TEXTURED SURFACES OF BIOCOMPATIBLE POLYMERIC MATERIALS

Abstract

Provided herein is a biocompatible polymeric material having a textured surface with an arithmetical mean height value (Sa) in a range of 0.1 to 10 ?m, a developed interfacial area ratio (Sdr) above 1.0, determined according to ISO 25178 using a Gaussian low pass S-filter with a nesting index value of at least 0.25 ?m. Also provided is a process for preparing such a biocompatible polymeric material using a microstructured template and a medical device that includes the biocompatible polymeric material.

Claims

1. A biocompatible polymeric material having a textured surface with an arithmetical mean height value (Sa) in a range of 0.1 to 10 ?m and a developed interfacial area ratio (Sdr) above 1.0, determined according to ISO 25178 using a Gaussian low pass S-filter with a nesting index value of at least 0.25 ?m.

2. The biocompatible polymeric material according to claim 1, wherein the textured surface comprises a density of peaks (Spd) above 1?10.sup.6 peaks/mm.sup.2.

3. The biocompatible polymeric material according to claim 1, wherein the textured surface has a texture aspect ratio (Str) above 0.6.

4. The biocompatible polymeric material according to claim 1, wherein the textured surface has a maximum height value (Sz) below 60 ?m.

5. The biocompatible polymeric material according to claim 1, wherein the biocompatible polymeric material comprises a polyester, a polyurethane, an organosilicon, or a polyolefin.

6. A process for preparing the biocompatible polymeric material according to claim 1, the process comprising: a) providing a microstructured template having a textured surface with an arithmetical mean height value (Sa) in a range of 0.1 to 10 ?m and a developed interfacial area ratio (Sdr) above 1.0, determined according to ISO 25178 using a Gaussian low pass S-filter with a nesting index value of at least 0.25 ?m, b) covering the textured surface of the microstructured template with a biocompatible polymeric material as a solid or with a liquid polymerizable mixture, c) conforming the biocompatible polymeric material to textural contours of the microstructured template using, in combination or separately, elevated temperature, pressure, or vacuum, d) when provided as the liquid polymerizable mixture, solidifying the biocompatible polymeric material, and e) separating the biocompatible polymeric material and the microstructured template.

7. A microstructured template for use in the process for preparing the biocompatible polymeric material according to claim 6, wherein the template has a textured surface with an arithmetical mean height value (Sa) in a range of 0.1 to 10 ?m and a developed interfacial area ratio (Sdr) above 1.0, determined according to ISO 25178 using a Gaussian low pass S-filter with a nesting index value of at least 0.25 ?m.

8. The template according to claim 7, wherein the textured surface has a density of peaks (Spd) above 1?10.sup.6 peaks/mm.sup.2.

9. The template according to claim 7, wherein the textured surface has a texture aspect ratio (Str) above 0.6.

10. The template according to claim 7, wherein the textured surface has a maximum height value (Sz) below 40 ?m.

11. The template according to claim 7, wherein the microstructured template comprises a sintered ceramic material.

12. (canceled)

13. (canceled)

14. (canceled)

15. A medical device comprising the biocompatible polymeric material according to claim 1.

16. The medical device according to claim 15, wherein an outer surface of the medical device comprises the textured surface.

17. The medical device according to claim 15, wherein an entire surface of the medical device comprises the textured surface.

18. The medical device according to claim 15, wherein the medical device is one of: a breast implant, a cardiac or cardiovascular implant, a surgical mesh, a neurostimulation lead, an ophthalmic implant, a urological implant, a drug delivery film, and a biosensor.

19. The template according to claim 11, wherein the sintered ceramic material comprises tricalcium phosphate (TCP).

Description

[0116] FIG. 1 shows scanning electron micrographs of the microstructured ceramic templates TCP1, TCP2 and TCP3.

[0117] FIG. 2 shows scanning electron micrographs of the SilkSurface? implant textured surface.

[0118] FIG. 3 shows scanning electron micrographs of microstructured and flat SPCU, PCU and PP polymer surfaces.

[0119] FIG. 4 shows scanning electron micrographs of microstructured PDMS surfaces.

[0120] FIG. 5 shows scanning electron micrographs of microstructured PDMS molds.

[0121] FIG. 6 is a representative histological micrograph depicting a typical fibrous capsule formed around polymer implants. The star denotes the implant, the solid arrow denotes the inflammatory layer (stained blue in the original color image), the short dashed arrow denotes the fibrous layer (stained fuchsin in the original color image), and the long dashed arrow denotes the total capsule thickness. The micrograph was captured at 10? magnification.

[0122] FIG. 7 are representative histological micrographs depicting typical fibrous capsules formed around microstructured and flat polymer implants at 4 weeks. Stars denote the implants, solid arrows denote inflammatory layers (stained blue in the original color images), and dashed arrows denote fibrous layers (stained fuchsin in the original color images). Micrographs were captured at 0.4? (top row) and 10? (bottom row) magnifications.

[0123] FIG. 8 are representative histological micrographs depicting typical fibrous capsules formed around flat PDMS (PDMS_F) and SilkSurface? (SS) implants at 4 weeks. Stars denote the implants, solid arrows denote inflammatory layers (stained blue in the original color images), and dashed arrows denote fibrous layers (stained fuchsin in the original color images). Micrographs were captured at 0.4? (top row) and 10? (bottom row) magnifications.

[0124] FIG. 9 are representative histological micrographs depicting typical fibrous capsules formed around microstructured and flat polymer implants at 17 weeks. Stars denote the implants. Micrographs were captured at 0.4? (top row) and 10? (bottom row) magnifications.

[0125] FIG. 10 are representative histological polarized micrographs depicting typical fibrous capsules formed around microstructured and flat polymer implants at 4 weeks. Stars denote the implants, and triangles denote the capsule immediately apposed to the implant surface. Micrographs were captured at 0.4? (top row) and 20? (bottom row) magnifications.

[0126] FIG. 11 are representative histological polarized micrographs depicting typical fibrous capsules formed around flat PDMS (PDMS_F) and SilkSurface? (SS) implants at 4 weeks. Stars denote the implants, and triangles denote the capsule immediately apposed to the implant surface. Micrographs were captured at 0.4? (top row) and 20? (bottom row) magnifications.

[0127] FIG. 12 are light micrographs depicting human macrophages cultured on flat and microstructured polymers for 21 days. Cells were stained with May-Gr?nwald-Giemsa staining, which appears blue and pink in the original color images. Micrographs were captured at 4? (top row) and 10? (bottom row) magnifications.

[0128] FIG. 13 are scanning electron micrographs depicting human macrophages cultured on flat and microstructured polymers for 21 days. Micrographs were captured at 1,000? (top row) and 2,500? (bottom row) magnifications.

[0129] FIG. 14 illustrates a process for forming a textured polymeric material according to one embodiment of the present invention.

[0130] FIG. 15 illustrates a process for forming a textured polymeric material according to one embodiment of the present invention.

[0131] FIG. 16 illustrates 3D surface scans produced by 3DLSCM analysis showing that microstructured features are present on a control film (left panel) and textured polymeric material according to one embodiment of the present invention (right panel).

