Systems and methods for improved delivery of osteoinductive molecules in bone repair

Abstract

Systems and methods for preparing synthetic osteoinductive bone grafts are provided in which a porous ceramic granule, which may be incorporated within a biocompatible matrix material, is loaded with an osteoinductive agent. Loading of granules is facilitated in some cases by the use of low-pH buffers and pre-treatments.

Claims

1. A kit for use in forming an implant, the kit comprising: a first composition including an acid etched calcium ceramic granule having a specific surface area greater than 30 m.sup.2/g, a pH less than 7, and an interconnected network of micropores defining at least one surface on an interior of the granule, wherein said micropores prevent multinucleated osteoclasts from penetrating into the interior of the granules; and a second composition comprising a solution comprising an osteoinductive protein and a buffering agent having a pKa between 2.3 and 4.5, the solution having a pH of less than 4.0, wherein said second composition is separate from the first composition.

2. The kit of claim 1, wherein the solution comprises: 25 mM Glutamic Acid; 2% (w/v) Glycine; and 1% (w/v) Sucrose.

3. The kit of claim 1, wherein the buffering agent is selected from the group consisting of glycine, lactic acid, formic acid, malic acid, malonic acid, glutamic acid, aspartic acid, citric acid, tartaric acid, fumaric acid, succinic acid and combinations thereof.

4. The kit of claim 1, further comprising an instructional material describing a method of making an osteoinductive implant comprising the step of: contacting the first composition with the second composition comprising the solution, thereby associating the osteoinductive protein with an interior surface of the granule and forming an osteoinductive implant.

5. The kit of claim 4, wherein the instructional material describes a method of treating a patient, comprising the step of: contacting a bony tissue of the patient with the osteoinductive implant.

6. The kit of claim 1, further comprising an applicator for contacting the first composition with the second composition comprising the solution.

7. The kit of claim 4, further comprising an applicator for applying the second composition comprising the solution to the first composition or for applying the implant to a site within or on a body tissue.

8. A kit for use in forming an implant, the kit comprising: a first composition including a calcium ceramic granule having a specific surface area greater than 30 m2/g, a pH less than 7, and an interconnected network of micropores defining at least one surface on an interior of the granule, wherein said micropores prevent multinucleated osteoclasts from penetrating into the interior of the granules; and a second composition comprising a solution comprising a buffering agent having a pKa between 2.3 and 4.5, the solution having a pH of less than 4.0; and a third composition comprising an osteoinductive protein wherein the first composition, second composition and third composition are maintained separately.

9. The kit of claim 8, wherein the osteoinductive protein is lyophilized.

10. The kit of claim 8, further comprising an applicator for placing the implant at a site within or on a body tissue or for contacting the composition or the osteoinductive protein with the solution.

11. The kit of claim 10, wherein the applicator is a syringe.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Certain embodiments of the present invention are illustrated by the accompanying figures. It will be understood that the figures are not necessarily to scale and that details not necessary for an understanding of the invention or that render other details difficult to perceive may be omitted. It will be understood that the invention is not necessarily limited to the particular embodiments illustrated herein.

(2) FIG. 1 shows in-situ retention of rhBMP-2 (% of initial dose, mean±SD) vs time (days) after injection with rhBMP-2/calcium phosphate matrix (CPM, 0.7 and 4.2 mg/mL) compared to 0.2 mg/mL rhBMP-2 delivered on an absorbable collagen sponge (ACS) and in buffer (0.7 mg/mL) in a rabbit ulna osteotomy model.

(3) FIGS. 2A through 2D show a comparison of callus formation in a nonhuman primate fibula osteotomy model in response to treatment with BMP-2 delivered in an absorbable collagen sponge (ACS) compared to treatment with BMP-2 delivered in a granulating calcium phosphate matrix (CPM) at 8 weeks. The BMP-2/ACS-treated osteotomy shown in panels A and B has a hollow callus architecture resulting from bone formation outside the microporous ACS carrier rather than within the ACS carrier. In contrast the BMP-2/CPM-treated repair shown in panels C and D has a much more uniform callus architecture resulting from the slower BMP-2 release profile.

(4) FIG. 3 shows in vitro loading/retention profiles (on/off rate) for BMP-2 delivered on a calcium deficient hydroxyapatite granule (CDHA) with high specific surface area compared to two BCP granules with low specific surface area. BMP was loaded onto the CaP granules in BMP buffer solution over a 24 hours period. The BMP-loaded CaP granules were then incubated in a solution containing 20% bovine serum to mimic exposure to serum proteins in vivo. High specific surface area CDHA granules loaded BMP-2 more efficiently compared to the low specific surface area BCP granules. In addition, more than 70% of the loaded BMP-2 remained bound to the granules over the subsequent 13-day period. In contrast, both low specific surface area BCP granules rapidly released a significant amount of BMP-2 in the first 24 hours and retained less than 20% of the initially loaded BMP-2 at 13 days.

