AU781196B2 - Method of producing a prosthetic device - Google Patents

Description

AUSTRALIA

Patents Act 1990 COMMONWEALTH SCIENTIFIC AND INDUSTRIAL RESEARCH ORGANISATION

ORIGINAL

COMPLETE SPECIFICATION STANDARD PATENT Invention Title: Method of producing a prosthetic device The following statement is a full description of this invention including the best method of performing it known to us:a METHOD OF PRODUCING A PROSTHETIC DEVICE Field of the Invention: This invention relates to prosthetic devices, particularly for use in vascular surgery, and to methods of producing same.

Background of the Invention: Conventional synthetic polymer vascular prostheses, typically comprising porous polyester, have been used successfully for a number of vascular procedures. The main disadvantage with these prostheses is their porosity, which results in bleeding at the time of implantation. It is therefore usually necessary to firstly coat or "seal" the prostheses so as to reduce the porosity. Originally, a process of "pre-clotting" the prosthesis was used. This involved soaking the prosthesis in the patient's own blood and was perceived 15 by some as a time consuming and cumbersome process that could increase the risk of infection.

A number of materials including albumin, collagen, gelatine, fibrin and alginate have been used as an alternate biological compatible component to pre-coat woven or knitted polyester to produce what has been colloquially referred to as biohybrids. The protein coating is intended to be a temporary cover or "sealant" that resorbs with time to generate a highly porous prosthesis that allows good tissue ingrowth. In addition, these coatings provide a means of surface passivation that can prevent early thrombogenic responses. Biohybrid vascular prostheses have therefore become widely popular due to their ease of handling and non-requirement for any preclotting prior to surgery. In addition these grafts can be readily used in clinical conditions that would preclude the use of pre-clotting in heparinised patients and those undergoing ruptured aneurysm resection).

Examples of commercially successful biohybrid vascular prostheses include the widely-used Hemashield graft (Meadox Medicals, New Jersey, USA), and the Hemaguard graft (InterVascular, Florida, USA), and the Tuscon graft (Tascon Medical Technology Corp., Irvine, CA, USA) which is similar to the Hemaguard graft. All of these use bovine Type I collagen as the coating. While many clinical studies using these prostheses have largely confirmed their safety, it has been recognised that they are prone to thrombosis. Further, there are concerns over the possibility of causing platelet activation and adverse immunogenic responses through the use of collagen as the sealant.

In recent times, considerable attention has been paid to the possibility of producing vascular prostheses in vivo by implantation of a polymer or silicone "mandrel", for example a silicone tube, usually covered by a polymer reinforcing mesh, into a suitable host such as a dog or sheep to produce a socalled fibrocollagenous tube (FCT). The technique relies upon the provocation in the host of a foreign body fibrous tissue response a wound healing response) resulting in encasement of the mandrel in collagen.

The mesh-containing FCTs may be regarded as true biosynthetic vascular prostheses integrating natural collagen with the synthetic polymer mesh and are distinct from the abovementioned synthetic grafts coated with collagen as sealants.

Currently, there is only one biosynthetic vascular prosthesis in use 15 clinically. This is the Omniflow Vascular ProsthesisTM (OVP) (BioNova International, North Melbourne, VIC, Australia), a glutaraldehyde-stabilised xenogenic FCT that is manufactured in sheep using a polyester covered silicone mandrel.' 3 After various stages of development, involving various mesh designs and stabilisation strategies, OVP I and, more recently, OVP II S 20 have been commercialised. This strategy of using a xenogenic host, a S* polyester reinforced mandrel support and an effective stabilisation process, has generated a device that has proved to be effective for peripheral arterial replacement. The use of glutaraldehyde stabilisation of the collagen minimises the extent of cross-species immunological reactivity to allow "off the shelf' transplantation into humans. The processing also allows the collagenous matrix to persist for long implantation times and provides a very durable and patent prosthesis.

4 For use as a small diameter 3-4 mm) vascular replacement for coronary artery replacement), the OVP has yet to be fully tested. The major concern here would be to generate a luminal surface that had good blood flow characteristics and was a deterrent for platelet activation and thrombosis, but which was favorable for endothelialisation. The present applicants have identified modifications to the manufacturing methods used for making re-inforced fibrocollagenous tubes which result in the production of biosynthetic vascular prostheses with inner surfaces that are less undulating and more uniformly covered with collagen. This effectively LII VU VV ~L ~V V~ V ~VVU UV~~ U -V- 3 protects the prostheses against potential deleterious polyester breakthrough of the inner surface and, as shown herein, the prostheses produce 100% patency in animal trials. The modifications identified by the present applicants thereby offer the possibility of producing,' for the first time, biosynthetic vascular prostheses which are suitable for small diameter vessel replacement, such as is needed for coronary applications.

Summary of the Invention: In a first aspect, the present invention provides a method for the production of a prosthetic device, said method comprising the steps of; preparing a structure comprised of a mandrel in supporting association with a mesh, said mesh comprising synthetic fibres coated and/or incorporating a substance which promotes collagenous tissue growth, the promotion of growth by said substance being other than growth promotion as 15 a result of any structural feature of the substance (ii) implanting said structure in a host animal at a site suitable for collagenous tissue growth thereon, (iii) allowing growth of collagenous tissue to occur on said structure until the mesh is, at least, wholly embedded within collagenous tissue, (iv) removing the structure with attached collagenous tissue from the host animal, and thereafter, and in either order, treating the collagenous tissue embedding said mesh with an agent to bring about collagen fibril cross-linking so as to increase the durability of said collagenous tissue, and 25 (vi) removing the mandrel from the collagenous tissue embedding said mesh.

In a second aspect, the present invention provides a prosthetic device Sproduced in accordance with the method of the first aspect.

