INJECTABLE SYNTHETIC PUR COMPOSITE

RELATED APPLICATIONS

[0001] This application claims priority from U.S. Provisional Application Serial Nos. 61/406,098, filed October 22, 2010, and 61/433,944, filed January 18, 201 1 , the entire disclosures of which are incorporated herein by this reference.

GOVERNMENT SUPPORT

[0002J This invention was made with support from the Rutgers-Cleveland Clinic

Consortium in the Armed Forces Institute of Regenerative Medicine, which is funded by Department of Defense (W81XWH-08-2-0034). This work was also supported by the National Science Foundation through a CAREER award to SAG (DMR084771 1 ), and by the Center for Military Biomaterials through the Department of Defense (W81XWH-04-2-0031).

[0003] The United State Government has certain rights to this invention.

TECHNICAL FIELD

[0004] Embodiments of the present invention include composites and methods of using the composites. Non-limiting examples include composites that comprise polyurethane, allograft, and, optionally, rhBMP-2, and methods for their use as functional biomaterials.

[0005] Other examples include putty and injectable bone void filler composites that can be used for bone healing and/or as weight-bearing applications.

[0006] The presently disclosed subject matter further relates to composites that comprise polyurethane, a synthetic substitute for allograft, and, optionally, rhBMP-2.

BACKGROUND

[0007] There is a compelling clinical need for functional biomaterials that are weight- bearing and actively remodel. For example, the treatment of tibial fractures is frequently complicated by delayed union and nonunion. The standard of care for treatment of displaced tibial plateau fractures (e.g., split and localized depression fractures) is internal fixation, which in some cases requires grafting with autologous bone to augment the internal fixation. Inadequate anatomical reduction of tibial plateau fractures has been associated with a high (30 - 80%) incidence of arthritic change in the knee. In order to eliminate the need for invasive internal fixation devices, the potential of calcium phosphate bone cements to maintain anatomical reduction of tibial plateau fractures has been investigated. In a retrospective analysis of 26 patients, 61% of patients treated with buttress plating and bone grafting experienced loss of reduction after one year compared to 23% of patients treated with calcium phosphate cement. Thus the bone cement preserved anatomical reduction, presumably due to its compressive strength exceeding that of the trabecular bone in the tibial plateau. However, the cement is not bio functional, since it does not extensively remodel and is not replaced by new bone.

[0008] Osteonecrosis of the femoral head, which typically leads to hip replacement at a young age (<40 years) and afflicts -15,000 new patients each year, is another orthopaedic condition where treatment with functional biomaterials could improve patient outcomes. Hip replacement outcomes are not satisfactory, with failure rates ranging from 10 - 50% after five years. Non-invasive techniques, such as core decompression and nonvascularized bone grafting, have been used to treat early-stage osteonecrosis before collapse of the femoral head necessitates hip replacement. However, the results are varied with a 60 - 80% success rate, and outcomes are generally better in patients with very early-stage disease. Therefore, a non-invasive therapy accomplishing more predictable outcomes is desirable.

[0009] Injectable, functionally weight-bearing biomaterials that both possess initial mechanical strength comparable to host bone and maintain their initial strength while actively remodeling to form new bone would transform clinical management of a number of orthopaedic conditions. Such a weight-bearing and biologically active biomaterial was not achievable using current technology, recommending a realistic near-term approach that separates the biological provisions for acceleration of bone wound healing from the provisions for function.

[0010] Injectable, functionally weight-bearing biomaterials are not currently available. Functionally weight-bearing biomaterials for treatment of bone defects ideally possess five qualities: ( 1 ) biocompatibility of the material and its breakdown products, (2) injectability to enable less invasive application and fill irregularly shaped defects, (3) weight-bearing properties with strength comparable to that of healthy host bone at the defect site, (4) support of rapid cellular infiltration and remodeling at a rate that does not inhibit bone repair, and (5) delivery of biologies with proper release kinetics to accelerate bone formation and remodeling.

Commercially available injectable materials marketed as bone void fillers include calcium phosphate-based bone cements, which are osteoconductive, have compressive strengths comparable to trabecular bone (e.g., 5 - 40 MPa), and have fast setting times (<15 min).

[0011] For instance, currently calcium phosphates are subject to brittle fracture and graft migration, potentially leading to infections and requiring additional surgeries for repair or removal. To accelerate cellular infiltration and remodeling, implantable scaffolds with interconnected pores have been extensively investigated. While interconnected pores have long been considered a prerequisite for integration of bone within a bio material, the pores

significantly diminish the initial load-bearing properties of the materials, rendering them largely unsuitable for weight-bearing devices. Resorbable polymers have been blended with ceramics to yield weight-bearing composite implants that integrate and resorb. However, these materials incorporated relatively low (e.g., 5 - 20 vol%) volumes of ceramic particles, which were completely embedded in the polymer matrix. The rate of remodeling was slow (<30% bony ingrowth after 4 years in a rabbit IM rod model) and scaled with the rate of polymer degradation. Furthermore, the particle size of the HA was generally <20 μιη, which is below the preferred size range for remodeling by creeping substitution.

[0012] There are remains a need for functional biomaterials that comprise synthetic allograft rather than bone allograft.

[0013] Thus while currently available biomaterials address individually the requirements of a functional weight-bearing biomaterial, there is no device available that possesses more than three of the five key characteristics.

DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 shows thin sections of allograft (64 vol%)/polyurethane composites implanted in bilateral femoral plug defects (5x1 1 mm) in rabbits stained with Sanderson's rapid bone stain. (A) Low magnification image showing cellular infiltration throughout the implant at 6 weeks. (B) - (C) Higher magnification images showing regions of residual allograft (A), cells (CI), residual polymer (P), and new bone formation (NB).

[0015] Figure 2 shows the Plexur™ technology platform. Plexur™ P and M, or weight- bearing injectable formulations.

[0016] Figure 3 shows the wet and dry compressive strength of of allograft/PUR pastes in accordance with embodiments of the present invention.

[0017] Figure 4 shows an embodiment of an allograft/polyurethane bone void filler injected into 15-mm calvarial defects in rabbits, illustrating that the bone void filler supports cellular infiltration and new bone formation.

[0018] Figure 5 shows μCT images of an explanted bone core with (left) and without (right) allograft/ PUR filling the defect. Top: Cross-sectional slices of the core. Bottom: 'cut-through' 3D renderings. Note that the composite is surrounded by host bone. Bone cores with the filled defect will be tested to determine the compressive strength of the composite after 6 and 12 weeks of remodeling.

[0019] Figure 6 shows data for PUR scaffolds incorporating 60 μg/ml rhBMP2 implanted into 6-mm femoral segmental defects in rats support cellular infiltration and new bone formation. (A) In vitro release kinetics measured for PUR scaffolds and collagen sponge. (B) μΟΤ data show that PUR scaffolds exhibiting a burst followed by sustained release yield higher bone volume compared to a collagen sponge (burst release) and PUR scaffolds without the burst release. Blue: 4 weeks, Red: 8 weeks. (C) 1.25X and (D) 20X images of PUPJBMP-2 histological sections stained with trichrome show formation of new blood vessels (BV) and bone (NB) at 8 weeks.

[0020] Figure 7 shows images of allograft/PUR + 420 g/ml rhBMP2 injected into 15-mm calvarial defects in rabbits supports new bone formation. NB: new bone, OB: osteoblasts, OC: osteoclasts, BV: blood vessels. In vitro release kinetics shows -20% release of BMP2 at 25 days.

[0021] Figure 8 shows (A) compressive properties and (B) density and swelling of allograft/PUR composites. [0022] Figure 9 shows data of the CATn analysis from a rabbit CSD calvaria study of PUR composite embodiments of the present invention.

[0023] Figure 10 shows μΟΤ images of allograft/PUR putty carrying rhBMP-2 injected into a 6x1 1 mm plug defect in the femoral condyle of NZW rabbits. Residual allograft particles are distinguished by their irregular shape and sharp corners.

[0024] Figure 11 shows (A) μΟΤ images of allograft/PUR composites injected into 6x1 1 mm plug defects in the femoral condyle of NZW rabbits. Residual allograft particles are distinguished by their irregular shape and sharp corners. (B) Storage (G', left axis, open circles) and loss (G", right axis, filled circles) shear moduli measured as a function of time for the injectable porous allograft/PUR composite, (working time = G' and G" intersection).

[0025] Figure 12 shows allograft/PUR + 80 μg/ml rhBMP2 injected into 15-mm calvarial defects in rabbits supports new bone formation. NB: new bone, OB: osteoblasts, OC: osteoclasts, BV: blood vessels.

[0026] Figure 13 shows the design of a NZW rabbit calvarial CSD study. (A) Table listing the study design. (B) Illustration of the rabbit calvarium showing the location of the defect. (C) Photograph of the empty defect. (D) Photographs of the CPC during injection (Dl) and cure (D2). (E) Photographs of the biocomposite during injection (El) and cure (E2).

[0027J Figure 14 shows rheological data measured for the non-setting form of embodiments of biocomposite to characterize the injectability. (A) Shear stress versus shear rate. Data were fitted to the Casson model (solid line) used to predict the rheological properties of solid-filled suspensions and to calculate the yield stress (arrow). (B) Viscosity versus shear rate.

[0028] Figure 15 shows stress-strain curves for an embodied biocomposite and CPC measured under compressive loads. The area under the curve represents the energy-to-failure of the material.

[0029] Figure 16 shows radiographs of the empty defect, CPC, biocomposite, and biocomposite + rhBMP-2 at 6 and 12 weeks.

[0030] Figure 17 shows the quantitative analysis new bone formation by analysis of radiographs for each treatment group at 6 weeks. (A) Relative density of the defect compared to the host bone. (B) Percentage area mineralized material in the defect. [0031J Figure 18 shows histological sections of (A) an empty defect and (B) a CPC-treated defect.

[0032] Figure 19 shows histological sections of the biocomposites at (A-C) 6 and (D-F) 12 weeks. (A&D) Low magnification (1.6X) image of the complete defect and host bone. (C&F) High magnification (18.4X) image showing blood vessels (BV), osteoblasts (OB) and osteoid (O), new bone (NB), and residual polymer (P).

[0033] Figure 20 shows histological sections of the biocomposites incorporating rhBMP-2. (A) Low magnification (1.6X) image of the complete defect and host bone. (B) High

magnification (18.4X) image showing blood vessels (BV), osteoblasts (OB) and osteoid (O), osteocytes (OC), new bone (NB), and cartilage (C). (C) High magnification (10X) image of a region near the upper surface of the biocomposite showing residual polymer (P), residual allograft particles (A), and new bone (NB).

[0034] Figure 21 shows histomorphometric analysis of calvarial defects. (A) Total bone (allograft and new bone) measured in the entire defect volume. (B) Image and schematic showing area of interest for high-magnification histomorphometric analysis required to distinguish allograft from new bone. (C) New bone, allograft, and polymer measured in the three representative areas progressing from the edge to the interior of the defect. New bone is significantly different (#) in Areas 2 (p < .03) and 3 (p < .02) for all treatment groups. Remaining polymer is significantly less (*) for the biocomposite at 12 weeks than at 6 weeks in Area 1 (p < .03).

[0035] Figure 22 shows stress-strain curves for the embodied biocomposites (BC) and calcium phosphate cement. (A) Compression. (B) Torsion.

[0036] Figure 23 shows μCT images of the empty defects and defects filled with the allograft bone particles, BC, BC + BMP-L, and BC + BMP-H at 6 and 12 weeks.

[0037] Figure 24 shows the analysis of total bone (BV/TV) in 36-mm axial cross sections as a function of distance from the cortex measured by μΟΤ. (A) 6 and (B) 12 weeks

[0038] Figure 25 shows low- (1.25x) and high- (20x) magnification images of histological sections of the BC, BC + BMP-L, and BC + BMP-H-treated defects at (A) 6 weeks and (B) 12 weeks. [0039] Figure 26 shows data of a histomorphometric evaluation of biocomposite-treated defects. (A) Area % new bone, (B) area % residual allograft, and (C) area % total bone as functions of rhBMP-2 dose at 6 and 12 weeks.

[0040] Figure 27 shows SEM images of PUR/p-TCP composites embodiments (A) in vitro and B) in vivo.

[0041] Figure 28 shows Faxitron data at 4 weeks for embodiments of PUR/p-TCP composites A) without rhBMP-2 B) with rhBMP-2.

[0042] Figure 29 shows μϋΤ data at 4 weeks for embodiments of PUR/p-TCP composites A) without rhBMP-2 B) with rhBMP-2.

[0043] Figure 30 shows μΟΤ data for various embodiments of the present invention comprising small (< 106 μπι) and large (106-500 μπι) mineralized bone particles.

DESCRIPTION OF EXEMPLARY EMBODIMENTS

[0044] Embodiments of the present invention include PUR/rhBMP-2/synthetic allograft composites. Further embodiments comprise other bioactive agents, aside from rhBMP-2 and various types of surface unmodified and modified synthetic allografts. More generally, the presently disclosed subject matter generally relates to polyurethane composites possess characteristics of functionally weight-bearing biomaterials. Two-component polyurethanes (PUR) are useful for injectable applications because they can be processed as a reactive liquid that subsequently cures in situ to form a solid composite. Furthermore, the isocyanate groups may react with proteins on the surface of the allograft particles to improve interfacial bonding. To accomplish the goal of fabricating a weight-bearing biomaterial that actively remodels, embodiments of the invention include compression-molded composites comprising allograft bone particles embedded in a two-component PUR binder. Thus, embodiments of composites in accordance with the present invention exhibit all five key characteristics of biocompatibility, injectability, weight-bearing properties, rapid cellular infiltration, and sustained release of biologies. [0045] One specific embodiment of the present invention is a biodegradable composite that comprises a polymer component such as polyurethane, an osteoconductive matrix; and an osteoinductor that includes a powdered growth factor.

[0046] In terms of the polymer component, synthetic polymers can be designed with properties targeted for a given clinical application. According to the present invention, polyurethanes (PUR) are a useful class of biomaterials due to the fact that they can be injectable or moldable as a reactive liquid that subsequently cures to form a porous composite. These materials also have tunable degradation rates, which are shown to be highly dependent on the choice of polyol and isocyanate components (Hafeman et ai, Pharmaceutical Research

2008;25( 10):2387-99; Storey et ai, J Poly Sci Pt A: Poly Chem 1994;32:2345-63; Skarja et ai, J App Poly Sci 2000;75: 1522-34). Polyurethanes have tunable mechanical properties, which can also be enhanced with the addition of bone particles and/or other components (Adhikari et ai, Biomaterials 2008;29:3762-70; Gorna et ai, J Biomed Mater Res Pt A 2003;67A(3):813-27) and exhibit elastomeric rather than brittle mechanical properties.

[0047] Polyurethanes can be made by reacting together the components of a two-component composition, one of which includes a polyisocyanate while the other includes a component having two or more hydroxyl groups {i.e., polyols) to react with the polyisocyanate. For example, U.S. Pat. No. 6,306,177, discloses a method for repairing a tissue site using

polyurethanes, the content of which is incorporated by reference.

[0048] It is to be understood that by "a two-component composition" it means a

composition comprising two essential types of polymer components. In some embodiments, such a composition may additionally comprise one or more other optional components.

[0049] In some embodiments, polyurethane is a polymer that has been rendered formable through combination of two liquid components (i.e. , a polyisocyanate prepolymer and a polyol). In some embodiments, a polyisocyanate prepolymer or a polyol may be a molecule with two or three isocyanate or hydroxyl groups respectively. In some embodiments, a polyisocyanate prepolymer or a polyol may have at least four isocyanate or hydroxyl groups respectively.

[0050] Synthesis of porous polyurethane results from a balance of two simultaneous reactions. Reactions, in some embodiments, are illustrated below in Scheme 1. One is a gelling reaction, where an isocyanates and a polyester polyol react to form urethane bonds. The one is a blowing reaction. An isocyanate can react with water to form carbon dioxide gas, which acts as a lowing agent to form pores of polyurethane foam. The relative rates of these reactions determine the scaffold morphology, working time, and setting time.

[0051] Exemplary gelling and blowing reactions in forming of polyurethane are shown in Scheme 1 below, where Ri , R2 and R3, for example, can be oligomers of caprolactone, lactide and glycolide respectively.

Blowing reaction

[0052] Biodegradable polyurethane scaffolds synthesized from aliphatic polyisocyanates been shown to degrade into non-toxic compounds and support cell attachment and proliferation in vitro. A variety of polyurethane polymers suitable for use in the present invention are known in the art, many of which are listed in commonly owned applications: U.S. Ser. No. 10/759,904 filed on January 16, 2004, entitled "Biodegradable polyurethanes and use thereof and published under No. 2005-0013793; U.S. Ser. No. 1 1/667,090 filed on November 5, 2005, entitled "Degradable polyurethane foams" and published under No. 2007-0299151 ; U.S. Ser. No.

12/298, 158 filed on April 24, 2006, entitled "Biodegradable polyurethanes" and published under No. 2009-0221784; all of which are incorporated herein by reference. Polyurethanes described in U.S. Ser. No. 1 1/336,127 filed on January 19, 2006 and published under No. 2006-0216323, which is entitled "Polyurethanes for Osteoimplants" and incorporated herein by reference, may be used in some embodiments of the present invention. [0053] Polyurethanes foams may be prepared by contacting an isocyanate-terminated prepolymer (component 1 , e.g, polyisocyanate prepolymer) with a hardener (component 2) that includes at least a polyol (e.g. , a polyester polyol) and water, a catalyst and optionally, a stabilizer, a porogen, PEG, etc. In some embodiments, multiple polyurethanes (e.g., different structures, difference molecular weights) may be used in a composite/composition of the present invention. In some embodiments, other biocompatible and/or biodegradable polymers may be used with polyurethanes in accordance with the present invention. In some embodiments, biocompatible co-polymers and/or polymer blends of any combination thereof may be exploited.

[0054] Polyurethane components used in preparing inventive composites may be selected from monomers, pre-polymers, oligomers, polymers, cross-linked polymers, partially

polymerized polymers, partially cross-linked polymers, and any combinations thereof. For example, a composition may include polyurethane precursors. In some embodiments, polyurethane precursors include polyisocyanates prepolymers and polyols. In certain

embodiments, polyisocyanates prepolymers may be prepared by reacting isocyanates with polyols. In certain embodiments, a polyol may include PEG.

[0055] Polyisocyanates or multi-isocyanate compounds for use in the present invention include aliphatic polyisocyanates. Exemplary aliphatic polyisocyanates include, but are not limited to, lysine diisocyanate, an alkyl ester of lysine diisocyanate (for example, a methyl ester or an ethyl ester), lysine triisocyanate, hexamethylene diisocyanate, isophorone diisocyanate (IPDI), 4,4'-dicyclohexylmethane diisocyanate (H12MDI), cyclohexyl diisocyanate, 2,2,4-(2,2,4)- trimethylhexamethylene diisocyanate (TMDI), dimers prepared form aliphatic polyisocyanates, trimers prepared from aliphatic polyisocyanates and/or mixtures thereof. In some embodiments, hexamethylene diisocyanate (HDI) trimer sold as Desmodur N3300A may be a polyisocyanate utilized in the present invention.

[0056] In some embodiments, polyols are polyester polyols. In some embodiments, polyester polyols may include poly(ethylene adipate), poly(ethylene glutarate), poly(ethylene azelate), poly(trimethylene glutarate), poly(pentamethylene glutarate), poly(diethylene glutarate), poly(diethylene adipate), poly(triethylene adipate), poly( l ,2-propylene adipate), mixtures thereof, and/or copolymers thereof. In some embodiments, polyester polyols can include, polyesters prepared from caprolactone, glycolide, D, L-lactide, mixtures thereof, and/or copolymers thereof. In some embodiments, polyester polyols can, for example, include polyesters prepared from castor-oil.

[0057] Polyurethanes used in accordance with the present invention can be adjusted to produce polymers having various physiochemical properties and morphologies including, for example, flexible foams, rigid foams, elastomers, coatings, adhesives, and sealants. The properties of polyurethanes are controlled by choice of the raw materials and their relative concentrations. For example, thermoplastic elastomers are characterized by a low degree of cross-linking and are typically segmented polymers, consisting of alternating hard (diisocyanates and chain extenders) and soft (polyols) segments. Thermoplastic elastomers are formed from the reaction of diisocyanates with long-chain diols and short-chain diol or diamine chain extenders. In some embodiments, pores in bone/polyurethanes composites in the present invention are interconnected and have a diameter ranging from approximately 50 to approximately 1000 microns.

[0058] Polyurethane prepolymers can be prepared by contacting a polyol with an excess (typically a large excess) of a polyisocyanate. The resulting prepolymer intermediate includes an adduct of polyisocyanates and polyols solubilized in an excess of polyisocyanates. Prepolymer can, in some embodiments, be formed by using an approximately stoichiometric amount of polyisocyanates in forming a prepolymer and subsequently adding additional polyisocyanates. The prepolymer therefore exhibits both low viscosity, which facilitates processing, and improved miscibility as a result of the polyisocyanate-polyol adduct. Polyurethane networks can, for example, then be prepared by reactive liquid molding, wherein the prepolymer is contacted with a polyester polyol to form a reactive liquid mixture (i.e., a two-component composition) which is then cast into a mold and cured.

[0059] Polyisocyanates or multi-isocyanate compounds for use in the present invention include aliphatic polyisocyanates. Exemplary aliphatic polyisocyanates include, but are not limited to, lysine diisocyanate, an alkyl ester of lysine diisocyanate (for example, the methyl ester or the ethyl ester), lysine triisocyanate, hexamethylene diisocyanate, isophorone diisocyanate (IPDI), 4,4'-dicyclohexylmethane diisocyanate (H ₁₂ MDI), cyclohexyl diisocyanate, 2,2,4-(2,2,4)-trimethylhexamethylene diisocyanate (TMDI), dimers prepared form aliphatic polyisocyanates, trimers prepared from aliphatic polyisocyanates and/or mixtures thereof. In some embodiments, hexamethylene diisocyanate (HDI) trimer sold as Desmodur N3300A may be a polyisocyanate utilized in the present invention. In some embodiments, polyisocyanates used in the present invention includes approximately 10 to 55% NCO by weight (wt %

NCO=100*(42/Mw)). In some embodiments, polyisocyanates include approximately 15 to 50% NCO.

[0060] Polyisocyanate prepolymers provide an additional degree of control over the structure of biodegradable polyurethanes. Prepared by reacting polyols with isocyanates, NCO- terminated prepolymers are oligomeric intermediates with isocyanate functionality as shown in Scheme 1. To increase reaction rates, urethane catalysts (e.g., tertiary amines) and/or elevated temperatures (60-90 °C) may be used (see, Guelcher, Tissue Engineering: Part B, 14 (1) 2008, pp 3-17).

[0061] Polyols used to react with polyisocyanates in preparation of NCO-terminated prepolymers refer to molecules having at least two functional groups to react with isocyanate groups. In some embodiments, polyols have a molecular weight of no more than 1000 g/mol. In some embodiments, polyols have a rang of molecular weight between about 100 g/mol to about 500 g/mol. In some embodiments, polyols have a rang of molecular weight between about 200 g/mol to about 400 g/mol. In certain embodiments, polyols (e.g., PEG) have a molecular weight of about 200 g/mol. Exemplary polyols include, but are not limited to, PEG, glycerol, pentaerythritol, dipentaerythritol, tripentaerythritol, 1 ,2,4-butanetriol, trimethylolpropane, 1 ,2,3- trihydroxyhexane, myo-inositol, ascorbic acid, a saccharide, or sugar alcohols (e.g., mannitol, xylitol, sorbitol etc.). In some embodiments, polyols may comprise multiple chemical entities having reactive hydrogen functional groups (e.g. , hydroxy groups, primary amine groups and/or secondary amine groups) to react with the isocyanate functionality of polyisocyanates.

[0062] In some embodiments, polyisocyanate prepolymers are resorbable. Zhang and coworkers synthesized biodegradable lysine diisocyanate ethyl ester (LDI)/glucose polyurethane foams proposed for tissue engineering applications. In those studies, NCO-terminated prepolymers were prepared from LDI and glucose. The prepolymers were chain-extended with water to yield biocompatible foams which supported the growth of rabbit bone marrow stromal cells in vitro and were non- immunogenic in vivo, (see Zhang, et al., Biomaterials 21 : 1247- 1258 (2000), and Zhang, et al , Tiss. Eng. , 8(5): 771 -785 (2002), both of which are incorporated herein by reference).

[0063] In some embodiments, prepared polyisocyanate prepolymer can be a flowable liquid at processing conditions. In general, the processing temperature is no greater than 60 °C. In some embodiments, the processing temperature is ambient temperature (25 °C).

[0064] Polyols utilized in accordance with the present invention can be amine- and/or hydroxy 1- terminated compounds and include, but are not limited to, polyether polyols (such as polyethylene glycol (PEG or PEO), polytetramethylene etherglycol (PTMEG), polypropylene oxide glycol (PPO)); amine-terminated polyethers; polyester polyols (such as polybutylene adipate, caprolactone polyesters, castor oil); and polycarbonates (such as poly(l ,6-hexanediol) carbonate). In some embodiments, polyols may be (1 ) molecules having multiple hydroxyl or amine functionality, such as glucose, polysaccharides, and castor oil; and (2) molecules (such as fatty acids, triglycerides, and phospholipids) that have been hydroxylated by known chemical synthesis techniques to yield polyols.