EXAMPLES

1: Fabrication and Characterization of Functional Surface Architectures

Example 1.ASample Fabrication

[0132] Microporous ceramic template materials were prepared according to methods previously described in the literature (DOI: 10.22203/eCM.v027a20). Briefly, calcium phosphate powders were synthesized by mixing calcium hydroxide and phosphoric acid (both from Fluka/Sigma-Aldrich, St. Louis, MO, USA) at a Ca/P molar ratio of 1.50. Grain size differences in the final ceramics were produced by carefully controlling the component reaction rates of the ceramic powder. The powders were mixed with diluted hydrogen peroxide (0.1%) (Merck, Darmstadt, Germany) and dried at room temperature to get microporous green bodies. The dry green bodies were subsequently sintered at temperatures ranging from 900-1,200? C. for 8 h to achieve variable microstructures ranging from relatively small at lower temperatures to relatively large at higher temperatures. Microporous discs (9 mm diameter?1 mm thickness) were machined from the ceramic bodies using a lathe and a diamond-coated saw microtome (Leica SP1600; Leica, Solms, Germany), then ultrasonically cleaned in deionized water and ethanol.

[0133] Molds of the desired template surface architecture were prepared using a soft-lithography PDMS casting approach. PDMS (Sylgard? 184; Dow Corning) was prepared at a ratio of 10:1 (w/w) base to curing agent, degassed under vacuum, then poured over the template ceramic discs. Cast PDMS was further degassed under vacuum to fully impregnate the template architecture with PDMS. The PDMS was cured at 70-80? C. for at least 2 hours, then carefully removed from the template disc. Optionally, the PDMS was cured at lower temperatures for longer periods of timee.g. room temperature for one weekto increase the elasticity and toughness of the cured PDMS mold and improve demolding characteristics. To create flat molds, PDMS solution was poured into a smooth polystyrene dish, as a flat template, and similarly cured and removed. All PDMS molds were ultrasonically cleaned in hydrochloric acid to dissolve any ceramic debris, then ultrasonically cleaned in pure ethanol.

[0134] Samples were prepared via hot embossing techniques using biocompatible thermoplastic polymers, such as the following: polycarbonate-based (poly)urethane (PCU; e.g. Carbothane? PC-3595A, Lubrizol Corporation), siliconized polycarbonate urethane (SPCU; e.g. ChronoSil? 85 AL 10% silicone, AdvanSource Biomaterials company), and polypropylene (e.g. Alfa Aesar). In the case of thermoplastic polymeric materials, thin sheets ranging in thickness from 1 to 2 mm, were produced by various standard approaches including solution casting (e.g. first dissolving the polymer in a relevant solvent to create a flowable polymer solution, casting the solution in a flat mold, then evaporating the solvent to achieve a flat sheet), heat molding (e.g. using a heated press to achieve a flat sheet), or extrusion. Discs were punched out of the flat sheets using a circular punch. To emboss the target ceramic microstructures, a polymer disc was sandwiched between flat or microstructured PDMS molds (produced above) in a custom made cylindrical PTFE chamber, and then uniform compressive pressure was applied to the molds on both sides using loaded pistons. Suitable pressure for hot embossing thermoplastic polymers was in the range of 5-50 bar. Embossing constructs were heated at temperatures above their Tg (e.g. 170-220? C.) in an oven for 60 minutes, then allowed to cool naturally to room temperature before carefully demolding. In this way, the desired surface architecture was transferred to both sides of the target material samples. Embossed sample discs were punched out having approximate dimensions of 8 mm diameter and 1 mm thickness.

[0135] Samples were also prepared via casting methods using biocompatible thermoset polymers, such as silicone (e.g. Sylgard? 184 PDMS). Depending on the polymer solution to be cast, pre-treatment of the microstructured mold was necessary to allow for demolding of the cast replica. In the present example, surface treatment and passivation of the mold was achieved by first plasma activating the surface of the PDMS mold in a commercially available plasma coater (e.g. Femto PCCE plasma coater, Dienter Electronic company) using the following settings: O.sub.2 gas, 0.5 mbar pressure, 100 W power, 60 seconds). After plasma activation, the textured PDMS mold was passivated in 100% ethanol under vacuum until fully evaporated. Other surface treatments could alternatively be applied to assist in demolding, such as silanation. In the present example, PDMS solution (Sylgard? 184, 10:1 w/w base to curing agent) was poured over the treated PDMS mold, then degassed in a vacuum for at least 30 minutes. Continuous pressure was applied, for example using a pressure chamber or weights, and cured at room temperature for 1 week. Alternatively, a strong vacuum could be applied and maintained during room temperature curing. Cured PDMS replicas were easily demolded from the textured PDMS molds using these methods with no adverse cytotoxic demolding agents. To create reference samples of a commercially available microstructured surface texture, the outer textured PDMS shell of a SilkSurface? breast implant (SS) (Motiva?, Establishment Labs) was used. Implant shell material was glued together using PDMS (Sylgard? 184, 10:1 w/w base to curing agent), with the textured side of the shell facing outward, and cured at 70? C. for at least 2 hours. Discs were punched out with approximate dimensions 8 mm diameter and 1 mm thickness. In this way, textured reference samples similarly possessed surface texture on both sides of the sample disc.

[0136] Sample discs were ultrasonically cleaned in pure ethanol for in vitro studies, and additionally sterilized in ethylene oxide gas before in vivo implantation.

Example 1.BSurface Texture Characterization

[0137] The method used to characterize surface texture was 3-D Laser Scanning Confocal Microscopy (3DLSCM). 3DLSCM was used to collect 3-D topographic data over a given area on a sample surface. The 3-D data were analyzed following the ISO 25178 standard to quantitatively characterize the surface microtextural features.

[0138] The surface texture of produced samples was analyzed using a commercially available Keyence VK-X210 series 3D Laser Scanning Confocal Microscope (consisting of a VK-X250K controller and a VK-X210 Measuring Unit). The controller emitted a measurement laser light source of 408 nm at 0.95 mW. The instrument manufacturer's software was used for data collection (VK Viewer version 2.8.1.0) and data analysis (VK Analyzer and MultiFile Analyzer version 1.3.1.120). VK Analyzer and MultiFile Analyzer software was capable of computing extracted characterization parameters in compliance with ISO 25178-2:2012.

[0139] The 3D surface Laser Scanning Confocal Microscope measured the surface heights of a specimen, and produced a map of surface height (z-directional or z-axis) versus displacement in the x-y plane. The surface map was then analyzed in the software according to ISO 25178-2:2012, from which the various areal surface texture parameterse.g. Sa, Sq, Sdr, Str, etcwere calculated. These parameters described key characteristics of the embodied surface textures.

[0140] The instrument was periodically calibrated according to the manufacturer's specifications.

[0141] To analyze the surface texture, a sample was mounted onto the microscope stage with the surface of the sample oriented orthogonally to the axis of the objective. Measurements were collected using the 150? APO (NA=0.95) objective lens provided with the instrument. This lens was selected based on the manufacturer's recommendation on the required spatial resolution, which in the case of the embodied microstructures was below 1 micron. Data was acquired using the acquisition software's Expert Mode wherein the following parameters were set: height scan range was set to encompass the height range of the sample (this can vary from sample to sample depending on the surface topography of each); Z-step size was set to 0.08 ?m for the 150? objective; laser intensity and detector gain were optimized for each sample using the autogain feature of the instrument control software (maximized reflected laser signal without causing detector saturation); laser double scan was set to Auto with ND filter set to 100%-only to perform when necessary as judged by the software. This last setting was found useful in reducing optical noise. Measurement mode was set to Surface Profile, Area set to Super-Fine (High-resolution), and images were collected with a resolution of 2048?1536 pixels.

