error from triplicate experiments. 2Errors represent standard errors from three independent production lots, tested in triplicate. 3Percentage composition based on total collagen from the hydroxyproline analysis only, encludes collagen III, collagen IV and soluble colla-gen. 4N.D. = Not detected. 5Percentage composition excludes heparin sulphate as this is included intotal GAG’s.
60
40
20
0Total
SolubleCollagen
CollagenIII
CollagenIV
SolubleElastin
InsolubleElastin
TotalSulphated
GAGs'
Figure 4: Low (left) and high (right) resolution SEM images of the Endoform template show-ing the relatively smooth luminal surface (left) and the complex collagenaceous matrix under-lying the basement membrane (right).
Figure 3: Representative images of Hoescht 33258 stained sections ofovine forestomach and Endoform template. Cells appear as intenseblue �uorescence while the matrix and connective tissues appear moredi�use. 20x Magni�cation.
Hodde, J. (2002). "Naturally occurring sca�olds for soft tissue repair and regeneration." Tissue Eng 8(2): 295-308.Lemer, M. L., D. C. Chaikin, et al. (1999). "Tissue strength analysis of autologous and cadaveric allografts for the pubovaginal sling." Neurourol Urodyn 18(5): 497-503.Morgan, A. S., T. McI�, et al. (2004). "Biomechanical properties of materials used in static facial suspension." Arch Facial Plast Surg 6(5): 308-10.Sclafani, A. P., S. A. McCormick, et al. (2002). "Biophysical and microscopic analysis of homologous dermal and fascial materials for facial aesthetic and reconstructive uses." Arch Facial Plast Surg 4(3): 164-71.Vural, E., N. McLaughlin, et al. (2006). "Comparison of biomechanical properties of alloderm and enduragen as static facial sling biomaterials." Laryngoscope 116(3): 394-6.Zerris, V. A., K. S. James, et al. (2007). "Repair of the dura mater with processed collagen devices." J Biomed Mater Res B Appl Biomater 83(2): 580-8.
FGF2 1.70 ± 1.38 1.47 ± 0.32 4.85 ± 0.84
Table 2: Minor biochemical components of the ovine forestomach, Endoform and OaSIS™. 1Errors represent stand- ard error from triplicate experiments. 2Errors represent stand- ard errors from three independent production lots, tested in triplicate.
Endoform Increases Cell Di�eren- tiation and VascularizationCell Di�erientation: In order to quantify cell proliferation, immunohistochemistry of biopsied tissue was conducted using the cell marker Ki67. Ki67 is a protein that is ex- pressed during all active phases of the cell cycle, and is therefore a useful marker of cell proliferation and cellular activity. Ki67 is not expressed in resting cells.
Formalin �xed biopsy tissues were mounted, sectioned and active cells werethen detected using a Ki67 primary antibody and staining developed using HRP-conjugated secondary antibody and DAB via an automated Bond Max™ AutomatedIHC/ISH Staining System (Leica Microsystems Instruments). Using ImageJ (NationalInstitute of Health) images were processed to quantify the number of Ki67-positivecells per frame (Figure 19).
Generally, cell proliferation spiked on days 7 and 14 for all four treatment groups.On both days 7 and 14, Endoform-based treatments had signi�cantly greater cell pro-liferation than OaSIS™-treated and untreated groups (P<0.01 one-way ANOVA usingGraphPad Prism). This proliferative phase resolved over time with Ki67-positive cellsreturning to ‘baseline’ on day 42 in all treatment groups.
Vascularization: In order to quantify the extent of endotheliazation and the develop-ment of vasculature in the healing wounds we used immunohistochemistry coupledwith digital quanti�cation methods. Fixed biopsy tissues were sectioned, mountedand stained with an anti-CD34 antibody at 1:100 dilution and visualised using anHRP-conjugated secondary antibody and DAB staining, before light counterstainingwith Mayer’s Haemotoxylin (Figure 20). Tissue biopsies taken at day 3 were excludedfrom this analysis given the absence of a clearly de�ned dermal layer. Stained slideswere digitally imaged at 40x magni�cation and the number of vessels quanti�edusing ImageJ.
The average total number of vessels count per frame is given in Figure 20, foreach of the four treatment groups over the course of the experiment (day 3 ex-cluded). There was a statistically signi�cant increase in the number of blood vessels inwounds treated with either of the Endoform templates, relative to untreated wounds.The increase in total blood vessels relative to untreated wounds was evident on days14 (P<0.01 FM and LFM), 28 (P<0.01 FM and LFM) and 42 (P<0.01 FM and P<0.05LFM). In comparison, OaSIS™ treatment did not increase the total number of bloodvessels relative to the untreated control.