Detailed disclosure of the Invention: The method of the invention provides a prosthetic device in the form of a conduit of a collagen mesh composite. The composite may include some cells connective or fibroblastic tissue) remnant from the collagenous tissue growth on the mandrel-mesh structure. The inner or lumen surface presents as a smooth surface with no mesh fibre breakthrough, making the 21/03 2005 MON 12:13 [TX/RX NO 6264] a005 m prosthetic devices suitable for use in coronary vascular applications where impedance of blood flow is to be avoided, 21/03 2005 MON 12:13 CTX/RX NO 6264] Z 0O06 The mandrel utilised in the method of the invention is typically in the form of a rod or tube and may be comprised of any inert synthetic material such as silicone or polyurethane. By varying the size of the mandrel, the internal diameter of the prosthetic devices may be varied to any suitable size for vascular applications <10 mm, including 6-8 mm for peripheral vascular applications, and 3-4 mm for coronary vascular applications).

The mesh utilised in the method of the invention is preferably comprised of all inert synthetic fibre such as polyester or polyurethane. It is preferred that the synthetic fibre mesh be woven or otherwise produced from yarn comprising 15 to 35 filaments. Although yarns comprising other numbers of filaments may also be used, it is believed that yarns comprising a filament number in this range can be treated to achieve a loading with the said substance sufficient for promoting collagenous tissue growth. The pattern of the mesh can be any of those known in the art. Several suitable 15 mesh patterns are described in Australian Patent No. 692365, the disclosure of which is to be regarded as incorporated herein by reference.

The substance for promoting collagenous tissue growth is preferably selected from cell growth promoters TGFP), cell adhesion proteins (e.g.

fibronectin) or ligands for cell adhesion proteins collagen), agents which 20 inhibit fibrin deposition heparin), and mixtures thereof. Particularly preferred, however, are heparin, collagen and mixtures thereof. Suitable collagen includes any of Types I to III, the interstitial collagens) as well as synthetic collagen having a primary structure based on the (Gly-X-Y).

repeating unit characteristic of collagen.

In the mandrel-mesh structure, the mesh is preferably fitted over the mandrel and secured at both ends, for example by knots. Before or following fitting of the mesh, the mesh is loaded with a solution or suspension of the substance for promoting collagenous tissue growth. To achieve high levels of loading through coating and/or incorporation by filling of interstices), the process involves an evacuation to first remove any air trapped within interstices. Preferably, the loading process involves; immersing the mesh in a wetting agent 1% Triton X-100), (ii) evacuating the wetted mesh to remove any air trapped in interstices, (iii) washing the evacuated mesh to remove excess wetting agent, and (iv) immersing the washed mesh in a solution or suspension of a substance which promotes collagenous tissue growth.

Where the substance to be coated and/or incorporated into the mesh is collagen, step (iv) is preferably performed with a 0.5 to 5 mg/ml solution of collagen ovine skin type I collagen). For heparin, step (iv) is preferably performed with a solution of 3000-10000 u/ml, more preferably 5000 u/ml, heparin. For gelatine, step (iv) is preferably performed with a solution of 5 to solution of gelatine.

Preferably, the mesh is coated and/or incorporates a substance which promotes collagenous tissue growth at an amount of less than about 1% by weight based upon the weight of said mesh. More preferably, the amount is in the range of 0.05 to 0.5% by weight based upon the weight of said mesh.

Most preferably, the amount is in the range of 0.1 to 0.2% by weight based upon tile weight of said mesh.

15 The mandrel-mesh structures are preferably implanted in an animal host dog or sheep) under the cutaneous trunci muscles for periods of about 5 to 20 weeks, more preferably 10 to 15 weeks.

The step of treating the collagenous tissue embedding the mesh so as to cause cross-linking of collagen fibrils, can be according to any of the 20 methods known in the art chemical treatments with aldehydes, epoxides, isocyanates and carbodiimides, and physical treatments such as photo-oxidation). The most preferred cross-linking treatment is treatment with glutaraldehyde 2 glutaraldehyde solution).

As mentioned above, the method of the invention provides a prosthetic device product in the form of a conduit of a collagen mesh composite. This conduit form can, however, be cut length-wise or flattened to produce a sheet which can be suitable for use as a hernia patch 5 or tape for orthopaedic applications, or a flat strip which can be suitable as a replacement for tendons or ligaments (particularly, the anterior cruciate ligament (ACL) which is the most commonly damaged knee ligament).

Transmission electron microscopy performed upon an example of a flat strip of this kind, has shown that the collagen fibres present are highly aligned along the lengthwise axis of the flat strip and thus ideally organised to withstand the mechanical stresses to which the flat strip will be exposed as a tendon or ligament prosthetic device.

It will be appreciated by persons skilled in the art, that sheets of a collagen mesh composite might also be produced by using a plate rather than a rod or tube-shaped mandrel in supporting association with a synthetic fibre mesh. It is therefore to be understood that the present invention extends to the use of mesh supports having shapes other than that of a tube or rod.

The terms "comprise", "comprises" and "comprising" as used throughout the specification are intended to refer to the inclusion of a stated step, component or feature or group of steps, components or features with or without the inclusion of a further step, component or feature or group of steps, components or features.

The invention will hereinafter be further described by way of the following, non-limiting examples and accompanying figures.

Brief description of the accompanying Figures: 15 Figure 1. Histology using H&E staining of initial test samples using different mesh modifications: No treatment gelatine treatment collagen treatment and heparin treatment (Bars 100 un.) Figure 2. SEM examination of initial test samples using different mesh S 20 modifications: No treatment gelatine treatment collagen treatment and heparin treatment (Bars 500 tm.) Figure 3. TEM examination of initial test samples using different mesh treatments showing collagen in the region immediately next to the polyester mesh: No treatment heparin treatment and collagen treatment (Bars 1 ipm.) Figure 4. Angiograms of vascular prostheses made using collagentreated or heparin-treated mesh at 6 months implantation showing no signs of occlusion or dilatatory changes.

Figure 5. Gross examination of early and late explants: 1 month explant of a prosthesis made using a collagen-treated mesh, showing smooth thrombus-free inner lining, and 6 month explant of a prosthesis made using a collagen-treated mesh, again showing a smooth thrombus-free inner surface and good tissue integration on the outside.

