WO2014207462A1

WO2014207462A1 - Biomedical scaffold

Info

Publication number: WO2014207462A1
Application number: PCT/GB2014/051941
Authority: WO
Inventors: Wanda Jadwija LEWIS; Gregory John Gibbons
Original assignee: The University Of Warwick
Priority date: 2013-06-27
Filing date: 2014-06-25
Publication date: 2014-12-31
Also published as: GB201311581D0; WO2014207462A8

Abstract

A tissue scaffold, the scaffold comprising units having interconnecting openings and a stable minimal surface extending between the openings, the surface being analogous to that which would be realised by a soap film surface extending between the same interconnecting openings.

Description

Biomedical Scaffold

The present invention relates to tissue scaffolds, a process for their manufacture, and methods of treating disease or disorders in a subject that involve the implantation of the scaffolds.

Considerable research has been reported in the use of polymeric, metallic and ceramic biomaterials for producing scaffolds. However, the ideal material, architecture of the scaffold, and fabrication technique for optimal tissue growth and, in particular, their suitability for optimal bone tissue regeneration have yet to be identified. While current solutions have met with varying successes, each material, scaffold architecture and/or technique exhibits limitations that must be addressed.

Ideally, scaffolds for bone tissue regeneration should 1) exhibit biocompatibility without causing an inflammatory response or foreign body/toxic reaction, 2) have closely matched mechanical properties when compared to native bone, 3) possess a mechanism to allow diffusion and/or transport of ions, nutrients, and waste, 4) provide a viable space for cells, 5) have pores which are minimally tortuous and 6) have curved cross-sections to mimic the pattern found in nature. Strong bonding with the host bone and vascular in-growth are equally desirable.

Restoration of bone function is dependent on the closely-matched mechanical properties of the scaffold to the native bone for instance the stiffness and permeability. This mechanical similarity is important as bone is primarily load bearing in function, with suitable load transfer necessary to regulate, adapt and remodel bone during the normal healing process. Additionally the architecture of the scaffolds (pore size, porosity and surface curvature) needed for favourable transport/diffusion of nutrients and waste is generally perceived as critical for achieving sustained cell proliferation and differentiation within the scaffolds, thereby affecting function and restoration of the regenerated tissue.

Current approaches to producing scaffolds include "stochastic techniques" that approximate biomorphic "natural" forms but produce sub optimal scaffold architecture, and "deterministic Solid Freeform Fabrication (SFF) techniques" that produce scaffold architecture that does not have a natural form. A common stochastic technique is particulate leaching, in which a soluble or thermally sensitive porogen is included in a material before it is cured and then removed after curing. The volume occupied by the porogen is then left void, forming pores within the material. This technique may have the disadvantage that it uses undesirable toxic solvents and imposes shape limitations on the resulting scaffold. An alternative stochastic technique starts with a natural or artificial sponge as a temporary support for chemical vapour deposition, metal plating or ceramic slurry coating. These

manufacturing techniques can be inflexible and can lead to inconsistent results. An SFF technique typically employs computerised fabrication to rapidly produce complex three-dimensional objects using data generated by computer aided design (CAD) systems, medical imaging modalities and digitizers. Existing CAD- based scaffolds are based on cubic lattices with straight edges and sharp turns or those derived from Boolean intersections of spheres and cylinders. SFF can be used to print bio-gels, ceramics or laser sintered metals. It can also be used indirectly to prepare a temporary, expendable structure for use in coating techniques of the prior art or for removal by heating for example. Neither provides a biomorphic environment suitable for cell attachment, migration and proliferation. There therefore exists a need for a geometry that provides a biomorphic environment while maintaining an adequate stiffness in the scaffold, for load sharing with the natural bone. In the past, particular minimal surfaces and especially Triply Periodic Minimal Surfaces (TPMS) have been considered suitable for tissue scaffolds. US 7718109B describes a tissue support structure derived from a curved minimal surface shape, that is one having a mean curvature of zero everywhere. This minimal surface shape is generated computationally using the equation:

cos X + cos Y +cos Z = o, which approximates to a Schwartz Minimal Surface (Schwartz MS) to generate a patch. The patch is then replicated using reflection rules to create a P-type surface- a symmetric structure with boundaries lying on a unit cube (Fig. 1). Made in polymeric material, the structure generated using this approach, however, has porosity around 50% and also low stiffness (too low to represent a bone material). The achieved porosity has the disadvantage that to minimize the amount of material to be removed from the body upon degradation, and to provide sufficient void space for the cells to grow, the porosity needs to be higher. Similarly, the scaffold stiffness needs to be increased to match that of human bone. So, to address both issues, the scaffold structure is obtained by evolving the fundamental patch sub-optimally to a non-zero, but constant surface curvature, giving a modulated 3D unit structure, which has its boundaries on the surface of a cube (Fig.2). The polymeric support structure is of the type which is seeded with cells which are then grown; the structure is implanted and later degrades leaving the regenerated tissue behind. Such a structure is at a higher energy state than a minimal surface, and is deprived of the advantage of zero mean curvature of the surface, known to facilitate tissue growth. The structures generated using the mathematical approach based on the Schwartz MS equation have not given rise to a utile tissue scaffold and particularly not a permanent structural scaffold such as that used for bone in-growth.

We have found that it is possible to generate a biomorphic tissue scaffold that uses forces to shape the structure, much like the form-finding processes used in nature. The structure is a Stable Minimal Surface (SMS) that gives scaffolds of suitable architecture and mechanical properties for tissue growth.

Accordingly, the invention provides a tissue scaffold, the scaffold comprising units having interconnecting openings and a SMS extending between the openings; the surface being analogous to that which would be realised by a soap film surface extending between the same interconnecting openings.

The scaffold is analogous to the surface that would be created if a soap film were created between assumed boundaries. Soap film surfaces are characterised by a minimum potential energy of surface tension, where the tension forces, which are constant everywhere, shape the surface. The surface of the invention is analogous to a soap film surface in that it is created using the principle of constant tension everywhere to arrive at a surface that preferably has a minimum area, is stable, and has zero mean curvature (i.e. curvatures at each point on the surface are equal and in opposite directions to each other). The surface of the invention is a minimum energy form. By using forces to shape the surface of the scaffold, the surface that results is analogous to those found in soap films. The scaffold comprises units which repeat to form a 3D array. The units preferably have six circular openings arranged orthogonally, joined by a surface of zero mean curvature. The six circular openings, arranged on the surface of a notional prism can be the same size or can be of different sizes (Figs. 3 and 4). The separation of opposite pairs of circular openings can be the same for all of the pairs in the unit or can be different. We believe, the tissue scaffold of the invention, by having an analogous soap film surface (a minimum energy form), will have the advantage that cells will need less energy to grow on such a surface and will therefore reproduce more readily. Further, such a surface will be characterised by a minimum weight with maximum stiffness.

SMS of zero mean curvature are found in nature; for instance, in the surface shape of the spines of sea urchins. These minimum energy forms, therefore, occur naturally.

We have found that the scaffolds of the invention can be made using a computational model that generates a whole soap film surface between given boundaries, i.e. the openings of the SMS unit.

Accordingly, a second aspect of the invention provides a method for making a tissue scaffold comprising units having interconnecting openings, the method comprising the steps of:

First setting characteristics for the unit; and then computationally generating a surface analogous to a soap film surface, to form the surface of the scaffold. This computational form-finding method answers the question of what would a soap-film surface look like when generated between assumed boundaries, in this case the openings of the interconnecting units of the tissue scaffold. This approach has the advantage of giving complete control over the design parameters, to ensure that the structure is generated to have an appropriate architecture for promoting tissue growth and providing appropriate scaffold stiffness while maintaining the SMS- minimum energy

characteristics.

Unlike stochastically generated scaffolds used commercially, the architecture of the SMS scaffold is highly organised, following the directions of principal stress, which bone growth is known to follow. In structures generated from a mathematical model, for instance based on modulating a patch, like those of the prior art (Fig. 2), the advantage of having a minimal surface is lost, once the characteristics e.g. porosity, are modified. A further difference arises when the surface characteristics of the scaffold, e.g. the macro-porosity and micro- porosity (porosity within the walls of the scaffold) as well as the stiffness are modified. The process according to the invention allows control over these characteristics, while still maintaining the stable minimal surface architecture. Preferably the invention provides a method for making a tissue scaffold comprising interconnected SMS units, comprising the steps of:

Setting the characteristics for the units, for instance the number, size and separation of openings in the units;