(5) FIG. 4 shows the in vitro retention profile (% of initial) for BMP-2 delivered in CDHA porous high SSA (specific surface area) granules, CDHA non-porous high SSA granules, macroporous calcium phosphate cement (CaP), 60:40 HA/TCP porous granules, 15:85 HA/TCP porous granules compared to an absorbable collagen sponge (ACS) as a function of time in days. High specific surface area CDHA granules with and without porosity and CaP cements had superior BMP in vitro retention compared to ACS and low SSA granules either alone or contained within a collagen sponge.

(6) FIG. 5 shows the in vivo BMP-2 retention profiles using an intramuscular implant model in rats for high SSA CDHA granules with a microporous and macroporous structure compared to high SSA CDHA granules without an internal pore structure and low SSA 60:40 HA:TCP granules with a porous structure. Low SSA granules demonstrated the lowest BMP-2 retention, and high SSA granules without a porous structure demonstrated only a slightly improved BMP-2 retention profile. However, high SSA granules with a porous architecture demonstrated superior BMP-2 retention with 50% of the initial protein still remaining for at least 12 days. The BMP-2 retention profile of the high SSA granules with a porous structure is markedly similar to that of the CaP cement where the BMP is entrapped within the cement matrix.

(7) FIG. 6 shows the in vitro retention profiles for BMP-2 delivered in high specific surface area (SSA) Calcium Deficient Hydroxyapatite (CDHA) granules with microporous and macroporous internal architecture compared to high SSA CDHA granules with only macroporous internal architecture and high SSA CDHA granules without internal architecture as a function of time in days. High SSA CDHA granules with both microporous and macroporous internal architecture loaded the highest percentage of the available BMP. High SSA CDHA granules with macroporous internal architecture loaded an intermediate percentage of the available BMP. High SSA CDHA granules without internal architecture loaded the smallest percentage of available BMP. The subsequent retention profiles were similar for all three granule types.

(8) FIGS. 7A and 7B show photomicrographs of granules loaded with BMP-based osteoinductive proteins without (A) and with (B) loading buffers according to certain embodiments of the invention. 7A shows that when fluorescently labeled BMP is delivered in a weakly buffered solution, the protein is restricted to the surface of the ceramic granules. In contrast, when fluorescently labeled BMP is delivered in a well-buffered, low pH solution, the protein is able to penetrate and localize to the interior surfaces of the ceramic granules.

(9) FIGS. 8A through 8F shows micrographs of untreated (A-C) and acid pre-treated (“etched”) (D-F) granules loaded with fluorescently labeled BMP in varying buffer compositions. In the first column (A&D), granules were loaded with protein in a low buffering capacity pH 4.0 buffer (“BVB-001 Buffer”; for composition see Table 2). In the second column (B&E), the granules were loaded with protein in an intermediate buffering capacity pH 3.5 buffer (“BVB-005 Buffer”). In the third column, the granules were loaded with protein in a high buffering capacity pH 3.0 buffer (“BVB-010 Buffer”). The distribution of protein shifts from being concentrated at the surface of the granules in A&D to a more uniform distribution as buffering capacity increases and pH decreases (i.e. the protein is distributed along pore surfaces near the centroid and near the exterior surface of the granules). The combination of etched granules and the BVB-010 buffer gave the most uniform distribution of protein.

(10) FIGS. 9A and 9B show fluorescent micrographs of granules loaded with fluorescently-labeled BMP protein. FIGS. 9C and 9D show quantification of the signal from fluorescently labeled BMP protein throughout the granules when delivered in a weak buffer (BVB-001 Buffer) or a strong buffer (1M Acetic Acid). FIGS. 9A and C demonstrate that in a weakly buffered system the vast majority of the BMP is confined near the exterior of the granules, whereas in a strongly buffered system the BMP is more uniformly distributed between both the centroid and the exterior. The ratio of peripheral to centroid fluorescence is lower (e.g. closer to 1:1) in the strong buffer with at least 30% of the total signal found in the centroid. FIG. 9D demonstrates that this is due, at least in part, to an increase in signal in the centroid of the granule.