Figure 6. Haemocompatibility of central explants assessed for the presence of endothelial cells by von Willebrand factor (vWf) immunofluorescent staining. 1 month explants of prostheses made using heparin-treated or collagen-treated mesh, showing continuous diffuse vWf staining. For comparison, 1 month explants of unmodified OVPII showing no staining at the central region and only patchy staining at the proximal anastomosis The inner blood flow surface appears on the right in all photomicrographs. Positive staining for endothelial cells appears as white on a black background. (Bars 100 pm.) Figure 7. Tissue integration of central explants assessed for the presence of new canine collagen type III using MvAb 1E7-D7/COL III immunofluorescent staining: 1 month explant of a prosthesis made using a collagen-treated mesh showing only capsular staining, 1 month explant of a prosthesis made using a heparin-treated mesh showing patchy deposits of new collagen type III, (C D) 3 and 6 month explants of prostheses, made using collagen-treated mesh, with new collagen type III throughout the vessel wall. The inner blood flow surface appears on the right in tie upper left corner in D) and the lower left corner in Positive staining for endothelial cells appears as white on a black background. (Bars 100 pin.) Figure 8. One month explant of a prosthesis made using a heparintreated mesh, showing minimal granulocyte infiltration. Staining with napthol AS-D chloroacetate, specific for granulocytes. The inner blood flow surface appears at the top and polyester fibre bundles are visible beneath.

(Bar 100 pim.) Transmission electron microscopy of a 6 month explant of a prosthesis made using a collagen-treated mesh showing well aligned collagen and no foreign body giant cells. A polyester filament appears in the bottom right corner. (Bar 100 pmn.) Example 1: Effects of mesh treatment on the structure of mandrel-grown biosynthetic vascular prostheses.

Materials and Methods: Conduit preparation Collagen-polymer composite conduits were made as previously described."" Briefly, 600 mm long mandrels, comprising 6 mm OD silicone tubing covered with modified polyester mesh, were implanted under the cutaneous trunci muscles of sheep. For initial trials, each mandrel was covered with a mesh made up from 4 individual 150 mm sections sewn together to give a single unit. The differently treated meshes that were used were rotated so that each occupied different positions along the mandrels.

Four samples of each treated mesh were implanted. After about 12 weeks, the mandrels and the new collagenous capsule that surrounded the mesh were removed from the sheep and were trimmed of excess tissue. In initial trials these samples were examined without further processing. In subsequent trials full length, 600 mm, mandrels were used. In these cases, the explanted tissue was sent to BioNova International (North Melbourne, VIC, Australia) for further processing by a glutaraldehyde-based stabilisation process that is used for manufacture of OmniflowTM Vascular Prostheses.' Where more than one mandrel was implanted in a single sheep, those that included the dexamethazone treatment were not implanted along with non-dexamethazone treated samples to avoid any systemic effects.

Biochemical treatments of the mesh 15 For mesh treatments, mesh samples were secured on the silicone S. tubing mandrels. The mesh, which comprises 27 filaments per yarn, as previously described, 4 was immersed at room temperature (RT) in 1% Triton X-100 for 10 min, and then, while still immersed, evacuated using a water pump to remove air trapped within the individual filaments of the yarn. .The S 20 wet mesh was washed in 3 changes of sterile water and then immersed in the .*test solution. After 3 hours at RT, the mesh covered mandrel was removed and air dried. Mandrels were sterilized by ethylene oxide at 55 0 C prior to implant. The treatments for the mesh were: No treatment; Pre-wetted in 1% Triton X-100 and rinsed, but no further treatment; 1 mg/ml ovine skin type I collagen in 10 mM sodium phosphate, 150 mM NaC1, pH 7.4 (PBS); 1 mg/ml heat denatured (50 0 C for 10 min) ovine skin type I collagen; 10% bovine B type gelatine, 170 Bloom, in water (Leiner Davis Gelatine, Sydney); 20% sheep serum in PBS; 10 mg/ml bovine serum albumin (BSA) in PBS (Sigma, St. Louis); 5,000 u/ml dexamethazone; and 5,000 u/ml heparin.

The treatments were selected to provide a range of potential biological responses. For example, the collagen and gelatine samples could potentially provide cell recruitment sites or could act as coagulants to limit fibrin deposition. On the other hand, heparin could reduce the fibrin deposition by its anticoagulant activity, and could also act as a ligand for cell adhesion proteins. Serum treatments could potentially minimise the inflammatory response to the polyester mesh, while the dexamethazone could provide short-term, local anti-inflammatory effects.

The amount of ovine skin type I collagen loaded onto the mesh by the treatment described above was determined by amino acid analysis after hydrolysis in 6M HCl at 105 0 C for 20 hours. The amount determined was about 0.16% by weight based upon the weight of the mesh.

Effectiveness of mesh treatment The effectiveness of mesh treatment was assessed using a model S° 15 protein. Mesh was coated and dried, as above, with 1 mg/ml FITC-labelled bovine serum albumin. The coated dry mesh was then immersed and gently agitated in PBS at 20 0 C, and samples examined at intervals from 2 hours up to 24 hours by fluorescence microscopy using a BioRad Lasersharp 500 Confocal imaging system, to determine the presence of remaining fluorescent FITC-protein.

Examination of evaluation samples For histological examination, samples from at least 6 points for each test treatment were cut and fixed in neutral buffered formalin and embedded in paraffin. Sections (7 pm) were cut and stained with haematoxylin and eosin or picro-sirius red, and were examined for the extent and uniformity of collagen cover of mesh. Measurements of depth of mesh cover by collagen were taken from a calibrated video image with 300 x magnification.

For scanning electron microscopy (SEM) examination, neutral buffered formalin fixed samples were washed in PBS, post fixed in 1%0 OsO for 1 hour, briefly rinsed in water and then taken through a graded series of aqueous ethanol solutions to dry ethanol. For drying, samples were infiltrated in a 50:50 mixture of dry ethanol:Peldri II (Ted Pella Inc., Redding, CA, USA) and then in 2 changes of 100% Peldri II. The Peldri II was allowed to solidify and was then sublimed off using vacuum. Samples were then sputter coated with gold and examined using a JEOL JSM T20 microscope.

For transmission electron microscopy (TEM) examination, neutral buffered formalin fixed tissue samples were further fixed in 2% (w/v) paraformaldehyde, 2% glutaraldehyde in PBS overnight and then washed in three changes of PBS. Samples were then post-fixed in 1% OsO, in PBS for 1 hour, followed by rinsing in distilled water. After washing in EtOH, they were then taken through a graded series of EtOH to dry EtOH, followed by an ascending resin/EtOH series into 100% Spurr's resin.