[0142] The Z-step size setting in the manufacturer's measurement software corresponds to the optical measurement resolution in the z-direction (i.e., depth/height of the sample surface). As the z-step size decreases, more variation in the height/depth of the sample surface texture can be resolved resulting in higher Sdr and Spd values. However, some of this signal may be generated by optical noise rather than actual depth/height variations in the surface texture. The manufacturer advises to use a minimum Z-step size of 0.08 ?m when the 150? objective and super-fine resolution are selected.

[0143] The Height Cut Level function is a proprietary feature in the manufacturer's analysis software that eliminates optical noise from the surface measurement. The manufacturer advises that Height Cut Level should always be set to medium when applying the 150? objective. When the Height Cut Level is off or set to low, optical noise can create artificially high Sdr and Spd values. However, when the Height Cut Level is too high (e.g., higher than medium), signal from actual surface features can be eliminated along with more optical noise, thereby lowering Sdr and Spd values.

[0144] Using the 150? objective, captured high-resolution images comprised a field of view of approximately 96?72 ?m, resulting in an x-y resolution of approximately 0.05 ?m/pixel.

[0145] Measurements using the 150? objective were collected using assembly mode, stitching together 9 individual images (3?3 format), resulting in a total image size of approximately 192?264 ?m.

[0146] If a larger field of view was required, e.g. 1 mm?1 mm, multiple scans, maintaining the same x-y resolution and z resolution, over the surface were collected and stitched together into a single image for analysis. Consequently, the required scale of areal field of view did not impact the accuracy or precision of the measurements. In other words, the calculated areal surface texture results were similar whether the analysis area was 10 square microns or 1 square millimeters.

[0147] The measurement file in MultiFile Analyzer software was opened. Manufacturer recommended image processing filters were applied in the Process Image software module to minimize measurement noise and maximize the quality of the surface data: (1) Reference Plane Settings, selecting All Areas; (2) Height Cut Level, medium setting. The surface height image in the surface texture analysis software was opened and S, F, and L filters were applied.

[0148] ISO 25178-2:2012 and ISO 25178-3:2012 describe a recommended filtration process.

[0149] The S-filter nesting index (cut-off) value for optical surfaces is user-defined and should be at least three times greater than the lateral (x-y) measurement resolution. In the case of the equipment and objective used, this measurement resolution is 0.05 ?m; therefore an S-filter greater than 0.15 ?m is appropriate. The nearest value possible in the manufacturer's analysis software is 0.25 ?m, so this value was selected for all analyses. As the S-filter nesting index value is increased, calculated areal parameters such as Sdr and Spd generally decrease because a greater amount of the high-frequency signal is filtered out of the measurement.

[0150] According to the ISO 25178-3: 2012 standard, the L-filter nesting index (cut-off) value for optical surfaces is also user-defined and should be five times as large as the coarsest surface feature to be quantified in the roughness measurement. Therefore, this value was set to at least 0.1 mm, to eliminate waviness from the roughness measurements.

[0151] The following filtering procedure was performed on each image: [0152] For 96?72 ?m images acquired using the 150? objective: 1) a Gaussian low pass S-filter with a nesting index value (cut-off) of 0.25 ?m; 2) an F-operation of plane tilt (auto) correction; and 3) a Gaussian high pass L-filter with a nesting index value (cut-off) of 0.1 mm. [0153] For assembled 192?264 ?m images, stitched together from 9 individual 96?72 ?m images acquired using the 150? objective: 1) a Gaussian low pass S-filter with a nesting index value (cut-off) of 0.25 ?m; 2) an F-operation of plane tilt (auto) correction; and 3) a Gaussian high pass L-filter with a nesting index value (cut-off) of 0.5 mm.

[0154] Filters were applied with end effect correction. This filtering procedure produced the S-L surface from which the areal surface texture parameters were calculated. The entire field of view was selected for measurement, and the areal surface roughness parameters were calculated by the Multifile Analyzer software based on the S-L surface. Filtering and parameter calculation was performed according to the according to ISO 25178-2:2012 and explanatory literature (DOI: 10.1007/978-3-642-36458-7_4). The mathematical derivations and descriptions of the surface texture parameterse.g. Sa, Sz, Sq, Ssk, Sku, Sdr, Spd, Sdq, Sal, and Strare published in ISO 25178-2:2012.

[0155] The surface textures of at least three different locations were scanned and analyzed. The texture values were averaged together and reported to the nearest 0.01 unit.

Example 1.0Scanning Electron Microscopy

[0156] Samples were sputter-coated with a nanolayer of gold, approximately 10 nm thick, and then imaged using a scanning electron microscope (SEM, e.g. Philips XL-30, JEOL IT200). Micrographs were captured at various magnification levels ranging from 500? to 5,000? with typical acceleration voltage of 10-15 keV.

Example 1.DWater Contact Angle

[0157] The water contact angle of produced samples was measured using a commercially available Drop Shape Analyzer instrument (Kruss). A droplet (4 ?l) of deionized water was dispensed by the instrument onto the sample surface and allowed to equilibrate for 30 seconds. The contact angle of the droplet was optically calculated via a 2-tangent algorithm in the software according to the manufacturer's instructions. Replicate samples were measured to confirm the results. Measurements were averaged and reported to the nearest 0.1 unit.

[0158] Table A shows the surface characteristics of the TCP template materials and the PDMS template materials; the areal surface texture values were measured using 3DLSCM equipped with a 150? objective.

[0159] The results as shown in Tables A and B, areal surface texture parameters of ceramic microstructures and polymer replicas were quantified according to the ISO 25178-2:2012 standard using 3DLSCM. In certain essential parameters, the areal surface texture parameters were substantially different than the comparative commercially available material derived from the outer shell of the textured SilkSurface? breast implant (simply, SilkSurface?, SS) and from the materials described in WO2019/118983.

[0160] Arithmetic Mean Height (Sa) of the ceramic templates TCP1, TCP2, and TCP3 ranges from 0.77 to 2.65 ?m mean values, illustrating that this parameter can be controlled by tuning the process conditions of the ceramic template production. Of particular importance is the capability of achieving low Sa values <1 ?m. Following similar trends, Root Mean Square Height (Sq) of TCP1, TCP2, and TCP3 ranges from 0.99 to 3.46 ?m mean values. Maximum peak-to-valley height (Sz) of TCP1, TCP2, and TCP3 ranges from 10.98 to 34.51 ?m mean values. Comparatively, mean Sa, Sq, and Sz values of SS are 2.11 ?m, 2.68 ?m, and 21.56 ?m, showing in particular that TCP1 presents substantially lower mean values for these parameters versus SilkSurface(D.

[0161] The template material used to create textured implant materials as disclosed in U.S. Pat. No. 10,595,979 is acellular dermal matrix basal membrane (ADM BM). According to the presented figures in that patent (16A-16D, 17A-17D, 18A-18D), ADM BM possesses surface features ranging from nanometer scale up to microscale. Although no filtering steps are described, it can be inferred from the surface profile plots that if a low-pass S-filter was applied, according to ISO 25178-3 (2012) and referenced standards such as the S-filter applied in the current examples (i.e., Gaussian S-filter with nesting index value=0.25 ?m), the contribution of the nano-scale features would be eliminated from the calculation of surface parameters. Based on this, one can reasonably expect that ADM BM, and the Silksurface? material which is the replica thereof, would not possess sufficiently complex surface architecture on the microscale to embody or produce via replication Sdr values greater than 1.0 or Spd values greater than 1?10.sup.6 peaks/mm.sup.2 while maintaining an Sa value of less than 3.0 ?m as described in the current invention.