(11) FIG. 10 illustrates the partitioning of the protein between a saline wash and a guanidine HCl extract following incubation of BMP with untreated or etched granules in varying buffer compositions. The protein in the saline wash is considered loosely bound to the granules, whereas the protein in the guanidine HCl extract is considered tightly bound to the granules. In untreated granules, the amount of loosely bound BMP decreases as the buffering capacity increases and the pH decreases (i.e., from “BVB-001” to the “BVB-010” buffer conditions). For etched granules a similar pattern was observed until the BMP was loaded in the high buffering capacity pH 3 buffer (BVB-010) in which case the fraction of loosely bound protein increased dramatically over that for the BVB-001 and BVB-005 conditions.

(12) FIG. 11 shows that in alkaline granules (e.g. pH 8.5) the BMP tends to be aggregated (high A320 values) and the extent of aggregation generally decreases as the strength of the buffer increases (BVB-001.fwdarw.BVB-012.fwdarw.BVB-010 Buffers). Additionally, in low pH granules (e.g. pH 5.2) the BMP tends to be less aggregated overall, thereby better enabling delivery to the interior of the granules.

(13) FIGS. 12A through 12D shows SEM pictures of the polished surfaces of a CDHA granule with complex internal architecture. The granule contains highly interconnected pores approximately 10 um in size. At the highest magnification individual CaP crystal spicules within the pores can be observed generating the high specific surface area characteristic of CDHA granules. The optimal in vivo BMP retention is likely due to BMP solution penetrating the microporous structure allowing BMP to bind to the internal CaP surfaces where it is protected from invading cell populations such as osteoclasts.

(14) FIGS. 13A through 13D illustrate the physical and chemical characteristics of granules of the present invention and demonstrates that these properties enable the ability to deliver BMPs in vivo at a well-controlled rate. FIGS. 13A-B show the structure of granules of the present invention demonstrating the interconnected pore structure via SEMs of the internal architecture (A) and the open surface architecture (B). FIG. 13C depicts fluoro-labeled BMP penetrating into the interior structure of the granules when delivered in a well buffered, pH 4 solution. FIG. 13D shows the BMP retention profile of preformed CaP granules of the present invention compared to BMP retention in a calcium phosphate cement, where the BMP is entirely trapped within the cement during the reaction. These data demonstrate that the present invention is able to mimic the BMP retention profile of a CaP cement by implementing the critical design features described herein.

DETAILED DESCRIPTION

(15) Osteoinductive Compositions

(16) Synthetic bone grafts (also referred to interchangeably as “implants” or “constructs”) utilizing the compositions of the invention generally include three components: an osteoconductive material, such as a calcium ceramic or other solid mineral body, an osteoinductive material such as a bone morphogenetic protein, and a biocompatible matrix such as a collagen sponge. As used herein, osteoconductive materials refer to any materials which facilitates the ingrowth or ongrowth of osteoblastic cells including osteoblasts, pre-osteoblasts, osteoprogenitor cells, mesenchymal stem cells and other cells which are capable of differentiating into or otherwise promoting the development of cells that synthesize and/or maintain skeletal tissue. In preferred embodiments of the present invention, the osteoconductive material is a granule comprising an osteoconductive calcium phosphate ceramic that is adapted to provide sustained release of an osteoinductive substance that is loaded onto the granule. In some cases, the granule includes interconnected, complex porous structures. Exemplary granules, which the inventors have found exhibit BMP binding and elution characteristics that are optimized for use in constructs, systems and methods of the present invention are described in International Patent Application Nos. PCT/CH2014/000085 and PCT/CH2015/000092 by Bohner et al. The entire disclosure of both of these applications is incorporated herein for all purposes.

(17) The granules are generally made of any suitable osteoconductive material having a composition and architecture appropriate to allow an implant of the invention to remain in place and to release osteoinductive material over time intervals optimal for the formation and healing of bone (e.g. weeks or months). While these characteristics may vary between applications, the granules generally include, without limitation, monocalcium phosphate monohydrate, dicalcium phosphate, dicalcium phosphate dehydrate, octocalcium phosphate, precipitated hydroxyapatite, precipitated amorphous calcium phosphate, monocalcium phosphate, alpha-tricalcium phosphate (α-TCP), beta-tricalcium phosphate (R-TCP), sintered hydroxyapatite, oxyapatite, tetracalcium phosphate, hydroxyapatite, calcium-deficient hydroxyapatite, and combinations thereof.