The resin was then cured for 3 days at 60°C. Sections were cut using a Reichart OM U3 ultramicrotome, stained in 3% uranyl acetate in EtOH for 20 minutes and in Reynold's lead citrate for 20 minutes, and then examined with a JEOL 100B microscope.' 0 Results cud Discussion: The method chosen for treating the mesh was effective, as judged by the slow rate of release of FITC-labelled BSA in test samples. The labelled 15 protein was found within the spaces between the filaments that make up the .polymer filament for the mesh, and was still present, albeit in reduced amount after 8 hours of washing.

The primary screen of potential treatments for the mesh showed that different treatments led to variations in the thickness of the collagenous tissue coverage of the polymer mesh (Figure Histology measurements enabled the average thickness of the collagen coverage of the polymer mesh to be estimated for each type of treatment (Table These data showed that for the mesh treated with the ovine type I collagen, there was a significant increase in depth of collagen cover compared to the control samples with no treatment of the mesh. The trends observed for the other mesh treatments (Table 1) suggested that the heparin treatment may also be leading to an increase in the depth of collagen cover of the mesh. There was little difference between the no treatment control and the detergent washed mesh, suggesting that the other changes reflect the specific modifications used in the treatment of the mesh, rather than the effects of the detergent treatment alone. In all cases, large errors were observed for the average depth of collagen coverage, due to variations in the extent of cover around the circumference of the tube within single sections, as previously observed.

0 Qualitative SEM data confirmed the histological data. For the samples with collagen and heparin treatments very low levels of breakthrough were observed (Figure This suggested that these treatments may provide additional collagen cover of the mesh. The dehydration process used in preparation of the SEM samples can make the tissue covering the mesh more fragile and lead to shrinkage. These effects can lead to additional breakthrough of the mesh that may not be present in fresh samples.

Although this is caused by the processing method, it has the value of indicating those samples that would be of greatest risk for breakthrough during processing and use of the vascular prosthesis. Nevertheless, for the mesh treated with collagen, only a limited number of mesh breakthroughs were observed (Figure 1C), comparable or slightly fewer than seen for the untreated control sample (Figure 1A), while for the mesh treated with heparin extremely few cases of breakthrough were observed (Figure 1D). For the other samples, for example, the mesh treated with gelatine, breakthrough was frequently observed (Figure 1B). Since for most samples the measured collagen thickness (Table 1) was apparently similar to the untreated control, 15 it is possible that the tissue architecture in the collagen layer that covers the mesh may vary between the different treatments. A further, noticeable feature of the samples where the mesh had been treated with heparin was that the additional collagen cover of the mesh led to a less undulating flow surface. This smoother flow surface was also observed, although to a lesser extent, for the sample where the mesh had been treated with collagen, but was not observed for other treatments.

Based on these preliminary data, further conduits (600 mm) were prepared using mesh treated with either heparin, collagen (the best treatments from the preliminary experiments), or dexamethazone, along with a no treatment control sample. After explant from the sheep, the tissue conduits were subjected to the abovementioned glutaraldehyde-based stabilisation process that is used for the manufacture of OmniflowTM Vascular Prostheses.

The improvement observed from the collagen and heparin treatments, based on the measurement of the mean collagen coverage of the mesh, was supported by histological evaluation of the extent of breakthrough (Table 2).

In these data, the number of breakthrough points seen by histology will be greater than would be present on the intact tube as the sectioning can fragment the sample. These data on the extent of mesh breakthrough were confirmed by SEM, which again showed the intact nature of the collagen cover for the samples where the mesh had been treated with collagen or heparin.

The tissue architecture of the conduit wall for the samples where the mesh was collagen or heparin treated, and of the control sample, was examined by TEM. These data (Figure 3) indicated that the collagen in the samples where the mesh had been treated was organized in a similar fashion to that found in the untreated control sample. Thus, in all cases the collagen of the main part of the wall of the vessel was highly organized and aligned parallel to the axis of the conduit, with only a limited number of cells being evident. The collagen bundles surrounding tihe polymer mesh were much finer, probably reflecting a high type III content,' and were interspersed through the individual filaments of the mesh fibers. More cells were observed, generally close to the mesh, than were seen in the outer layer of the device. Tihe inner surface of the conduits, which had been in contact with the silicone niandrel, consisted of loose connective tissue with a lower collagen content, and with remnants of a cellular layer close to the surface.

There was no clear variation observed between the 3 samples, with the observed structural characteristics consistent with those previously reported for OmniflowTM Vascular Prosthesis." Samples from each of these same 3 mesh treatments were also examined to see if there were differences in the distribution of the cells associated with the tissue that formed the conduit. The substantial majority of tile cells present were fibroblastic in nature; few, if any inflammatory cells were observed. Thus, in all cases, there were fewer cells in the outer wall of the conduit. An increased nunmber of cells were observed at the depth of the polymer mesh layer, although the sample fields examined did not include the polymer fiber bundles. A noticeable increase in the number of cells observed at the inside wall of the device, which had originally been in contact with the silicone mandrel, was also observed, consistent with TEM studies.

Conclusion: The process used in making biosynthetic vascular prostheses with mandrel-mesh structures is sufficiently flexible to allow modification and hence changes in properties of the material. Improvement in the extent of collagen coverage leading to diminished levels of mesh breakthrough and the provision of a less undulated blood flow surface, was achieved by treating the mesh with collagen or heparin.

Table 1. Thickness of collagenous tissue covering the polyester mesh for different pre-treatmeiits of the mesh.