[0065] Polyols used in the present invention may be polyester polyols. In some

embodiments, polyester polyols may include polyalkylene glycol esters or polyesters prepared from cyclic esters. In some embodiments, polyester polyols may include poly(ethylene adipate), poly(ethylene glutarate), poly(ethylene azelate), poly(trimethylene glutarate),

poly(pentamethylene glutarate), poly(diethylene glutarate), poly(diethylene adipate), poly(triethylene adipate), poly( l ,2-propylene adipate), mixtures thereof, and/or copolymers thereof. In some embodiments, polyester polyols can include, polyesters prepared from caprolactone, glycolide, D, L-lactide, mixtures thereof, and/or copolymers thereof. In some embodiments, polyester polyols can, for example, include polyesters prepared from castor-oil. When polyurethanes degrade, their degradation products can be the polyols from which they were prepared from.

[0066] In some embodiments, polyester polyols can be miscible with prepared prepolymers used in reactive liquid mixtures {i.e. , two-component composition) of the present invention. In some embodiments, surfactants or other additives may be included in the reactive liquid mixtures to help homogenous mixing.

[0067) The glass transition temperature (Tg) of polyester polyols used in the reactive liquids to form polyurethanes can be less than 60 °C, less than 37 °C (approximately human body temperature) or even less than 25 °C. In addition to affecting flowability at processing conditions, Tg can also affect degradation. In general, a Tg of greater than approximately 37 °C will result in slower degradation within the body, while a Tg below approximately 37 °C will result in faster degradation.

[0068] Molecular weight of polyester polyols used in the reactive liquids to form

polyurethanes can, for example, be adjusted to control the mechanical properties of

polyurethanes utilized in accordance with the present invention. In that regard, using polyester polyols of higher molecular weight results in greater compliance or elasticity. In some embodiments, polyester polyols used in the reactive liquids may have a molecular weight less than approximately 3000 Da. In certain embodiments, the molecular weight may be in the range of approximately 200 to 2500 Da or 300 to 2000 Da. In some embodiments, the molecular weight may be approximately in the range of approximately 450 to 1800 Da or 450 to 1200 Da.

[0069] In some embodiments, a polyester polyol comprise poly(caprolactone-co-lactide-co- glycolide), which has a molecular weight in a range of 200 Da to 2500 Da, or 300 Da to 2000 Da.

[0070] In some embodiments, polyols may include multiply types of polyols with different structures, molecular weight, properties, etc.

[0071] Additional Components. In accordance with the present invention, two-component compositions {i.e., polyprepolymers and polyols) to form porous composites may be used with other agents and/or catalysts. Zhang et al. have found that water may be an adequate blowing agent for a lysine diisocyanate/PEG/glycerol polyurethane (see Zhang, et al. , Tissue Eng. 2003 (6): 1 143-57) and may also be used to form porous structures in polyurethanes. Other blowing agents include dry ice or other agents that release carbon dioxide or other gases into the composite. Alternatively, or in addition, porogens (see detail discussion below) such as salts may be mixed in with reagents and then dissolved after polymerization to leave behind small voids.

[0072] Two-component compositions and/or the prepared composites used in the present invention may include one or more additional components. In some embodiments, inventive compositions and/or composites may includes, water, a catalyst (e.g. , gelling catalyst, blowing catalyst, etc.), a stabilizer, a plasticizer, a porogen, a chain extender (for making of

polyurethanes), a pore opener (such as calcium stearate, to control pore morphology), a wetting or lubricating agent, etc. (See, U.S. Ser. No. 10/759,904 published under No. 2005-0013793, and U.S. Ser. No. 1 1/625, 1 19 published under No. 2007-0191963; both of which are incorporated herein by reference).

[0073] In some embodiments, inventive compositions and/or composites may include and/or be combined with a solid filler (e.g., carboxymethylcellulose (CMC) and hyaluronic acid (HA)). For example, when composites used in wound healing, solid fillers can help absorb excess moisture in the wounds from blood and serum and allow for proper foaming.

[0074] In certain embodiments, additional biocompatible polymers (e.g., PEG) or copolymers can be used with compositions and composites in the present invention.

[0075] Water may be a blowing agent to generate porous polyurethane-based composites. Porosity of bone/polymer composites increased with increasing water content, and

biodegradation rate accelerated with decreasing polyester half-life, thereby yielding a family of materials with tunable properties that are usefull in the present invention. See, Guelcher et al., Tissue Engineering, 13(9), 2007, pp2321-2333, which is incorporated by reference.

[0076] In some embodiments, an amount of water is about 0.5, 1 , 1.5, 2, 3, 4 5, 6, 7, 8, 9, 10 parts per hundred parts (pphp) polyol. In some embodiments, water has an approximate rang of any of such amounts.

[0077] In some embodiments, at least one catalyst is added to form reactive liquid mixture (i.e., two-component compositions). A catalyst, for example, can be non-toxic (in a

concentration that may remain in the polymer).

[0078] A catalyst can, for example, be present in two-component compositions in a concentration in the range of approximately 0.5 to 5 parts per hundred parts polyol (pphp) and, for example, in the range of approximately 0.5 to 2, or 2 to 3 pphp. A catalyst can, for example, be an amine compound. In some embodiments, catalyst may be an organometallic compound or a tertiary amine compound. In some embodiments the catalyst may be stannous octoate (an organobismuth compound), triethylene diamine, bis(dimethylaminoethyl)ether,

dimethylethanolamine, dibutyltin dilaurate, and Coscat organometallic catalysts manufactured by Vertullus (a bismuth based catalyst), or any combination thereof.

[0079] In some embodiments, a stabilizer is nontoxic (in a concentration remaining in the polyurethane foam) and can include a non-ionic surfactant, an anionic surfactant or combinations thereof. For example, a stabilizer can be a polyethersiloxane, a salt of a fatty sulfonic acid or a salt of a fatty acid. In certain embodiments, a stabilizer is a polyethersiloxane, and the concentration of polyethersiloxane in a reactive liquid mixture can, for example, be in the range of approximately 0.25 to 4 parts per hundred polyol. In some embodiments, polyethersiloxane stabilizer are hydro lyzable.

[0080] In some embodiments, the stabilizer can be a salt of a fatty sulfonic acid.

Concentration of a salt of the fatty sulfonic acid in a reactive liquid mixture can be in the range of approximately 0.5 to 5 parts per hundred polyol. Examples of suitable stabilizers include a sulfated castor oil or sodium ricinoleicsulfonate.

[0081] Stabilizers can be added to a reactive liquid mixture of the present invention to, for example, disperse prepolymers, polyols and other additional components, stabilize the rising carbon dioxide bubbles, and/or control pore sizes of inventive composites. Although there has been a great deal of study of stabilizers, the operation of stabilizers during foaming is not completely understood. Without limitation to any mechanism of operation, it is believed that stabilizers preserve the thermodynamically unstable state of a polyurethane foam during the time of rising by surface forces until the foam is hardened. In that regard, foam stabilizers lower the surface tension of the mixture of starting materials and operate as emulsifiers for the system. Stabilizers, catalysts and other polyurethane reaction components are discussed, for example, in Oertel, Gunter, ed., Polyurethane Handbook, Hanser Gardner Publications, Inc. Cincinnati, Ohio, 99-108 (1994). A specific effect of stabilizers is believed to be the formation of surfactant monolayers at the interface of higher viscosity of bulk phase, thereby increasing the elasticity of surface and stabilizing expanding foam bubbles.

[0082] To prepare high- molecular-weight polymers, prepolymers are chain extended by adding a short-chain (e.g., <500 g/mol) polyamine or polyol. In certain embodiments, water may act as a chain extender. In some embodiments, addition of chain extenders with a functionality of two (e.g. , diols and diamines) yields linear alternating block copolymers.

[0083] In some embodiments, inventive compositions and/or composites include one or more plasticizers. Plasticizers are typically compounds added to polymers or plastics to soften them or make them more pliable. According to the present invention, plasticizers soften, make workable, or otherwise improve the handling properties of polymers or composites. Plasticizers also allow inventive composites to be moldable at a lower temperature, thereby avoiding heat induced tissue necrosis during implantation. Plasticizer may evaporate or otherwise diffuse out of the composite over time, thereby allowing composites to harden or set. Without being bound to any theory, plasticizer are thought to work by embedding themselves between the chains of polymers. This forces polymer chains apart and thus lowers the glass transition temperature of polymers. In general, the more plasticizer added, the more flexible the resulting polymers or composites will be.

[0084] In some embodiments, plasticizers are based on an ester of a polycarboxylic acid with linear or branched aliphatic alcohols of moderate chain length. For example, some plasticizers are adipate-based. Examples of adipate-based plasticizers include bis(2- ethylhexyl)adipate (DOA), dimethyl adipate (DMAD), monomethyl adipate (MMAD), and dioctyl adipate (DOA). Other plasticizers are based on maleates, sebacates, or citrates such as bibutyl maleate (DBM), diisobutylmaleate (DIBM), dibutyl sebacate (DBS), triethyl citrate (TEC), acetyl triethyl citrate (ATEC), tributyl citrate (TBC), acetyl tributyl citrate (ATBC), trioctyl citrate (TOC), acetyl trioctyl citrate (ATOC), trihexyl citrate (THC), acetyl trihexyl citrate (ATHC), butyryl trihexyl citrate (BTHC), and trimethylcitrate (TMC). Other plasticizers are phthalate based. Examples of phthalate-based plasticizers are N-methyl phthalate, bis(2- ethylhexyl) phthalate (DEHP), diisononyl phthalate (DINP), bis(n-butyl)phthalate (DBP), butyl benzyl phthalate (BBzP), diisodecyl phthalate (DOP), diethyl phthalate (DEP), diisobutyl phthalate (DIBP), and di-n-hexyl phthalate. Other suitable plasticizers include liquid polyhydroxy compounds such as glycerol, polyethylene glycol (PEG), triethylene glycol, sorbitol, monacetin, diacetin, and mixtures thereof. Other plasticizers include trimellitates (e.g. , trimethyl trimellitate (TMTM), tri-(2-ethylhexyl) trimellitate (TEHTM-MG), tri-(n-octyl,n- decyl) trimellitate (ATM), tri-(heptyl,nonyl) trimellitate (LTM), n-octyl trimellitate (OTM)), benzoates, epoxidized vegetable oils, sulfonamides (e.g. , N-ethyl toluene sulfonamide (ETSA), N-(2-hydroxypropyl) benzene sulfonamide (HP BSA), N-(n-butyl) butyl sulfonamide (BBSA- NBBS)), organophosphates (e.g. , tricresyl phosphate (TCP), tributyl phosphate (TBP)), glycols/polyethers (e.g., triethylene glycol dihexanoate, tetraethylene glycol diheptanoate), and polymeric plasticizers. Other plasticizers are described in Handbook of Plasticizers (G. Wypych, Ed., ChemTec Publishing, 2004), which is incorporated herein by reference. In certain embodiments, other polymers are added to the composite as plasticizers. In certain particular embodiments, polymers with the same chemical structure as those used in the composite are used but with lower molecular weights to soften the overall composite. In other embodiments, different polymers with lower melting points and/or lower viscosities than those of the polymer component of the composite are used.

[0085] In some embodiments, polymers used as plasticizer are poly(ethylene glycol) (PEG). PEG used as a plasticizer is typically a low molecular weight PEG such as those having an average molecular weight of 1000 to 10000 g/mol, for example, from 4000 to 8000 g/mol. In certain embodiments, PEG 4000, PEG 5000, PEG 6000, PEG 7000, PEG 8000 or combinations thereof are used in inventive composites. For example, plasticizer (PEG) is useful in making more moldable composites that include poly(lactide), poly(D,L-lactide), poly(lactide-co- glycolide), poly(D,L-lactide-co-glycolide), or poly(caprolactone). Plasticizer may comprise 1 - 40% of inventive composites by weight. In some embodiments, the plasticizer is 10-30% by weight. In some embodiments, the plasticizer is approximately 10%, 15%, 20%, 25%, 30% or 40% by weight. In other embodiments, a plasticizer is not used in the composite. For example, in some polycaprolactone-containing composites, a plasticizer is not used.

[0086] In some embodiments, inert plasticizers may be used. In some embodiments, a plasticizer may not be used in the present invention. [0087] Porosity of inventive composites may be accomplished using any means known in the art. Exemplary methods of creating porosity in a composite include, but are not limited to, particular leaching processes, gas foaming processing, supercritical carbon dioxide processing, sintering, phase transformation, freeze-drying, cross-linking, molding, porogen melting, polymerization, melt-blowing, and salt fusion (Murphy et ai, Tissue Engineering 8(l):43-52, 2002; incorporated herein by reference). For a review, see Karageorgiou et al, Biomaterials 26:5474-5491 , 2005; incorporated herein by reference. Porosity may be a feature of inventive composites during manufacture or before implantation, or porosity may only be available after implantation. For example, a implanted composite may include latent pores. These latent pores may arise from including porogens in the composite.

[0088] Porogens may be any chemical compound that will reserve a space within the composite while the composite is being molded and will diffuse, dissolve, and/or degrade prior to or after implantation or injection leaving a pore in the composite. Porogens may have the property of not being appreciably changed in shape and/or size during the procedure to make the composite moldable. For example, a porogen should retain its shape during the heating of the composite to make it moldable. Therefore, a porogen does not melt upon heating of the composite to make it moldable. In certain embodiments, a porogen has a melting point greater than about 60 °C, greater than about 70 °C, greater than about 80 °C, greater than about 85 °C, or greater than about 90 °C.

[0089] Porogens may be of any shape or size. A porogen may be spheroidal, cuboidal, rectangular, elonganted, tubular, fibrous, disc-shaped, platelet-shaped, polygonal, etc. In certain embodiments, the porogen is granular with a diameter ranging from approximately 100 microns to approximately 800 microns. In certain embodiments, a porogen is elongated, tubular, or fibrous. Such porogens provide increased connectivity of pores of inventive composite and/or also allow for a lesser percentage of the porogen in the composite.

[0090] Amount of porogens may vary in inventive composite from 1% to 80% by weight. In certain embodiments, the plasticizer makes up from about 5% to about 80% by weight of the composite. In certain embodiments, a plasticizer makes up from about 10% to about 50% by weight of the composite. Pores in inventive composites are thought to improve the osteoinductivity or osteoconductivity of the composite by providing holes for cells such as osteoblasts, osteoclasts, fibroblasts, cells of the osteoblast lineage, stem cells, etc. Pores provide inventive composites with biological in growth capacity. Pores may also provide for easier degradation of inventive composites as bone is formed and/or remodeled. In some embodiments, a porogen is biocompatible.

[0091] A porogen may be a gas, liquid, or solid. Exemplary gases that may act as porogens include carbon dioxide, nitrogen, argon, or air. Exemplary liquids include water, organic solvents, or biological fluids (e.g. , blood, lymph, plasma). Gaseous or liquid porogen may diffuse out of the osteoimplant before or after implantation thereby providing pores for biological in-growth. Solid porogens may be crystalline or amorphous. Examples of possible solid porogens include water soluble compounds. Exemplary porogens include carbohydrates {e.g., sorbitol, dextran (poly(dextrose)), starch), salts, sugar alcohols, natural polymers, synthetic polymers, and small molecules.

[0092] In some embodiments, carbohydrates are used as porogens in inventive composites. A carbohydrate may be a monosaccharide, disaccharide, or polysaccharide. The carbohydrate may be a natural or synthetic carbohydrate. In some embodiments, the carbohydrate is a biocompatible, biodegradable carbohydrate. In certain embodiments, the carbohydrate is a polysaccharide. Exemplary polysaccharides include cellulose, starch, amylose, dextran, poly(dextrose), glycogen, etc.

[0093] In certain embodiments, a polysaccharide is dextran. Very high molecular weight dextran has been found particularly useful as a porogen. For example, the molecular weight of the dextran may range from about 500,000 g/mol to about 10,000,000 g/mol, preferably from about 1 ,000,000 g/mol to about 3,000,000 g/mol. In certain embodiments, the dextran has a molecular weight of approximately 2,000,000 g/mol. Dextrans with a molecular weight higher than 10,000,000 g/mol may also be used as porogens. Dextran may be used in any form (e.g. , particles, granules, fibers, elongated fibers) as a porogen. In certain embodiments, fibers or elongated fibers of dextran are used as a porogen in inventive composites. Fibers of dextran may be formed using any known method including extrusion and precipitation. Fibers may be prepared by precipitation by adding an aqueous solution of dextran (e.g., 5-25% dextran) to a less polar solvent such as a 90- 100% alcohol (e.g. , ethanol) solution. The dextran precipitates out in fibers that are particularly useful as porogens in the inventive composite. Once the composite with dextran as a porogen is implanted into a subject, the dextran dissolves away very quickly. Within approximately 24 hours, substantially all of dextran is out of composites leaving behind pores in the osteoimplant composite. An advantage of using dextran in a composite is that dextran exhibits a hemostatic property in extravascular space. Therefore, dextran in a composite can decrease bleeding at or near the site of implantation.

[0094] Small molecules including pharmaceutical agents may also be used as porogens in the inventive composites. Examples of polymers that may be used as plasticizers include polyvinyl pyrollidone), pullulan, poly(glycolide), poly(lactide), and poly(lactide-co-glycolide). Typically low molecular weight polymers are used as porogens. In certain embodiments, a porogen is poly(vinyl pyrrolidone) or a derivative thereof. Plasticizers that are removed faster than the surrounding composite can also be considered porogens.

[0095] In some embodiments, the osteoconductive matrix may be a particulate material, inorganic material, synthetic materials including synthetic allografts, bone allografts, and combinations thereof. The elements that make up an osteoconductive matrix may not always be mutually exclusive. For example, particular material may also be inorganic material, and so forth.

[0096] The osteoconductive matrix may comprise additional particulate materials. These materials may be any type of additional components comprising inorganic materials and/or other bone substitute materials (i.e. , compositions similar to natural bone such as collagen, biocompatible polymers, osteoinductive agents, other commercial bone graft products, any composite graft, etc.), may be utilized in the present invention. Inorganic materials, including but not limited to, calcium phosphate materials, and other bone substitute materials, may also be exploited for use as particulate inclusions in the inventive composites. Exemplary materials utilized in accordance with the present invention include aragonite, dahlite, calcite, amorphous calcium carbonate, vaterite, weddellite, whewellite, struvite, urate, ferrihydrite, francolite, monohydrocalcite, magnetite, goethite, dentin, calcium carbonate, calcium sulfate, calcium phosphosilicate, sodium phosphate, calcium aluminate, calcium phosphate, hydroxyapatite, a- tricalcium phosphate, dicalcium phosphate, β-tricalcium phosphate, tetracalcium phosphate, amorphous calcium phosphate, octacalcium phosphate, and BIOGLASS™, a calcium phosphate silica glass available from U.S. Biomaterials Corporation. Substituted calcium phosphate phases are also contemplated for use with the invention, including but not limited to fluorapatite, chlorapatite, magnesium-substituted tricalcium phosphate, and carbonate hydroxyapatite. In certain embodiments, the inorganic material is a substituted form of hydroxyapatite. For example, hydroxyapatite may be substituted with other ions such as fluoride, chloride, magnesium, sodium, potassium, and groups such as silicates, silicon dioxides, carbonates, etc. Additional calcium phosphate phases suitable for use with the invention include those disclosed in U.S. Patents RE 33, 161 and RE 33,221 to Brown et al.; 4,880,610; 5,034,059; 5,047,031 ; 5,053,212; 5, 129,905; 5,336,264; and 6,002,065 to Constantz et al; 5, 149,368; 5,262, 166 and 5,462,722 to Liu et al ; 5,525, 148 and 5,542,973 to Chow et ai , 5,717,006 and 6,001 ,394 to Daculsi et ai , 5,605,713 to Boltong et al, 5,650,176 to Lee et al , and 6,206,957 to Driessens et al, and biologically-derived or biomimetic materials such as those identified in Lowenstam HA, Weiner S, On Biomineralization, Oxford University Press, 1989; each of which is incorporated herein by reference.

[0097] In certain embodiments the osteoconductive matrix may comprise particulate components that are bone particles. Any kind of bone and/or bone-derived particles may be used in the present invention. In some embodiments, bone particles employed in the preparation of bone particle-containing composites are obtained from cortical, cancellous, and/or

corticocancellous bone. Bone particles may be obtained from ariy vertebrate. Bone may be of autogenous, allogenic, and/or xenogeneic origin. In certain embodiments, bone particles are autogenous, that is, bone particles are from the subject being treated. In other embodiments, bone particles are allogenic {e.g. , from donors). In certain embodiments, the source of bone may be matched to the eventual recipient of inventive composites {i.e., the donor and recipient are of the same species). In certain embodiments, bone particles are obtained from cortical bone of allogenic origin. In certain embodiments, bone particles are obtained from bone of xenogeneic origin. Porcine and bovine bone are types of xenogeneic bone tissue that can be used individually or in combination as sources for bone particles and may offer advantageous properties. Xenogenic bone tissue may be combined with allogenic or autogenous bone.

[0098] U.S. Patent Publication No. 2010/01 12032 Al , the entire contents of which are incorporated herein by reference, describe various bone particles, their characteristics, and possible surface modifations that can be used with embodiments of the present invention.

(0099] Bone particles can be formed by any process known to break down bone into small pieces. Exemplary processes for forming such particles include milling whole bone to produce fibers, chipping whole bone, cutting whole bone, grinding whole bone, fracturing whole bone in liquid nitrogen, or otherwise disintegrating the bone. Bone particles can optionally be sieved to produce particles of a specific size range. Bone particles may be of any shape or size.

Exemplary shapes include spheroidal, plates, shards, fibers, cuboidal, sheets, rods, oval, strings, elongated particles, wedges, discs, rectangular, polyhedral, etc.

[0100] In some embodiments, bone particles have a medium or mean diameter about 1200 microns, 1 100 microns, 1000 microns, 900 microns, 800 microns, 700 microns, 600 microns, 500 microns, 400 microns, 300 microns, 200 microns, 100 microns, etc. In some embodiments, diameters of bone particles are within a range between any of such sizes. For example, medium or mean diameters of bone particles have a range from approximately 100 microns to approximately 1000 microns.

[0101] As for irregularly shaped bone particles, recited dimension ranges may represent the length of the greatest or smallest dimension of the particle. As examples, bone particles can be pin shaped, with tapered ends having an average diameter of from about 100 microns to about 500 microns. As will be appreciated by one of skill in the art, for injectable composites, the maximum particle size will depend in part on the size of the cannula or needle through which the material will be delivered.

[0102] In some embodiments, particle size distribution of bone particles utilized in accordance with the present inventions with respect to a mean value or a median value may be plus or minus, e.g. , about 10% or less of the mean value, about 20% or less of the mean value, about 30% or less of the mean value, about 40% or less of the mean value, about 50% or less of the mean value, about 60% or less of the mean value, about 70% or less of the mean value, about 80% or less of the mean value, or about 90% or less of the mean value.

[0103] In some embodiments, bone particles have a median or mean length of about 1200 microns, 1 100 microns, 1000 microns, 900 microns, 800 microns, 700 microns, 600 microns, 500 microns, 400 microns, 300 microns, 200 microns, 100 microns, etc. In some embodiments, about 70, about 80 or about 90 percent of bone particles possess a median or mean length within a range of any of such sizes.

[0104] For bone particles that are fibers or other elongated particles, in some embodiments, at least about 90 percent of the particles possess a median or mean length in their greatest dimension in a range from approximately 100 microns to approximately 1000 microns. Particles may possess a median or mean length to median or mean thickness ratio from at least about 5: 1 up to about 500: 1 , for example, from at least about 50: 1 up to about 500: 1 , or from about 50: 1 up to about 100: 1 ; and a median or mean length to median or mean width ratio of from about 10: 1 to about 200: 1 and, for example, from about 50: 1 to about 100: 1. In certain embodiments, bone particles are short fibers having a cross-section of about 300 microns to about 100 microns and a length of about 0.1 mm to about 1 mm.

[0105] Processing of bone to provide particles may be adjusted to optimize for the desired size and/or distribution of bone particles. The properties of resulting inventive composites {e.g. , mechanical properties) may also be engineered by adjusting weight percent, shapes, sizes, distribution, etc. of bone particles or other particles. For example, an inventive composite may be made more viscous and load bearing by including a higher percentage of particles.

[0106] U.S. Patents 5,899,939; 5,507,813; 6, 123,731 ; 6,294,041 ; 6,294, 187; 6,332,779; 6,440,444; and 6,478,825; the contents of all of which are incorporated herein by reference, describe methods for preparing composites including allogenic bone for use in orthopedic applications.