Computationally generating a soap film surface using Rhino Membrane, GSA Oasys, or any other form-finding software that, through an iterative process, gains convergence towards a soap film surface extending between the openings of the units;

Giving the surface a thickness and saving the output as an STL (stereolithography) file, or any other SFF-compatible format;

Varying the thickness of the units to give a family of SMS units;

Taking that file and using finite element analysis (FEA) software to test for appropriate strength and stiffness in the units to match that of natural bone;

Connecting the units to make a continuously repeating structure expressed as an .STL ,

Polygon File Format (.PLY), Additive Manufacturing Format (.AMF) or any other file format compatible with SFF techniques and reproducing the structure by laser melting a metallic powder in a high resolution SFF process, or any other capable SFF processes, to produce the scaffold.

Preferably the circular openings of the units lie on the faces of a prism. More preferably the openings have a minimum opening size of 0.45mm, moving upwards to 0.5mm and higher. Preferably, the openings in a cube or a rectangular prism are separated in the vertical direction by a distance of imm.

The porosity of the scaffold of the invention is preferably greater than 50%, more preferably from 60% to 90%.

Preferably the Young's modulus of the scaffolds of the invention is from 2 to 5 GPa, The strength of the scaffolds of the invention is preferably from 80 to 150 MPa, more preferably 90 to 110 MPa.

Preferably the scaffolds are made from a biocompatible metal, or alloy, such as titanium, by a process of laser melting or other focused energy source melting such as electron beam melting. Laser melting is characterised by the use of a focused laser to melt and selectively, layer by layer, consolidate a powder starting material into a 3D structure. Preferably the scaffolds are made using a powder as a starting material although other material forms could be used, for instance filament or wire or other materials such as biocompatible metals, and metallic/bioceramic composites.

Focused energy source melting and in particular, laser melting (an SFF process), is preferred as a manufacturing technique as it makes it possible to include micro porosity in the surface of the units. This has the advantages that it assists gas and fluid transfer throughout the scaffold and provides anchorage for the growing cells.

The invention will now be described with reference to the accompanying drawings in which:

Figure l (Prior art) shows a Triply periodic P-type unit surface showing

a) Fundamental patch generated from approximation to Schwartz minimal surface equation: cos X + cos Y + cos Z =o;

b) P-type unit surface characterised by zero mean curvature;

Figure 2 (Prior art) shows modulated 'P-unit' cell obtained by evolving a fundamental patch at non-zero mean curvature;

Figure 3 (current invention) shows

a) Initial meshed surface constructed from individual patches, prior to the form- finding process;

b) Perspective views (mesh and surface) of the form-found SMS unit within a rectangular prism, according to the invention-a soap film surface;

c) Elevation view and dimensions of the form-found SMS unit of Fig. 3b;

d) the SMS scaffold of the invention generated from the unit of Fig. 3c;

Figure 4 (current invention) shows

a) Perspective view of the form-found SMS unit within the unit cube, according to the invention-a soap film surface;

b) Elevation view and dimensions of a form-found SMS unit of Fig. 4a. Aspect ratio

( ring separation to ring diameter) is 2.17, which is a limit at which a soap film surface will form;

c) The SMS scaffold of the invention, generated from the unit of Fig. 4b; Figure 5 shows a perspective view of a manufactured sample of the SMS scaffold of Fig. 4. The sample is manufactured using laser melting SFF in biocompatible titanium or its alloys, has nominal wall thickness of 70 microns, and porosity of 75%. Sample size: 10x10x10 mm cube;

Figure 6 (prior art) shows a sample of a commercial scaffold (75% porosity) manufactured in titanium using a stochastic process (particulate leaching) leading to randomly distributed pores. Sample size: 10x10x10 mm. Results of in-vitro studies, drawing comparisons between the SMS and this commercial product, are given in Fig. 7; Figure 7 shows results of in-vitro studies: cell (osteoblast) growth given by the mean of 3 samples and showing the standard error of the estimate (SEM). Findings: cells grow on the SMS scaffold (Fig. 5) from day 1 and there is 200% more of them just after 7 days from seeding, compared to the commercial product (Fig. 6). Cells started to die after 21 days;

Figure 8 shows results (mean ± SEM) of in-vitro studies: osteoblast matrix (collagen and calcium) production on day 21. Findings: 400% more calcium and 40% more collagen found on the SMS scaffold (Fig. 5) compared to the commercial product (Fig.6).