Treatment Collagen thickness (jim) (mean SEM) No treatment 27.1 3.8 Pre-wetted and rinsed 32 7 5.8 Ovine type I collagen 49.3.. 7.7 Denatured ovine type I collagen 33.1 7.6 15 Gelatine 38.2 7.3 Sheep serum 43.7 7.8 Bovine serum albumin 38.7 8.9 Dexamethazone 33.8 6.7 H-eparin 45.5 7.1 Unpaired Student's t-test was p~erformled on each set of data relative to no treatment standard Significant difference be tween mocan s, p<0.05 (t=2.44, df=I11). *Nunibe rs ini brackets represent total numiber of cross-sectional samples examined from 4 replicate miaterials; the cross-sectional thlickness of each sammple was averaged froin jindividlual replicate counts ranging from 18-30 icaS UreHemIs aroun md t11 lepolyester bundles.

Table 2. Thickness of collagenous tissue covering the polyester mesh for selected treatments of the mesh prior to implantation of the mandrel.

Treatment Collagen thickness (pum) Percent breakthrough* mean SEM mean SEM No treatment 35.5 8.3 9.6 6.9 Ovine type I collagen 66.6 16.6 3.0 1.2 Dexamethazone 42.9 5.1 10.5 3.6 Heparin 85.4" 9.0 1.6 1.1 Unpaired Student's t-lest was performed on each set of data relative to no treatment standard significant difference between means, p<0.05 (t=3.67, df=9).Numibers in brackets are as in Table 1.

Percent breakthrough defines tie degree of polyester mesh which has perturbed the inner surface of *o lithe vessel. Each sample representing replicates as shown in brackets was examined over the entire cross-sectional lumen and included on average between 18 to 30 polyester bundles for each section.

Example 2: In vivo evaluation of mesh treated mandrel-grown biosynthetic vascular prostheses.

Materials and Methods: Manufacture of Modified Mandrel Grown Vascular Prostheses Mandrel-grown collagen-polymer composite conduit prostheses were manufactured as described in example 1.

Two separate mesh treatments were made as described in Example 1.

Samples of mesh-covered mandrels were pre-soaked in 1% Triton X-100, washed thoroughly and then treated with either 1 mg/ml ovine collagen type I (Col-M) or 50,000 units heparin (Hep-M). All mandrels were 600 mm long and the nominal internal diameter was 6 mm.

Prior to implantation into dogs, the collagen-polyester composite conduit prostheses were stored in 50% ethanol.

Animal Model The animal model used for evaluation of the mandrel-grown collagenpolymer composite conduit prostheses has been described in detail previously for the assessment of the OVP I and OVP II.

4 7 A total of 12 dogs were used and one implant (either Col-M or Hep-M) was performed on any one animal. Each adult greyhound dog weighing 20 to 30 kg was implanted with a 100 to 120 mm long prosthesis as a single aorto-left-external iliac bypass graft with ligation of the terminal aorta. The anastomoses were mm long end-to-side using a continuous suture of 6-0 polypropylene. No anticoagulant or anti platelet agents were administered during the study.

Angiograms were performed to assess patency levels. Explants of either type of prosthesis (Col-M or Hep-M) were removed at 1, 3 and 6 months (two dogs for each time point) and assessed for thrombus formation, aneurysmal or dilatory changes, and patency. For conventional haematoxylin and eosin 15 staining and immunohistological assessment, the prostheses were perfused in situ at explant with saline and then removed for storage at -20 0

C.

Antibodies Used in Explant Analysis .9 Two areas were investigated in this study and the results were compared with data obtained from OVP I or OVP II described previously.

4 7.8 S 20 These were the inner blood contacting surface, which was assessed for the degree of endothelialisation using a rabbit anti-von Willebrand factor (RavWf) antibody (Dako Corporation, Ca), and the degree of tissue augmentation, which was assessed using a monoclonal antibody (MAb) specifically reactive with dog collagen but not ovine collagen type III (1E7- D7/COL III)." The reactivity of both these antibodies have been shown in previous work to be specific for von Willebrand factor and dog collagen type

III.

Imnunohistological and Histological Assessment For immunohistological analyses with the antibodies described, 6-8 pm sections of frozen tissue were cut using a freezing microtome and reacted with 1/200 dilution of RavWf antibody or undiluted culture supernatant of MAb 1E7-D7/COL III as described previously.

7 Reactivity was visualised using affinity-purified, fluorescein isothiocyanate-conjugated, sheep anti-rabbit or anti-mouse antibodies (Silenus Laboratories, Melbourne, VIC, Australia) diluted 1/50 in PBS. Sections were examined using a Leitz Ortholux II microscope with fluorescence observed with a Bio-Rad Lasersharp lMRC-500 confocal fluorescence imaging system. As controls in all studies, primary antibodies against vWf and collagen type III were omitted and replicate sections were stained with only the secondary fluorescein isothiocyanate-labelled antibodies.

Conventional H&E staining was performed on samples of all explants to qualitatively assess the degree of wound healing and tissue reactivity.

Neutrophils and macrophages were specifically detected using enzyme histochemistry to target cellular esterases. Unfixed, frozen sections were incubated with either napthol AS-D chloroacetate (specific for granulocytes) or a-napthyl acetate (specific for macrophages and histiocytes) in the presence of a stabilised diazonium salt, according to the suppliers instructions (Sigma Diagnostics Kits 91A and C, Sigma, USA).

Samples of explants from all time points were obtained from proximal, 5. central and distal regions of the prostheses. For the histological, 15 immunohistological and cellular evaluation studies, only data from central regions are reported for direct comparison of performance with known data from OmniflowTM Vascular Prostheses.7 12 Results: 20 Animal Patency **All prostheses were assessed for function and patency by regular palpation of the femoral pulses and by angiograms during the period of implantation. At 6 months, all prostheses were 100% patent; angiograms of prostheses made using both ovine collagen type I-treated mesh (Col-M) or heparin-treated mesh (Hep-M) showed a fully functional blood vessel with no signs of obstruction or aneurysmal deterioration (Figure Gross examination of explants of Col-M or Hep-M prostheses showed an even, smooth, shiny inner blood surface which was unobstructed and thrombusfree (Figure There was considerable tissue integration on the outer adventitial zone, and this increased with time (Figure Imnunohistological Explant Analysis Two biological components were evaluated in the explants, using the described antibody probes, as indicators for the degree of haemocompatibility of the blood surface and the degree of new tissue augmentation within the vessel wall.