Suture Retention Strength
Sutute retention testing is a practical clinicalconsideration that determines the resistance of a biomaterial to suture ‘pull-out’. The test protocol is similar to uniaxial strength, however in this instance one edge of the test material is clamped, while the other is secured to the opposing clamp via a suture placed through the test material. Lamination of the Endoform increased the resis- tant of the product to suture pull-out (Figure 26). A comparative analysis of Endoform prducts and the dural repair products Durarepair™ and Dura- Guard™ is shown in Table 6 and Figure 27. Taking into account the relative thicknesses of the four products via normalized suture reten- tion the four products had equivalent potential to resist suture pull-out.
20181614121086420
1-ply 2-ply 3-ply 4-plyEndoform Product
Figure 26. Comparison of the suture retention strength of multi-ply Endoform products. Samples of multi-ply Endoform products were tested for suture retention according to ANSI/AAMI VP20–1994 Guidelines for Cardiovascular Implants Vascu- lar Prostheses Measured in Newton’s. Sutures were made in 4 cm x 2.5 cm samples, using suture with a bite-depth of 2 mm. Load at failure was recorded using a Instron 5800 series electromechanical tester, �tted with a 100 N load cell using an advance rate of 100 mm/min. Load at failure was de�ned as a 90% reduction in the observed load. The free end of the sample was held in a 25 mm vice grip, while the suture was attached to the opposing clamp via a stainless hook. Error bars repre- sent the standard error of at least six samples.
90
80
70
60
50
40
30
20
10
04-Ply OFM Alloderm Strattice Surgisis
Material
Figure 28. Comparison of the normalized suture retention strength of 4-ply Endo- form product and commercially available implant matrices. Errors represent standard errors from at least �ve samples, or from published data. No error reported for Surgisis™. Alloderm™ and Strattice™ tested with a bite depth of 10 mm, 4-ply Endoform and Surgisis™ tested with a bite depth of 2 mm.
30
25
20
15
10
5
01-Ply Endoform 2-Ply Endoform Dura-Guard Durepair
Material
Figure 27. Comparison of the normalized suture retention strength of Endoform products and dural repair products. Errors represent standard errors from �ve independant samples, or from published data.
Dura-Guard™(Zerris, James et al.2007)Durarepair™(Zerris, James et al.2007)
10.02±1.35 0.40±0.01 25.1±3.4
12.38±2.10 0.50±0.02 24.76±5.19
Mesynthes Limited, 69 Grace�eld Road, Lower Hutt 5040, New Zealand.
Ovine forestomach EndoformEndoform is prepared using a propri-
etory processing method to decellularizeand delaminate ovine forestomach tissue.
Total
(m g/g ± SE)1 %composition
(m g/g ± SE)2 %composition
(m g/g ± SE)1
%composition 3
of Endoform and OaSIS™ were quanti�ed usingestablished techniques.
Endoform was shown to be a ‘rich’ biological
Epithelium Histology was used to demonstrate e�ectiveremoval of the epithelial and muscle layers(Figure 1). Immunohistochemistry stainingfor collagen IV and laminin demonstrated
Collagen
SolubleCollagen
Collagen
142.7 ± 9.7 40.5 821.0. ± 9.0 86.4 629.7 ± 39.7 80.6
55.3 ± 3.21 -3 71.2 ± 16.3 -3 48.7 ± 5.3 -3
matrix containing connective, adhesion and signal-ing proteins, all of which play critical roles in celladhesion, di�erientation and proliferation. Thematrix was mainly composed of collagen, includingtypes III and IV. Elastin was present in high concen-ECM the existence of a partially intact basement
III61.3 ± 0.7 -3 196.8 ± 6.9 -3 171.5 ± 11.1 -3
trations as well as GAG’s and �bronectin. On the
In Vivo E�cacy - Porcine Model of Wound Healing
On day 0 of the study, a total of 20 full thickness 20 mm diameter wounds were surgically created on the back of a 6 week old anaesthetised
Endoform Persistence Endoform - Day 7
40x 40x
at day 7 (indicated by black arrows, Figure 10). Cells Endoform - Day 14were clearly visible within the exogenous ECM colla-gen sca�old at day 7. Both the Endoform matrices and
Biophysical Characterization
%!