[0162] For the materials described in WO2019/118983 the determined values for Sa are below 3.0 ?m and the values for Sdr are below 1.0 when a filter is applied during the performance of the surface roughness test using a Keyence 3D surface Laser Scanning Confocal Microscope according to ISO 25178-2:2012. The results of the tests are given in Table 1 at p.15 of WO2019/118983.

[0163] The first two lines in Table 1 show the surface roughness of samples A and B that were determined using no filter.

[0164] However, as soon as a filter is applied during the determination of the surface roughness, as is the case for the samples tested according to the invention, the Sdr values for samples A and B are below 1.0.

[0165] Surface skewness (Ssk) of the ceramic templates TCP1, TCP2, and TCP3 ranges from ?0.35 to ?0.66 indicating a negative skew of surface features. This means that the surfaces comprise relatively more valleys (i.e. pores) than peaks. When Ssk is >0, this means the surface comprises relatively more peaks than valleys. Surface kurtosis (Sku) of the templates TCP1, TCP2, and TCP3 ranges from 3.40 to 4.68 indicating a surface of relatively sharp sloping peaks rather than gradual bumps. This is meaningful in reducing tribological friction of a surface. The literature teaches that when Sku increases and Ssk becomes more negative, the coefficient of friction is minimized. Comparatively, Ssk and Sku of SilkSurface? are 0.16 and 3.28, suggesting relatively higher coefficient of friction versus the ceramic templates according to the literature (DOI: 10.1080/10402004.2016.1159358).

[0166] The surface developed interfacial ratio (Sdr) of the ceramic templates TCP1, TCP2, and TCP3 ranges from 1.57 to 21.10 mean values, indicating an increase of surface area ranging from 157% to 2,110% for the TCP templates versus a perfectly flat surface. Sdr is the quantification of the percentage of additional surface area contributed by the texture as compared to an ideal plane the size of the measurement region, and in this way relates to the complexity of the surface. An Sdr value of 1.00 means a measured surface exhibits 100% more surface area than a perfectly flat material of the same measurement area. The surface peak density (Spd) of the ceramic templates TCP1, TCP2, and TCP3 ranges from, on average, 2.09?10.sup.6 to 3.03?10.sup.6 peaks/mm.sup.2, showing the extremely high density of surface peaks of the templates. Depending on the application, a low Spd may result in higher localized contact stresses resulting in possible pitting and debris generation. In applications involving sliding components, a high density of surface peaks are needed to reduce friction while maintaining a reasonable load distribution. The surface root mean square gradient (Sdq) of the ceramic templates TCP1, TCP2, and TCP3 ranges from, on average, 2.03 to 8.07. Sdq is a general measurement of the slopes which comprise the surface and may be used to differentiate surfaces with similar average roughness. The Sdq of a perfectly flat surface is 0. Sdq is affected both by texture amplitude and spacing. Thus for a given Sa, a wider spaced texture will result in a lower Sdq value than a surface with the same Sa but finer spaced features. Taken together, Sdr, Spd, and Sdq collectively represent the complexity of the surface. In comparison, the mean Sdr, Spd, and Sdq values of SilkSurface? are 0.77, 0.73?10.sup.6 peaks/mm.sup.2, and 1.45all substantially lower than any of the ceramic templates. Collectively, these values show that the SilkSurface? surface is substantially less complex than the ceramic templates.

[0167] For the materials described in WO2019/118983, the determined values for Sdr are below 1.0 and the determined values for Spd are below 1.00?10.sup.6 peaks/mm.sup.2 when a filter is applied during the performance of the surface roughness test using a Keyence 3D surface Laser Scanning Confocal Microscope according to ISO 25178-2:2012. The results of the tests are given in Table 1 at p.15 of WO2019/118983. The first two lines in Table 1 show the surface roughness of samples A and B that were determined using no filter. However, as soon as a filter is applied during the determination of the surface roughness, as is the case for the samples tested according to the invention, the Sdr and Spd values for samples A and B are below 1.0 and 1.00?10.sup.6 peaks/mm.sup.2, respectively.

[0168] Surface autocorrelation length (Sal) of the ceramic templates TCP1, TCP2, and TCP3 ranges from, on average, 7.48-11.57 ?m. Sal is a quantitative measurement of the minimum distance along the surface between two locations whose textures that are statistically different from each other, i.e. minimal correlation. The surface texture aspect ratio (Str) of the ceramic templates TCP1, TCP2, and TCP3 ranged from, on average, 0.66 to 0.85, indicative of isotropic surface textures. Str is a measure of the spatial isotropy or directionality (also, lay) of the surface texture. For a surface with perfect directionality, Str=0.00; for a surface that is perfectly isotropic, Str=1.00. In comparison, the mean Sal and Str values of SilkSurface? are 21.73 and 0.78.

[0169] Representative SEM micrographs of ceramic templates, as shown in FIG. 1, demonstrate that these surfaces comprise an isotropic distribution of interconnected, spheroidal grains and pores arising from the specific conditions of the production process. In general, TCP1 possesses the smallest features-both grains and poresversus TCP2 and TCP3, which are substantially smaller than 1 ?m on average. Despite the randomly occurring network of fused grains and pores visible at high magnification, the surface textures all visually appear to be uniformly distributed and structured at lower magnifications, evident of high production purity and consistency. The high complexity, peak density, and specific surface area can also be appreciated for all ceramic templates, particularly in high magnification micrographs. In comparison to other known microstructures in the art, there is no machined direction array pattern or discernible demarcation between the granular protrusions and interdigitated porous valleys. The topographical features are both randomly and homogeneously distributed over the surface at both high and low magnifications.

[0170] In contrast, representative SEM micrographs of SilkSurface?, as shown in FIG. 2, demonstrate this surface comprises a random array of protruding hills and irregular depressions of varying nodular and crescent form and size. It is evident that the complexity of the surface is relatively low compared to the ceramic template surfaces, particularly from the apparent areal density of protrusive and receding elements, as well as the perceivable interfacial area. The irregularity of the surface feature morphology is also comparatively high versus the ceramic templates, as is visually apparent in the high magnification images. At low magnifications, the SilkSurface? surface appears heterogenous and isotropic.

[0171] The microstructured ceramic and flat template surface structures are replicated in various medical polymers including SPCU, PCU, PP, and PDMS using both hot-embossing and casting methods. As shown in the SEM micrographs (FIG. 3, 4), replication of ceramic template surface microstructurein terms of feature size, shape, and arrangementoccurs consistently across the various polymers with good fidelity. The template ceramic surface microstructures are particularly well replicated in PDMS (FIG. 4). Replication of microstructured templates is enabled by high-fidelity PDMS molds produced by soft lithography; similar grain and pore feature sizes are observed with the inverse form of the originating templates (FIG. 5). In concert with these results, areal surface texture parameters as measured by 3DLSCM (Table B) confirm that generally the key surface texture parameter values-Sa, Sz, Sq, Ssk, Sku, Sdr, Spd, Sdq, Sal, and Str-are well preserved and within reasonable range of those of the ceramic templates.