Neo-intimal blood suiface: No substantial differences were observed between the Col-M or Hep-M prostheses. At 1 month, samples from the central regions of both types of explants showed near continuous, albeit diffuse, staining for von Willebrand factor (Figure 6, A In certain areas, patchy staining could also be seen near the adventitial side of the vessel wall, indicative of early neovascularisation in the implants. Staining of the inner intimal layer was occasionally a little patchy in some samples from the Col- M explants. In addition, at this early time point, the staining was not indicative of a single layer of cells, but more representative of a dynamic process of cellular growth. This diffuse staining disappeared with time, presumably as the newly forming intimal matrix and migrating endothelial cells stabilised. Nonetheless, the degree of this early type of endothelialisation in both types of prostheses was good. For comparative purposes, samples of 1 month explants of unmodified OVP II from a similar 15 animal study are also shown in Figure 6. Unlike the modified prostheses, no antibody staining indicative of endothelialisation is apparent in the central regions of these explants (Figure 6C). In these explants, only patchy endothelialisation was found at the proximal anastomotic regions of the prostheses (Figure 6D). At 3 and 6 months, all explants from both types of 20 modified prostheses showed uniform and consistent monolayer staining similar to that observed with standard OVPII explants at similar time points.

Vessel wall: For either Col-M or Hep-M prostheses, tissue augmentation of the vessel wall was barely evident at 1 month after implantation. In general, at this early time point, dog collagen type III could be detected only in the capsular tissue which formed around the outside of the implant (Figure 7A). At 1 month, all samples from Col-M explants showed complete absence of infiltrating new collagen type III, while the occasional sample from Hep-M explants showed limited patchy deposition of new dog collagen type III (Figure 7B). At 3 months there was good evidence for tissue augmentation in both explant types, with substantial new collagen type III in most explants (Figure 7C); by 6 months this was complete with uniform detection of new dog collagen type III throughout the entire vessel wall including in and around the PET mesh (Figure 7D). Collagen type VI and type I were also detected in these prostheses (data not shown). Deposition of collagen type VI was similar to that of type III, while collagen type I was detected later. No fibrous capsule formation was evident around either prosthesis, indicative of a sound wound healing response.

Histological and Quantitative Assessment Examination of samples of both types of explants showed cell infiltration that varied with time. At 1 month each of the Col-M or Hep-M explants were similar with marked cellular infiltration. The degree of cellular infiltration had decreased at 3 months to a similar extent in both explant types; at 6 months there was a substantial reduction in total cell numbers which was particularly more pronounced in the Hep-M explants compared with the Col-M explants. By and large, cellular infiltration occurred from the adventitia of the prosthesis and, by 3 months, the degree of infiltration was uniform throughout the vessel wall. By 6 months, the density of cells had dropped substantially although the distribution was fairly uniform. The large majority of the infiltrating cells were fibroblasts 15 that did not stain with the esterase histochemical stains. At 1 month, which shows the most cell numbers, these comprised a mixture of new canine as well as pre-existing ovine fibroblasts. Granulocytes and macrophages were rare relative to fibroblasts. At 1 month, there was a modest scattering of macrophages and granulocytes. Macrophages from both types of explants were present patchily throughout the entire vessel, and the localization within the vessel varied between different samples. Granulocytes were even less evident, but those present were localised either along the inner luminal surface and occasionally within some regions of the media of Hep-M explants (Figure 8a) or, in the case of the Col-M explants, were confined to the polyester fibre bundles. At 3 months, the distribution of granulocytes and macrophages was similar to the 1 month time point, although the numbers were even fewer. At 6 months, there was only the occasional macrophage detected with the esterase histochemical stains.

Using TEM, the tissue response at this stage reflected a very mature rate of healing in both types of explants. Consistent with the quantitative histology and collagen immunohistochemical analyses, the vessel walls were low in cell numbers and displayed excellent collagen structure (Figure 8B), indicative of a mature granulation healing response, not of a fibrotic response. This acellular pattern was seen even adjacent to the polyester fibre bundles where there was no sign of foreign body giant cells (Figure 8B).

Discussion: Grafted blood vessels, like all implants, evoke a foreign body reaction arising from the adventitial surface that can either mature into a tissue capsule or facilitate tissue integration. In the case of both modified vascular prostheses, the tissue response exemplified one of a good wound healing response with no fibrosis or tissue encapsulation. Examination of the vessel wall of both prostheses showed a rapid cellular response and good tissue augmentation with new host collagen type III. The rate of deposition of collagen was indeed similar to that seen with explants of the OVP 12.13 as well as with other collagen-based materials used in hernia replacement.

5 In these studies, type III collagen and, in particular, type VI collagen, was found to be a very early marker of tissue integration and augmentation, associated with a good wound healing response to the implant. Inflammatory cells, mainly macrophages with some granulocytes were present at 1 month, but 15 disappeared rapidly with the healing response. The majority of cells infiltrating through the wall from the outer adventitia were fibroblasts, and the rate and numbers of cells correlated well with the deposition of new collagen throughout the wall.

In previous evaluations of the performance of OVP I and OVP II, examination of the central regions of the prostheses was chosen to make an accurate judgement on the structural durability of the vessel wall and also the extent of endothelial cell coverage away from the anastomotic regions.

Mid-sectional tissue of both prostheses, explanted at 3 and 6 months, revealed a single layer of endothelial cells identified by staining for von Willebrand factor. Indeed, although the sample size was low for both modified grafts, endothelialisation in the central region was detected by 1 month, albeit that the diffuse pattern of staining reflected a more complex, atypical picture of cell migration and growth. Explants of standard unmodified OVP II generally showed no signs of any endothelial coverage at this early stage. In fact, at 1 month in the OVP, endothelial cells are only seen patchily at the anastomotic regions, and again the immunohistological picture depicts a complex cellular process. Von Willebrand factor staining seen along the intimal lining in these modified prostheses at the early time point appeared as a layered structure. The staining detects a product of endothelial cells and it is possible at this early time point that endothelial cells have tried to adhere and failed. At this early time point in these limited samples, the staining pattern could be reflecting signs of a pseudo-intima rather than a true neo-intima formation. With time this diffuse staining disappears and all prostheses were free of thrombus formation and patent.