[0107] Bone particles utilized in accordance with the present inventions may be

demineralized, non-demineralized, mineralized, or anorganic. In some embodiments, bone particles are used "as is" in preparing inventive composites. In some embodiments, bone particles are defatted and disinfected. An exemplary defatting/disinfectant solution is an aqueous solution of ethanol. Other organic solvent may also be used in the defatting and disinfecting bone particles. For example, methanol, isopropanol, butanol, DMF, DMSO, diethyl ether, hexanes, glyme, tetrahydrofuran, chloroform, methylene chloride, and carbon tetrachloride may be used. In certain embodiments, a non-halogenated solvent is used. A defatting/disinfecant solution may also include a detergent (e.g., an aqueous solution of a detergent). Ordinarily, at least about 10 to about 40 percent by weight of water (i.e., about 60 to about 90 weight percent of defatting agent such as alcohol) should be present in the defatting/disinfecting solution to produce optimal lipid removal and disinfection within the shortest period of time. An exemplary concentration range of a defatting solution is from about 60 to about 85 weight percent alcohol, for example, about 70 weight percent alcohol.

[0108] In some embodiments, bone particles are demineralized. Bone particles can be optionally demineralized in accordance with known. and/or conventional procedures in order to reduce their inorganic mineral content.

[0109] In an exemplary defatting/disinfecting/demineralization procedure, bone particles are subjected to a defatting/disinfecting step, followed by an acid demineralization step. An exemplary defatting/disinfectant solution is an aqueous solution of ethanol.

[0110] In alternative embodiments, surfaces of bone particles may be lightly demineralized according to the procedures in our commonly owned U.S. Patent Application, U.S. S.N.

10/285,715, filed November 1 , 2002, published as U.S. Patent Publication No. 2003/0144743, on July 31 , 2003, the contents of which are incorporated herein by reference. In certain

embodiments, bone particles are subjected to a process that partially or totally removes their initial organic content to yield mineralized and anorganic bone particles, respectively. Different mineralization methods have been developed and are known in the are (Hurley, et al. , Milit. Med. 1957, 101 - 104; Kershaw, Pharm. J. 6:537, 1963; and U.S. Patent 4,882, 149; each of which is incorporated herein by reference).

[0111] Mixtures or combinations of one or more of the foregoing types of bone particles can be employed. For example, one or more of the foregoing types of demineralized bone particles can be employed in combination with non-demineralized bone particles, i.e., bone particles that have not been subjected to a demineralization process, or inorganic materials. The amount of each individual type of bone particle employed can vary widely depending on the mechanical and biological properties desired. Thus, in some embodiments, mixtures of bone particles of various shapes, sizes, and/or degrees of demineralization may be assembled based on the desired mechanical, thermal, chemical, and biological properties of a composite. A desired balance between the various properties of composites (e.g. , a balance between mechanical and biological properties) may be achieved by using different combinations of particles. Suitable amounts of various particle types can be readily determined by those skilled in the art on a case-by-case basis by routine experimentation.

[0112] The differential in strength, osteogenicity, and other properties between partially and fully demineralized bone particles on the one hand, and non-demineralized, superficially demineralized bone particles, inorganic ceramics, and other bone substitutes on the other hand can be exploited. For example, in order to increase the compressive strength of an osteoimplant, the ratio of nondemineralized and/or superficially demineralized bone particles to partially or fully demineralized bone particles may favor the former, and vice versa. Bone particles in composites also play a biological role. Non-demineralized bone particles bring about new bone in-growth by osteoconduction. Demineralized bone particles likewise play a biological role in bringing about new bone in-growth by osteoinduction. Both types of bone particles are gradually remodeled and replaced by new host bone as degradation of the composite progresses over time. Thus, the use of various types of bone particles can be used to control the overall mechanical and biological properties, (e.g. , strength, osteoconductivity, and/or osteoinductivity, etc.) of osteoimplants.

[0113] In some embodiments, the osteoconductive matrix includes blends. These blends include at least one of TCP, bioactive glass, aragonite, dahlite, calcite, amorphous calcium carbonate, vaterite, weddellite, whewellite, struvite, urate, ferrihydrite, franco lite,

monohydrocalcite, magnetite, goethite, dentin, calcium carbonate, calcium sulfate, calcium phosphosilicate, sodium phosphate, calcium aluminate, calcium phosphate, hydroxyapatite, a- tricalcium phosphate, dicalcium phosphate, tetracalcium phosphate, amorphous calcium phosphate, octacalcium phosphate (OCP), fluoroapatite, chloroapatite, magnesium-substituted tricalcium phosphate, carbonate hydroxyapatite, cortical-cancellous bone chips, and derivatives thereof.

[0114] The terms "putty", "injectable filler", "bone void filler", "moldable composition", and the like, as used herein, refer to the various embodiments of PUR composites. These composites may all be compositionally equivalent, and, for example, may comprise PUR, allograft and/or synthetic allograft, and a bioactive agent, such as rhBMP-2. For certain embodiments, there is a physical distinction between moldable or putty composites versus injectable or bone void filler composites.

[0115] Putty composites refer to composites that generally lend themselves to being moldable. Putties therefore tend to have a relatively higher initial viscocity, which is imparted by having a high relative concentration of allograft, synthetic allograft, or combinations thereof. While putties may be injected, one of ordinary skill in the art would not characterize them as being flowable, but instead would see them as being moldable composites that may be hand or machine molded to retain a particular shape around a bone injury site. Putty may also refer to composites that have relatively high osteoconductive solid particulate content (e.g., > 45wt% - 55wt%), such as allograft or synthetic allograft. Putties may be molded to retain a particular shape. Specific putties are also capable of functioning as weight-bearing composites.

[0116] On the other hand, bone void fillers, injectable composites, and the like generally lend themselves to being injected, for instance through a syringe, into or onto an injury site. Thus, bone void fillers may be injected into and swell to fill a bone injury site. Bone void fillers tend to have relatively low osteoconductive solid particulate content (e.g., < 45wt% - 55wt%).

[0117] Therefore, while the terms putty, bone void filler, and the like may be used interchangeably herein to refer to embodiments of composites of the present invention, the terms may at times also be used to describe the physical traits of a particular composite. That said, those of ordinary skill in the art will appreciate that terms such as putty and bone void filler refer to general characteristics of a composite, and there exists no hard line between the two.

Accordingly, certain embodiments of the present invention may be described as being both a putty and a bone void filler, since putties and bone void fillers are not necessarily mutually exclusive. [0118] Various embodiments of the present invention relate to composites for delivering osteotherapeutic material and/or delivery mechanisms for an osteotherapeutic material. Certain compositess according to various embodiments of the present invention may comprise mixtures of a physiologically acceptable biodegradable carrier, an osteoinductive material, and/or an osteoconductive material. The compositions may thus be applied to promote formation of new bone. Other embodiments of the present invention relate to the preparation of compositions and methods of using such compositions.

[0119] The term "osteotherapeutic material", as used herein, refers to a material that promotes bone growth, including, but are not limited to, osteoinductive, osteoconductive, osteogenic and osteopromotive materials. Further, osteotherapeutic materials, or factors, include: bone morphogenic protein ("BMP") such as BMP 2, BMP 4, and BMP 7 (OP 1); demineralized bone matrix ("DBM"), platelet-derived growth factor ("PDGF"); insulin-like growth factors I and II; fibroblast growth factors ("FGF's"); transforming growth factor beta ("TGF-beta."); platelet rich plasma (PRP); vescular endothelial growth factor (VEGF); growth hormones; small peptides; genes; stem cells, autologous bone, allogenic bone, bone marrow, biopolymers and bioceramics.

[0120] The term "osteoinductor", as used herein, refers to a material that has the capability of inducing ectopic bone formation. Osteoinductive materials include: DBM; BMP 2 (including rhBMP-2); BMP 4; and BMP 7.

[0121] The term "osteoconductor", as used herein, refers to a material that does not have the capability of ectopic bone formation, but provides the surface for the osteoblast cells to adhere, proliferate, and/or synthesize new bone. Osteoconductor(s) may form an osteoconductor matrix, which is a gathering of one or more types of osteoconductors. Osteoconductive materials include (but are not limited to): cortical-cancellous bone chips ("CCC"); hydroxy apatite ("HA");

tricalcium phosphate ("TCP"); bioactive glass such as Bioglass 45S5; mixtures of

HA/TCP/bioactive glass; other calcium phosphates; calcium carbonate; calcium sulfate;

collogen; DBM; other allograft material; and other synthetic allografts. One or more

osteoconductors may comprise an "resorbable osteoconductive matrix". [0122] The term "synthetic substitute", "synthetic allograft", and the like, as used herein, refer to any substances that can be used in the place of allograft bone in the PUR composites described herein. Examples of synthetic substitutes include tricalcium phosphate (TCP) and bioactive glasses, such as Bioglass 45 S5. Synthetic substitutes may be incorporated in a modified state, wherein they are chemically bound to a modifying substance, or in an unmodified state. For example, certain embodiments described herein utilize β-TCP in its unmodified state to fabricate PUR/TCP composites. Thus, the terms "osteoconductor", "synthetic allograft", and the like may in certain circumstances be used interchangeably herein.

[0123] The term "bioactive glass", as used herein, refers to a group of glass-ceramic biomaterials that may be surface reactive. Certain bioactive glasses, comprises Si0 ₂ , Na ₂ 0, CaO P2O5, and combinations thereof. An example of bioactive glass is Bioglass, including Bioglass 42S5. Various characteristics of bioactive glass make embodiments of bioactive glass suitable for use in PUR composites. For illustrative non-limiting purposes only, below are composition (wt%), structure, and index of bioactivity for various bioactive glasses.

[0124] Certain bioactive glass may comprise less than 60 mol% S1O ₂ , high Na20 and CaO content, and a high CaO/T^Os ratio.

[0125] The term "osteogenic factor", as used herein, refers to a material that supplies and supports the growth of bone healing cells. Osteogenic materials include autogenous cancellous bone, bone marrow, periosteum, and stem cells. Certain embodiments of the present invention may comprise osteogenic factor(s).

[0126] The term "sustained release", as used herein, refers to a substance, such as an osteoinductive substance, that is released from a composite in a controlled fashion for a period of time so that effective amounts of the substance may be maintained over that period of time. Sustained release does not imply that a substance is released at a constant rate over the period of time. For example, a composite that comprises osteoinductors and degrades over a certain period of time will release osteoinductors as it degrades. This release is sustained in that it is controlled by the degradation rate of the composite. Furthermore, it is sustained because a substance may be released in an environment for a longer period than would otherwise be possible by inserting the raw substance directly into that environment.

[0127] The term "subject", as used herein, refers to a target of administration. The subject of the herein disclosed subject-matter may be any organism, including vertebrates, such as a mammals, fish, birds, reptiles, or amphibian. Thus, the subject of the herein disclosed subject matter may be a human or non human. A subject may be unicellular or multicellular. Veterinary therapeutic uses are provided in accordance with the presently disclosed subject matter. The term "subject" does not denote a particular age or sex. Adult and newborn subjects, as well as fetuses, whether male or female, are intended to be covered.

[0128] As such, the presently disclosed subject matter provides for administration to mammals such as humans and non-human primates, as well as those mammals of importance due to being endangered, such as Siberian tigers; of economic importance, such as animals raised on farms for consumption by humans; and/or animals of social importance to humans, such as animals kept as pets or in zoos. Examples of such animals include but are not limited to:

carnivores such as cats and dogs; swine, including pigs, hogs, and wild boars; ruminants and/or ungulates such as cattle, oxen, sheep, giraffes, deer, goats, bison, and camels; rabbits, guinea pigs, and rodents. Also provided is the treatment of birds, including the treatment of those kinds of birds that are endangered and/or kept in zoos, as well as fowl, and more particularly domesticated fowl, i.e., poultry, such as turkeys, chickens, ducks, geese, guinea fowl, and the like, as they are also of economic importance to humans. Thus, also provided is the treatment of livestock, including, but not limited to, domesticated swine, ruminants, ungulates, horses (including race horses), poultry, and the like.

[0129] Further still, embodiments of the present invention may comprise "osteopromoters", or materials that enhance or accelerate the natural cascade of bone repair such as PRP, FGF's, TGF-beta., PDGF, VEGF.

[0130] The presently disclosed subject matter further relates to low porosity (<10%) compression- molded PUR composites with high allograft content (e.g., 68 vol% allograft). Such compositions have support rapid cellular infiltration as early as 6 weeks when implanted in 6x1 1mm plug defects in rabbit femurs (Figure 1A). Higher magnification images show regions of allograft resorption, cellular infiltration, polymer degradation, and new bone formation

(Figure 1B-C). Without being bound by theory or mechanism, the rapid cellular infiltration observed for composites with high allograft content approaching the random close-packing (RCP) limit of 64 vol% has been attributed to the percolated morphology characterized by multiple particle-particle contacts, which presents a continuous pathway for cells to infiltrate the composite via creeping substitution. These observations challenge the current tissue engineering paradigm emphasizing that inter-connected pores are essential for rapid cellular infiltration.

[0131] Embodiments of allograft/PUR and synthetic/PUR composites may exhibit tough mechanical properties and undergo plastic deformation. For example, allograft/PUR

embodiments may have compressive and bending strengths at yield exceeding 150 MPa and 50 MPa, respectively, and yield strains≡5%. Biodegradable PURs synthesized from lysine-derived polyisocyanates have been shown to undergo tunable degradation to non-toxic breakdown products. Furthermore, additional embodiments of the present invention, sustained release of recombinant human bone morphogenetic protein (rhBMP2, a osteoinductive molecule) from PUR scaffolds has been shown to enhance new bone formation. Synthetic allograft composites also strong. For instance, certain PUR/Bioglass embodiments are particularly strong, exhibiting yield strength and Young's modulus of about 60 MPa and about 2200 MPa, respectively.

[0132J Without being bound by theory, the inventors believe that for high allograft (>40 vol%) injectable composites, strength will decrease and the rate of cellular infiltration will increase with increasing allograft concentration. For compression-molded allograft PLGA composites, the rate of cellular infiltration decreases with decreasing allograft fraction as it falls below the RCP limit of 64 vol%, and the extent of cellular infiltration is minimal at 45 vol%. However, at volume fractions approaching the RCP limit, the compressive strength of injectable allograft/PUR composites decreases with increasing allograft concentration (Figure 8A) due to packing defects in the composite, as evidenced by the increase in swelling and decrease in density with increasing allograft content (Figure 8B). Thus for high-allograft injectable composites, biological activity increases and mechanical properties decrease with increasing allograft content, suggesting that intermediate allograft concentrations are necessary to balance the rate of remodeling and mechanical strength.

[0133] Embodiments of the present invention include injectable, functional allograft/PUR or synthetic-allograft/PUR composites incorporating rhBMP2 with initial mechanical strengths targeted to two specific orthopedic indications, reconstruction of tibial plateau fractures and early-stage osteonecrosis of the femoral head. Without being bound by theory or mechanism, the resorption of allograft creates pores in the composite, and the resulting release of rhBMP2 into the newly formed pores enhances new bone formation, thereby maintaining the initial strength.

[0134] Embodiments of the present invention can potentially transform clinical management of fractures and osteonecrosis by providing surgeons with functional biomaterials that improve patient outcomes.

[0135] Currently collagens and gels are popular platforms for the delivery of rhBMP-2; however, there are few weight-bearing injectable platforms for recombinant human bone morphogenetic protein. Embodiments of AMBP/PUR, TCP/PUR, Bioglass/PUR, and the like systems are both biocompatible and have remodeling capabilities in vivo. Embodiments comprise both non-porous and porous platforms. In the rabbit distal femur model, compression molded AMBP/PUR implants have shown that at the random closed packing limit (64 vol%) osteoclast-mediated resorption of the AMBP phase provides a pathway for cellular infiltration. In the rabbit calvaria critical size defect (CSD) model, injectable AMBP/PUR bone void filler (BVF) composites (-50% porosity) have demonstrated modest new bone formation around the perimeter of the implant. [0136] Without being bound by theory or mechanism, mechanical strength may decrease and the rates of cellular infiltration and remodeling will increase with increasing initial porosity. An initial porosity between 15 and 45 vol% may provide the desired balance between the rate of remodeling and mechanical strength. In other embodiments, desired remodeling rates may be obtained when the volumetric fraction of pores and osteoconductive matrix is close to the random close packing limit for sphere (e.g., 64%).

[0137] For certain embodiments of injectable bone void filler, the cured composite will have a porosity of about 20-50%. Furthermore, these embodiments may initially have 45wt% oseoconductive matrix (e.g., allograft, synthetic allograft), but once the composite has reacted, cured, and expanded, the osteoconductive matrix may only comprise about 16 vol% to about 26 vol% of the composite. Thus, these composites have a sum of osteoconductive matrix and pores that comprise about 46 vol% to about 66 vol% of the composite. Similarly, certain embodiments comprising relatively higher concentrations of osteoconductive matrix swell less during curing. These specific embodiments may comprise about 50-60wt% osteoconductive matrix and less than about 5 vol% pores. Therefore, porosity may range from about 0vol% to about 50vol% for embodiments of composites, but for specific embodiments optimal new bone growth is observed when the sum of the porosity and osteoconductive matrix volumetric fraction are about 60 vol% to about 70 vol% of the cured composite.

[0138] Embodiments of injectable allograft/PUR composite incorporating about 34 vol% allograft (the maximum level possible that is injectable through a syringe) support cellular infiltration and remodeling as early as 3 weeks in rat femoral condyle plug defects and at 6 weeks in rabbit calvarial defects. In certain embodiments, the combination of a moderate (about 34 vol%) concentration of allograft particles and porosity (about 15 - 45 vol%) provides both sufficient mechanical support and rapid cellular infiltration and remodeling.

[0139] The present inventors have shown that the compressive strength of PUR composites can be tuned to targeted values by varying the initial porosity. Since mechanical properties increase with increasing allograft content at moderate (e.g., <40 vol%) volume fractions, the allograft content may be held at the maximum level (34 vol%) that preserved injectability through a standard syringe in order to maximize compressive strength. By varying the concentration of triethylene diamine catalyst, the working and setting times were adjusted to 5 and 12 minutes, respectively, which are comparable to values reported for calcium phosphate cements. Figure 3 shows the wet compressive modulus and yield strength of allograft/PUR composites fabricated from a 300 g eq ^"1 polyester triol as a function of porosity. These data show that the compressive strength of the allograft/PUR void filler may exceed that of, for example, trabecular bone in the tibial plateau (5.8 MPa) at porosities <45 vol%.

[0140] Furthermore, it has been found that remodeling proceeds from the external surface to the interior through the process of creeping substitution, the limited remodeling of current allograft devices is conjectured to be due in part to their low specific surface area. To overcome this limitation, embodiments of the present invention include embedding small allograft particles (e.g., 100 - 500 μιη) in a porous polymer binder to increase the rate of remodeling by increasing the specific surface area. It has been found for certain embodiments that particle sizes less than 100 μιη do not result in optimal remodeling, and particularly so for particles smaller than 50 μηι. Without being bound by theory or mechanism, this is believed to be due to the fact that osteoclasts are unable to efficiently recognize particles smaller than 100 μπι, and therefore these relatively small particles may cause an inflammatory response. Particles larger than about 500 μιη also result in less than optimal remodeling for embodiments of the present invention.

[0141] The presently-disclosed subject matter also relates to remodeling of composites incorporating calcium phosphate fillers. Due to its relatively low volume fraction (<18 vol%), the calcium phosphate was completely embedded in polymer, thus the rate of remodeling was very slow and scaled with the rate of polymer degradation. Furthermore, the particle size of the calcium phosphates was generally <20 μπι, which is below the preferred size range for remodeling by creeping substitution. To address these limitations that result in slow remodeling, the rate of remodeling was accelerated by incorporating > 100 μπι allograft bone particles and a modest (e.g., <50%) amount of pores with mean size 180±90 μιτι. Likewise, unmodified TCP particles may be used and have relatively large particle sizes, particularly with respect to surface modified TCP particles, and allow remodeling by creeping substitution.

[0142] The presently-disclosed subject matter further relates to cellular infiltration and remodeling of allograft/PUR composites injected into, for example, 15-mm calvarial defects in NZW rabbits. Figure 4 shows a photograph of the material after injection and an x-ray image taken at sacrifice at 6 weeks. The low magnification image (top) of the composite shows extensive cellular infiltration into the interior of the composite and new bone formation near the host bone interface. Higher magnification of regions near the host bone interface (bottom) shows active remodeling, including allograft resorption, osteoid, and new bone formation. Un- remodeled allograft particles embedded in polymer matrix are also evident. Histological sections taken at 12 weeks show that >80% of the polymer had resorbed (data not shown). These observations show that resorption of the allograft creates pores into which cells subsequently migrate, thereby presenting an alternative pathway by which cells can infiltrate the composite.

[0143] Certain preferred embodiments of the present invention exhibit relatively small to no reductions in mechanical strength during remodeling, and rhBMP2 is incorporated accelerate re- mineralization. Specific embodiments include allograft/PUR and synthetic/PUR composites used to deliver rhBMP2 locally from the site of an injury. For certain embodiments this accelerates re- mineralization, reduces the volume of the demineralized zone, and maintains compressive strength at values within 80% of the initial value.

[0144] Certain embodiments comprise rhBMP-2, and of these certain embodiments, some have been found to reduce or eliminate resoption gaps formed when re-mineralization lags behind resorption in non-rhBMP2, or other osteoinductive material, embodiments. Specifically, delivery of rhBMP2 from PUR scaffolds and injectable allograft/PUR composites or synthetic allograft/PUR composites enhances new bone formation in rat and rabbit models of bone regeneration. Without being bound by theory or mechanism, it is anticipated that certain embodiments of the present invention that release of rhBMP2 into pores created by resorption of allograft particles will reduce the time period between resorption and remineralization, resulting in more consistent mechanical properties and fewer resorption gaps.

[0145] Composites may be prepared by reactive liquid molding. rhBMP2 is added as a labile powder to the hardener component of the reactive PUR. Embodiments of the present invention are capable of incorporating rhBMP2 as a powder, which allows for both easy and highly tunable application of rhBMP2. Embodiments of the present invention do not require that rhBMP2 be encapsulated in, for example, PLGA microspheres in order obtain optimal release kinetics. Embodiments of the present invention comprise PUR degradation rates and release kinetics of rhBMP-2 in powder form that are equivalent to or superior to the kinetics obtainable with encapsulated rhBMP-2.

[0146] Addition of rhBMP2 as a labile powder may results in a burst followed by a sustained release for >21 days, which may promote the most extensive bone formation. The labile powder approach is the simplest to use in a clinical environment.

[0147] The presently disclosed subject matter also relates to composites comprising PUR a synthetic allograft substitute and, optionally, rhBMP-2. Synthetic allograft substitutes include osteoconductors, and may comprise at least one of TCP, bioactive glass, cortical-canellous bone chips, hydroxyapatite, other osteoconductors, as well as combinations and derivatives thereof. Synthetic allograft substitutes include both tricalcium phosphate (TCP) and Bioglass 45S5 (BG). Certain high allograft embodiments incorporate BG rather than TCP. IN certain embodiments the Bioglass is modified with polycapro lactone (PCL), and results in composites having dramatically improved mechanical properties over PUR/allograft or PUR/unmodified-BG embodiments. The TCP may be used in an unmodified state. In certain embodiments TCP is incorporated into low allograft embodiments in which strength is not a primary concern.

[0148] These synthetic substitutes (e.g. osteoconductors) may be used interchangeably, in combination with other synthetic substitutes, or in combination with bone allograft. With regard to TCP, the term TCP, as used herein, refers to β-TCP, TCP, or combinations thereof.

[0149] Further embodiments comprise PUR/synthetic/rhBMP2 composites. Composites having rhBMP-2 have been discussed above. To reap the benefits of both the bioactivity from the rhBMP-2 and mechanical and other benefits of the synthetic substitutes, these elements may be combined. Thus, certain embodiments may comprise PUR/TCP/rhBMP-2 composites,

PUR/modified-TCP/rhBMP-2 composites, PUR/BG/rhBMP-2 composites, PUPJmodified- BG/rhBMP-2 composites, and the like.

[0150] For specific embodiments of the present invention, it has been found that TCP is substantially equivalent to or superior to PCL-TCP. This finding is unexpected because previous PUR composites were typically superior and/or feasible for use only if they comprised a modified form of TCP. One superior and unexpected benefit that arises from using unmodified- TCP is that TCP has a larger size (e.g., about 100-300 μπι) than modified-TCP, such as PCL- TCP (e.g., <100 μηι). These larger particles give PUR/TCP composites increased remodeling capabilities. Certain embodiments comprise about 45wt% TCP.

[0151) The terms "modified" and "unmodified", as used herein, refers to whether substances are modified by another substance in any manner. For example, bioactive glass, such as Bioglass, may be modified if its surface is modified by functionalizing the surface of the bioactive glass particles with a modifying substance, such as silane coupling angent 3-aminopropyl- trietoxysiline. Similarly, TCP may be modified by functionalizing TCP with a modifying substance, such as polycaprolactone. Thus, the term modified generally refers to substances that are surface functionalized or have additional components relative to the initial unmodified substance.