[0172] Water contact angle measurement of microstructured and flat polymers (Table B) showed that increasing microstructural roughness results in increasing hydrophobicity for all the polymer types tested. This result is important because highly hydrophobic surfaces have been shown to resist microbial adhesion and biofilm formation.

TABLE-US-00001 TABLE A Sa Sz Sq Spd Spc Sal Material ?m ?m ?m Ssk Sku Sdq Sdr 1/mm.sup.2 1/mm Str ?m CE A SilkSurface?, 2.11 21.56 2.68 0.16 3.28 1.45 0.77 0.73E+06 12385.67 0.78 21.73 SS 1a TCP1 0.77 10.98 0.99 ?0.66 4.68 2.03 1.57 2.91E+06 17545.75 0.84 7.48 2a TCP2 1.57 16.98 2.00 ?0.35 3.40 3.17 3.43 2.09E+06 26058.67 0.85 8.44 3a TCP3 2.65 34.51 3.46 ?0.54 4.59 8.07 21.10 3.03E+06 99736.09 0.66 11.57

TABLE-US-00002 TABLE B Water contact Sa Sz Sq Spd Spc Sal angle Material ?m ?m ?m Ssk Sku Sdq Sdr 1/mm.sup.2 1/mm Str ?m degrees, ? CE A SilkSurface?, 2.11 21.56 2.68 0.16 3.28 1.45 0.77 0.73E+06 12385.67 0.78 21.73 126.2 SS CE B PDMS_F 0.01 0.40 0.01 3.22 63.38 0.04 0.00 5.59E+06 500.46 0.22 2.05 123.5 CE C SPCU_F 0.02 1.06 0.03 2.51 43.35 0.04 0.00 2.90E+06 280.12 0.26 35.63 111.7 CE D PCU_F 0.02 0.95 0.03 1.23 17.92 0.04 0.00 2.62E+06 266.85 0.28 37.24 104.1 CE E PP_F 0.64 8.19 0.81 ?0.32 3.69 0.12 0.01 2.21E+05 518.59 0.77 26.13 110.0 1b SPCU_1 0.80 9.08 1.04 ?0.89 4.47 1.77 1.19 2.44E+06 13016.68 0.92 7.19 127.8 1c PCU_1 0.86 11.40 1.08 ?0.61 4.07 1.72 1.12 2.22E+06 12353.98 0.91 9.66 116.6 1d PP_1 1.09 10.19 1.36 ?0.52 3.07 2.03 1.53 2.84E+06 17958.12 0.43 12.58 123.4 1e PDMS_1 1.22 12.36 1.53 ?0.61 3.45 1.66 1.04 1.74E+06 13120.45 0.65 8.97 n.d. 2b SPCU_2 1.75 17.42 2.21 ?0.30 3.15 3.06 3.24 2.04E+06 25328.43 0.90 12.03 129.2 2c PCU_2 1.55 21.22 2.00 ?0.74 4.05 3.13 3.30 1.66E+06 30462.04 0.75 10.83 118.1 2d PP_2 1.53 16.25 1.92 ?0.21 3.04 2.91 2.95 2.27E+06 23650.70 0.89 8.03 133.4 3b SPCU_3 2.43 23.46 3.01 ?0.26 2.83 3.64 4.32 1.69E+06 29521.25 0.81 12.20 130.6 3c PCU_3 2.63 25.01 3.31 0.00 3.05 4.20 5.62 1.67E+06 29460.49 0.94 14.32 121.0 3d PP_3 2.46 27.88 3.09 0.00 3.05 5.48 9.54 2.39E+06 44100.88 0.93 11.68 138.2 3e PDMS_3 2.42 27.23 3.08 0.37 3.45 5.79 10.89 1.88E+06 56375.68 0.81 12.13 n.d.

2: In Vivo Performance

Example 2.AIn Vivo Implantation and Histological Processing

[0173] Discs were implanted in the dorsal subcutaneous fat layer of Bama minipigs. With the permission of the local ethic committee [SYXK (|I|) 2019-189-AS2019-045], surgical implantation was performed under general anesthesia (with intravenous injection of pentobarbital sodium, 30 mg/kg body weight) and sterile conditions on Bama miniature pigs (n=5; 6 months old, 20?5 kg weight, both male and female). In brief, following anesthesia, longitudinal skin incisions were made on the back beside the spine, and subcutaneous tissue pouches, spaced >2 cm apart, were subsequently made in the fat layer using a scalpel on both sides of the skin incision. Sterile disc implants (diameter=8 mm, thickness=1 mm) were inserted into the subcutaneous pocketsone implant per pocketusing forceps. Pockets were sutured tightly closed using resorbable PGA suture (4-0), and the skin incision was finally sutured with silk suture (4-0). The skin wounds were sterilized with iodine and penicillin was intramuscularly injected for 3 consecutive days to prevent infection. A second surgical implantation identical to the first was performed 13 weeks after the first surgical operation. Four weeks after the second surgical operation, the animals were sacrificed with an overdose of sodium pentobarbital and implants were harvested with surrounding tissues. The harvested samples with implanted durations of 4 weeks and 17 weeks were fixed in 10% buffered formalin, dehydrated in a graded ethanol series, embedded in poly(methyl methacrylate) (PMMA, Cool-Set, Aorigin, Chengdu, China). Sections (?20 ?m thick), oriented cross-sectionally through the discs, were made with histological diamond saw (SAT-001, Aorigin, Chengdu, China) and stained with methylene blue/basic fuchsin or hematoxylin/eosin. The stained sections were then digitized with scanning microscopy (Austar43, AiMco, Xiamen, China) at 10? magnification for histological evaluation and histomorphometry. To view the collagen, sections were scanned to polarized overview images using a slide scanner (Konica Minolta Elite 5400 II, Japan) and polarized film.

Example 2.BCapsule Thickness and Composition

[0174] The thickness of the fibrous capsules and comprising capsule tissue layers at 4 and 17 weeks was measured using AxioVision software (Carl Zeis, version 4.9.1) at a magnification level of 5?. Thickness measurements were made at 15 different locations surrounding each implant from a representative section, arising from three histological sections per implant. Generally, two distinct tissue layers were evident within the capsule: the inflammatory layer, mainly made up of mononuclear leukocytes (e.g. macrophages), and the denser fibrous layer, mainly made up of fibrous tissue oriented in a parallel direction to the surface of the implant. Total capsule thickness measurements were made perpendicular from the material surface outward to the point where the capsule ceased and native tissue such as fat or dermal tissue began. Inflammatory layer thickness measurements were similarly made but only to the transition point between the loose inflammatory layer and the denser fibrous layer of the capsule. The dense fibrous layer was calculated as the difference between the total capsule thickness and the inflammatory layer thickness (e.g. total capsule thickness minus inflammatory layer thickness).

Example 2.0Capsule Density Measurement

[0175] The tissue density of the fibrous capsules formed at 4 and 17 weeks was characterized by measuring the chromatic saturation levels in the histological sections corresponding to stained collagenous tissue. ImageJ software was used for all image processing and analysis. Similar methods have been previously described in the literature (Chen Y, et al. Int J Clin Exp Med 2017; 10(10):14904-14910). The chromatic saturation of a particular histologically stained tissue section (e.g. using a chromogenic dye) can be linked to the areal density of that tissue given that more densely packed tissue will bind more chromogen than less densely packed tissue of the same area and thickness.