The SMS units shown in Figures 4 and 5 are surfaces of minimum energy generated by computational form-finding. The surfaces have equal and opposite curvature at every point, giving them a zero mean curvature. The units have six orthogonal circular openings, the size of which may be chosen to give desirable characteristics to the structure, for instance the porosity of the structure.

The invention will now be further described with reference to the above Figures in the following example. Example

Stage 1. Methodology of generating SMS forms

The generation of the SMS units involves a process of form-finding, which is captured in Figures 3, and 4. Initially, a smooth surface (not Schwartz minimal surface) is generated between boundary rings of chosen diameter and separation. This initial surface, prior to form- finding, is assembled from smooth surface patches, represented by a mesh of chosen density, as shown in Fig. 3a. This step constitutes the setting of the characteristics of the unit, described earlier. Subsequently, the surface is prescribed a constant value of tension (as in a thin soap film) and as this tension is incompatible with the arbitrarily assumed initial geometry, the surface will not be in static equilibrium. At this point, the process of iterative adjustments of surface geometry begins, as part of the form-finding methodology. Iterative computations of surface geometry are carried out until the state of static equilibrium (minimum energy) is reached, i.e., the surface has achieved a configuration corresponding to the condition of constant surface tension. The final results of the form-finding computations are SMS units which are unique surfaces, as shown in Figs 3b, 3c, 4a and 4b. The SMS scaffolds assembled from these units are illustrated in Figs 3d and 4 c, respectively.

The SMS scaffold taken to manufacture (described in Part 3 of this example) is that shown in Fig.4, as it represents a structure that is at the limit at which a soap-film surface can form. In other words, if the aspect ratio (ring separation to ring diameter) of this fully symmetric structure exceed 2.17, the soap film surface breaks into flat, disc- type surfaces. This limit (aspect ratio) has been established computationally.

Prior to the manufacture, the SMS surface, as shown in Fig 4a is given thickness to generate a 3D solid body, and the geometric data is stored in the .STL or any SFF- compatible format. Stage 2. Manufacture of the SMS scaffold

The 3D data describing the SMS surface was taken from Stage 1 as an .STL file, being a triangulated representation of the 3D solid body. This file represents one unit cell of the SMS scaffold (Fig. 4a). The 3D scaffold data was formed by generating copies of the unit cell along the three Cartesian coordinates (x,y,z). In this example, the unit cell was replicated 10 times along each axis to generate a scaffold containing 1000 unit cells. The 1000 cells were unified to generate one 3D solid representing the scaffold, and the .STL file for this 3D solid was employed for manufacture. This data manipulation was performed using commercial software such as, but not exclusively, Materialise Magics (Materialise N.V), designed to process .STL files. The scaffold was manufactured using a commercial SFF system (EOSint M280 - EOS GmbH), although any SFF system with an appropriate capability could be utilised. The .STL file representing the scaffold was post-processed to generate a machine specific build file, which is a series of slices through the scaffold from the bottom to the top. Within each slice, all the information required by the SFF system is provided, including the laser power and the laser scan trajectories. Following the data in each slice, the SFF system melted a biocompatible titanium alloy powder (Ti 6A1 4V), with each layer being successively melted onto the previous one, with a 70 micron laser spot size. The layer thickness used was 2omicrons, and this layer thickness was also used in preparation of the data for slicing.

After manufacture, the scaffold had any remaining un-melted powder removed from it using compressed air and cleaning using an ultrasonic bath. The scaffolds were then supplied for in-vitro cell culturing and were sterilised prior to cell-seeding.

Stage 3. Testing of the SMS scaffold against a commercial product

In- vitro cell culturing studies have been carried using the SMS scaffold shown in Fig. 5 and a commercial product described in Fig. 6. The results, given in Figures 7 and 8 show the SMS scaffold outperforming the commercial product. In the case of cell growth (Fig. 7), predicted clinical outcomes would be: earlier integration of the scaffold with the bone, reduced hospitalisation times, physiotherapy and societal costs.

In the case of matrix production (Fig. 8), we have a healthy ratio of calcium to collagen for the SMS scaffold, which would lead to a stronger bone as calcium is primarily responsible for bone strength.

Stage 4. Measurement of SMS porosity

The porosity of the tissue scaffold of Figure 4 was measured using a magnetic suspension balance in a vacuum chamber. It gave a porosity of 75%.