[0152] Specific embodiments of bone void filler TCP/PUR composites incorporate TCP in an unmodified state. For these specific embodiments, the presence of unmodified TCP does not pose any issues, as the TCP does not need to be grafted and therefore does not unduly weaken the composite's strength. Further specific embodiments of the present invention comprise injectable bone void fillers comprising less than 50wt% TCP and, optionally, rtiBMP-2. Thus, certain embodiments comprising TCP offer the advantage of having a simple system does not require the modification of TCP and the more optimal particle sizes of TCP versus modified TCP.

[0153J Certain embodiments comprise about 10wt% osteoconductive matrix, about 15wt% osteoconductive matrix, about 20wt% osteoconductive matrix, about 25wt% osteoconductive matrix, about 30wt% osteoconductive matrix, about 35wt% osteoconductive matrix, about 40wt% osteoconductive matrix, about 45wt% osteoconductive matrix, about 50wt%

osteoconductive matrix, about 55wt% osteoconductive matrix, about 60wt% osteoconductive matrix, about 65wt% osteoconductive matrix, about 70wt% osteoconductive matrix, about 75wt% synthetic substitute, about 80wt% osteoconductive matrix, about 85wt% osteoconductive matrix, or about 90wt% osteoconductive matrix. Of course, the osteoconductive matrix may comprise bone particles, inorganic particles, any osteoconductive particles or substances, and combinations thereof. [0154] Additional embodiments comprise PUR composites that include combinations of synthetic allograft (e.g., inorganic particles) and bone allograft. These embodiments may comprise similar total allograft content as the strictly synthetic or bone allograft embodiments, but the ratio of allograft to synthetic allograft may be varied to any extent. For instance, PUR composites may comprise bone allograft and synthetic allograft in a ratio of 1 :99 to a ratio of 99: 1. Other embodiments comprise a bone allograft and synthetic allograft in a ratio of about 10wt%, about 20wt%, about 30wt%, about 40wt%, about 50wt%, about 60wt%, about 70wt%, about 80wt%, or about 90wt%. Thus, embodiments may comprise mixtures of synthetic and non- synthetic allograft to meet the limitations of a particular circumstance.

[0155] The presently disclosed subject matter also relates to method for synthesizing PUR composites that utilize synthetic substitutes, such as TCP and Bioglass. Synthetic substitutes may be used in the same wt% that allograft is used in for a particular embodiment. Synthetic substitutes have the superior and unexpected advantage of not suffering from transient resorption to the extent that PUR/allograft embodiments may. Furthermore, synthetic substitute and PUR composites present fewer biological concerns when combined with a growth factor, such as rhBMP-2.

[0156] Certain synthetic/PUR embodiments may be embodied as putties or as bone void fillers. They may be pre-molded and/or injectable, and that may also comprise rhBMP-2.

Examples of materials that can be incorporated include chemotactic factors, angiogenic factors, bone cell inducers and stimulators, including the general class of cytokines such as the TGF-beta superfamily of bone growth factors, the family of bone morphogenic proteins, osteoinductors, and/or bone marrow or bone forming precursor cells, isolated using standard techniques.

Sources and amounts of such materials that can be included are known to those skilled in the art.

[0157] Embodiments of composites may also comprise biomolecules, small molecules, and bioactive agents to, for example, stimulate particular metabolic functions, recruit cells, or reduce inflammation. For example, nucleic acid vectors, including plasmids and viral vectors, that will be introduced into the patient's cells and cause the production of growth factors such as bone morphogenetic proteins may be included in a composite. Biologically active agents include, but are not limited to, antiviral agent, antimicrobial agent, antibiotic agent, amino acid, peptide, protein, glycoprotein, lipoprotein, antibody, steroidal compound, antibiotic, antimycotic, cytokine, vitamin, carbohydrate, lipid, extracellular matrix, extracellular matrix component, chemotherapeutic agent, cytotoxic agent, growth factor, anti-rejection agent, analgesic, antiinflammatory agent, viral vector, protein synthesis co-factor, hormone, endocrine tissue, synthesizer, enzyme, polymer-cell scaffolding agent with parenchymal cells, angiogenic drug, collagen lattice, antigenic agent, cytoskeletal agent, mesenchymal stem cells, bone digester, antitumor agent, cellular attractant, fibronectin, growth hormone cellular attachment agent, immunosuppressant, nucleic acid, surface active agent, hydroxyapatite, and penetraction enhancer. Additional exemplary substances include chemotactic factors, angiogenic factors, analgesics, antibiotics, anti-inflammatory agents, bone morphogenic proteins, and other growth factors that promote cell-directed degradation or remodeling of the polymer phase of the composite and/or development of new tissue (e.g. , bone). RNAi or other technologies may also be used to reduce the production of various factors.

[0158] In some embodiments, inventive composites include antibiotics. Antibiotics may be bacteriocidial or bacteriostatic. An anti-microbial agent may be included in composites. For example, anti- viral agents, anti-protazoal agents, anti-parasitic agents, etc. may be include in composites. Other suitable biostatic/biocidal agents include antibiotics, povidone, sugars, and mixtures thereof. Exemplary antibiotics include, but not limit to, Amikacin, Gentamicin, Kanamycin, Neomycin, Netilmicin, Streptomycin, Tobramycin, Paromomycin, Geldanamycin, Herbimycin, Loravabef, etc. (See, The Merck Manual of Medical Information - Home Edition, 1999).

[0159] Inventive composites may also be seeded with cells. In some embodiments, a patient's own cells are obtained and used in inventive composites. Certain types of cells {e.g. , osteoblasts, fibroblasts, stem cells, cells of the osteoblast lineage, etc.) may be selected for use in the composite. Cells may be harvested from marrow, blood, fat, bone, muscle, connective tissue, skin, or other tissues or organs. In some embodiments, a patient's own cells may be harvested, optionally selected, expanded, and used in the inventive composite. In other embodiments, a patient's cells may be harvested, selected without expansion, and used in the inventive composite. Alternatively, exogenous cells may be employed. Exemplary cells for use with the invention include mesenchymal stem cells and connective tissue cells, including osteoblasts, osteoclasts, fibroblasts, preosteoblasts, and partially differentiated cells of the osteoblast lineage. Cells may be genetically engineered. For example, cells may be engineered to produce a bone morphogenic protein.

[0160] To enhance biodegradation in vivo, composites of the present invention can also include different enzymes. Examples of suitable enzymes or similar reagents are proteases or hydrolases with ester-hydrolyzing capabilities. Such enzymes include, but are not limited to, proteinase K, bromelaine, pronase E, cellulase, dextranase, elastase, plasmin streptokinase, trypsin, chymotrypsin, papain, chymopapain, collagenase, subtilisin, chlostridopeptidase A, ficin, carboxypeptidase A, pectinase, pectinesterase, an oxireductase, an oxidase, or the like. The inclusion of an appropriate amount of such a degradation enhancing agent can be used to regulate implant duration.

[0161] While surface of inventive composite will be mixed somewhat as the composite is manipulated in implant site, thickness of the surface layer will ensure that at least a portion of the surface layer of the composite remains at surface of the implant. Alternatively or in addition, biologically active components may be covalently linked to the bone particles before

combination with the polymer. For example, silane coupling agents having amine, carboxyl, hydroxyl, or mercapto groups may be attached to the bone particles through the silane and then to reactive groups on a biomolecule, small molecule, or bioactive agent.

EXAMPLES

[0162] The presently-disclosed subject matter is further illustrated by the following specific but non-limiting examples.

[0163] Example 1

[0164] Polyester triol, LTI-PEG prepolymer, and allograft synthesis and characterization.

[0165] Poly^-caprolactone-co-glycolide-co-DL-lactide) triols with an equivalent weight of 300 g eq ^"1 and a backbone comprising 60 wt% caprolactone, 30% glycolide, and 10% lactide (T6C3G1L300) are synthesized using known techniques. Preliminary experiments have shown this polymer undergoes approximately 80% degradation after 12 weeks in vivo. Appropriate amounts of dried glycerol and μ-caprolactone (Aldrich), glycolide and DL-lactide

(Polysciences), and stannous octoate (Aldrich, 0.1 wt-%) are mixed in a 100-ml flask and heated under an argon atmosphere with mechanical stirring to 140°C for 24h. The triol is washed with hexane and characterized by NMR, OH number, and GPC. An LTI-PEG prepolymer is synthesized by charging lysine triisocyanate (LTI, Osteotech) to a 50mL flask, adding PEG200 (Aldrich, 200 g mol ^"1 , 2: 1 mol LTIrmol PEG) dropwise under intense stirring at 60°C, and reacting overnight. The hydroxy 1 number of the polyester triol and %NCO of the prepolymer is measured by titration (Metrohm Titrino) and molecular weight by gel permeation

chromatography (Waters Alliance). Mineralized allograft bone particles (Osteotech) is prepared by comminuting debrided and cleaned cortical bone in a mill, sieving (106-500 μηι diameter), defatting in 70% denatured alcohol for >1 h, washing with sterile deionized water, lyophilizing for >6 h at -35 °C, and vacuum-drying for >12 h at 35 °C and 500 mtorr. Lyophilized bone particles are treated with supercritical carbon-dioxide at 105°C for >25 min, packed under argon, and gamma- irradiated at 25-35 KGy.

[0166] Composite fabrication.

[0167] Defatted allograft bone particles are mixed with prepolymer, polyester triol, and catalyst solution, mixed for 60s, and cast into a mold as described previously. The catalyst solution is prepared as a mixture of 10% triethylene diamine catalyst (TEDA, Aldrich) in dipropylene glycol, and is added at sufficient concentration to yield a working time of 3 - 5 min and a cure time of 10 - 12 min. A sufficient amount of LTI-PEG prepolymer is added to yield an index of 1 15 (15% excess of isocyanate equivalents). The reactive mixture is injected into molds and cured at 37°C for 24h. Composites are incubated in PBS for one week, and the mass swelling ratio (Q _m ) is calculated as the ratio of the wet and dry weights of the composite. Composition is determined by FT-IR (Bruker), and the free NCO is reported as the ratio of area under the NCO peak (2230 cm ^'1 ) to that under the C=0 stretching vibration (1760 cm ^'1 ) peak. The density of the scaffolds is determined gravimetrically, and the porosity, defined as the volume fraction pores, is calculated from the composite foam density. Scanning electron microscope (SEM) micrographs (Hitachi S-4200) are used to determine pore size. In vitro degradation is determined by incubating specimens in PBS at 37°C for up to 36 weeks and measuring the mass loss weekly.

[0168] In vitro (initial) mechanical properties.

[0169] Specimens for compression, torsional, and flexural testing are incubated in PBS at 37°C for 24h prior to testing. Cylindrical compression specimens (6mm D x 12mm H) are loaded at 25 mm/min by the platens of a material testing system (Bionix 858, MTS). Upon converting the force vs. displacement to engineering stress vs. engineering strain, the modulus of elasticity (linear slope), yield strength (stress at 0.2% offset), and energy-to-failure (area under curve) is recorded. For torsion testing, each 'hour glass' specimen (gauge region: 10 mm long x 2 mm diameter) is twisted at 40deg/s until failure. The torque vs. twist data is then be converted to a shear stress (μ) vs strain (μ) curve using i=daiL and μ=[Θ^ΤΛ1θ) + 3T]/2, where Θ is the angle of twist in radians, a is the radius of specimen, L is the gauge length of the specimen, and T is the torque. Ultimate torque and torsional modulus is measured as the maximum torque endured by the specimen and the slope of the initial linear portion of the curve, respectively. The bending strength and modulus of elasticity is determined from 3-point bending tests in which

parallelepipeds (40mm x 4mm x 2mm) are loaded at 3mm/rriin using a bench-top material testing system (Dynamight, Instron). Peak force and stiffness are converted to the material properties using the flexural equations from beam theory. Dynamic mechanical properties (Ε', E", and tan μ) of 13.5mmx25mmx2mm slabs are measured in 3-point bending mode (TA Instruments Q800 DMA). Both frequency (0.1-10 Hz) and temperature (-50-150°C) sweeps are performed to determine the viscoelastic properties of the composites.

[0170] Remodeling of allograft/PUR composites in a rabbit femoral condyle model.

[0171] Composites are prepared as described previously and injected into unicortical bilateral plug defects in the femoral condyles of NZW rabbits. Allograft bone particles, catalyst solution, polyester triol, and LTI-PEG prepolymer are irradiated using a dose of approximately 25 kGY. Glycopyrrolate is administered at 0.01 mg/kg IM followed by ketamine at 40 mg/kg IM. Bilateral defects of approximately 6.1 mm diameter by 1 1 mm in depth are drilled in the metaphysis of the distal femurs of each rabbit. Composites from each treatment group (Table 2) is subsequently injected into each defect. Treatment groups for each composite are dispersed randomly among the rabbits. Rabbits are euthanized at the appropriate time points using Fatal- plus (2.2 mL/10 kg) intra-venously. After sacrifice, femurs are extracted and placed in a 1 X phosphate buffer solution for 2 hours followed by dehydration in a series of ethanol and fixation in 10% formalin for 3 weeks. Testing of biomaterial composites found this difference or greater in maximum strength measures when the porosity was increased from 15% to 30%.

[0172] Ex vivo μΟΎ is used to quantify the volume of new bone in the defect volume for the composites. Cross sectional contiguous μCT images of the entire defect are acquired at 70 kV and 1 14 mA with an isotropic voxel size of 30 μηι on a Scanco μCT40 (Scanco Medical AG, Switzerland). A volume of interest comprising the entire defect is selected for analysis as defined by the perimeter of intact host bone around the defect site. Scanco software is used to determine fractional bone volume (BV/TV), which is used as the primary endpoint, and the architecture of the bone as published previously by the Col. Additional analyses includes quantification of the mineralization void volume, average thickness between the resorbing implant and the new bone that is filling the defect, and, when possible, analysis of allograft volume.

[0173] Following μΟΤ analysis, all specimens are dehydrated and embedded in MMA for non-decalcified histology. Central, 4-6 μπι thick sections are cut and stained with H&E to assess inflammation, Safranin-O/Fast Green for cartilage, and Masson's trichrome for new bone and implant volume. The areas of inflammation, new bone, fibrous tissue, cartilage, and implant relative to the defect area are measured at low magnification using Osteomeasure software (Osteometries, Decatur, GA). The volume of the demineralized zone in the composite is also measured relative to defect area, new bone area, and implant area.

[0174] An 8-mm trephine tool attached to a drill-press is used to core filled defects in rats. The ends of the cored defects are ground on silicon carbide paper to make them parallel. The cylindrical specimen of host bone surrounding the composite is imaged by μCT (Figure 5) to verify the integrity of the filled defect and quantify both the BV/TV and apparent volumetric mineral density of the specimen. Following hydration in PBS, explanted specimens are placed between two compression platens and loaded at 25 mm/min. Force is recorded from an appropriately sized load cell and displacement is recorded from an extensometer attached to the platens (data collection at 50 Hz). The resulting force-displacement curve is converted to an engineering stress-engineering strain curve using the initial cross-sectional area of the specimen and the gage of the extensometer. Apparent modulus (slope of the linear portion of the curve), apparent yield strength (stress at proportional limit), and apparent peak strength (maximum stress is recorded).

[0175] Statistical analysis includes one-way ANOVA to test dose-dependent effects of the factors (either initial porosity or rhBMP-2) on compressive strength; BV/TV; and the areas of demineralized tissue, allograft, residual polymer, and new bone formation within the implant. Individual differences among groups at each time period are determined by the Fisher protected least significant difference test for multiple comparisons with significance established at p<0.05.

[0176] Remodeling of allograft/PUR composites depends on several parameters, including allograft volume fraction, initial porosity, and polymer composition. For allograft contents <50 vol%, the number of mechanical defects resulting from allograft particle-particle contacts is minimal, and therefore mechanical strength increases with increasing allograft volume fraction. Therefore, to maximize initial strength, the allograft content was selected as the highest possible level that supports injection through a syringe (34 vol%). Raman, μΟΤ, and histomorphometry show enhanced new bone formation, accelerated PUR degradation, and minimized volume of the resorption front at higher porosity. However, initial strength decreases with increasing porosity. Therefore, in one embodiment of the present invention, an intermediate porosity of -30% provides the necessary balance between biological and mechanical requirements. Data have shown that the composition of the polymer does not significantly affect the remodeling process if it is biocompatible and biodegradable. The present inventors have polyurethanes synthesized from an LTI-PEG prepolymer and poly^-caprolactone(60%)-co-glycolide(30%)-co-DL- lactide(10%)) triol (300 g eq ^"1 ) and investigated in the rabbit studies. This polymer degrades to -80% of its initial mass to non-toxic decomposition products after 12 weeks in a rabbit calvarial defect model. Without being bound by theory or mechanism, this degradation rate is suitable, but if necessary the degradation rate is decreased by varying the composition of the polyester triol or decreasing its equivalent weight. Alternatively, to increase the degradation rate, the equivalent weight of the polyester triol is increased, or a triol sensitive to MMP-mediated degradation is synthesized. [0177] Example 2

[0178] This Example shows that embodiments of the present invention are capable of producing a sustained release of rhBMP2 from PUR scaffolds, which increases new bone formation relative to a collagen sponge in a rat femoral segmental defect model. rhBMP2 delivered from a collagen sponge (INFUSE® Bone Graft, Medtronic) is an FDA-approved therapy for posterior-lateral spine fusion, tibial fractures, and specific craniofacial applications. The collagen sponge delivery system results in a bolus release of growth factor in the first 24 - 48 hours, but a number of studies have suggested that sustained release of rhBMP2 is more effective for promoting new bone formation. To modulate the release kinetics, rhBMP2 (60 μg/ml) was incorporated in PUR scaffolds by either direct addition as a labile powder or by encapsulation in large (L) or small (S) PLGA microspheres prior to incorporation in the scaffold. The labile powder (PUR/BMP2) formulation resulted in a burst followed by a sustained release of rhBMP2 up to day 21 (Figure 6A). Encapsulation of rhBMP2 in ~1 μηι PLGA (50/50 L/G, M„~50,000 g/mol) microspheres prior to incorporation in the PUR scaffolds essentially eliminated the burst release.

[0179] The in vitro bioactivity of rhBMP2 released from PUR scaffolds was comparable to that of fresh rhBMP2, thereby demonstrating that this approach produced sustained release of active rhBMP2 over a 20-day period. To investigate the effects of release kinetics on healing in a critical size defect, PUR scaffolds incorporating rhBMP2 were implanted in 6-mm segmental femoral defects in Sprague-Dawley rats. After both 4 and 8 weeks implantation time,

PUR/rhBMP2 scaffolds exhibited significantly more new bone formation compared to the collagen+rhBMP2 control (Figure 6B). However, PUR scaffolds with no rhBMP2 and

PUR PLGA-S-BMP2 scaffolds (slow release) showed only minimal new bone formation.

Histological sections of the PUR/BMP2 scaffolds show cellular infiltration, new bone formation, and blood vessel formation (Figure 6C and D). These results suggest that both a burst and sustained release of rhBMP2 are desirable for new bone formation, which is consistent with our study in a rat femoral plug model. [0180] Two-component polyurethanes enable customization using added biologies (e.g., growth factors and/or antibiotics) at the point of care. Allograft/PUR composites incorporating 420 μg/ml rhBMP2 into 15-mm were injected rabbit calvarial defects. The in vitro release kinetics show lower cumulative release (-20%, Figure 7) at 25 days compared to the high (-90%) porosity PUR scaffolds (-70%, Figure 6). However, the in vivo release kinetics are conjectured to be considerably faster due to the resorption of allograft particles, which creates new pores into which rhBMP2 can diffuse from the polymer. Histological sections at 6 weeks show extensive new bone formation along the upper surface of the composites and near the host bone interface (Figure 7). In many animals, new bone had completely bridged the upper surface of the defect. Higher magnification images (20X and 40X) show active bone remodeling by osteoblasts (OB) and osteoclasts (OC), as well as formation of new blood vessels. Interestingly, the rate of polymer degradation was higher compared to the samples without rhBMP2, as evidenced by the absence of a significant amount of polymer at 6 weeks. In contrast, the collagen+rhBMP2 samples exhibited no significant new bone formation (comparable to the negative control).

[0181] Example 3

[0182] Synthesis of allograft/PUR composites incorporating rhBMP2.

[0183] Briefly, rhBMP2 is mixed with a solution incorporating 20: 1 heparin:rhBMP2 and 100: 1 trehalose and lyophilized to yield a dry powder, which is subsequently added to the hardener component of the PUR prior to mixing with the prepolymer and allograft particles. Three replicate scaffold samples (-50 mg) containing 2.5 μg rhBMP-2 are immersed in 1 ml release medium (μ-ΜΕΜ incorporating 1 % BSA). The medium is refreshed every 24 h to minimize degradation of the growth factor. The rhBMP-2 concentration in the releasates is determined using a Human BMP-2 Quantikine ELISA kit (R&D systems).

[0184] rhBMP2 release kinetics from allograft/PUR + rhBMP2 composites.

[0185] Considering that the resorption of allograft particles has been shown to create new pores for cellular infiltration, the release kinetics from allograft/PUR+rhBMP2 composites is higher in vivo compared to in vitro. rhBMP2 is labeled with radioactive iodine ( ¹²⁵ I) using IODO-BEADS Iodination Reagent (Pierce Biotechnology, Rockford, IL) in accordance with

125 previously published techniques. IODO-beads containing approximately 1 mCi Na I is incubated in 1 ml of reaction buffer for 5 min under room temperature, followed by addition of 50 μg rhBMP2 to the reaction solution and incubation for another 25 min. The solution is then removed from the IODO-BEADS reaction tube and the Iodine- labeled rhBMP2 ( ¹²⁵ I-rhBMP2) is separated in a Sephadex disposable PD-10 desalting column (Sigma-Aldrich). Eluted fractions are collected and a Cobra II Autogamma counter (Packard Instrument Co, Meridien, CT) adapted to determine the fractions containing the ^{l 25} I-rhBMP2. The ¹²⁵ I-labeled growth factor is combined with non-labeled rhBMP-2 (1 :5 hot-cold ratio) and trehalose (100: 1 trehalose:rhBMP2 ratio), lyophilized, and mixed with the hardener component prior to mixing with the allograft bone particles and LTI-PEG prepolymer. To measure the in vivo release kinetics, the radioactive composites are injected into femoral defects and the release measured using a Cobra II

Autogamma counter as previously described. Activity is measured over four 1 -min periods and is repeated weekly while the rabbits are under sedation.

[0186] The Medtronic-recommended dose is 420 μg/ml for use with the collagen sponge. The data (Figure 7) show that the sustained release achieved with the PUR delivery system results in more bone formation relative to the collagen sponge, which justifies investigation of a lower dose. Therefore, the two doses selected are 100 and 420 μg/ml.

[0187] Example 4

[0188] Composites were prepared by reactive liquid molding of defatted allograft bone particles (100-500 μπι), LTI-PEG prepolymer, polyester triol, and catalyst mix using previously described techniques. The concentration of allograft particles are varied from 47 - 57 vol%. Composites with <45 vol% allograft do not support extensive cellular infiltration, and composites with >57 vol% allograft are not cohesive, have weak compressive strength, and cannot be injected through a 2.3mm trocar. Composites are injected into 6-mm bilateral plug defects in the femurs of NZW rabbits, and calcium phosphate bone cement are investigated as a clinical control. |0189] Rates of allograft resorption, cellular infiltration, new bone formation, polymer degradation, and biomechanical properties are measured as described herein. Biomechanical properties are measured for specimens cored from the femoral condyle. Preferred embodiments have rhBMP2 incorporated in the composites to accelerate re-mineralization.

[0190] Data show that high allograft content (>45 vol%) composites exhibit compressive mechanical properties comparable to those of trabecular bone in the femoral head. The compressive strength and modulus of allograft/PUR composites as a function of vol% allograft are plotted in Figure 8A. For the 56.7 vol% allograft composite, the bending strength was 13.8+1.7 MPa. It is important to note that the composites in Figure 8 were synthesized from a 100 g eq ^"1 polyester triol. At allograft loading <57 vol%, the initial compressive properties exceed those of trabecular bone in the femoral head (17.5 MPa), and the yield strain is >5%. Preliminary experiments have shown that the defatted allograft bone particles react with the isocyanate-functional prepolymer, and that increasing the reactivity of the allograft through surface-demineralization does not increase the mechanical properties. At allograft loadings >57 vol%, there are significant defects in the composite since the loading is approaching the RCP limit. As a result, the density decreases and the swelling increases with increasing allograft content >57 vol%, resulting in reduced strength.

[0191] Data show that low-porosity allograft/PUR composites support cellular infiltration and remodeling in a rabbit femoral plug model. This Example shows substantial changes in opposite directions over a narrow range of allograft concentrations near the RCP limit.

Therefore, embodiments of the present invention identify and include the optimum allograft content that effectively balances the mechanical and biological requirements.

[0192] Example 5

[0193] This Example demonstrates that in vivo resorption of allograft particles accelerates the formation of pores, which may then be infiltrated by rhBMP-2.