[0176] Representative images of the fibrous capsule were captured at 22? digital magnification at three different locations around the implant surface. Representative images of the dermal tissue were similarly captured at 22? magnification at three locations in the same histological section as a reference for the standard chromatic saturation of collagenous tissue. Image acquisition was conducted using standard digital slide scanner software (e.g. HD Scanner software) and exported in 24-bit RGB TIFF format. Acquired images were imported into ImageJ and color deconvoluted into RGB channels using the Color Deconvolution plugin (version 2). In this way, the red color channel corresponding to fuchsin-stained collagenous tissue could be independently analyzed. The deconvoluted red-channel of each image was first converted into HSB (hue-saturation-brightness) format so the saturation values of each image could then be measured using the Measure function in ImageJ. The Measure function provides a measured output of mean (average) pixel intensity on a scale from 0 to 255, corresponding to standard 8-bit 256 shade grayscale; in this context, a pixel intensity of 255 represents complete saturation and 0 represents complete unsaturation. Mean saturation levels of replicate locations within the same capsule were averaged and normalized against the average of mean saturation levels measured in the dermal tissue of the same histological section. In this way, differences in section thickness as well as differences in animal tissue structure were normalized in the measurements. Averaged normalized chromatic saturation values of microstructured samples in each material group were then divided by the mean color saturation values of respective flat samples within that same material group, resulting in relativized, normalized chromatic saturation values (e.g. the normalized chromatic saturation values of SPCU_1 and SPCU_3 each divided by the normalized chromatic saturation value of SPCU_F; the normalized chromatic saturation values of PCU_1 and PCU_3 each divided by the normalized chromatic saturation value of PCU_F; the normalized chromatic saturation values of PP 1 and PP 3 each divided by the normalized chromatic saturation value of PP_F; the normalized chromatic saturation values of SS divided by the normalized chromatic saturation value of PDMS_F). Resulting relativized, normalized chromatic saturation values are thus reported as a relative percentage vs. respective flat values.

Example 2.DResults of the In Vivo Performance

[0177] Microstructured and flat polymeric sample discs were implanted in the dorsal subcutaneous tissue of Bama mini-pigs to study their tissue response, according to the study design outlined in Table C. The study was designed to evaluate the biological response to ceramic surface microstructure replicated in different polymers, made up of different material chemistries. For each polymer cluster (e.g. SPCU groups, PCU groups, PP groups), sample groups comprised either TCP1, TCP3, or a flat surface structure. Flat PDMS (PDMS_F) and textured SilkSurface? (SS), which is composed of PDMS (DOI: 10.1093/asj/sjx150), were also included as relevant benchmarks in the art. In this way, the effects of surface structure and surface chemistry were decoupled so that the effects of surface structure could be more universally appraised. Relative comparisons could be made within polymer groups to discern the specific impact of surface structure on tissue response, specifically: SPCU_1 and SPCU_3 versus SPCU_F; PCU_1 and PCU_3 versus PCU_F; PP_1 and PP_3 versus PP_F; SS versus PDMS_F.

[0178] The subcutis, located below the dermis, is principally made up of vascularized adipose tissue and is a relevant location to study the foreign body response to biomaterials. Moreover, the Bama minipig is a well-accepted species model for understanding the human immune system. Implanted samples were recovered after 4 and 17 weeks implantation and processed for histology as described in the methods. The fibrous capsule formed around the implants was measured in stained histological tissue sections using specialized software. The fibrous capsule formed around the implants due to the foreign body response is generally found to comprise two distinct tissue layers, as depicted in FIG. 6: (1) the inflammatory layer, principally composed of mononuclear leukocytes and occasional multinucleated giant cells immediately apposed to the implant surface, and (2) the fibrous layer, principally composed of fibroblastic cells oriented in a parallel manner to the surface of the implant. Typically, the inflammatory layer is well vascularized and more densely populated by cells, while the fibrous layer is less vascularized, and less densely populated by cells. The fibrous capsule was quantified in terms of total thickness, as well as constituent inflammatory and fibrous layers as shown in FIG. 6 to characterize the capsule tissue morphology as well as composition.

[0179] At 4 and 17 weeks, the thickness of capsules formed around microstructured and flat polymeric samples were approximately similar within the same polymer compositional groups (e.g. SCPU_1 vs. SPCU_3 vs. SPCU_F) (Table C). Comparison between SS and PDMS_F were also made because SS is also composed of PDMS. In agreement with the other polymer groups, there was no apparent difference in average capsule thickness between SS and PDMS_F groups.

[0180] There were, however, differences in capsule composition, as defined by inflammatory and fibrous layer thicknesses, in response to the surface structure. At 4 weeks, sample groups incorporating microstructured surfaces exhibited substantially higher mean inflammatory layer thickness versus the respective flat reference groups (e.g. from Table C: the mean inflammatory layer thickness of PCU_1 and PCU_3 was roughly 100% greater than PCU_F). This systematic effect of surface structure, demonstrated across all evaluated polymer chemistries, points to a powerful, universal role of surface microstructure directing the immune response. A similar relationship in inflammatory layer thickness was observed for SilkSurface, which was ?83% greater than flat PDMS (Table C). FIGS. 7 and 8 present representative micrographs of histological sections stained with basic fuchsin/methylene blue, which depict these global differences in inflammatory layer thickness at 4 weeks. In concert with increased inflammatory layer thickness, mean fibrous layer thickness decreased by up to ?45% in response to surface microstructure at 4 weeks for all polymer clusters (e.g. SPCU, PCU, and PP). For example, mean fibrous layer thickness for SPCU_1 and SPCU_3 ranged from 152-161 microns versus 252 microns for SPCU_F (Table C). In comparison, mean fibrous layer thickness of SS implants decreased by ?15% versus flat PDMS_F. FIGS. 7 and 8 present representative micrographs of histological sections stained with basic fuchsin/methylene blue, which depict these global differences in fibrous layer thickness at 4 weeks.

[0181] By 17 weeks, the inflammatory layer thickness systematically decreased for all microstructured groups by similar proportions relative to their 4 week values, illustrating healthy restitution of temporal inflammation to a lower, equilibrium state. Fibrous layer thickness also equilibrated to similar levels within polymer groups, irrespective of surface structure as shown in Table C.

[0182] In the literature, it is known that capsule thickness alone may not be a determining factor of maladies such as capsular contracture. Compositional characteristics of the capsule such as capsular collagen density and alignment may also play important roles. To probe the capsule composition deeper, two different methods were employed: chromatic saturation measurement and polarized light microscopy. Chromatic saturation measurement was conducted using image analysis software and corresponds to the amount of histologic stain taken up by the tissue section during staining. Qualitatively, it is apparent that densely organized dermal tissue, which is principally composed of collagen, stains deep fuchsia following staining with basic fuchsin and methylene blue dyes. By normalizing the chromatic saturation level of the capsule fibrous layer to that of neighboring dermal tissue, the relative collagen content of the capsule can be quantitatively measured from density of chromatic saturation. Moreover, polarized light microscopy is often used in the art to analyze collagen density and alignment, based on the birefringence properties of collagen's morphology. Under polarized light, collagen reflects the polarized light and can therefore be visualized.