The mechanism and nature of the endothelial cell coverage on these biological-synthetic composite prostheses is unclear, like most synthetic counterparts. It is probable that such cell coverage in the central region occurs by ingrowth from the anastomotic regions, although this appears to be very rapid. In addition, continuous von Willebrand staining is not always seen. The findings suggest that the newly developed modified surface allows for a rapid endothelialisation from transmural ingrowth as a consequence of neovascularisation in the vessel wall. Adhesion and proliferation of endothelial cells to a vascular surface depends on the nature of the extracellular matrix (ECM). With these biological prostheses there is substantial ECM which allows for good endothelial cell growth. Where no 15 ECM is present as is the case with synthetic prostheses, the surfaces have been modified with a variety of biological components including fibronectin, collagens, laminin, naturally-produced ECM and combinations of synthetic substrates to promote endothelial cell coverage.

20 Conclusion: The use of mandrel-mesh structures wherein the mesh has been treated with either collagen or heparin treatment allows for the production of biosynthetic vascular prostheses that are excellent blood vessels for 5 to 6 mm replacements, with 100% patency at 6 months in the standard dog model used for assessment of vascular performance. The mesh treatments appear not to have had any deleterious effects on the overall performance of the prostheses and, like OVP II, there were no signs of thrombosis or aneurysms.

Further, the prostheses provide a uniform flat collagen-rich surface that should be an ideal substrate for seeding endothelial cells that are required for a totally non-thrombotic surface.

References: 1. V. Ketharanathan, United States of America Patent No. 4,319,363.

2. V. Ketharanathan, and B.A. Christie, "Glutaraldehyde-tanned ovine collagen conduits as vascular xenografts in dogs," Arch. Surg., 115, 967-969 (1980).

3. B. Christie, V. Ketharanathan, and L.J. Perloff, "Patency rates of minute vascular replacements. The glutaraldehyde modified mandrel grown conduits," J. Surg. Res., 28, 519-532 (1980).

4. J.A. Werkmeister, V. Glattauer, T.A. Tebb, J.A.M. Ramshaw, G.A.

Edwards, and G. Roberts, "Structural stability of long-term implants of a collagen-based vascular prosthesis," J. Long-term Effects Med. Implants, 1, 107-119 (1991).

J.A. Werkmeister, G.A. Edwards, F. Casagranda, J.F. White, and J.A.M.

S 15 Ramshaw, "Evaluation of a collagen-based biosynthetic material for the repair of abdominal wall defects", J. Biomed. Mater. Res., 39, 429-436, 1998.

6. G.A. Edwards and G. Roberts, "Development of an ovine collagenbased composite vascular prosthesis," Clinical Materials, 9, 211-223 (1992).

7. J.A. Werkmeister, J.F. White, and J.A.M. Ramshaw, "Evaluation of the Omniflow collagen-polymer vascular prosthesis," Med. Prog. Technol., 231-242 (1994).

8. J.A.M. Ramshaw, J.A. Werkmeister, and G.A. Edwards, "Tissuepolymer composite vascular prostheses," in Encyclopedic Handbook of Biomaterials and Bioengineering, Wise, Trantolo, Altobelli, D.E., Yaszemski, Gresser, and Schwartz, Marcel Dekker Inc., pp 953-978, (1995).

9. J.A.M. Ramshaw, D.E. Peters, J.A. Werkmeister, and V. Ketharanathan, "Collagen organization in mandrel-grown vascular grafts", J. Biomed.

Materials Res., 23, 649-660, (1989).

10. J.F. White, J.A. Werkmeister, G.A. Edwards, and J.A.M. Ramshaw, "Structural analysis of a collagen-polyester composite vascular prosthesis," Clinical Materials, 14, 271-276 (1993).

11. J.A. Werkmeister, and J.A.M. Ramshaw, "Multiple antigenic determinants on type III collagen," Biochem 274, 895-898 (1991).

12. J.A. Werkmeister, J.F. White, G.A. Edwards, and J.A.M. Ramshaw, "Early performance appraisal of the Omniflow II Vascular Prosthesis as an 22 indicator of long-term function," J. Long-term Effects Med. Implants, 5, 1-10 (1995).

13. J.A. Werklneister, T.A. Tebb, J.F. White, and J.A.M. Ramshaw, "Monoclonal antibodies to type VI collagen demonstrate new tissue augmentation of a collagen-based biomaterial implant," J. Histochem.

Cytochem., 41, 1701-1706 (1993).

*e 23 It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

e* *a

Claims (33)

1. A method for the production of a prosthetic device, said method comprising the steps of; preparing a structure comprised of a mandrel in supporting association with a mesh, said mesh comprising synthetic fibres coated and/or incorporating a substance which promotes collagenous tissue growth, the promotion of growth by said substance being other than growth promotion as a result of any structural feature of the substance, (ii) implanting said structure in a host animal at a site suitable for collagenous tissue growth thereon, (iii) allowing growth of collagenous tissue to occur on said structure until the mesh is, at least, wholly embedded within collagenous tissue, (iv) removing the structure with attached collagenous tissue from *the host animal, and thereafter, and in either order, treating the collagenous tissue embedding said mesh with an S: agent to bring about collagen fibril cross-linking so as to increase durability of said collagenous tissue, and (vi) removing the mandrel from the collagenous tissue embedding Ssaid mesh.

2. A method according to claim 1, wherein said prosthetic device is in the form of a conduit of a collagen mesh composite defining a lumen, said lumen 25 having a smooth surface with no mesh fibre breakthrough. 0

3. A method according to claim 2, wherein the diameter of the lumen is less than about 10 mm.

4. A method according to claim 3, wherein the diameter of the lumen is about 6-8 mm, A method according to claim 3, wherein the diameter of the lumen is less than about 4 mm. 21/03 2005 MON 12:13 [TX/RX NO 6264] 007

6. A method according to any one of the preceding claims, wherein the mesh is a synthetic fibre mesh produced from yarn comprising 15-35 filaments.

7. A method according to any one of the preceding claims, wherein the substance for promoting collagenous tissue growth is selected from cell growth promoters, cell adhesion proteins, ligands for cell adhesion proteins, and agents which inhibit fibrin deposition, and mixtures thereof.