(18) Osteoinductive materials generally include peptide and non-peptide growth factors that stimulate the generation of osteoblasts from populations of pre-cursor cells. In some embodiments, the osteoinductive material is a member of the transforming growth factor beta (TGF-β) superfamily such as BMP-2, BMP-3, BMP-4, BMP-5, BMP-6, BMP-7, BMP-9, or a designer BMP such as the BMP-GER or BMP-GER-NR chimeric BMPs described in U.S. Pat. No. 8,952,131 by Berasi et al. entitled “Designer Osteogenic Proteins,” the entire disclosure of which is hereby incorporated by reference for all purposes. In other embodiments, the osteoinductive material is a fibroblast growth factor, insulin-like growth factor, platelet-derived growth factor, a small molecule, a nucleotide, a lipid, or a combination of one or more of the factors listed herein.

(19) The third component of implants (also referred to herein as “constructs”) according to the present invention is the biocompatible matrix, which can be any suitable biocompatible material which (a) when used in concert with the granules, exhibits sufficient rigidity and/or column strength to withstand the loads placed upon it when implanted, (b) which does not cause excessive inflammation (i.e. inflammation sufficient to inhibit or prevent the formation of new bone or the healing of a broken bone), inhibit the proliferation of osteoblasts, or otherwise interfere with the activity of the granules and/or the osteoinductive material, and (c) has sufficient cohesion over an appropriate interval to permit the deposition of new bone within a defined area. In addition, the biocompatible matrix is optionally degradable and/or osteoconductive. The biocompatible matrix is, in various embodiments, hyaluronic acid (HA), and functionalized or modified versions thereof, collagen, whether animal or recombinant human, gelatin (animal or recombinant human), fibrin, chitosan, alginate, agarose, self-assembling peptides, whole blood, platelet-rich plasma, bone marrow aspirate, polyethylene glycol (PEG) and derivatives thereof, functionalized or otherwise cross-linkable synthetic biocompatible polymers including poly(lactide-co-glycolide), poly(caprolactone), poly(lactic acid), poly(glycolic acid), poloxamers and other thermosensitive or reverse-thermosensitive polymers known in the art, and copolymers or admixtures of any one or more of the foregoing.

(20) Technical Considerations for Implant Design

(21) Implants of the invention, which include the osteoinductive materials, granules and biocompatible matrices as described above, generally have characteristics which are tailored to the facilitation of bone growth and healing and which are not exhibited by currently available synthetic bone grafting materials. One important distinguishing characteristic of the implants described herein is that they retain and release osteoinductive materials over intervals sufficient to induce bone formation in humans.

(22) BMPs induce bone formation primarily by stimulating differentiation of osteoblast progenitors either resident at the site of repair in the bone or in the surrounding soft tissue. Physiological bone repairs are stimulated by the release of picogram/femtogram amounts of BMPs stored in the mineral phase of bone and from newly synthesized BMPs secreted by bone progenitor cells at the site of the repair. These two sources in concert with negative regulators maintain BMP concentrations at the site of repair at physiological levels for the appropriate amount of time to induce a successful bone repair.

(23) Exogenous BMPs are, ideally, delivered in constructs which elute BMP in amounts and over intervals that mimic the physiological BMP response. It should be noted, however, that the administration of much larger pharmacological BMP concentrations is generally required to achieve physiological concentrations of BMPs at the cellular level and to maintain the physiological concentrations for the appropriate amount of time. This is due to a combination of factors that are not totally understood. Without wishing to be bound by any theory, one factor driving the need for super-physiological BMP concentrations in these constructs may be the inability of exogenous BMP to mimic the efficiency of physiological local release of endogenous BMPs from bone and newly formed endogenous BMPs from cells. In addition, rhBMPs are generally insoluble at physiological pH, so (again, not wishing to be bound by any theory) much of the exogenously delivered BMP may not be biologically available.

(24) The amount of exogenous rhBMP required to stimulate bone repair appears to be species dependent. Empirical data suggests that lower concentrations of exogenous rhBMPs are required to stimulate bone formation in small animals such as rodents and rabbits compared to larger animals including dogs, sheep and goats. Nonhuman primates and humans appear to require the highest concentrations of exogenous rhBMPs to stimulate bone repair. For example, the FDA approved concentration of rhBMP-2 delivered in an absorbable collagen sponge (ACS) for bone repair in dogs is 0.2 mg/mL compared to 1.5 mg/mL in people. Again, the factors contributing to this difference in required exogenous rhBMP concentration are not clearly understood, but those of skill in the art will understand that inter-species differences must be considered in evaluating findings in animal models for its applicability to human patients.