[0183] Using both methods, it was observed that the microstructured polymeric surfaces reduce the density of fibrous capsules versus flat surfaces at both 4 and 17 weeks. Quantitatively, chromatic saturation analysis showed that capsule fibrous density was significantly reduced in microstructured groups versus their respective flat reference groupsin some cases by more than 50% (e.g. at 4 weeks, PCU 1=46% and PCU 3=45%; both relativized versus the normalized chromatic saturation level of PCU_F at 4 weeks) (Table C). This difference in fibrous capsule density is evident in the histological micrographs presented in FIG. 9, in that the chromatic saturation (i.e. richness of color) of the histological stain is clearly greater for the flat implant versus that of the microstructured implants. Qualitatively, these results were confirmed using polarized light microscopy, which showed that the capsular tissue layer closest to the microstructured surfaces is appreciably devoid of collagen (FIG. 10). This modulation of tissue response and soft tissue repair demonstrate important benefits of the embodied microstructured surfaces for reducing capsular contracture of polymeric medical implants. In comparison, the capsule density of SilkSurface? implants, according to chromatic saturation, was on average 20% greater than flat PDMS at 4 weeks (i.e. SS=120% relativized versus the normalized chromatic saturation level of PDMS_F at 4 weeks; Table C). This result was further corroborated by polarized light microscopy which showed a similar collagen dense layer immediately apposed and highly aligned to the implant surface for both SS and PDMS_F implants (FIG. 11). Taken together, these results illustrate a unique benefit of the embodied microstructured architectures of the present invention versus the flat and the SilkSurface? surface textures to more potently reduce capsular contracture.

TABLE-US-00003 TABLE C Capsule Thickness and Density Measurements Capsule Density Capsule Density (normalized (normalized chromatic chromatic 4 weeks saturation 17 weeks saturation Total Inflammatory Fibrous relative Total Inflammatory Fibrous relative Thickness Layer Layer to flat) Thickness Layer Layer to flat) Example Material ?m ?m ?m % ?m ?m ?m % CE F PDMS_F 209 39 170 270 33 237 CE G SilkSurface?, 217 71 146 120 n.d. n.d. n.d. n.d. SS CE H SPCU_F 336 107 229 348 98 250 CE I PCU_F 189 50 138 243 64 179 CE J PP_F 266 82 184 318 53 265 1f SPCU_1 330 170 161 64 335 93 242 70 1g PCU_1 183 101 83 46 286 66 220 45 1h PP_1 278 118 160 43 354 84 270 66 3f SPCU_3 334 181 152 66 368 86 282 74 3g PCU_3 215 102 113 45 272 78 194 59 3h PP_3 245 107 138 48 354 84 270 67 n.d. = not determined

3: In Vitro Performance

Example 3.ACell Culture

[0184] Human CD14+ monocytes (Cat. #2W-400A, Poeitics? Lonza) were cultured on the surface of microstructured and flat sample discs for up to 25 days. Discs were placed in suspension culture 48-well plates (one disc per well) and fixed to the bottom of the well using silicone O-rings. Immediately prior to seeding cells, discs were incubated in 500 ?l culture medium for up to 2 hours at 37 C, 5% CO.sub.to equilibrate the discs to culture conditions. For the purpose of this macrophage study, the culture medium consisted of the following ingredients: 25 ng/mL human recombinant macrophage colony stimulating factor (M-CSF; Cat. #300-25-10UG, Peprotech), 10% (v/v) fetal calf serum (FCS, Cat. #SH30080.03, HyClone GE Health), 1% (v/v) penicillin streptomycin (P-S, Gibco), all contained in Minimum Essential Medium-? (MEM-?, Cat. #BE02-002F, Lonza). On the day of seeding, cells were gently thawed and resuspended in culture medium, according to the manufacturer's instructions, then further diluted in culture medium to a concentration of 550,000 cells in 500 ?l. From this cell stock, 500 ?l cell suspension was carefully pipetted on the surface of each disc, then the containing multiwell plates were transferred to culture incubators and left undisturbed for the cells to attach to the discs. Culture media was completely removed and replaced with 500 ?l fresh, prewarmed media every 3-4 days. The removed culture media was collected on medium refreshment days-denoted, conditioned mediacentrifuged to remove cellular material, and immediately stored at ?80? C. until further use.

[0185] Separately, human dermal fibroblasts (Cat #: CC-2511 Lonza) were cultured in tissue culture treated 96-well plates for up to 7 days. Culture medium used for fibroblast cultures consisted of 10% FCS, 1% P-S, and MEM-?. Leading up to the study, cells were expanded in T75 culture flasks and subpassaged between 4 and 6 times upon reaching 80% confluence to create a cell stock with sufficient cell number for the study. On the day of seeding, cells were trypsinized in 2.5% trypsin-EDTA and resuspended at a density of 80,000 cells/ml. Cells were seeded in each well at a density of 8,000 cells in 100 ?l medium. After 1 day of culture, culture media was completely removed and replaced with 50 ?l fresh, prewarmed culture media per culture well. An additional 50 ?l of prewarmed macrophage-derived conditioned media was supplemented to each culture well. After 2 days of culture, an additional 50 ?l of prewarmed macrophage-derived conditioned media was supplemented to each culture well. After 3 days of culture, an additional 50 ?l of prewarmed macrophage-derived conditioned media and 50 ?l of prewarmed fresh culture media was supplemented to each culture well, without any media removal. The culture was maintained through 7 days culture. In this way, human fibroblasts were exposed to a sustained dose of secreted paracrine factors produced by macrophages cultured on the sample discs in the experiment described above. Cells were cultured in incubators maintained at 37? C., 5% CO.sub.2.

[0186] The cell studies were repeated at least twice to ensure reproducibility and consistency of results.

Example 3.BCell Viability

[0187] Cell viability was measured using the PrestoBlue Cell Viability assay (Cat. #A13261, Invitrogen) following the manufacturer's instructions. On the day of measurement, culture media was removed from the culture well and replaced with 250 ?l 5% (v/v) PrestoBlue prewarmed media, diluted in fresh culture medium (10% FCS, 1% P-S, MEM-?). Cells were incubated in PrestoBlue media for 2 hours at 37? C., 5% CO.sub.2 to ensure sufficient signal. The incubated PrestoBlue media was then individually sampled to the wells of a black, opaque 96-well plate (100 ?l sample per well) for fluorescent measurement using a Clariostar Plus multimode plate reader (560 nm/590 nm, excitation/emission wavelengths). Duplicate samples per culture well were measured, and averaged across replicate cell-culture wells. Mean relative fluorescent unit values, rounded to the nearest 1 unit, were reported.

Example 3.0May-Gr?nwald-Giemsa Staining

[0188] May-Gr?nwald-Giemsa (MGG) staining is a Romanowsky-type, polychromatic stain routinely used for hematology and diagnostic cytopathology. At specific culture time-points, cell culture media was fully removed, cells were washed with neutral phosphate buffered saline (PBS) containing calcium and magnesium. (Cat. #14040141, Gibco), and fixed in 2.5% glutaraldehyde diluted in neutral PBS for at least one hour at room temperature (RT). Fixative was removed, and cells were further fixed and permeabilized by incubating in methanol for 10 minutes at RT. Methanol was removed, and cells were stained according to the following protocol: 5 minute incubation in May-Grunwald solution (diluted 1:1 v/v in pH 6.8 PBS; Cat. #MG500, Sigma-Aldrich); 3 minute incubation in. pH 6.8 PBS buffer; 30 minute incubation in Giemsa solution (diluted 1:9 v/v in pH 6.8 PBS buffer; Cat. #48900, Sigma-Aldrich); washed 3 minutes in pH 6.8 PBS buffer; dried. All staining steps were conducted at RT. Imaging of the stained cells was conducted using a Nikon SMZ25 stereomicroscope.