[0194] rhBMP2 is mixed with a solution incorporating 20: 1 heparin:rhBMP2 and 100: 1 trehalose :rhBMP2 and lyophilized to yield a dry powder, which are subsequently added to the hardener component of the PUR prior to mixing with the prepolymer and allograft particles. [0195] In vitro release kinetics are measured by ELISA as described previously. In vivo release kinetics of rhBMP2 are higher compared to in vitro kinetics due to osteoclast-mediated resorption of allograft particles which creates pores in the composite over time.

[0196] Example 6

[0197] The following Example shows data in connection with embodiments of the present invention.

[0198] The present inventors have investigated the effects of the rhBMP-2 dose on remodeling of allograft/PUR composites injected into 6 1 1 mm plug defects in the femoral condyle of NZW rabbits. Three doses were evaluated: 0, 100, and 420 (the recommended dose for rabbits) mg/ml. The settable putty was prepared from a lysine triisocyanate (LTI)- polyethylene glycol (PEG) prepolymer, polyester polyol, allograft bone particles (AMBP), triethylene diamine (TED A) catalyst in a dipropylene glycol (DPG) carrier, and rhBMP-2. The rhBMP-2 was mixed with trehalose and heparin, and freeze-dried to produce a powder. The polyester polyol backbone was composed of 60% caprolactone, 30% glycolide, and 10% lactide and had a molecular weight of 900 g mol ^"1 (6C3G1L900). Polyol, AMBP, catalyst solution, and LTI-PEG prepolymer were added to a mixing cup and mixed for 90 seconds. The filler content (AMBP and rhBMP-2 powder) was maintained constant at 70 wt% for each treatment group. The resulting paste was then added to the rhBMP-2 vial and mixed for 60 seconds. Bilateral plug defects approximately 6 mm in diameter by 1 1 mm in depth were drilled in the metaphysis of the distal femurs of each rabbit. AMBP PUR putty from each treatment group was injected into the defects. The setting time was approximately 10 minutes. After 6 or 12 weeks, the rabbits were sacrificed and the femurs removed. Faxitron LX-60 X-ray and mCT40 systems were used to acquire images of the femurs. The wet (i.e., after 24h incubation in saline) compressive strength of the composites cured in vitro ranged from 27.2 to 33.2 MPa and was not dependent on the concentration of rhBMP-2. Figure 10, immediately below, shows 2D mCT scans of the composites at each time point and dose of rhBMP-2. The images reveal evidence of allograft resorption and new bone formation in all treatment groups due to creeping substitution of the allograft component. Composites carrying rhBMP-2 reveal less allograft (appearing as large, dense, irregularly shaped white particles in the images) in the center of the putty. New bone formation also appears to be enhanced by rhBMP-2. Interestingly, the high (420 mg/ml) rhBMP- 2 dose treatment group showed an unpredictable response, with some composites almost completely remodeled at 12 weeks (top) and others showing extensive resorption (bottom).

[0199] These data suggest that the low-porosity allograft/PUR composite putty is an efficient carrier for rhBMP-2, and that an optimum rhBMP-2 dose exists at which predictable healing can be achieved. Significantly, the data show that the optimum dose is likely less than the recommended dose for the absorbable collagen sponge carrier that yields a bolus release of drug.

[0200] Example 7

[0201] In this Example a composite was synthesized using the procedure described in the other Examples, keeping AMBP content was maintained constant at 45 wt% for each treatment group. Bilateral plug defects approximately 6 mm in diameter by 1 1 mm in depth were drilled in the metaphysis of the distal femurs of each rabbit. AMBP/PUR putty from each treatment group was injected into the defects. The setting time was approximately 10 minutes, and the porosity of the composites ranged from 27 - 30%. After 8 weeks, the rabbits were sacrificed and the femurs removed. Faxitron LX-60 X-ray and mCT40 systems were used to acquire images of the femurs. The images reveal evidence of allograft resorption and new bone formation due to cellular migration and creeping substitution of the allograft component. These data show that the injectable porous allograft/PUR composites remodel in the rabbit femoral condyle model.

[0202] The rheological profiles are shown in Figure 11B. Storage (G\ left axis, open circles) and loss (G", right axis, filled circles) moduli were measured under shear conditions as a function of time for the injectable porous allograft/PUR composite. The working time is determined by the intersection between G' and G". Sample 1 (blue) showed a working time of 7.4 minutes, while Sample 2 (red) showed a working time of 1 1.6 min. These precise rheological measurements of the working time are consistent with the gel point previously measured as the time at which the composite no longer flows out of a syringe. We have also investigated the composition of the components that leach out of the composites after injection. Allograft/PUR composites were mixed and injected into a solution of buffer at 2 and 20 minutes post-mixing, and allowed to incubate in buffer for 24h. The composition of the leachate (i.e., the buffer solution) was determined using NMR and gel permeation chromatography (GPC). The leachates were found to contain both unreacted polyester triol and dipropylene glycol (DPG), which is the carrier for the tertiary amine catalyst (triethylene diamine, TEDA). Both of these components have very low toxicity and do not accumulate in the body. Furthermore, no evidence of LTI, prepolymer, or TEDA was found in the leachates, suggesting that essentially all the NCO equivalents reacted in the composite. Additionally, these observations suggest that the TEDA is not released in a burst, but rather is released slowly over time as the composite degrades. It is important to note that TEDA is cleared from the body in urine and the total concentration in the composite is still at least an order of magnitude below the LD50. Thus it is anticipated that the slow release of TEDA from the composites will be cleared from the body and will not reach toxic levels. These results are consistent with the ISO 10993 systemic and cytotoxicity tests, which showed no toxic effects of leachates obtained from the composites.

[02031 Example 8

[0204] This example shows low porosity injectable (reactive-allograft-bone/polyurethane) composites incorporating rhBMP-2. These embodiments enhance bone remodeling in a in a rabbit femoral plug model.

[0205] Embodiments of comprise allograft bone/polyurethane (PUR) non-porous composite putties which provides a release mechanism of recombinant human bone morphogenetic protein- 2 (rhBMP-2) responsive to the surrounding cellular environment. The interactions between the filler surface and the polymeric matrix are investigated as a tool to reinforce the composites. rhBMP-2 was included in the formulation to enhance the osteogenic properties of low porosity injectable composites. The effects of rhBMP-2 dose on new bone formation at 6 and 12 weeks were investigated in a rabbit model.

[0206] The settable putty comprised a lysine isocyanate-polyethylene glycol prepolymer, polyester polyol, allograft bone (AMBP), amine catalyst, and rhBMP-2. The filler content of the composite putty was maintained at 70 wt%. To study the filler-matrix interactions, the surface of the allograft bone was a) demineralized (SD),or b) protected with 4-methoxyphenyl isothiocyanate (PROT); the compressive mechanical properties of the corresponding composites were compared. Two doses of rhBMP-2 were used: 1 10 and 440 μg/ml. The cure time was approximately 10 minutes. Bilateral defects (6 mm diameter by 1 1 mm in depth) were drilled in the metaphysis of the distal femurs of NZW rabbits. AMBP/PUR biocomposite from each treatment group was injected into the defects. A μΟΤ40 system was used to acquire images of the femurs. Histological ground sections were stained with Sanderson's rapid bone stain counterstained with Van Gieson.

[0207] AMBP/PUR biocomposites exhibited compressive strengths (27.2-33.2MPa) comparable to trabecular bone. No significant differences between the mechanical properties of AMBP and SD were identified; the PROT samples had mechanical properties three times lower than the AMBP composites. This observation suggested that AMBP reinforced the material by creating chemical bonds between the filler and the matrix. Histological sections of the biocomposite without rhBMP-2 after 6 and 12 weeks of implantation revealed extensive cellular infiltration and new bone deposition, while μΟΤ images were characterized by extensive remodeling with negligible resorption gaps. Incorporation of rhBMP-2 enhanced new bone formation relative to the biocomposite without rhBMP-2, as evidenced by the presence of less AMBP. However, approximately 30% of the samples incorporating a high dose of rhBMP-2 displayed extensive areas of osteoclast-mediated resorption at 6 or 12 weeks. In this Example the high dose was the recommended dose for rabbits, suggesting that the release mechanism of rhBMP-2 from the biocomposite may reduce the minimum effective dose required to enhance bone healing.

[0208] A conclusion was that AMBP had a sufficient density of reactive groups in the surface which promoted extensive interfacial binding with the matrix and reinforcement of the composite. Release of rhBMP-2 corresponding to 25% of the recommended dose enhanced remodeling of the material, while some of the composites showed resorption gaps at the high dose of rhBMP-2 corresponding to the recommended dose. Thus the allograft/polymer biocomposites of the present invention is a promising approach for developing injectable biomaterials that maintain their initial mechanical properties during remodeling. [0209] Example 9

[0210] This Example is directed to an exemplary injectable allograft bone/polymer composite bone void filler of the present invention. Among other things, this embodiment may be used for repairing calvarial defects.

[0211] Injectable MBP/PUR composite void fillers are composed of lysine triisocyanate (LTI), poly(e-caprolactone-co-glycolide-co-lactide) triol, rabbit mineralized bone particles (RMBP, 100-500 μιη), and Infuse rhBMP-2. The appropriate amounts of the triol, RMBP, and LTI-PEG prepolymer were added to a 10 mL cup and hand-mixed for 90 seconds. A 0.25 mL scoop was used to transfer approximately 0.38 g of the mixture into the vial of rhBMP-2, and the appropriate amount of catalyst solution (5% triethylene diamine in dipropylene glycol) was added to the vial. The components were mixed for 1 minute followed by loading and injection from a 1 mL syringe. The target bone content was 47 wt%, and the target porosity was 30%. A critical-sized rabbit calvarial defect study was designed to study the enhanced remodeling capability of the composites with the incorporation of rhBMP-2. A 15-mm circular defect was cut in the calvaria of New Zealand white rabbits. The volume of the defects was measured to be -0.5 mL. Thus, a volume of 0.25 mL of MBP/PUR/rhBMP-2 composites was injected into the defect to allow for expansion.

[0212] The MBP/PUR/rhBMP-2 composites expanded to fill the entire defect volume. After 10 min, the foams had cured and become tack-free, completely dampening the pulsation of the dura. Compressive modulus and strength values range from 173-444 MPa and 4.4-9.5 MPa, respectively, which are in the range required to withstand pulsatile forces from the dura. ² The wounds were subsequently closed and the rabbits were closely monitored until all vital signs were normal. Radiographs and histological sections of MBP/PUR composites without rhBMP2 showed -2-4 mm of new bone ingrowth after 6 weeks implantation time in vivo. MBP/PUR composites incorporating rhBMP-2 showed bridging of the defect and extensive new bone formation. MBP/PUR composites exhibit suitable mechanical properties and remodeling for repair of calvarial defects, and are an effective delivery system for rhBMP-2. [0213) Example 10 - Injectable BVF for Rabbit Calvarial

[0214] Materials

[0215] The materials were obtained as discussed in Example 6.

[0216] Preparation of rhBMP-2

[0217] A solution of rhBMP-2 (1.5 μg/mL) was prepared by reconstituting rhBMP-2 powder per mixing instructions provided with the Infuse kit. The solution was aliquoted into vials to achieve 80 μg/mL of active rhBMP-2 dose in each sample. The vials were frozen at -80 C and lyophilized to achieve a powder.

[0218] Synthesis of the injectable biocomposite

[0219] An index of 125 was targeted to produce a biocomposite with a porosity of 47% upon injection. The TEDA catalyst was blended with DPG to yield a 10% solution of TEDA.

Hydroxyl equivalents from the polyester triol, the DPG carrier, and water were included in the index calculation:

NCO Eq

INDEX=100 x ^

OH Eq (Triol) + OH Eq (Water) + OH Eq (DPG)

[0220] The appropriate amounts of polyester triol, allograft (45 wt%), and LTI-PEG prepolymer were added to a mixing cup and mixed for 90 seconds. The resulting paste was then added to the rhBMP-2 vial followed by the addition of TEDA. After mixing for 60 seconds, the biocomposite (BC) was poured in between parallel plates for rheological characterization, or injected into either molds for mechanical testing or into rabbit calvarial defects.

[0221] Rheological properties

[0222] The rheological properties of non-setting samples were determined using a TA Instruments AR-2000ex rheometer. Samples were prepared without catalyst, poured between two 25 mm diameter parallel plates, and compressed to a gap of 1000 μιη. The material was allowed to flow between the plates to cover the whole area and excess material was removed. The samples were then subjected to a dynamic frequency sweep (0.1 to 100 rad sec ^'1 ) at 25°C with controlled strain amplitude of 0.02%. A Cox Merz transformation was applied to the dynamic data to obtain the steady state viscosity (Pa*s) and shear stress (Pa) as a function of shear rate (s ¹ ). The shear stress versus shear strain data were fit to the Casson model. [0223] Mechanical properties

[0224] Cylindrical specimens with a 6mm diameter were prepared by injecting the materials into a plastic mold. Samples with approximate height of 12mm (n=4) were hydrated for 24 hours in PBS and then tested for compression using an MTS 898 equipped with a 13 kN load cell. The samples were preloaded to 12N, followed by compression at a constant strain rate of 25%min ^~ ' until failure. Load and displacement were recorded and transformed to stress and strain using the initial sample cross-sectional area and height respectively. The stress-strain curve was used to determine the Young's modulus, compressive strength (maximum stress), yield stress and strain, and energy-to-failure (area under the curve calculated at the yield point) of the samples.

[0225] Rabbit Study

[0226] As shown in Figure 13, four treatment groups were evaluated in this animal study using skeletally mature New Zealand white rabbits at two time points, 6 and 12 weeks. An empty defect was included as the negative control, and the injectable calcium phosphate cement (CPC) was used as the clinical control. The effects of rhBMP-2 delivered from the biocomposite were also investigated at 6 weeks. Following standard practices for aseptic surgery, a full- thickness calvarial defect was prepared in the parietal bones using a 15-mm surgical trephine for rabbits as described previously (Figure 13B). Briefly, upon the surgical exposure of the cranium, MicroAire surgical hand piece with a brass trephine was used to create the critical size defect (CSD) of 15mm during copious saline irrigation (Figure 13C). The cranial cap was carefully removed to separate the attached dura from the underside of the cap. Pressure with sterile gauze was applied to stop bleeding. The defects were treated by injection of the CPC (Figure 13D) or biocomposite (Figure 13E) according to the pre-determined randomization scheme. Soft tissues were closed in layers using resorbable 3-0 Dexon sutures to create 2 sets of continuous sutures. The animals were euthanized at the given endpoints.

[0227] Radiographic Analysis

[0228] Radiographs were acquired using a Faxitron MX20 X-ray Digital System (Faxitron X-ray Corporation, Wheeling, Illinois) for each calvarium after extraction. The images were captured at 25kV at a 15 second exposure time and imported into the Faxitron DR Software (Version 3.2.2). For quantification, the images were exported as a BITMAP file using window levels 1396/184. CTAn software vl . l 1 , (Skyscan, Kontich, Belgium) was used to analyze the % defect area coverage and relative X-ray attenuation through the defect thickness for each treatment group. A region identical to the size of the defect created during the original study was outlined on each x-ray and automated thresholding was performed within this region using the Otsu method across all samples to determine the mineralized tissue within the defect. The percent of the defect area filled by the mineralized tissue was measured as a ratio of the pixels of gray above the threshold to the total number of pixels in the defect area. The relative x-ray attenuation through the defect was determined as the ratio of the mean grayscale level of the mineralized tissue within the defect to the mean grayscale value of the mineralized tissue of the surrounding host bone.

[0229] Histology and histomorphometry.

[0230] The calvaria were placed in a solution of 10% neutral buffered formalin followed by a series of ethanol dehydrations. The specimens were then embedded in methyl/butyl

methacrylate. The resulting blocks were then sectioned using an Exakt system, producing 75- micron sections. The sections were stained with Sanderson's rapid bone stain counterstained with van Gieson. Bone was stained red with osteocytes, osteoblasts and osteoclasts stained dark blue, residual polymer stained black, red blood cells stained turquoise and other cells stained a lighter blue. Quantifying the residual material (CPC or polymer), allograft bone, and new bone formation required the use of high magnification. Therefore, three zones progressing from the edge of the defect to the center region were examined at 40X magnification with and without polarizing the light. The edge of the defect was determined by visualizing (at 40X

magnification) and then marking the disruption of the linear pattern of the calvarial bone and cells resulting from the surgical creation of the defect. To differentiate between the new bone and the residual allograft the allograft bone was quantified in these zones by meeting the following three criteria: (1 ) acellular, (2) angular in shape, and (3) illuminated under polarized light. In addition, the total amount of bone in the defect area was quantified using a stitched image taken with an Olympus camera (DP71) at 10X magnification (Microscope Olympus SZX16). Adobe Photoshop (CS3) was utilized to stitch the images together and to complete the histomorphometry (Version 7.0.1). Histomorphometry data was obtained by using a color thresholding and an image layering technique to quantify the pixels of each layer and compare it to the total pixels in the area of interest.

[0231] Injectability of biocomposites

[0232] The working time of the biocomposite, as measured as the time after which the material could not be injected from the syringe, was 4.5 min. The tack-free time, corresponding to the time when the material did not stick to a metal spatula, was 12 min. Solid-filled suspensions typically exhibit a yield stress, which is the pressure that must be applied to initiate flow of the material. The viscosity data (Figure 14B) show that the biocomposite is shear- thinning, and the viscosity at 5 s ^"1 is 170 Pa*s.

[0233] Compression properties of the CPC and biocomposite

[0234] The porosity was 47 vol% and the pore size was 177 ± 90 μηι, resulting in an allograft fraction of 17.5 vol%. Representative stress-strain curves for the biocomposite and CPC measured under compression are shown in Figure 15. The CPC failed due to brittle fracture at 1.0D0.2% strain and exhibited compressive strength of 15.9±3.4MPa. In contrast, the biocomposite exhibited plastic behavior and did not fracture at strains up to 50%. The yield strength of the biocomposite was 4.06±0.03 MPa, above which the material continued to undergo plastic deformation up to 1 1.4±2.3% strain. The energy-to-failure, which is approximated by the area under the stress-strain curve, was 2971121 kJ m ^"3 for the CPC and 3122±404 kJ m ^"3 for the biocomposite.

[0235] Injection of the CPC and biocomposites in calvarial defects

[0236] During the surgical procedure, no treatment, the CPC or one of the biocomposite groups was injected in the defect, which had a volume of approximately 0.5 mL. A total of 0.25 mL of the biocomposite was used to fill the defect as it expanded in volume during cure. After cure, the both the CPC and biocomposite showed good contact with host bone. Some of the defects treated with the CPC developed cracks immediately after cure, which were observed before closure of the wound, while no cracks were observed for the biocomposites.

[0237] Radiographic Analysis [0238] Radiographs (Figure 17) of the negative control defects showed minimal bone formation near the edges of the defect at both 6 and 12 weeks, as anticipated for a CSD.

Consistent with observations during surgery, x-rays of the CPC treatment group showed cracking of the material. Bone ingrowth was observed around the perimeter of the biocomposite treatment groups with traces of bone in the center. X-ray images (Figure 16) of the BC+rhBMP- 2 group suggested a substantial increase in new bone formation within the defect relative to the other treatment groups.

[0239] In Figure 17A, the relative density (as approximated by the radio-opacity of the defect relative to the host bone) calculated using the CTAn software is plotted for each treatment group. While the CPC showed significantly higher relative density (p<0.02) compared to the other treatment groups at 6 weeks, the majority of the mineral content measured derived from residual hydroxyapatite and not new bone formation. There were no significant differences in relative density between the biocomposite treatment groups (p=0.08). Figure 17B shows the area % mineralization (as approximated by the percentage of the defect filled with tissue having density comparable to that of the host bone) for each treatment group. As expected, there was significantly less mineralized tissue in the negative control compared to the other treatment groups (pO.0001). In addition, the percent defect area covered was significantly greater in the CPC and BC+rhBMP-2 groups compared to the BC only group (p<0.05). However, since CTAn analysis cannot differentiate between calcium phosphate, allograft, or new bone within the mineralized tissue, differences between the CPC and BC+rhBMP-2 groups were not significant.

[0240] Histology and histomorphometry

[0241] Histological sections indicate that there were no adverse responses to any of the treatment groups used in this study. As expected, a fibrous scar filled the untreated defect at both time points (Figure 18A). The CPC treatment groups (Figure 18B) showed appositional bone growth around the surface and between the cracks of the material as evident by the

mineralization stained in pink. This pattern was the same for both the 6 and 12 week CPC groups. However, there was no cellular infiltration into the cement. Figure 19A-C shows a representative histological section of a biocomposite sample at the 6 week time point. Cells, stained light blue, migrated into pores initially present in the material due to the foaming reaction as well as those resulting from resorption of the allograft bone particles. Near the host bone/biocomposite interface, new bone lined with osteoid (stained light green) formed within the pores of the material. There was a moderate amount of residual polymer (stained black) remaining within the biocomposite. Representative histological sections at 12 weeks for the biocomposite treatment group (Figure 19D-F) showed extensive polymer degradation as well as new bone formation.

[0242] Representative histological sections of the BC+rhBMP-2 treatment group (Figure 20) revealed extensive bone growth around the composite as well as throughout the pores of the material. A higher magnification view of a region near the lower surface of the defect (Figure 20B) shows both intra-membranous and endochondral new bone formation, as evidenced by the presence of cartilage (C). Areas of active remodeling characterized by osteoid (O) and osteoblasts (OB) lining the surface of the bone are evident, as well as formation of new blood vessels (BV). A higher magnification view of a region near the upper surface of the defect (Figure 20C) shows residual allograft particles (A), residual polymer (black), and new bone formation (NB). Bridging of bone across the defect can also been seen in this histological section. While all rhBMP-2-treated defects showed new bone spanning the upper surface of the defect, calvarial defects in 5/10 animals in the BC+rhBMP-2 group had completely bridged with new bone at 6 weeks, which was significantly greater compared to the other treatment groups (p < 0.0009), in which complete bridging was not observed in any of the defects.

[0243] Figure 21 A shows the total area% of allograft and new bone measured over the entire defect for the biocomposite treatment groups. Total bone (the sum of residual allograft and new bone) in the biocomposites was greater than that in the negative control at both time points, and increased from 6 to 12 weeks. The addition of rhBMP-2 to the biocomposites resulted in significantly more bone at 6 weeks compared to the biocomposite at both 6 and 12 weeks without rhBMP-2. To determine how much of the total bone was newly formed versus residual allograft, three areas progressing from the edge to the interior of the defect (Figure 21B) were analyzed at high magnification. As shown in Figure 21C, new bone formation was highest in Area 1 (near the host bone interface) for all three treatment groups. However, the BC + rhBMP- 2 group had significantly higher new bone formation in the interior areas 2 and 3. The area% of allograft was <2% for all groups in all three areas, suggesting that most of the total bone in Area 1 of the biocomposite was newly formed and not allograft. At 6 weeks, the area % of new bone in areas 2 and 3 was comparable to the area% of allograft. However, at 12 weeks and in the biocomposites with rhBMP-2, the amount of new bone in the interior areas exceeded that of residual allograft. As anticipated, the polymer decreased significantly from an initial value of 24 - 29 area% at week 6 to 18 area% at week 12 (the difference was significant only for area 1 ). Interestingly, while the residual polymer was lower at 6 weeks in the presence of rhBMP-2, the difference was not significant, suggesting that delivery of this relatively low amount of rhBMP-2 does not substantially affect the degradation rate of the polymer.

[0244] The biocomposites exhibited handling properties, including working and setting times, that are comparable to those reported for CPCs. After injection, the biocomposites expanded to fill the defects and hardened to form a tough elastomeric solid that did not fail mechanically throughout the healing process, in contrast to the CPC that exhibited brittle fracture after cure. The mechanical integrity of the materials observed in vivo was consistent with their in vitro mechanical properties, as evidenced by the order of magnitude higher energy-to-failure of the biocomposites compared to the CPC. As early as 6 weeks, cells had infiltrated the biocomposites, resulting in new bone formation near the host bone/biocomposite interface, while the CPC showed minimal cellular infiltration. rhBMP-2 added to the biocomposites enhanced new bone formation, resulting in a bridge of bone covering the upper surface of the defect as well as new bone formation throughout the interior of the biocomposite.

[0245] The PUR biocomposite was shear thinning over a physiologically relevant range of shear rates (0.01 - 10 s ^"1 ). The initial viscosity of the biocomposite at 5 s ^"1 was 170 Pa*s. The relatively higher viscosity of the biocomposites is due in part to the higher viscosity of the liquid PUR components (21 Pa*s compared to 10 ^"3 Pa*s for water). The biocomposites showed a yield stress of only 2.1 Pa. Thus, PUR biocomposites may present handling advantages compared to CPCs due to their higher initial viscosity, which minimizes filter pressing and extravasation, and relatively low yield stress, which requires a smaller initial force to inject the material.