1200
1000
800
600
400
200
04-Ply Endoform Alloderm Strattice Surgisis
Material
Figure 22. Ball burst strength of ECM-based products, normalized to thickness.Data taken from Table 3. Errors represent standard errors from at least �vesamples, or from published data. No thickness error reported for Surgisis™.
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.001-ply 2-ply 3-ply 4-ply
Endoform Product
We compared the biaxial strength of the 4-ply Endoform laminate with published ball burst data for the commercial products,Alloderm™ (LifeCell Corporation), Strattice™ (LifeCell Corporation), and Surgisis™ (Cook Biotechnology) (Table 3 and Figure 22). Usingthis analysis, we demonstrated that the relative biaxial strengths of the four products were statistically similar.
MaximumCompressionLoad (N±SE)
Thickness (mm)
NormalizedMaximum
CompressionLoad
(N/mm±SE)4-ply Endoform™ 361.5±24.9 0.47±0.01 773.7±68.6Alloderm™(Boguszewski,Dyment et al.2008)Strattice™(Boguszewski,Dyment et al.2008)Surgisis™(package insert)
1781.5±80.2 1.9±0.1 937±106.0
1059.7±181.8 1.49±0.07 711.2±155.4
440±81 0.76 578.9
Modulus ofElasticity(Youngs’)(GPa±SE)
Yield Stress(MPa)
Thickness(mm±SE)
Alloderm™(Lemer,Chaikin et al. 1999) N.D
Alloderm™(Morgan, McI� et
16.79±2.10 N.D
al. 2004)Alloderm™ (Choe,Kothandapani etal. 2001)Alloderm™ (Vural,
N.D 10.55±2.37 0.9±0.1
N.D 7.20±2.56 1.00±0.05
Table 3. Comparison of ball-burst properties the 4-ply Endiform and commer-cial implant products. Errors represent standard errors from at least �ve sam-ples, or from published data. No thickness error reported for Surgisis™.
4-ply Endoform™ 0.06±0.01 11.97±1.16 0.47±0.01
McLaughlin et al.2006)Alloderm™ (Gouk,Lim et al. 2008)Alloderm™(Sclafani,McCormick et al.2002)
N.D 15.25±7.13 1.34±0.05
0.014±0.0015 7±1 0.45±0.05
0.001±0.002 8.64±3.31 1.89±0.30
The uniaxial strength properties and thicknesses of the 1-
Collagen IV
Ovine forestomach Endoform1Ovine
Forestomach2Endoform(m g/g ± SE)
1OaSIS™(m g/g ± SE)
Laminin 6.30 ± 0.24 9.65 ± 4.54 6.00 ± 0.30
Endoform - Day 42
40x
OaSIS™- Day 42
40x
1-PlyEndoform
2-PlyEndoform
DuraDerm Dura-Guard Durarepair
Product
Modulus ofElasticity (Youngs’)
(GPa±SE)Yield Stress
(MPa)Thickness(mm±SE)
1-ply Endoform™ 0.04±0.01 10.15±1.81 0.25±0.012-ply Endoform™ 0.05±0.01 9.77±1.68 0.31±0.01DuraDerm™(Sclafani,McCormick et al. 2002)Dura-Guard™ (Zerris,James et al. 2007)Durarepair™ (Zerris,James et al. 2007)
0.002±0.009 6.27±4.20 1.4±0.2
0.08±0.02 13.5±3.3 0.400±0.001
0.07±0.01 22.7±2.8 0.50±0.02
Laminin
220
200
180
160
20
15
10
140
120
100
80
5
0Fibronectin Laminin FGF2
Endoform OaSIS™
Figure 2: Representative immunohistochemistry images of collagen IV and laminin stained ovine forestomach and Endoformtemplate. Immunostaining appeared as red-brown. Basement membrane is indicated with black arrows. 10x Magni�cation.
Ovine forestomach Endoform
Matrix
Cell nucelii
Endoform OaSIS™
Figure 5: Comparison of the major biochemical components of the matrices, Endoform and OaSIS™. Errors for Endoform represent thestandard error from three independent production lots, tested in triplicate. Errors for ovine forestomach and OaSIS™ represent stand-ard errors from triplicate experiments. Figure 6: (insert) Comparison of the minor biochemical components of the matrices, Endoformand OaSIS™. Errors for Endoform represent the standard error from three independent production lots, tested in triplicate. Errors forovine forestomach and OaSIS™ represent standard errors from triplicate experiments.