Example 3.DScanning Electron Microscopy

[0189] Scanning electron microscopy (SEM) of cultured cells was performed as follows. At specific culture time-points, cell culture media was fully removed, cell-seed samples were washed wdth neutral phosphate buffered saline (PBS) containing calcium and magnesium (Cat. #14040141, Gibco), and fixed in 2.5% glutaraldehyde diluted in neutral PBS for at least one hour at RT. Fixative was removed, and samples were further dehydrated in a graded series of ethanol (e.g. 50%, 60%, 70%, 80%, 90%, 100%, 100% ethanol). Samples were further dehydrated in an automated critical point dryer (e.g. Lei ca EM CPD300 machine) and affixed to conductive metal pins using carbon tape. Samples were sputter-coated with a nanolayer of gold, approximately 10 nm thick, and then imaged using a scanning electron microscope (SEM, e.g. Philips XL-30, JEOL IT200). Micrographs were captured at various magnification levels ranging from 500? to 5,000? with typical acceleration voltage of 10-15 keV.

Example 3.EResults of the In Vitro Performance

[0190] The behavior and response of human macrophages cultured on the surface of microstructured and flat polymeric discs was characterized in vitro. Human CD14+ monocytes were used as a relevant model for studying the foreign body response, in that monocytes differentiate into macrophages upon adhesion with a substrate, and macrophages are the body's primary line of defense against foreign bodies such as medical implants. Macrophages exist in nearly every tissue of the body and are the progeny of monocytes arising from the bone marrow.

[0191] Microstructured and flat polymer discs were prepared as described in the methods and used for the in vitro study according to Tables D and E. Culture on microstructured samples was shown to profoundly change the macrophage morphology at 21 days based on MGG staining (FIG. 12) and SEM (FIG. 13). From MGG staining, it was observed that macrophage shape on flat samples was predominantly roundish, and well spread; in comparison, macrophages cultured on microstructured surfaces were more irregularly shaped, less round, and generally less spread. It is particularly noted that macrophages cultured on microstructured surfacesespecially surfaces with TCP1 microstructure, e.g. SPCU_1, PCU_1, and PP_1tended to create more cell-cell junctions, linked by cellular filopodia connections. This resembled a more activated state. In SEM, differences were also noted. Specifically, macrophages cultured on flat tended to be round and well spread, sometimes 100s of microns in length, suggesting cells have fused to form giant cells. It is typical for macrophages cultured on flat surfaces to have a spheroid apical morphology resembling a doorknob. In comparison, macrophages cultured on micro structured surfaces appeared generally less spread out. On TCP1 microstructured surfaces (e.g. SPCU_1, PCU_1, and PP_1), macrophage size was generally less than 60 micron across; on TCP2 microstructured surfaces (e.g. SPCU_2, PCU_2, and PP_2), macrophage size was generally less than 100 micron across. Most cells cultured on TCP1 microstructured surfaces possessed a spheroid apical morphology resembling a doorknob, while cells cultured on TCP2 microstructured surfaces less often possessed this morphology but rather appear flatter. Taken together, these results indicate that microstructured polymeric surfaces affect macrophage morphology and phenotype in distinct ways, and that the specific size scale of microstructural features plays a role in this.

[0192] Macrophage viability after culture for 3, 10, and 21 days on microstructured and flat polymeric discs was measured using the PrestoBlue assay (Table D). No consistent differential effects of the surface structure on cell viability across the various polymers tested were discernible. These results indicate that microstructured polymeric surfaces are not cytotoxic to cells, nor do they promote more or less cell adhesion than respective flat surfaces.

[0193] During macrophage culture on microstructured and flat polymer discs, conditioned medium containing macrophage-secreted paracrine factors was collected and supplemented into fibroblast cultures to study the effects on fibroblasts response. This type of in vitro model is often used in the art to study the interplay between macrophages and fibroblasts in the foreign body response and fibrosis. Fibroblasts are the principal constructor and resident of the fibrous capsule that forms around a foreign body; moreover, given the relevant paracrine signals, fibroblasts can differentiate into myofibroblasts which are the main contractile agent causing capsular contracture. It is known in the literature that when fibroblasts are proliferating-evidenced by high viabilitythey are not also differentiating into myofibroblasts (DOI: 10.1096/fj.03-0699com); therefore, it is useful to study fibroblast proliferation, as a function of viability, to determine their propensity to promote capsular contracture.

TABLE-US-00004 TABLE D Macrophage viability as measured by the PrestoBlue assay (values: mean relative fluorescent units) Example Material Day 3 Day 10 Day 21 CE K PDMS_F 39465 58130 49725 CE L SPCU_F 32306 43429 58983 CE M PCU_F 27778 35132 45324 CE N PP_F 57465 62681 67889 1i SPCU_1 26642 43557 54571 1j PCU_1 18026 21083 48580 1k PP_1 53515 91454 101282 2i SPCU_2 31572 62264 72934 2j PCU_2 12788 28737 40926 2k PP_2 46328 95592 96663

[0194] Fibroblast viability was measured after culture for 7 days using the PrestoBlue assay (Table E). During the culture period, fibroblasts were exposed to macrophage-derived conditioned medium (MP CM) as described here above in the methods. Cell viability is considered to be a proxy measure of cell proliferation because higher cell viability correlates with higher cell proliferation. After treatment with MP CM derived from microstructured polymer surfaces, mean fibroblast viability was generally higher than from MP CM treatment derived from flat surfaces. This effect was evident for all polymer groups, illustrating a conserved response irrespective of material chemistry. This effect was also consistent for MP CM collected at various macrophage culture time points, e.g. after 3 days, 10 days, 17 days, or 21 days. The promotive effect on fibroblast viability was most pronounced and consistently evident from MP CM treatment derived from TCP1 microstructured polymer surfaces (e.g. SPCU_1, PCU_1, and PP 1). Fibroblast viability generally also increased following treatment with MP CM derived from TCP2 microstructured polymer surfaces (e.g. SPCU_2, PCU_2, and PP 2) versus MP CM derived from flat polymer surfaces (e.g. SPCU_F, PCU_F, and PP_F), although to a lesser extent. While fibroblast viability increases following treatment with MP CM derived from microstructured surfaces, a positive effect on capsular contracture can be expected based on the literature. These results point to a protective paracrine effect elicited by macrophages cultured on microstructured surfaces, with increasing potency as the scale of microstructural features decreases (e.g. TCP1 vs. TCP2).

TABLE-US-00005 TABLE E Fibroblast viability as measured by the PrestoBlue assay (values: mean relative fluorescent units) Example Material Day 3 Day 10 Day 17/21 CE O SPCU_F 78723 49419 69009 CE P PCU_F 126561 120306 113934 CE Q PP_F 126111 125663 117034 1l SPCU_1 98344 97572 105326 1m PCU_1 132957 133250 123498 1n PP_1 138106 136937 121054 2l SPCU_2 75142 73229 91261 2m PCU_2 132798 124230 119085 2n PP_2 133313 129640 121162