[0246] The biocomposite set to form a hard solid within 10 minutes of injection, which offers the advantage of wound closure shortly after placing the material. The biocomposites did not reveal evidence of cracking or fragmentation either immediately after cure or at the time of explantation. The superior mechanical integrity of the biocomposite is attributed to its tougher mechanical properties, having an energy-to-failure measured under compression of 3122 + 404 kJ m ^"3 compared to 297 + 121kJ m ^'3 for the CPC. Taken together, these observations suggest that the biocomposite may be more effective at providing early protection to the brain during the early stages of the healing process.

[0247] Rapid cellular infiltration and remodeling is another desirable attribute of injectable bone grafts. Histological sections (Figure 19) showed extensive cellular infiltration for all of the biocomposite groups. In the present study the volume fraction of allograft in the cured biocomposite was 17.4 vol%. In the rabbit calvarial defect and athymic rat femoral plug studies, the combination of pore and allograft volumes were 61.4 and 59.7 vol%, respectively. Thus the rapid cellular infiltration of these biocomposites is consistent with the notion that cellular infiltration and remodeling proceed independent of polymer degradation when the sum of the pore and osteoconductive matrix volumes approaches 64 vol%, the random close-packing (RCP) limit for spheres.

[0248] The polymer, which was initially present at 36 vol%, had degraded to 24 - 29 area% at week six and 18 area% at week 12. These data suggest that the polymer had degraded by 19 - 33% at week 6 and 50% at week 12, which is in reasonable agreement with an in vitro study reporting 10% and 45% mass loss of the polymeric scaffold having the same composition at 6 and 12 weeks, respectively. In the previous in vitro study, the tensile strength and modulus of the scaffolds decreased to <20% of their initial values after 8 weeks of degradation time in vitro. When rhBMP-2 was added to the biocomposites, histological sections showed a bridge of new bone covering the upper surface of the implant as well as new bone formation throughout the defect. While the area % polymer was less in the biocomposites + rhBMP-2 group compared to the biocomposite group at 6 weeks, the differences were not significant. Thus rhBMP-2 released from the matrix at doses equal to 20% of that recommended for the ACS carrier does not significantly affect the degradation of the PUR phase.

[0249] The improvement in new bone formation at sub-optimal doses was attributed to the more sustained release of rhBMP-2 from the PUR carrier, compared to the bolus release (>30%) of rhBMP-2 from the collagen sponge. Extensive vascular formation in the defect was observed in the biocomposites incorporating rhBMP-2. In addition to its osteoinductive and angiogenic effects, rhBMP-2 also stimulates osteoclast activity. Thus rhBMP-2 released from the biocomposites can accelerate the resorption of allograft bone particles and the consequent infiltration of cells and growth of new bone in the newly formed pores.

[0250] Example 1 1 - Putty Rabbit Femur

[0251] This Example shows the remodeling of injectable ABP/PUR biocomposites in a NZW rabbit femoral condyle plug defect model. Reducing the volume fractions of allograft particles from 67 vol% to 57 vol% were used to slow the rate of cellular infiltration, resulting in more balanced remodeling. The potential for rhBMP-2 to enhance new bone formation and support balanced remodeling in the low-porosity biocomposites is also shown.

[0252] Materials

[0253] See previous Example.

[0254] Preparation of rhBMP-2

[0255] The rhBMP-2 was supplied as a solution comprising 35% acetonitrile and0.1% TFA. A separate acetonitrile/TFA solution was prepared containing a ratio of 10: 1 of

trehalosedehydrate:heparin sodium. The rhBMP-2 and trehalose mixtures were combined such that the ratio of rhBMP-2 to trehalose was 1 : 125. The resulting mixture was distributed in glass vials and frozen at -80°C in preparation for freeze-drying, which produced a powder.

[0256] Synthesis ofAMBP/PUR Putty

[0257] The method of Example 12 was implemented, but the target index was 130 and the catalyst concentration was 5500 ppm. The filler content (AMBP and rhBMP-2 powder) was maintained at 70 wt% for each putty treatment group, and rhBMP-2 was utilized at low (100 μg/mL) and high (400 μg/mL) concentrations. The resulting reactive paste had a tack-free (i.e., cure) time of approximately 10 minutes.

[0258] Mechanical properties

[0259] Cylindrical samples of each treatment group were prepared for mechanical testing. The reactive paste was transferred into cylindrical plastic cups and a 1 -pound weight (20.7 psi) was placed on the material for 10 minutes. The resulting cylinders were placed in a vacuum oven at 37°C overnight and removed from the plastic cups. After cure, the cylinders were removed from the cups and cut using a Buehler saw to produce 6 mm x 12 mm cylinders. Three different formulations were synthesized. After 24 hours of hydration in phosphate buffered saline (PBS), the rods were tested using a MTS 898 using compression.

[0260] Animal Study

[0261] Forty-two New Zealand White (NZW) rabbits weighing between 3.8 and 4.1 kg were used in this study. All surgical and care procedures were carried out under aseptic conditions per the approved IACUC protocol. The AMBP/PUR putty components were gamma irradiated using a dose of approximately 25 kGY. Glycopyrrolate was administered at 0.01 mg/kg IM followed by ketamine at 40 mg/kg IM. Bilateral defects of approximately 6 mm diameter by 1 1 mm in depth were drilled in the metaphysis of the distal femurs of each rabbit. AMBP/PUR plugs from each treatment group were subsequently injected into each defect using a 1 mL syringe. Treatment groups for each composite were dispersed randomly among the rabbits. The rabbits were euthanized at both 6 and 12 week time points using Fatal-plus (2.2 mL/10 kg) intravenously.

[0262] μΟΤ Analysis

[0263] A / CT40 (SCANCO Medical, Basserdorf, Switzerland) was used to acquire images of the biocomposites prior to implantation and of the extracted femurs post implantation at 6 and 12 weeks. Briefly, μΟΤ scans were performed at 70 kV source voltage and 1 14 μΑ source current and at a spatial resolution of 36 μιη and manufacturer provided software was used to reconstruct axial images. The reconstructed image stack was converted from a TIFF to a BMP format using ImageJ (U. S. National Institutes of Health, Bethesda, Maryland)and reoriented to be perpendicular to the axis of the femoral condylar defect created using the cortical borders of the defect site for alignment. Images were calibrated using hydroxyapatite phantoms with densities of 100- 800 mgHA/cc. μΟΤ thresholding was performed to include only ossified tissues (defined as a density greater than 346 mgHA/cc)by using the Otsu algorithm applied across all the samples in the entire study. The region of interest was registered in each sample using the defect boundary to define a 6 mm circular section in the first slice at the outer cortical surface of the femur and proceeding to a depth of 1 1 mm to generate a cylindrical region of interest. The bone area within the 6 mm circular cross sections were calculated for each section from the femoral cortex to the interior of the defect space and grouped by treatment and time point to determine spatial trends in bone regeneration. The bone volume (BV), tissue mineral density (TMD) and multiple 3D morphometric descriptors of trabecular organization and connectivity within this region of interest were computed using CTAn software (Skyscan, Aartselaar, Belgium). The relative distribution of trabecular thickness was calculated as the relative percentage of trabeculae in each range of thickness for the biocomposites prior to implantation and after implantation at 6 and 12 weeks to evaluate bone remodeling response within the grafts.

[0264] Histology and histomorphometry

[0265] The femora were placed in a solution of 10% formalin for two weeks followed by a series of ethanol dehydrations. After fixation, the femurs were embedded in poly(methyl methacrylate)and 200-μπι sections were cut from the resulting blocks using an Exakt band saw. The sections were then ground and polished using an Exakt grinding system to less than 100 μ ^η ι and stained with Sanderson's rapid bone stain counterstained with van Gieson. Residual allograft bone particles stained light brown, residual polymer stained black, new bone stained pink with dark blue osteocytes within the matrix, red blood cells stained torquiose, and other cells stained a lighter blue. Residual allograft bone particles and new bone formation were quantified in an area of interest 1.5 cm high x 6 cm wide located in the center of the defect. Images were taken at 40X magnification with an Olympus camera (DP71) using a Microscope Olympus SZX16 microscope with and without polarizing the light. To differentiate between the new bone and the residual allograft the allograft bone was quantified by meeting the following three criteria: (1) acellular, (2) angular in shape, and (3) illuminated under polarized light.

Metamorph was utilized to complete the histomorphometry (Version 7.0.1). Histomorphometry data was obtained by using a color thresholding and an image layering technique to quantify the pixels of each layer and compare it to the total pixels in the area of interest.

[0266] Mechanical properties

[0267] Representative compression and torsion stress-strain curves measured for the biocomposites are shown in Figure 22 and compared to the triphasic calcium phosphate cement ProDense . The Young's modulus, yield stress, yield strain, and energy-to-failure (area under the curve) for the biomaterials are presented in Table 1. There were no significant differences between the mechanical properties of each treatment group as the strength and modulus values ranged from 24.2-28.1 MPa and 357.3-503.0 MPa, respectively.

Table 1. Mechanical properties of the injectable biocomposite with no rhBMP-2 (BC) and the calcium phosphate cement (CPC) measured under compressive and torsional loads. Data are reported as the mean ± SEM.

[0268] μΟΤΩαία

[0269] Representative μCT images of the biocomposites and control groups after 6 and 12 weeks implantation are presented in Figure 23. Minimal new bone formation, primarily in the region of the femoral cortex, was observed for the empty and AMBP-treated groups at 6 and 12 weeks. This data suggests that these defects did not heal and that the allograft had resorbed without the polymer binder to maintain its structure. All putty treatment groups showed evidence of resorption of allograft particles (irregularly shaped bright white particles with sharp edges) and remodeling. Incorporation of rhBMP-2 in the putty appeared to enhance remodeling of the composites.

[0270] The total bone content was measured by μCT for each treatment group as shown in Figure 24. BV/TV was measured for radial cross-sections as function of distance from the cortex is shown in Figure 24A primarily to indicate spatial regions along the defect where changes in bone volume are occurring between week 6 and week 12. In the empty treatment group at both 6 and 12 weeks, the average volume of regenerated bone in the first 2 mm from the exterior surface adjacent to the femoral cortex was greater than the average volume of regenerated bone per section within the remainder of the defect (interior 9 mm depth) adjacent to the trabecular marrow. From 6 to 12 weeks, the AMBP treatment group showed a decrease in sectional bone volume immediately adjacent to the femoral cortex indicating severe localized resorption. The AMBP/PUR biocomposite with no rhBMP-2 showed very little variation in bone distribution from the cortex to the defect interior at both 6 and 12 weeks. At the low rhBMP2 dose, from 6 to 12 weeks an increase in bone volume was observed, whereas at the high rhBMP-2 dose bone volume in the interior adjacent to the trabecular marrow decreased with very little change in the bone volume adjacent to the femoral cortex (Figure 24A).

[0271] Volume-average BV/TV and TMD are shown in Figure 24B. The BV/TV within the empty and AMBP groups was significantly lower (pO.01 ) than the BV/TV in the biocomposite groups both with and without rhBMP-2 at both time points. No significant change was observed in the BV/TV within any treatment group from 6 to 12 weeks. While there was no significant change in the TMD between all 5 treatment groups at both 6 and 12 weeks, it was observed that there was a significant increase between pre-implantation TMD and post implantation TMD at 6 and 12 weeks for the biocomposite groups (no rhBMP2 p<0.01 , l OOmg rhBMP2 p=0.05, 400mg rhBMP2 p<0.03).

[0272] At week 0 (representing the pre-implantation architecture), a majority of the trabeculae have an average thickness of 190 mm and a very narrow distribution. All biocomposite groups (with or without rhBMP-2) show a much broader trabecular thickness distribution with a mean thickness of 330 mm after in vivo implantation for 6 or 12 weeks. However, while the group with no rhBMP-2 shows almost no variation in thickness distribution between 6 and 12 weeks, greater changes in the percentage of trabeculae in each thickness range are observed between 6 and 12 weeks for the biocomposites loaded with rhBMP2. In the lower dose ( 100 μg rhBMP-2) group, the distribution becomes uniformly broader from 6 to 12 weeks, indicating an increase in thickness of some trabeculae as well as a percentage of trabeculae with lower thickness which could be attributed to either resorption or the initialization of new ossification. In the higher dose (400 μgrhBMP-2) group, an increase in the larger trabeculae is observed with little change in the smaller trabeculae indicating a stronger appositional growth trend from 6 to 12 weeks. A rhBMP-2 dose dependent increase in bone remodeling was seen between 6 and 12 weeks with the higher dose exhibiting greater remodeling.

[0273] Histology and histomorphometry

[0274] Histological sections of the empty and allograft-treated defects show minimal new bone formation, which is consistent with the μΟΤ data. In contrast, histological sections of the biocomposite treatment groups (Figure 25) reveal evidence of cellular infiltration (C), allograft (A) resorption, and new bone formation (NB). High magnification views show regions of active remodeling, osteoid formation, and appositional growth of new bone on residual allograft particles, suggesting that the biocomposites remodel by creeping substitution.

[0275] Histomorphometric analysis of the area of interest (AOI) is shown in Figure 26 and also shows that rhBMP-2 is accelerating remodeling.

[0276] Discussion

[0277] Incorporation of rhBMP-2 enhanced new bone formation at 12 weeks relative to the biocomposite without rhBMP-2, as evidenced by the presence of fewer allograft bone particles (irregularly shaped white particles). Similar results were observed at 6 weeks. The initial release of rhBMP-2 from the AMBP/PUR composites stimulates the differentiation of osteoprogenitor cells to osteoblasts, which subsequently regulate osteoclast differentiation through production of Receptor Activator for Nuclear Factor Β ligand (RANKL). In addition to its role of indirect regulation of osteoclasts through RANKL, rhBMP-2 can also directly stimulate osteoclast differentiation, and the concentration of rhBMP-2 must be maintained below a threshold to prevent excessive resorption.

[0278] Example 12

[0279] This Example shows the effects of stoichiometry and catalyst concentration on the reactivity, injectability, and biocompatibility of injectable PUR/allograft bone biocomposites. The biocompatibility of the biocomposites as well as the inflammatory response was evaluated in a rabbit femoral condyle plug defect model at 8 and 16 weeks. [0280] Materials

[0281] See previous Examples. Bovine mineralized bone particles (B-MBP) were obtained from Medtronic, Inc (Minneapolis, MN). All other reagents were purchased from Aldrich. DPG was dried over 4 A sieves before use. TED A was dissolved in a 10% (w/v) solution with dry DPG. Excess organic material (e.g. fat) was removed from B-MBPs with a chloroform/acetone solution before use. The B-MBPs were then sieved to include only 105-500 μιη particles.

[0282] Component synthesis

[0283] The hydroxyl (OH) number of the polyester triol was measured by titration according to ASTM D4274-99 Method C, and the molecular weight was determined by gel permeation chromatography (GPC, Water Breeze). The polyester triol was composed of 60% ε- caprolactone, 30% glycolide, and 10% D,L-lactide monomers. Before use the polyol was washed with hexane dried under vacuum at 80° C for 24 h. The %NCO of the prepolymer was measured by titration according to ASTM D2572-97. The prepolymer was maintained at 4° C under argon prior to use. Water content for all components was determined by Karl Fischer (KF) titration with a 798 MPT Titrino with a 10 mL burette (Metrohm). Briefly, 0.5-5.0 g of material was dissolved in dry methanol. Hydranal-Composite 2 (Sigma-Aldrich), a stock KF reagent, was used to titrate the samples.

[0284] Synthesis of biocomposites

[0285] Biocomposites (BCs) were prepared by adding the polyester triol, catalyst solution, and B-MBP (45 wt%) to a mixing cup, in which they were hand mixed for 30 seconds before adding the prepolymer and hand mixing for an additional 45 seconds. The composite was then loaded into a syringe. The study design in summarized in Table 2.

T6C3G 1 L900 (wt%) 30.1 28.8 22.3 21 .3

DPG (wt%) 2.3 4.4 2.3 4.4

Table 2. Biocomposite formulat

[0286] Two catalyst weight percentages (0.50 and 0.25 wt%) and two index values (108 and 195) were utilized. The index characterizes the stoichiometry and is the ratio of isocyanate (NCO) equivalents in the prepolymer to the sum of the hydroxyl (OH) equivalents in the polyester triol and water.

[0287] A TR-FTIR analysis of the reacting system

[0288] Attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) measurements were conducted with a Seagull Variable Angle Reflection Accessory (Harrick Scientific) applied to a Tensor 27 FTIR instrument (Bruker). A ZnSe hemispherical crystal (Harrick Scientific) was utilized to obtain time-resolved ATR spectras. For each reaction characterized, spectra were taken every 20 to 60 seconds at a resolution of 4 cm ^" ' and 56 scans per spectra. Briefly, to obtain the spectral profiles for the reactions of the biocomposites, a given composite was synthesized and placed on a sample holder in direct contact with the bottom of the ZnSe crystal. To derive the spectral profiles for the individual component reactions of the biocomposites, the components were mixed with the prepolymer and catalyst only. The isocyanate peak (2270 cm ^"1 ) was deconvoluted and integrated using a MATLAB program and a calibration curve was used to correlate integrated peak values with known concentrations of isocyanate (described fully in the Supplemental Data). The analysis was completed in triplicates for each reaction analyzed.

[0289] Porosity as a function of water concentration

[0290] To determine porosity as a function of water content, biocomposites were prepared with 0- 1.0 wt% added water and porosity was measured gravimetrically. Briefly, each 0.5 g batch of biocomposites was injected via a large diameter syringe into cylindrical molds where they were allowed to react overnight at room temperature. Triplicate slices of the cylinders were cut from the fully reacted biocomposites and measured with calipers to determine the volume. Scanning electron microscopy (SEM, Hitachi S-4200) micrographs were obtained and analyzed for pore size using MetaMorph 7.1 Image Analysis software (MDS Analytical Technologies). The mass of each slice was used to obtain the density, and the measured density was compared to the theoretical density to calculate the porosity.

[0291] In vitro porosity

[0292] Porosity measurements were completed for biocomposites with index values of 108 and 195 with either 0.50 or 0.25 wt% catalyst. Biocomposites were injected, immediately after mixing, into 2 mL of deionized water and allowed to react overnight in an incubator at 37° C. At least three cylindrical cores were taken from each sample and analyzed gravimetrically to obtain porosity as described above.

[0293] Rheology of curing biocomposites

[0294] The Theological profiles during cure were measured for each biocomposite in situ with an AR 2000ex rheometer with a Rheology Advantage AR Controller (TA Instruments). Continuous oscillation measurements were conducted at 1 Hz and 1% strain with 25 mm disposable parallel plates and a 1 mm gap. Measurements for each configuration of

biocomposites were taken with either dry conditions or submerged in water via a submersion assembly kit (TA Instruments). Initial viscosities (η ',) and working times (gel points, r _w ) were tabulated.

[0295] Characterization of intermediates leached from the reactive biocomposites

[0296] In order to determine whether cytotoxic reactive intermediates leach from the biocomposite during cure, in vitro leaching experiments were performed. Briefly, 2.5 g of each biocomposite were injected into an empty vial, and after 2 min (after mixing was started) 5 ml PBS was added to the vial. For the second time point, 2.5 g of each biocomposite were injected into a sample cup and transferred to a vial filled with 5 mL of PBS after 45 min post-mixing. For the leachate cytotoxicity experiments (Section 2.9), leachates were collected in a-minimum essential medium (a-MEM) with 10% (v/v) fetal bovine serum (FBS) and 1 % (v/v)

penicillin/streptomycin using the procedure described above. The samples were maintained at 37°C for 72 hours, at which time the PBS was removed, the pH measured, and the samples subsequently lyophilized and weighed. After reconstitution in solvent, the composition of the residue was characterized by NMR. The spectra were compared to those of the pure components in the biocomposites to determine the presence of individual components in the leachates. [0297] The cytotoxicity of the leachates from the biocomposites was measured using MC3T3-E1 embryonic mouse osteoblast precursor cells in vitro. Cells were seeded in a 96-well plate with a density of 5x 10 ³ cells per well and cultured in a-MEM with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) penicillin/streptomycin in a C0 ₂ incubator with 5% C0 ₂ at 37° C. The concentration of the leachates varied from 6.15% (16X dilution with serum medium) to 100% (IX). The culture medium was changed every 2 days. Trypsin-EDTA was used for dissociation of MC3T3-E1 cells. The cells were analyzed for viability using a Live/Dead Viability kit (Invitrogen). The assay was completed as recommended by the manufacturer's instructions. Cells were analyzed after 24 hours exposure to the leachate solution. Triplicates for each group were analyzed with control groups treated with blank PBS. All experiments were conducted in accordance with ISO- 10993-5.

[0298] In vivo biocompatibility and new bone formation in a rabbit femoral condyle plug defect model

[0299] Eighteen New Zealand White (NZW) rabbits weighing between 4.0 and 5.4 kg were used in this study. All surgical and care procedures were carried out under aseptic conditions per the approved IACUC protocol. The components of the biocomposites were gamma-irradiated using a dose of approximately 25 kGY. Glycopyrrolate was administered at 0.01 mg/kg IM followed by ketamine at 40 mg/kg IM. Bilateral cylindrical defects of approximately 5 mm diameter by 1 1 mm in depth were drilled in the metaphysis of the distal femurs of each rabbit under copious sterile saline irrigation using a trephine in a MicroAire handpiece. Materials from each treatment group were subsequently injected into each defect using a syringe, made flush with the cortical surface and allowed to harden. Closure was attained using a 3-layered approach comprising muscle, fascia, and subcuraneous 3-0 Vicryl sutures. Skin glue was applied topically to maintain closure. Treatment groups for each composite were dispersed randomly among the rabbits. The rabbits were euthanized at both 8 and 16 week time points using Fatal-plus ( 1 mL/4.5 kg) intravenously.

[0300] μΟΤ analysis

[0301] See previous Examples.

[0302] Histology [0303] Harvested femoral condyles were fixed in 10% neutral buffered formalin at room temperature for one week. The samples were then decalcified in hydrochloric acid, dehydrated and embedded in paraffin. After embedding, the samples were then sectioned onto slides at 4 microns thick and stained using hematoxylin/eosin (H & E stain). Ground sections were also prepared by immersing the femora in a solution of 10% formalin for two weeks followed by a series of ethanol dehydrations. After fixation, the femurs were embedded in poly(methyl methacrylate) and 200- m sections were cut from the resulting blocks using an Exakt band saw. The sections were then ground and polished using an Exakt grinding system to less than 100 m and stained with Sanderson's rapid bone stain counterstained with van Gieson. Residual allograft bone particles stained light brown, residual polymer stained black, new bone stained pink with dark blue osteocytes within the matrix, red blood cells stained turquoise, and other cells stained a lighter blue.

[0304] Thin (5 μηι) histological sections stained with H&E were evaluated using a subjective scoring system. Inflammation, granulation tissue, reactive bone formation, marrow edema, and synovitis were all evaluated on a scale of 0-5 (0=normal, l=minimal, 2=mild, 3=moderate, 4=marked, and 5=severe). The samples were also given an overall effect score in which the following criteria were evaluated: (1) no discernable defect, (2) visible circular defect surrounded by a complete to vague circle of bone filled with fat and marrow elements, and (3) a circular area of dropout, surrounded by proteinaceous fluid and a thin rim of new bone with minimal to mild inflammation.

[0305] Results

[0306] Reactivity of PUR biocomposites

[0307] The individual components of the biocomposite (polyester triol, DPG, B-MBP, and water) were analyzed for their reactivity with the NCO-terminated prepolymer. The conversion of NCO equivalents in the prepolymer was monitored in situ by ATR-FTIR, which was analyzed to obtain the second-order rate constants for each reaction at each catalyst level. The same technique was applied for the overall reaction of the biocomposite at two different indices and catalyst levels. [0308] Concentration of NCO equivalents versus time for the overall biocomposite reaction, polyester triol concentration, DPG concentration, and B-MBPs concentrations, at high (0.50 wt%) and low (0.25 wt%) catalyst concentration, were monitored. Each of the reactions was found to follows a second-order mechanism as anticipate, and thus the slope of line is equivalent to the rate constant for each of the reactions. Water has the highest reactivity compared to the other reactions, regardless of catalyst concentration. The polyester triol is approximately 20 times less reactive than water for the higher catalyst level, while the DPG is approximately 3 times less reactive than the triol. The reactivity of the B-MBPs is the lowest of all the components at both catalyst levels.

[0309] Based on the rate constants of the individual components, a kinetic model was developed to predict the overall reactivity of the biocomposites. The equivalent balance equations were then solved to calculate the concentration profiles of each component as a function of time.

[0310] Using the fitted rate constants and the initial concentration of equivalents, the overall conversion of NCO equivalents in the biocomposite was plotted and compared to the

experimental values. Due to difficulties associated with accurately measuring the concentration of water in the polyester triol, catalyst solution, and prepolymer, the initial water concentration was used as a fitting parameterThe water concentrations measured by titration varied from 15 to 36% of the fitted values. While the water conversion approaches 100% after approximately 10 - 20 minutes, the conversions of the other components are less than 100%, and decrease with decreasing index.