Boguszewski, D. V., N. A. Dyment, et al. (2008). Biomechanical Comparison of Abdominal Wall Hernia Repair Materials. ASME 2008 Summer Bioengineering Conference Marriott Resort, Marco Island, Florida, ASMEChoe, J. M., R. Kothandapani, et al. (2001). "Autologous, cadaveric, and synthetic materials used in sling surgery: comparative biomechanical analysis." Urology 58(3): 482-6.Gouk, S. S., T. M. Lim, et al. (2008). "Alterations of human acellular tissue matrix by gamma irradiation: histology, biomechanical property, stability, in vitro cell repopulation, and remodeling." J Biomed Mater Res B Appl Biomater 84(1): 205-17.
HyaluronicAcid
DNA Total Lipid
1200
1100
1000
900
** **
**
70
60
50
** **
**** **
*800
700
600
500
400
300
**40
30
20
10200
100
00
7 14 28 423 7 14 28 42
DayFM LFM OaSIS Untreated
Figure 19: Quanti�cation of cell proliferation during wound healing. (top left). Representative image ofKi67-stained tissue section (40x). (top right). A representative ’mask’ of the same Ki67-stained image gener-ated using ImageJ and used to quantify the number of Ki67-positve cells. (bottom). The total number of
Day
FM LFM OaSIS Untreated
Figure 20: Quanti�cation of blood vessels in treated wounds during the time course. (top left). Representa-tive image of a CD34-stained tissue section. (top right). The same image split into a HDAB mask usingImageJ. Blood vessels appear black after deconvolution of the image and are counted accordingly.
Ki67-positive cells in three 40x frames taken from the epithelial layer and three 40 x frames taken from theregenerating dermal layer were counted using IHC and digital methods. Each of the four treatment groupswere sampled at the days indicated. **P<0.01 signi�cance relative to untreated control using one-way
(bottom). Average total number of blood vessels counted per frame, analyzed for each tissue biopsy. Errorbars represent standard errors from the 20 biopsies analyzed for each treatment group at the time pointsindicated. **P<0.01 and *P<0.05 signi�cance relative to untreated control using one-way ANOVA.
ANOVA.
SutureRetention- MaximumLoad (N±SE)
Thickness(mm±SE)
NormalizedSuture
Retention(N/mm)
4-ply Endoform™ 16.0±1.3 0.47±0.01 34.2±2.8 2Alloderm™(Boguszewski, Dymentet al. 2008)Strattice™(Boguszewski, Dyment
135.2±11.1 1.9±0.13 71.2±10.71 10
59.9±2.3 1.49±0.07 40.2±3.43 10et al. 2008)Surgisis™ (OasisProduct Insert) 13.7±3.2 0.76 18.0 2
Table 7. Comparison of the suture retention strength of Endoform products and commercially avail-able implant matrices. Errors represent standard errors from at least �ve samples, or from publisheddata. No error reported for Surgisis™.
1-ply Endoform™ 4.7±0.4 0.25±0.01 18.9±1.52-ply Endoform™ 7.1±0.5 0.31±0.01 23.2±1.6
Table 6. Comparison of the suture retention strength of Endoform products andcommercially available dural repair matrices. Errors represent the standard errorof �ve samples, or published data.
The comparative study was extended to determinethe performance of Endoform products relative to com-mercially available implant products, Alloderm™, Strat-tice™ and Surgisis™. The analysis is summarized in Table7 and Figure 28. The 4-ply Endoform laminate productwas signi�cantly more resistant to suture pull-out thanthe Surgisis™ product (normalized suture retentionstrengths, 34.2±2.8 and 18.0 N/mm, respectively). Thereported suture retention strengths of both Alloderm™and Strattice™ have employed a modi�cation to the
Bite Depth(mm)
ASTM standard, using a bite-depth of 10 mm (Table 5).The normalized suture retention strengths of theseproducts was 71.2±10.71 and 40.2±3.43 N/mm, respec-tively. In comparison, the normalized suture retentionstrength of the 4-ply Endoform product was 34.2±2.8N/mm, using a bite-depth of 2 mm. We suspect that thesuture retention strength of the 4-ply product wouldequivalent to both Alloderm™ and Strattice™ if all threeproducts were tested in side-by-side experiments usingequivalent bite-depths. This notion is supported by theobservation that the 4-ply Endoform has similar yieldstress under uniaxial testing (see Table 5), and that thethree products have similar normalized ball-burststrengths (Table 3).
SutureRetention- MaximumLoad (N±SE)
Thickness(mm±SE)
NormalizedSuture
Retention(N/mm)