The measurement was carried out using 2 samples, which were weighed in air (Wa) and in vacuum (Wv) using a magnetic suspension balance. The difference in weight (Wa- Wv) equals buoyancy, which, in turn, equals the density of air (dA) times the volume of solid in the sample (Vs). Knowing the difference (Wa-Wv) and also the density of air (dA) , we can find the volume of solid (Vs) in the sample. Then subtracting this from the overall sample volume (Vo) we get the volume of air (Va). Va divided by the overall sample volume (Vo) gives porosity. Thus

Wa-Wv = dA *Vs

from which we can find the volume of solid (Vs) in the samples. Thus:

Vs= (Wa-Wv)/dA.

The overall volume of the sample, (Vos) was calculated from the linear dimensions of the cube samples. The volume of air voids in the sample is: Va= Vos-Vs;

This volume of air relative to the sample volume, expressed as percentage, is the porosity: Va/Vos*ioo. This method of estimating porosity gave us the value of 75%, and is accurate to about 2%.

Claims

1. A tissue scaffold, the scaffold comprising units having interconnecting openings and a stable minimal surface extending between the openings, the surface being analogous to that which would be realised by a soap film surface extending between the same interconnecting openings.

2. A tissue scaffold as claimed in claim ι wherein the surface is created using the principle of constant tension everywhere to arrive at a surface that has a minimum area, is stable, and has zero mean curvature.

3. A tissue scaffold as claimed in claim 1 or claim 2 wherein the surface of the scaffold is a minimum energy form.

4. A tissue scaffold as claimed in any preceding claim having a porosity of at least 70% measured by magnetic suspension balance in a vacuum chamber.

5. A tissue scaffold as claimed in any preceding claim wherein the openings of the units lead to a repeatable pattern.

6. A tissue scaffold as claimed in claim 5 wherein each unit has a plurality of openings arranged on the notional surface of a prism.

7. A tissue scaffold as claimed in claim 5 or claim 6 wherein the ratio of the separation of the openings to the diameter of the opening is less than or equal to 2.17.

8. A tissue scaffold as claimed in any preceding claim wherein the openings are of different sizes.

9. A tissue scaffold as claimed in any preceding claim having a strength of 80 to 150 MPa and a Young's Modulus of 2 to 5 GPa.

10. A method for making a tissue scaffold having a stable minimal surface comprising the steps of :

Setting characteristics for the scaffold; and Computationally generating a stable minimal surface with those characteristics to form the scaffold.

11. A method for making a tissue scaffold comprising units having interconnecting openings, the method comprising the steps of:

First setting characteristics for the unit; and then

Computationally generating a surface analogous to a soap film surface to form the surface of the scaffold.

12. A method as claimed in claim 10 wherein the scaffold comprises units

interconnected by openings in the units, the characteristics including the size, number and separation of the openings.

13. A method as claimed in claim 12 wherein the ratio of the separation of the openings to the diameter of the opening is less than or equal to 2.17.

14. A method as claimed in claim 10 wherein the computation generates a surface analogous to a soap-film surface between the openings in the units.

15. A method for making a tissue scaffold comprising interconnected SMS units, comprising the steps of:

Computationally generating a soap film surface using Rhino Membrane, or GSA Oasys, or any other form-finding software that, through an iterative process of computation, gains convergence towards a soap film surface extending between the openings of the units;

Giving the surface a thickness and saving the output as an .STL (stereolithography) file, or any other SFF-compatible format;

Varying the thickness of the units to give a family of SMS units;

Taking that file and using an FEA software to test for appropriate strength and stiffness in the surface to match that of natural bone;

Connecting the units to make a continuously repeating structure expressed as an .STL , Polygon File Format (.PLY), Additive Manufacturing Format (.AMF) or any other file format compatible with SFF techniques and Reproducing the structure by laser melting a metallic powder in a high resolution SFF system, or any other capable SFF process, to produce the scaffold.

16. A scaffold as claimed in claim 1 made from a biocompatible metal, alloy or ceramic, preferably from titanium.

17. A scaffold as claimed in claim 1 wherein the units have micro porosity (porosity within the wall of the surface).

18. A scaffold as claimed in claim 1 for use in bone.

19. A method of treating a void in the bone of a subject by implanting the tissue scaffold of claim 1 to allow bony in-growth.