[0311] Effect of water on biocomposite porosity

[0312] Reaction of the NCO-terminated prepolymer with water yields carbon dioxide gas, which acts as a blowing agent resulting in the formation of pores. The porosity of the biocomposites as a function of total waterincreases with water concentration up to a plateau value of 60 vol% independent of the catalyst level or index. Using SEM, it was observed that biocomposites at 0.2, 0.4, and 1.0 total wt% water have pore diameter, porosity, and

interconnectivity increase with water concentration. While the pores are predominantly closed at the lower water concentrations, they appear to be more interconnected at 1.0 wt% water. It was observed, pore diameter is independent of index and catalyst concentration and increases with total water concentration, but the differences are not significant.

[0313] Under in vivo conditions, water from the wound bed can diffuse into the

biocomposite, resulting in increased expansion and porosity. The effects of water diffusion were simulated in an in vitro test where the biocomposites are reacted in an aqueous environment. At the lower catalyst concentration, both indices yield biocomposites with porosities of 48 - 55%, and at the higher catalyst level the index 210 biocomposite results in 50% porosity. Thus the porosities obtained under wet cure exceed those obtained from dry cure (9 - 20%). In contrast, the index 1 15 biocomposite has a porosity of 22%, which is comparable to the 17% porosity measured for the biocomposite cured under dry conditions with no added water. These observations suggest that diffusion of water from the wound bed can significantly increase expansion, particularly at the low catalyst concentration and high index.

[0314] Characterization and cytotoxicity of leachates in vitro

[0315] NMR spectra for the leachates from I0-C 1 biocomposite injected into PBS at 2 and 45 minutes after mixing were compared to spectra for the individual components to determine which components were leaching from the reactive polymer at time points corresponding to the cream (2 min) and tack-free (45 min) stages of cure. The other biocomposites had nearly identical spectral profiles to that of I0-C1. The peak at 2.3 ppm associated with the proton adjacent to the carbonyl group in the polyester appears in the spectra of the leachates collected at 2 and 45 min, suggesting that unreacted polyester triol had leached into the medium. Similarly, the peak at 1.0 ppm associated with the protons on the methylene carbon group in DPG also appear in the leachates at both time points, indicating that unreacted DPG had diffused into the medium. In contrast, the prepolymer was uniquely distinguished by a series of peaks above 6 ppm, none of which appeared in the spectra for any of the leachates, suggesting that the prepolymer did not leach into the medium. Gravimetric analysis of the leached biocomposites revealed a 0.2 - 1.2% mass loss due to diffusion of individual components from the

biocomposites into the buffer. The pH of the leachates recovered at 2 and 45 minutes varied from 6.6 to 6.8 compared to the initial value of 7.35. [0316] MC3T3-E 1 murine osteoprogenitor cells were treated with leachates from the biocomposites collected at 2 and 45 min and diluted with serum medium such that the final concentration of leachates ranged from 6.25% (16X dilution) to 100% (I X dilution). Cells were cultured for 24 h. Leachate dose-response curves measured for leachates collected at 2 and 45 min reveal the anticipated sigmoidal shape. Furthermore, three of the eight treatment groups showed cytotoxicity, which is defined as <70% viability. For the three treatment groups showing cytotoxicity, the dilution factors required to render the culture medium non-cytotoxic varied from 1.60 - 1.81. For a specific biocomposite composition and dilution factor, the percentage of viable cells was generally higher for leachates collected at 45 min (except for the I X dilution for U -Cl), which is consistent with the notion that the concentration of leachates was lower at 45 min due to the higher conversion. Using cell culture on tissue culture polystyrene stained with calcein, and a mesenchymal phenotype was observed when treated with leachates from biocomposite I0-C1 collected at 2 and 45 min and diluted 8X with serum medium. Control cells treated with PBS show a similar morphology.

[0317] Rheological properties of biocomposites.

[0318] Working times varied from 7 - 29 min, and decreased with increasing catalyst concentration and index. Similarly, initial viscosities ranged from 90 - 900 Pa*s, and increased with increasing catalyst concentration and index. Working time and initial viscosity measured under wet conditions were generally within 15% of values measured for dry conditions.

[0319] In vivo inflammatory response and remodeling in a rabbit femoral plug model

[0320] Results from the histological scoring of H&E sections at both 8 and 16 weeks show that the defects in the control (empty defect) treatment group had a central area of fat and hematopoietic elements surrounded by a variably vague circle of bone and trabeculae. There appeared to be very little inflammation within the control group. At 8 weeks, defects treated with the I0-C 1 biocomposite consisted of nonviable bone fragments surrounded by osteoclasts, osteoblasts, new trabeculae, and marrow elements. There was mild to moderate inflammation and edema surrounding and within the area of the defect. At 16 weeks, the biocomposite group defect contained no non- viable bone fragments, fewer new trabeculae, and decreased inflammation and edema. [0321] Discussion

[0322] The biocompatibility of the I0-C1 formulation that showed the most predictable performance under wet conditions was evaluated in a rabbit femoral condyle plug defect model at 8 and 16 weeks, which showed cellular infiltration, new bone formation, complete resorption of the polymer at 16 weeks, and a mild inflammatory response.

[0323] The water reaction is referred to as the blowing reaction due to the production of carbon dioxide, which creates pores in the biocomposite. Thus the water reaction can be exploited to generate >50 μπι pores in the biocomposite to accelerate cellular infiltration. To balance the requirements for both mechanical strength and cellular infiltration, expansion of the biocomposites must be controlled such that porosity < 55 vol%. The porosity of the I0-C0, I I - C0, and Il -Cl with no added water increased from 8 - 20% under dry conditions to 50 - 55% under wet conditions. These observations suggest that formulations I0-C0, I I -CO, and Il -C l undergo unpredictable expansion in vivo.

[0324] Over-expansion due to diffusion of external water may also be mitigated by the choice of catalyst. Due to the cytotoxicity of heavy metal urethane catalysts (e.g., dibutyl tin dilaurate), tertiary amine catalysts have been investigated for synthesis of biodegradable polyurethanes. While tertiary amines are known to catalyze both the gelling (polyester triol) and blowing (water) reactions, TEDA is known as one of the strongest amine gelling catalysts.

However, despite the relatively strong gelling activity of TEDA, the water reaction was the fastest for both catalyst levels. More potent gelling catalyst with moderate toxicity, such as ferric acetylacetonate, is anticipated to limit the effects of water in biocomposites.

[0325] While formulation I0-C 1 minimized the effects of external water on expansion, the lower index resulted in a lower conversion of polyester and DPG. Incomplete conversion may result in network defects that reduced the mechanical properties of the cured polymer. The rheology data suggest that after 60 min the biocomposites have formed a crosslinked network.

[0326] At the early stages of the curing process before the gel time, the NCO conversion is low (e.g., 10 - 20% at 2 min), and thus leaching of reactive intermediates is anticipated to occur. However, neither prepolymer nor TEDA was identified in NMR spectra leachates at any conditions, and only a small amount (e.g., <2%) of polyester and DPG were leached from the biocomposites at both 2 and 45 min. These observations are consistent with the predictions of the kinetic model, which showed that the conversion of polyester triol and DPG were highest in the II -C I materials (96% and 68%, respectively) and lowest in the II -CO materials (35% and 47%, respectively). When diluted 8: 1 with fresh buffer, the leachates had no adverse effect on cell viability. Thus injectable PUR biocomposites do not pose the risk of releasing cytotoxic catalysts, solvents, or reactive intermediates to the surrounding tissue.

[0327] Lowering the index and increasing the catalyst concentration may allow one to limit porosity can be limited to below 30% when injected into an aqueous environment, ensuring good mechanical and wound healing properties. The reacting biocomposites have been found to leach low amounts of non-cytotoxic products during the curing process.

[0328] Example 13

[0329] Delivery of rhBMP-2 combined with allograft may result in transient resorption, β- Tricalcium phosphate (β -TCP) is a biocompatible, resorbable ceramic that has been used effectively as a substitute for allograft bone. In the present Example, the ability of an injectable PUR/ -TCP composite with rhBMP-2 to heal bone defects is shown.

[0330] The biodegradable polyurethane was synthesized from a lysine triisocyanate (LTI) prepolymer and polyethylene glycol (PEG), a polyester triol (900 g/mol), and triethylene diamine catalyst. The prepolymer, polyester, and β-TCP were mixed in a mixing cup by hand for 60 seconds. The mixture was then transferred to a vial containing the lyophilized rhBMP-2 powder, the catalyst added, and the resulting paste hand-mixed for an additional 60 seconds.

Biocomposites were injected into 8-mm critical-size calvarial defects in rats. Animals were sacrificed at 4 weeks and new bone formation evaluated by radiographs, μCT, histology, and histomorphometry. Treatment groups included the biocomposite containing 45% TCP with and without 200 μg/mL rhBMP-2. Pores were generated by the reaction of water in the biocomposite with the NCO-terminated prepolymer, resulting in the formation of carbon dioxide gas. The porosity of materials cured in vitro was compared to that of samples injected in vivo

gravimetrically and using SEM analysis. [0331] The porosity of bone grafts is important for control of rhBMP-2 release and cellular infiltration. SEM images of biocomposites cured under in vitro (A) or in vivo (B) conditions are shown in Figure 27. The porosity of biocomposites injected in vivo ranged from 40 - 50%, which was comparable to that of materials cured in vitro. Thus, the effects of diffusion of water from the wound bed under in vivo conditions did not adversely affect cure of the biocomposite, resulting in predictable cure. Representative μΟΤ images taken at 4 weeks (Figure 28) show that the injected biocomposite completely filled the defect for samples with and without rhBMP-2 (n=13 per group). In the biocomposites without rhBMP-2, there is evidence remodeling near the perimeter of the graft in contact with host bone, as suggested by the increased density (white color) near the host bone interface. Addition of rhBMP-2 resulted in both new bone formation as well as bridging of the defect with new bone at 4 weeks (Figure 29B). These observations suggest that the biocomposites supported cellular infiltration and remodeling, and that rhBMP-2 enhanced healing.

[0332] Example 14

[0333] In this Example, the effects of Bioglass 45 S5 (BG) surface modification on the bioactivity and mechanical properties of PUR/BG composites are investigated. Prior to reaction with the PUR binder, BG particles were functionalized with the silane coupling agent 3- aminopropyl-trietoxysilane (APTES), which has been shown to increase the mechanical compressive strength of BG, as well as surface grafting of polycapro lactone (PCL) to enhance interfacial bonding.

[0334] Materials & Methods

[0335] Surface modification of BG with APTES and PCL (from ε -caprolactone monomer and Sn(Oct) ₂ catalyst) was based on known protocols. A flat BG disk model was utilized to evaluate the effect of the surface modifications on the properties of the BG used in the PUR composite. The disks were characterized by contact angle and X-ray photoelectron spectroscopy (XPS). In order to test bioactivity, BG disks were immersed in simulated body fluid (SBF) for various amounts of time over a 14 day period. Bioactivity was assessed by measuring the formation of hydroxyl carbonate apatite using wide-angle X-ray diffraction, scanning electron microscopy, and energy-dispersive x-ray spectroscopy (EDS). Cylindrical biocomposites were prepared from a lysine triisocyanate- poly(ethylene glycol) prepolymer, triethylene diamine catalyst, PCL triol (Mn -300 g/mol), and BG (46.3 volume %). Mechanical testing was completed in compression mode.

[0336] Results & Discussion

[0337] The presence of APTES and PCL (via Sn(Oct) ₂ ) on the surface of the BG disks was detected via XPS based on the presence of the Nl s (5.02 at. %) and Sn3d (0.35 at. %) spectra, respectively. A change in advancing contact angle compared to unmodified bioglass was observed. The contact angles for unmodified, silanized, and PCL surface-modified BG disks were 19 ± 3°, 45 ± 3°, and 66 ± 1.73°, respectively. EDS was used to calculate the Ca/P ratio and compared to the value of apatite (1 .67), as an indicator for complete coverage of the surface with apatite. As shown in Table 3, Ca/P for unmodified BG at 7 days was smaller compared to PCL- modified

Time (days) Unmod. BG PCL-mod. BG

0 ίΓδΟ 5l37

7 2.09 2.19

14 1.67 1.66

Ratio based on Ca and P atomic percent, obtained from EDS

Table 3. Ca/P of BG disk surface after immersion in SBF

[0338] From this result, it appears that the bioactivity of PCL-modified BG was slightly delayed compared to unmodified bioglass. PUR composites incorporating unmodified BG exhibited an ultimate yield strength and Young's modulus of 4.01 ± 0.53 MPa and 46.26 ± 2.97 MPa, compared to 58.49 ± 5.32 MPa and 2185.71 ± 422.75 MPa, respectively, for PCL-modified BG composites. Thus, the overall mechanical properties of PUR/BG composites are dramatically improved with the use of PCL-modified BG particles compared to unmodified BG. The increase in strength is attributed to improved adhesion between the BG and PUR phases due to the similar contact angles (66° for the PCL-modified BG compared to 66° for the PUR phase). Furthermore, the OH-terminated PCL chains are anticipated to react with the NCO groups in the LTI-PEG prepolymer, resulting in increased covalent binding. A dose-response experiment is ongoing to identify the surface coverage of grafted PCL that maintains the desired bioactivity of the BG particles while also attaining the mechanical properties required for functionally weight-bearing bone grafts.

[0339] BG particles modified by treatment with APTES and subsequent PCL grafting exhibit contact angles comparable to that of the lysine-derived PUR binder. Grafted PCL increased the compressive modulus and strength of PUR/BG composites by an order of magnitude, and only slightly delayed biomineralization in vitro by 7 days. However, by 14 days, the Ca/P ratio of the mineralized surface layer on PCL-modified BG disks was comparable to that of HA (1.67), suggesting that while surface modification delays the rate of apatite formation on the BG surface when in SBF, it does not block the bioactivity of the material.

[0340) Example 15

[0341] This Example illustrates that for certain embodiments of the present invention remodeling is superior when the composite comprises allograft particles, or any other synthetic substitute particles, having a particle size of at least 100 μιη.

[0342] Shown in Figure 30 are μCTscans of both injectable, porous embodiments (47wt% allograft) and moldable embodiments (67wt% allograft) used in a 5mm defect in a rabbit distal femoral condyle at week 6. The polymers tested are shown in Table 4, below.

Table 4. Tested embodiments for μCT analysis of remodeling.

[0343] The μΟΎ images in Figure 30 depict that remodeling for certain embodiments is superior in embodiments comprising larger particles sizes. Thus, for certain embodiments of the present invention is preferable to include a resorbable osteogenic matrix, whether it be allograft or synthetic allograft, that has relatively large particles sizes, including, for example, particles sizes of at least about 100 μηι.

[0344] The invention thus being described, it will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope or spirit of the invention. Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the Specification, including the Example, be considered as exemplary only, and not intended to limit the scope and spirit of the invention.

[0345] While the following terms are believed to be well understood by one of ordinary skill in the art, definitions are set forth herein to facilitate explanation of the presently-disclosed subject matter. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the presently- disclosed subject matter belongs. Although many methods, devices, and materials similar or equivalent to those described herein can be used in the practice or testing of the presently- disclosed subject matter, representative methods, devices, and materials are now described.

[0346] Following long-standing patent law convention, the terms "a", "an", and "the" refer to "one or more" when used in this application, including the claims. Thus, for example, reference to "a composite" includes a plurality of such composites, and so forth.

[0347] Unless otherwise indicated, all numbers expressing quantities of ingredients, properties such as reaction conditions, and so forth used herein are to be understood as being modified in all instances by the term "about." Accordingly, unless indicated to the contrary, the numerical parameters set forth in the herein are approximations that may vary depending upon the desired properties sought to be determined by the present invention.

[0348] As used herein, the term "about," when referring to a value or to an amount of mass, weight, time, volume, concentration or percentage is meant to encompass variations in some embodiments of ±20%, in some embodiments of ±10%, in some embodiments of ±5%, in some embodiments of±l %, in some embodiments of ±0.5%, and in some embodiments of±0.1 % from the specified amount, as such variations are appropriate to perform the disclosed method. It is also understood that there are a number of values disclosed herein, and that each value is also herein disclosed as "about" that particular value in addition to the value itself. For example, if the value "10" is disclosed, then "about 10" is also disclosed. It is also understood that each unit between two particular units are also disclosed. For example, if 10 and 15 are disclosed, then 1 1 , 12, 13, and 14 are also disclosed.

REFERENCES

[0349] Throughout this application, various publications are referenced. All such references, specifically including those in the lists below, are incorporated herein by reference.

1. Haidar, Z.S., R.C. Hamdy, and M. Tabrizian, Delivery of recombinant bone

morphogenetic proteins for bone regeneration and repair. Part A: Current challenges in BMP delivery. Biotechnol Lett, 2009. 31(12): p. 1817-24.

2. Gautschi, O.P., S.P. Frey, and R. Zellweger, Bone morphogenetic proteins in clinical applications. ANZ J Surg, 2007. 77(8): p. 626-31.

3. McKay, W.F., S.M. Peckham, and J.M. Badura, A comprehensive clinical review of

recombinant human bone morphogenetic protein-2 (INFUSE Bone Graft). Int Orthop, 2007. 31(6): p. 729-34.

4. Han, D., et al., Optimal delivery systems for bone morphogenetic proteins in orthopedic applications should model initial tissue repair structures by using a heparin-incorporated fibrin-fibronectin matrix. Med Hypotheses, 2008. 71(3): p. 374-8.

5. Smith, D.M., et al., Bone morphogenetic protein 2 therapy for craniofacial surgery. J Craniofac Surg, 2008. 19(5): p. 1244-59.

6. Miyazaki, M., et al., A porcine collagen-derived matrix as a carrier for recombinant human bone morphogenetic protein-2 enhances spinal fusion in rats. Spine J, 2009. 9(1 ): p. 22-30. 7. Garrison, K.R., et al., Clinical effectiveness and cost-effectiveness of bone morphogenetic proteins in the non-healing of fractures and spinal fusion: a systematic review. Health Technol Assess, 2007. 11(30): p. 1 - 150, iii-iv.

8. Dumas, J.E., et al., Synthesis, characterization, and remodeling of weight-bearing

allograft bone/polyurethane composites in the rabbit. Acta Biomater. 6(7): p. 2394-406.

9. Dumas, J.E., et al., Synthesis and characterization of an injectable allograft

bone/polymer composite bone void filler with tunable mechanical properties. Tissue Eng Part A. 16(8): p. 2505-18.

10. Bonzani, I.C., et al., Synthesis of two-component injectable polyurethanes for bone tissue engineering. Biomaterials, 2007. 28(3): p. 423-433.

1 1. Shah, M.M., M.D. Smyth, and A.S. Woo, Adverse facial edema associated with off-label use of recombinant human bone morphogenetic protein-2 in cran ial reconstruction for craniosynostosis. Case report. J Neurosurg Pediatr, 2008. 1(3): p. 255-7.

1. Ueeck BA. Penetrating injuries to the face: delayed versus primary treatment- considerations for delayed treatment. J oral Maxillofac Surg. 2007 Jun;65(6): 1209- 14.

2. Shermak MA, Wong L, Inoue N, Nicol T. Reconstruction of complex cranial wounds with demineralized bone matrix and bilayer artificial skin. J Cranio fac Surg. 2000

May; l l (3):224-31.

3. Lemperle SM, Calhoun CJ, Curran RW, Holmes RE. Bony healing of large cranial and mandibular defects protected from soft-tissue interposition: A comparative study of spontaneous bone regeneration, osteoconduction, and cancellous autografting in dogs. Plast Reconstr Surg. 1998 Mar; 101 (3):660-72.

4. Khan Y, Yaszemski MJ, Mikos AG, Laurencin CT. Tissue engineering of bone: material and matrix considerations. J Bone Joint Surg Am. 2008 Feb;90 Suppl 1 :36-42.

5. Chim H, Gosain AK. Biomaterials in craniofacial surgery experimental studies and clinical application J Cranio fac Surg. 2009;20:29-33.

6. Tessier P. Autogenous bone grafts taken from the calvarium for facial and cranial applications. Clin Plast Surg. 1982 Oct;9(4):531 -8. 7. Wan DC, Aalami 00, Wang Z, Nacamuli RP, Lorget F, Derynck R, et al. Differential gene expression between juvenile and adult dura mater: a window into what genes play a role in the regeneration of membranous bone. Plast Reconstr Surg. 2006 Sep 15; 1 18(4):851 -61.

8. Smith DM, Cooper GM, Mooney MP, Marra KG, Losee JE. Bone morphogenetic protein 2 therapy for craniofacial surgery. J Craniofac Surg. 2008 Sep; 19(5): 1244-59.

9. Lew TA, Walker JA, Wenke JC, Blackbourne LH, Hale RG. Characterization of craniomaxillofacial battle injuries sustained by United States service members in the current conflicts of Iraq and Afghanistan. J oral Maxillofac Surg. 2009 Jan;68(l ):3-7.

10. Kauvar DS, Wolf SE, Wade CE, Cancio LC, Renz EM, Holcomb JB. Burns sustained in combat explosions in Operations Iraqi and Enduring Freedom (OIF/OEF explosion burns). Burns. 2006 Nov;32(7):853-7.

1 1. Schmitz JP, Hollinger JO, Milam SB. Reconstruction of bone using calcium phosphate bone cements: A critical review. J Oral Maxillofac Surg. 1999;57: 1 122-6.

12. Elshahat A, Shermak MA, Inoue N, Chao EY, Manson P. The use of Novabone and Norian in cranioplasty: a comparative study. J Craniofac Surg. 2004 May; 15(3):483-9.

13. Moghadam HG, Sandor GK, Holmes HH, Clokie CM. Histomorpho metric evaluation of bone regeneration using allogeneic and alloplastic bone substitutes. J oral Maxillofac Surg. 2004 Feb;62(2):202- 13.

14. Clokie CM, Moghadam H, Jackson MT, Sandor GK. Closure of critical sized defects with allogenic and alloplastic bone substitutes. J Craniofac Surg. 2002 Jan; 13( l ): l 1 1-21 ;

discussion 22-3.

15. Goldberg CS, Antonyshyn O, Midha R, Fialkov JA. Measuring Pulsatile Forces on the Human Cranium. J Craniofacial Surgery. 2005; 16(1): 134 - 9.

16. Bohner M. Designing ceramics for injectable bone graft substitutes. In: Vernon BL, editor. Injectable Biomaterials: Science and Applications. Philadelphia: Woodhead Publishing; 201 1.

17. Spector J, Luchs J, Mehera B, Greenwald J, Smith L, Longaker M. Expression of bone morphogenetic proteins during membranous bone healing. Plastic Reconstr Surg. 2001 ; 107: 124- 34. 18. Boerckel JD, olambkar YM, Dupont KM, Uhrig BA, Phelps EA, Stevens HY, et al. Effects of protein dose and delivery system on BMP-mediated bone regeneration. Biomaterials. 201 1 Aug;32(22):5241 -51.

19. Brown KV, Li B, Guda T, Perrien DS, Guelcher SA, Wenke JC. Improving Bone Formation in a Rat Femur Segmental Defect by Controlling Bone Morphogenetic Protein-2 Release. Tissue Eng Part A. 201 1 Apr 3.

20. Smith DM, Afifi AM, Cooper GM, Mooney MP, Marra KG, Losee JE. BMP-2-based repair of large-scale calvarial defects in an experimental model: regenerative surgery in cranioplasty. J Cranio fac Surg. 2008 Sep; 19(5): 1315-22.

21. Haidar ZS, Hamdy RC, Tabrizian M. Delivery of recombinant bone morphogenetic proteins for bone regeneration and repair. Part A: Current challenges in BMP delivery.

Biotechnol Lett. 2009 Dec;31 (12): 1817-24.

22. McKay WF, Peckham SM, Badura JM. A comprehensive clinical review of recombinant human bone morphogenetic protein-2 (INFUSE Bone Graft). Int Orthop. 2007 Dec;31(6):729- 34.

23. Carter TG, Brar PS, Tolas A, Beirne OR. Off-Label Use of Recombinant Human Bone Morphogenetic Protein-2 (rhBMP-2) for Reconstruction of Mandibular Bone Defects in Humans. J oral Maxillofac Surg. 2008;66: 1417-25.

24. Herford AS, Boyne PJ. Reconstruction of Mandibular Continuity Defects With Bone Morphogenetic Protein-2 (rhBMP-2). J oral Maxillofac Surg. 2008;66:616-24.

25. Panetta NJ, Gupta DM, Longaker MT. Bone tissue engineering scaffolds of today and tomorrow. J Craniofac Surg. 2009 Sep;20(5): 1531-2.

26. Hollister SJ, Lin CY, Saito E, Schek RD, Taboas JM, Williams JM, et al. Engineering craniofacial scaffolds. Orthod Craniofac Res. 2005 Aug;8(3): 162-73.

27. Dumas JE, Zienkiewicz K, Tanner SA, Prieto EM, Bhattacharyya S, Guelcher S.

Synthesis and Characterization of an Injectable Allograft Bone/polymer Composite Bone Void Filler with Tunable Mechanical Properties. Tissue Eng Part A. 2010 Mar 10; 16(8):2505- 18. 28. Dumas JE, Davis TE, Yoshii T, Nyman J, Holt GE, Perrien DS, et al. Synthesis of Allograft Bone/Polymer Composites and Evaluation of Remodeling in a Rabbit Femoral Condyle Model. Acta Biomaterialia. 2010;6:2394-406.