8. A method according to claim 7, wherein the substance for promoting collagenous tissue growth is selected from heparin, collagen and mixtures thereof.

9. A method according to any one of the preceding claims, wherein the 15 mesh is coated and/or incorporates a substance which promotes collagenous *tissue growth at an amount of less than about 1.0 by weight based upon the weight of said mesh.

10. A method according to claim 9, wherein the mesh is coated and/or S 20 incorporates a substance which promotes collagenous tissue growth at an amount in the range of 0.05 to 0.5% by weight based upon the weight of said mesh.

11. A method according to claim 9, wherein the mesh is coated and/or 25 incorporates a substance which promotes collagenous tissue growth at an amount in the range of 0.1 to 0.2% by weight based upon the weight of said mesh.

12. A method according to any one of the preceding claims, wherein step (ii) consists of implanting said structure in a host animal under the cutaneous trunci muscles for about 5-20 weeks.

13. A method according to claim 12, wherein step (ii) consists of implanting said structure in a host animal under the cutaneous trunci muscles for about 10-15 weeks. 21/03 2005 MON 12:13 [TX/RX NO 6284] @008

14. A method according to any one of the preceding claims, wherein step consists of treating the collagenous tissue with 2% glutaraldehyde solution. 15, A method according to claim 1 further comprising the step of; (vii) cutting or flattening the collagenous tissue embedding said mesh to produce a prosthetic device in the form of a sheet.

16. A method according to claim 15, wherein said prosthetic device has smooth surfaces with no mesh fibre breakthrough,

17. A method according claim 15 or 16, wherein the mesh is a synthetic fibre mesh produced from yarn comprising 15-35 filaments. S 15 18. A method according to any one of preceding claims 15 to 17, wherein the substance for promoting collagenous tissue growth is selected from cell growth promoters, cell adhesion proteins, ligands for cell adhesion proteins, and agents which inhibit fibrin deposition, and mixtures thereof.

19. A method according to claim 18, wherein the substance for promoting collagenous tissue growth is selected from heparin, collagen and mixtures thereof. A method according to any one of claims 15 to 19, wherein the mesh is 25 coated and/or incorporates a substance which promotes collagenous tissue growth at an amount of less than about 1.0 by weight based upon the weight of said mesh.

21. A method according to claim 20, wherein the mesh is coated and/or incorporates a substance which promotes collagenous tissue growth at an amount in the range of 0.05 to 0.5% by weight based upon the weight of said mesh.

22. A method according to claim 20, wherein the mesh is coated and/or incorporates a substance which promotes collagenous tissue growth at an 21/03 2005 MON 12:13 [TX/RX NO 6264] a009 4IlY. VQ L.xlv -AA VI V oUU. sVOU nAX Mi o w f-lV V 27 amount in the range of 0.1 to 0.2% by weight based upon the weight of said mesh.

23. A method according to any one of claims 15 to 22, wherein step (ii) consists of implanting said structure in a host animal under the cutaneous trunci muscles for about 5-20 weeks.

24. A method according to claim 23, wherein step (ii) consists of implanting said structure in a host animal under the cutaneous trunci muscles for about 10-15 weeks. A method according to any one of claims 15 to 24, wherein step (v) consists of treating the collagenous tissue with 2% glutaraldehyde solution.

26. A method according to claim 1 further comprising the step of; (vii) cutting or flattening the collagenous tissue embedding said Smesh to produce a prosthetic device in the form of a flat strip.

27. A method according to claim 26, wherein said prosthetic device has smooth surfaces with no mesh fibre breakthrough.

28. A method according claim 26 or 27, wherein the mesh is a synthetic fibre mesh produced from yarn comprising 15-35 filaments.

29. A method according to any one of preceding claims 26 to 28, wherein the substance for promoting collagenous tissue growth is selected from cell growth promoters, cell adhesion proteins, ligands for cell adhesion proteins, and agents which inhibit fibrin deposition, and mixtures thereof. A method according to claim 29, wherein the substance for promoting collagenous tissue growth is selected from heparin, collagen and mixtures thereof.

31. A method according to any one of claims 26 to 30, wherein the mesh is coated and/or incorporates a substance which promotes collagenous tissue 21/03 2005 MON 12:13 [TX/RX NO 6264] @010 LIIUO VJ L&.Xv FZLAA 0± .3 000. UOO r0 a I L L *frlv J J 28 growth at an amount of less than about 1.0 by weight based upon the weight of said mesh.

32. A method according to claim 31, wherein the mesh is coated and/or incorporates a substance which promotes collagenous tissue growth at an amount in the range of 0.05 to 0.5% by weight based upon the weight of said mesh.

33. A method according to claim 31, wherein the mesh is coated and/or incorporates a substance which promotes collagenous tissue growth at an amount in the range of 0.1 to 0.2% by weight based upon the weight of said mesh.

34. A method according to any one of claims 26 to 33, wherein step (ii) 15 consists of implanting said structure in a host animal under the cutaneous trunci muscles for about 5-20 weeks,

35. A method according to claim 34, wherein step [ii) consists of Simplanting said structure in a host animal under the cutaneous trunci muscles for about 10-15 weeks.

36. A method according to any one of claims 26 to 35, wherein step (v) consists of treating the collagenous tissue with 2% glutaraldehyde Ssolution.

37. A prosthetic device produced in accordance with the method of any S.one of the preceding claims.

38. A prosthetic device produced in accordance with the method of any one of claims 1 to 14, wherein the prosthetic device is suitable for use in vascular replacement.

39. A prosthetic device produced in accordance with the method of any one of claims 15 to 25, wherein the prosthetic device is suitable for use as a hernia patch. 21/03 2005 MON 12:13 [TX/RX NO 6264] l 011 ,It VV rZAt. V4 V CUUJ QVV 29 A prosthetic device produced in accordance with the method of any one of claims 26 to 36, wherein the prosthetic device is suitable for use in tendon or ligament replacement. C* C C 0e C CC C* C C C C. C C C C. CC C C CC. C CCC. C C C C C C C C C C C. C C 21/03 2005 MON 12:13 [TX/RX NO 6264) IAJ012