29. Li JQ, Salovey R. Model Filled Polymers: The Effect of Particle Size on the Rheology of Filled Poly(methyl methacrylate) Composites. Polym Eng Sci. 2004;44:452-62.

30. Baroud G, Cayer E, Bohner M. Rheological characterization of concentrated aqueous beta-tricalcium phosphate suspensions: the effect of liquid-to-powder ratio, milling time, and additives. Acta Biomater. 2005 May; l (3):357-63.

31. An YH, Friedman RJ. Animal models of bone defect repair. In: An YH, Friedman RJ, editors. Animals Models in Orthopaedic Research. Boca Raton: CRC Press; 1999.

32. Otsu N. A threshold selection method for gray level histogram. IEEE Trans Syst Man Cybern. 1978;SMC-9(l)(62-66).

33. Bohner M. Calcium orthophosphates in medicine: from ceramics to calcium phosphate cements. Injury. 2000;31 (Supplement 4):S-D37-47.

34. Lewis G. Percutaneous Vertebroplasty and Kyphoplasty for the Stand-Alone

Augmentation of Osteoporosis-Induced Vertebral Compression Fractures: Present Status and Future Directions. J Biomed Mater Res Part B : Appl Biomater. 2007;81B:371 -86.

35. Clarkin OM, Boyd D, Madigan S, Towler MR. Comparison of an experimental bone cement with a commercial control, Hydroset™. J Mater Sci: Mater Med. 2009;20: 1563 - 70.

36. Bohner M, Baroud G. Injectability of calcium phosphate pastes. Biomaterials.

2005;26( 13): 1553-63.

37. Cherng A, Takagi S, Chow LC. Effects of hydroxypropyl methylcellulose and other gelling agents on the handling properties of calcium phosphate cement. J Biomed Mater Res. 1997 Jun 5;35(3):273-7.

38. Gilardino MS, Cabiling DS, Bartlett SP. Long-term follo.w-up experience with carbonated calcium phosphate cement (Norian) for cranioplasty in children and adults. Plastic and reconstructive surgery. 2009 Mar; 123(3):983-94.

39. Torquato S, Truskett TM, Debenedetti PG. Is Random Close Packing of Spheres Well Defined? Phys Rev Let. 2000;84( 10):2064-7. 40. Adhikari R, Gunatillake PA, Griffiths I, Tatai L, Wickramaratna M, Houshyar S, et al. Biodegradable injectable polyurethanes: synthesis and evaluation for orthopaedic applications. Biomaterials. 2008 Oct;29(28):3762-70.

41. Malinin TI, Carpenter EM, Temple HT. Particulate bone allograft incorporation in regeneration of osseous defects; importance of particle sizes. Open Orthop J. 2007; 1 : 19-24.

42. Hasegawa S, Ishii S, Tamura J, Furukawa T, Neo M, Matsusue M, et al. A 5-7 year in vivo study of high-strength hydroxyapatite/poly(L-lactide) composite rods for the internal fixation of bone fractures. Biomaterials. 2006;27: 1327-32.

43. Hafeman AE, Zienkiewicz KJ, Zachman AL, Sung HJ, Nanney LB, Davidson JM, et al. Characterization of the degradation mechanisms of lysine-derived aliphatic poly(ester urethane) scaffolds. Biomaterials. 201 1 Sep 21 ;32(2):419-29.

44. Li B, Yoshii T, Hafeman AE, Nyman JS, Wenke JC, Guelcher SA. The effects of rhBMP-2 released from biodegradable polyurethane/microsphere composite scaffolds on new bone formation in rat femora. Biomaterials. 2009 Dec;30(35):6768-79.

45. Donath K, Laass M, Gunzl HJ. The histopathology of different foreign-body reactions in oral soft tissue and bone tissue. Virchows Arch A Pathol Anat Histopathol. 1992;420(2): 131 -7.

46. Takaoka K, Koezuka M, Nakahara H. Telopeptide-depleted bovine skin collagen as a carrier for bone morphogenetic protein. J Orthop Res. 1991 Nov;9(6):902-7.

47. Haidar ZS, Hamdy RC, Tabrizian M. Delivery of recombinant bone morphogenetic proteins for bone regeneration and repair. Part B: Delivery systems for BMPs in orthopaedic and craniofacial tissue engineering. Biotechnol Lett. 2009 Dec;31 (12): 1825-35.

48. Miyazaki M, Morishita Y, He W, Hu M, Sintuu C, Hymanson HJ, et al. A porcine collagen-derived matrix as a carrier for recombinant human bone morphogenetic protein-2 enhances spinal fusion in rats. Spine J. 2009 Jan-Feb;9(l):22-30.

49. Sheehan JP, Sheehan JM, Seeherman H, Quigg M, Helm GA. The safety and utility of recombinant human bone morphogenetic protein-2 for cranial procedures in a nonhuman primate model. J Neurosurg. 2003 Jan;98( l ): 125-30. 50. Kolambkar YM, Dupont KM, Boerckel JD, Huebsch N, Mooney DJ, Hutmacher DW, et al. An alginate-based hybrid system for growth factor delivery in the functional repair of large bone defects. Biomaterials. 201 1 Jan;32(l):65-74.

51. Geiger M, Li RH, Friess W. Collagen sponges for bone regeneration with rhBMP-2. Adv Drug Deliv Rev. 2003 Nov 28;55(12): 1613-29.

52. Santos MI, Unger RE, Sousa RA, Reis RL, Kirkpatrick CJ. Crosstalk between osteoblasts and endothelial cells co-cultured on a polycaprolactone-starch scaffold and the in vitro development of vascularization. Biomaterials. 2009 Sep;30(26):4407- 15.

53. Unger RE, Sartoris A, Peters K, Motta A, Migliaresi C, Kunkel M, et al. Tissue-like self- assembly in cocultures of endothelial cells and osteoblasts and the formation of microcapillary- like structures on three-dimensional porous biomaterials. Biomaterials. 2007 Sep;28(27):3965- 76.

54. Szpalski M, Gunzburg R. Recombinant human bone morphogenetic protein-2: a novel osteoinductive alternative to autogenous bone graft? Acta Orthop Belg. 2005 Apr;71(2): 133-48.

55. Jensen ED, Pham L, Billington CJ, Jr., Espe K, Carlson AE, Westendorf JJ, et al. Bone morphogenic protein 2 directly enhances differentiation of murine osteoclast precursors. J Cell Biochem. 2010 Mar l ; 109(4):672-82.

56. Okamoto M, Murai J, Yoshikawa H, Tsumaki N. Bone morphogenetic proteins in bone stimulate osteoclasts and osteoblasts during bone development. J Bone Miner Res. 2006

Jul;21(7): 1022-33.

57. Belfrage O, Flivik G, Sundberg M, Kesteris U, Tagil M. Local treatment of cancellous bone grafts with BMP-7 and zoledronate increases both the bone formation rate and bone density. Acta Orthop. 201 1 Apr;82(2):228-33.

58. Schwartz Z, Somers A, Mellonig JT, Carnes DL, Jr., Wozney JM, Dean DD, et al.

Addition of human recombinant bone morphogenetic protein-2 to inactive commercial human demineralized freeze-dried bone allograft makes an effective composite bone inductive implant material. J Periodontol. 1998 Dec;69(12): 1337-45. 59. McGee MA, Findlay DM, Howie DW, Carbone A, Ward P, Stamenkov R, et al. The use of OP-1 in femoral impaction grafting in a sheep model. J Orthop Res. 2004

Sep;22(5): 1008- 15.

1. Anderson DD, Van Hofwegen C, Marsh JL, Brown TD. Is elevated contact stress predictive of post-traumatic osteoarthritis for imprecisely reduced tibial plafond fractures? J Orthop Res. 201 1 Jan;29(l):33-9.

2. Russell TA, Leighton RK. Comparison of autogenous bone graft and endothermic calcium phosphate cement for defect augmentation in tibial plateau fractures. A multicenter, prospective, randomized study. J Bone Joint Surg Am. 2008 Ocf;90(10):2057-61.

3. Simpson D, Keating JF. Outcome of tibial plateau fractures managed with calcium phosphate cement. Injury. 2004 Sep;35(9):913-8.

4. Hall JA, Beuerlein MJ, McKee MD. Open reduction and internal fixation compared with circular fixator application for bicondylar tibial plateau fractures. Surgical technique. J Bone Joint Surg Am. 2009 Mar 1 ;91 Suppl 2 Pt 1 :74-88.

5. Schwartz Z, Goldstein M, Raviv E, Hirsch A, Ranly DM, Boyan BD. Clinical evaluation of demineralized bone allograft in a hyaluronic acid carrier for sinus lift augmentation in humans: a computed tomography and histomorphometric study. Clin Oral Implants Res. 2007 Apr; 18(2):204- l l .

6. Bo ner M. Designing ceramics for injectable bone graft substitutes. In: Vernon BL, editor. Injectable Biomaterials: Science and Applications. Philadelphia: Woodhead Publishing; 201 1.

7. Friedman CD, Constantino PD, Takagi S, Chow LC. BoneSource™ hydroxyapatite cement: a novel biomaterial for craniofacial skeletal yissue engineering and reconstruction. J Biomed Mater Res. 1998;43(4):428-32.

8. Chim H, Gosain AK. Biomaterials in craniofacial surgery experimental studies and clinical application J Craniofac Surg. 2009;20:29-33. 9. Gasparini G, Boniello R, Moro A, Tamburrini G, Di Rocco C, Pelo S. Cranial reshaping using methyl methacrylate:technical note. J Craniofac Surg. 2009;20: 184-90.

10. Moreira-Gonzalez A, Jackson IT, Miyawaki T, Barakat K, DiNick V. Clinical outcome in cranioplasty: critical review in long-term follow-up. J Craniofac Surg. 2003; 14: 144-53.

1 1. Cammisa FP, Jr., Lowery G, Garfin SR, Geisler FH, Klara PM, McGuire RA, et al. Two- year fusion rate equivalency between Grafton DBM gel and autograft in posterolateral spine fusion: a prospective controlled trial employing a side-by-side comparison in the same patient. Spine (Phila Pa 1976). 2004 Mar 15;29(6):660-6.

12. Chan C, Thompson I, Robinson P, Wilson J, Hench L. Evaluation of Bioglass/dextran composite as a bone graft substitute. Int J Oral Maxillofac Surg. 2002 Feb;31 (l ):73-7.

13. Chazono M, Tanaka T, Komaki H, Fujii K. Bone formation and bioresorption after implantation of injectable beta-tricalcium phosphate granules-hyaluronate complex in rabbit bone defects. J Biomed Mater Res A. 2004 Sep 15;70(4):542-9.

14. Urban RM, Turner TM, Hall DJ, Inoue N, Gitelis S. Increased bone formation using calcium sulfate-calcium phosphate composite graft. Clin Orthop Relat Res. 2007 Jun;459: l 10-7.

15. Gitelis S, Urban RM, Turner TM, Heck R, Parameswarart AD, editors. Outcomes in the Treatment of Benign Bone Lesions Using an Engineering Ceramic: Preclinical and Clinical Results. Materials and Processes for Medical Devices Conference; 2009 August 10 - 12, 2009; Minneapolis, M .

16. Ikenaga M, Hardouin P, Lemaitre J, Andrianjatovo H, Flautre B. Biomechanical characterization of a biodegradable calcium phosphate hydraulic cement: a comparison with porous biphasic calcium phosphate ceramics. J Biomed Mater Res. 1998 Apr;40(l): 139-44.

17. Urban RM, Turner TM, Hall DJ, Infanger S, Cheema N, Lim TH. Healing of large defects treated with calcium sulfate pellets containing demineralized bone matrix particles. Orthopedics. 2003 May;26(5 Suppl):s581 -5.

18. Greish YE, Brown PW. Phase evolution during the formation of stoichiometric hydroxyapatite at 37.4 degrees C. J Biomed Mater Res B Appl Biomater. 2003 Oct

15;67(l):632-7. 19. Wagoner Johnson AJ, Herschler BA. A review of the mechanical behavior of CaP and CaP/polymer composites for applications in bone replacement and repair. Acta Biomater. 201 1 Jan;7(l): 16-30.

20. Dumas JE, Davis TE, Yoshii T, Nyman J, Holt GE, Perrien DS, et al. Synthesis of Allograft Bone/Polymer Composites and Evaluation of Remodeling in a Rabbit Femoral Condyle Model. Acta Biomaterialia. 2010;6:2394-406.

21. Otsu N. A threshold selection method for gray level histogram. IEEE Trans Syst Man Cybern. 1978;SMC-9(l )(62-66).

1. Khan, Y., et al., Tissue engineering of bone: Material and matrix considerations. Journal of Bone and Joint Surgery-American Volume, 2008. 90A: p. 36-42.

2. Legeros, R.Z., A. Chohayeb, and A. Schulman, APA TITIC CALCIUM PHOSPHATES POSSIBLE DENTAL RESTORATIVE MATERIALS. Journal of Dental Research, 1982. 61(SPEC. ISSUE): p. 343.

3. Chim, H. and A.K. Gosain, Biomaterials in Craniofacial Surgery Experimental Studies and Clinical Application. Journal of Craniofacial Surgery, 2009. 20(1 ): p. 29-33.

4. Moreira-Gonzalez, A., et al., Clinical outcome in cranioplasty: Critical review in long- term follow-up. Journal of Craniofacial Surgery, 2003. 14(2): p. 144- 153.

5. Friedman, CD., et al., BoneSource (TM) hydroxyapatite cement: A novel biomaterial for craniofacial skeletal tissue engineering and reconstruction. Journal of Biomedical Materials Research, 1998. 43(4): p. 428-432.

6. Gasparini, G., et al., Cranial Reshaping Using Methyl Methacrylate: Technical Note.

Journal of Craniofacial Surgery, 2009. 20( 1): p. 184- 190.

7. Bohner, M., Designing ceramics for injectable bone graft substitutes, in Injectable Biomaterials: Science and Applications, B.L. Vernon, Editor. 201 1 , Woodhead Publishing: Philadelphia.

8. Wagoner Johnson, A.J. and B.A. Herschler, A review of the mechanical behavior of CaP and CaP/polymer composites for applications in bone replacement and repair. Acta Biomaterialia, 201 1. 7(1 ): p. 16-30. Chan, C, et al., Evaluation of Bioglass/dextran composite as a bone graft substitute. International Journal of Oral and Maxillofacial Surgery, 2002. 31(1): p. 73-77.

Chazono, M., et al., Bone formation and bioresorption after implantation of injectable beta-tricalcium phosphate granules-hyaluronate complex in rabbit bone defects. Journal of Biomedical Materials Research Part A, 2004. 70A(4): p. 542-549.

Schwartz, Z., et al., Clinical evaluation of demineralized bone allograft in a hyaluronic acid carrier for sinus lift augmentation in humans: a computed tomography and histomorphometric study. Clinical Oral Implants Research, 2007. 18(2): p. 204-21 1. Cammisa, F.P., et al., Two-year fusion rate equivalency between Grafton (R) DBM gel and autograft in posterolateral spine fusion. Spine, 2004. 29(6): p. 660-666.

Dumas, J.E., et al., Synthesis, characterization, and remodeling of weight-bearing allograft bone/polyurethane composites in the rabbit. Acta Biomaterialia, 2010. 6(7): p. 2394-2406.

Bennett, S., et al., Initial biocompatibility studies of a novel degradable polymeric bone substitute that hardens in situ. Bone, 1996. 19(1 ): p. S 101-S 107.

Bonzani, I.C., et al., Synthesis of two-component injectable polyurethanes for bone tissue engineering. Biomaterials, 2007. 28(3): p. 423-433.

Ertel, S.I., et al., IN- VITRO STUDY OF THE INTRINSIC TOXICITY OF SYNTHETIC SURFACES TO CELLS. Journal of Biomedical Materials Research, 1994. 28(6): p. 667- 675.

Pons, F., et al., Effect of toluene diisocyanate and its corresponding amines on viability and growth of human lung fibroblasts in culture. Cell Biology and Toxicology, 1999. 15(5): p. 333-340.

Mishra, P.K., et al., Isocyanates induces DNA damage, apoptosis, oxidative stress, and inflammation in cultured human lymphocytes. Journal ^' of Biochemical and Molecular Toxicology, 2008. 22(6): p. 429-440.

Dumas, J.E., et al., Synthesis and Characterization of an Injectable Allograft Bone/Polymer Composite Bone Void Filler with Tunable Mechanical Properties. Tissue Engineering Part A, 2010. 16(8): p. 2505-2518. Hinrichsen, G., Polyurethane handbook (2nd ed.). Edited by G. Oertel, Hanser Publishers, Munich 1993, 770 pp., DM 358, ISBN 3-446-17198-3. Acta Polymerica, 1994. 45(5): p. 398-398.

Hafeman, A., et al., Injectable biodegradable polyurethane scaffolds with release of platelet-derived growth factor for tissue repair and regeneration. Pharm Res, 2008. 25(10): p. 2387-99.

Hoven, V.P., et al., Surface-charged chitosan: Preparation and protein adsorption. Carbohydrate Polymers, 2007. 68(1): p. 44-53.

GluckHirsch, J.B. and J.L. okini, Determination of the molecular weight between crosslinks of waxy maize starches using the theory of rubber elasticity. Journal of Rheology, 1997. 41(1): p. 129-139.

Bagley, E.B., The seperation of elastic and visous effects in polymer flow. Transactions of the Society of Rheology, 1961 : p. 355-368.

Guan, J.J., et al., Biodegradable poly(ether ester urethane)urea elastomers based on poly (ether ester) triblock copolymers and putrescine: synthesis, characterization and cytocompatibility. Biomaterials, 2004. 25(1): p. 85-96.

Ortel, G., Polyurethane Handbook. 1994: Hanser Gardner Publications.

Guelcher, S.A., et al., Synthesis and in vitro biocompatibility of injectable polyurethane foam scaffolds. Tissue Engineering, 2006. 12(5): p. 1247- 1259.

Adolph, E.J., et al., Injectable biodegradable polyurethane scaffolds for wound healing. Submitted, 201 1.

Adhikari, R., et al., Biodegradable injectable polyur ethanes: synthesis and evaluation for orthopaedic applications. Biomaterials, 2008. 29(28): p. 3762-70.

Dumas, J.E., et al., Synthesis and Characterization of an Injectable Allograft Bone/polymer Composite Bone Void Filler with Tunable Mechanical Properties. Tissue Eng Part A, 2010. 16(8): p. 2505-18.

Bennett, S., et al., Initial biocompatibility studies of a novel degradable polymeric bone substitute that hardens in situ. Bone, 1996. 19(1 , Supplement): p. 101 S- 107S. Buckley, M.J. and E.J. Beckman, Adhesive Use in Oral and Maxillofacial Surgery. Oral and Maxillofacial Surgery Clinics of North America, 2010. 22(1): p. 195- 199.

Zhang, J.-Y., et al., A new peptide-based urethane polymer: synthesis, biodegradation, and potential to support cell growth in vitro. Bio materials, 2000. 21 : p. 1247- 1258.

Hafeman, A.E., et al., Characterization of the degradation mechanisms of lysine-derived aliphatic poly (ester urethane) scaffolds. Bio materials, 201 1. 32(2): p. 419-29.

Elliott, S.L., et al., Identification of biodegradation products formed by L-phenylalanine based segmented polyurethaneureas. Journal of Biomaterials Science Polymer Edition, 2002. 13(6): p. 691 -71 1.

Bonzani, I.C., et al., Synthesis of two-component injectable polyurethanes for bone tissue engineering. Biomaterials, 2007. 28: p. 423-433.

Adhikari, R., et al., Biodegradable injectable polyurethanes: Synthesis and evaluation for orthopaedic applications. Biomaterials, 2008. 29(28): p. 3762-3770.

Steven, F.S.J., D. S., Purification and amino acid composition of monomeric and polymeric collagens. . Biochem. J., 1967. 104: p. 534-536.

Lu, Q.-W., T.R. Hoye, and C.W. Macosko, Reactivity of common functional groups with urethanes: Models for reactive compatibilization of thermoplastic polyurethane blends. Journal of Polymer Science Part A: Polymer Chemistry, 2002. 40(14): p. 2310-2328. Szycher, M., Szycher's Handbook of Polyurethanes. 1999, Boca Raton: CRC Press LLC. Kar, A., Pharmaceutical Drug Analysis. 2nd ed. 2005, Daryaganj, New Delhi: New Age International Limited.

Zhang, J.Y., et al., A new peptide-based urethane polymer: synthesis, biodegradation, and potential to support cell growth in vitro. Biomaterials, 2000. 21(12): p. 1247- 1258. Fogler, H.S., Elements of Chemical Reaction Engineering. 4th Ed. ed. 1999, Upper Saddle River, NJ: Pearson Education.

Sivak, W.N., et al., Catalyst-dependent drug loading of LDI-glycerol polyurethane foams leads to differing controlled release profiles. Acta Biomaterialia, 2008. 4(5): p. 1263- 1274. Silva, A.L. and J.C. Bordado, Recent developments in polyurethane catalysis: Catalytic mechanisms review. Catalysis Reviews-Science and Engineering, 2004. 46(1): p. 31 -51. Semsarzadeh, M.A. and A.H. Navarchian, Effects of NCO/OH ratio and catalyst concentration on structure, thermal stability, and crosslink density of poly(urethane- isocyanurate). Journal of Applied Polymer Science, 2003. 90(4): p. 963-972.

Gorna, K. and S. Gogolewski, Preparation, degradation, and calcification of biodegradable polyurethane foams for bone graft substitutes. Journal of Biomedical Materials Research Part A, 2003. 67A(3): p. 813-827.

Storey, R.F., J.S. Wiggins, and A.D. Puckett, Hydrolyzable poly(ester-urethane) networks from L-lysine diisocyanate and D,L-lactide/e-caprolactone homo- and copolyester triols. Journal of Polymer Science Part A: Polymer Chemistry, 1994. 32( 12): p. 2345-2363.

Wang, Z., et al., Preparation and rapid degradation of nontoxic biodegradable polyurethanes based on polyflactic acid) -poly (ethylene glycol)-poly(lactic acid) and l- lysine diisocyanate. Polymer Chemistry, 201 1. 2(3): p. 601 -607.

Caracciolo, P., F. Buffa, and G. Abraham, Effect of the hard segment chemistry and structure on the thermal and mechanical properties of novel biomedical segmented poly(esterurethanes). Journal of Materials Science: Materials in Medicine, 2009. 20(1): p. 145-155.

Guan, J., et al., Biodegradable poly(ether ester urethane)urea elastomers based on polyfether ester) triblock copolymers and putrescine: synthesis, characterization and cytocompatibility. Biomaterials, 2004. 25( 1): p. 85-96.

Hafeman, A., et al., Injectable Biodegradable Polyurethane Scaffolds with Release of Platelet-derived Growth Factor for Tissue Repair and Regeneration. Pharmaceutical Research, 2008. 25(10): p. 2387-2399.

retlow, J.D., et al., Injectable Biomaterials for Regenerating Complex Craniofacial Tissues. Advanced Materials, 2009. 21(32-33): p. 3368-3393. . Timmer, M.D., et al., In vitro cytotoxicity of injectable and biodegradable poly(propylene fumar ate) -based networks: Unreacted macromers, cross-linked networks, and degradation products. Biomacromolecules, 2003. 4(4): p. 1026-1033.

. Dumas, J.E., et al., Synthesis of Allograft Bone/Polymer Composites and Evaluation of Remodeling in a Rabbit Femoral Condyle Model. Acta Biomaterialia, 2010. 6: p. 2394- 2406. i, B., et al., Biomaterials, 2009, 30(35), 6768-6779.

elfrage, O., et al., Acta Orthopaedica, 201 1 , 82(2), 228-233.

Bashoor-Zedah, M., Baroud, G. and Bohner, M., Biomaterials, 201 1 , 32(27), 6362-6373. D. Simpson, J. F. Keating, Injury, 2004, 35, 913.

J. Dumas, T. Davis, G. Holt, T. Yoshii, D. Perrien, J. Nyman, T. Boyce, S. Guelcher, Acta Biomaterialia, 2010, 6, 2394.

J. B laker, A. Boccaccini, Expert review of medical devices, 2005, 2, 303.

E. Verne, C. Vitale-Brovarone, E. Bui, C. Bianchi, A. Boccaccini, Journal of Biomedical Materials Research, 2008, 90, 981 .

C. Kunze, T. Freier, E. Helwig, B. Sandner, D. Reif, A. Wutzler, H. Radusch, Biomaterials, 2003, 24, 967.

Q. Chen, R. Rezwan, D. Armitage, S. Nazhat, A. Boccaccini, J Mater Sci: Mater Med, 2006, 17, 979.

G. Jiang, I. Jones, C. Rudd, G. Walker, Journal of Applied Polymer Science, 2009, 114, 658. T. Kokubo, H. Takadama, Biomaterials, 2006, 27, 2907.