(25) Similarly, the interval over which BMPs must be delivered to tissues varies among species: BMP residence time for repairs in rodents and rabbits can be as short as several days due to their rapid intrinsic rate of bone formation, while nonhuman primates and human patients generally requires several weeks BMP residence time. While not wishing to be bound by any theory, the longer interval observed in primates and humans appears to be related to the amount of time for the healing process to transition from an initial catabolic inflammatory phase caused by the surgery or trauma to an anabolic phase involving the migration and differentiation of osteoblast progenitors and associated new blood vessel units to support the fusion/repair process. Short BMP residence time optimal for rodents may not maintain physiological BMPs levels for a sufficient amount of time to stimulate bone repair in animals with slower bone formation rates. Conversely, BMP may not be released in sufficient amounts from a carrier with a longer retention profile to stimulate optimal bone formation in animals with rapid intrinsic bone formation rates.

(26) As one example, the residence time of BMPs delivered locally in buffer solution to a repair site is extremely short, and even when relatively large amounts of BMP are delivered in solution, an adequate bone response is only stimulated in rodent models. For applications in non-human primates and human patients, an extended-release carrier is preferably used to localize BMP to sites of treatment for a period of weeks.

(27) One strategy for providing extended local BMP release is to utilize carriers that mimic the binding of BMP to endogenous extracellular matrix. As one example, collagenous carriers exhibit longer BMP residence times than BMP solutions, due (without being bound to any theory) to the intrinsic binding properties of BMP to extracellular matrix components including endogenous collagen. Ceramic carriers including calcium phosphate matrices (CPM) can further extended the duration over which BMPs are released from the matrix. The release of BMP from ceramic carriers may require the same osteoclastic resorption observed in release of BMP from bone. Based on this unique property, implants comprising ceramic components embedded within composite carriers, as are used in the present invention, may be superior vehicles for BMP delivery compared to other naturally occurring and synthetic biomaterials.

(28) In addition to efficacy considerations, controlling the release of BMPs from a carrier is also important for patient safety. With respect to the avoidance of trabecular bone resorption, the rapid release of BMP within metaphyseal bone or into the trabecular bone associated with interbody fusions where the endplates are penetrated results in rapid upregulation of osteoblast precursor cells in a location where there are also significant numbers of osteoclast precursor cells. As a result of normal cross talk between these two cell types, sufficient mature osteoclasts are generated to cause transient resorption of trabecular bone prior to bone formation. This phenomenon may be partially responsible for osteolysis sometimes observed when BMPs are used in interbody fusions and metaphyseal bone repairs.

(29) Designing Granules for Optimal BMP Delivery

(30) Previous experience with calcium phosphate cements demonstrated incorporation of BMP within the cement itself provides appropriate BMP retention (FIG. 1) for optimal bone formation in a variety of indications including segmental repair, diaphyseal repairs and metaphyseal repairs. However, calcium cement hardening is highly dependent on the conditions present during hardening, and the physical properties of hardened cements, such as porosity, etc, may vary significantly among individual applications. This variability prompted an investigation of other, more reliable carrier materials such as preformed calcium ceramic granules.

(31) Without wishing to be bound by any theory, optimal BMP retention profiles observed using granules or other carriers of the present invention (FIG. 4 and FIG. 5), is in part the result of cell-based osteoclast resorption of BMP incorporated within the carrier material itself. When the carrier is a cement, BMP is incorporated within the cement by using a BMP solution to hydrate the cement powder precursors. During the cementing reaction BMP binds with a high affinity to the surfaces of the rapidly forming poorly crystalline hydroxyapatite and is trapped within the cement as well as on the cement surface. BMP is then gradually released, presumably in soluble form, by the low pH environment created by the osteoclasts during the cement resorption process. Osteoclasts have to completely resorb the cement to release all of the contained BMP.

(32) In contrast, when BMP is added to currently available preformed calcium phosphate cement particles after the cementing reaction has occurred, it tends to concentrate on the outer surfaces of such particles. Consequently, BMP retention is significantly reduced (i.e. the release tends to be much faster) as osteoclasts can rapidly access and release almost all of the BMP without the need for significant granule resorption

(33) Relatedly, when BMP is incorporated into calcium cements, it is released more predictably and consistently than is observed when BMP is added to currently available CaP granules. Without wishing to be bound by any theory, it is believed that, again, the consistency of release from CaP cements is due in part to the involvement of osteoclast-mediated resorption in the release process. And again, the consistent release profiles observed in CaP cements is not easily replicated using preformed granules.

(34) The target BMP retention profile for a BMP utilizing preformed CaP granules in people, based on previous studies evaluating rhBMP-2 delivered in a calcium phosphate matrix (CPM) discussed previously, requires minimal burst release in the first 24 hours, a half-life of approximately 1 week and a mean residence time of approximately 2 weeks. The ability of a preformed CaP granule composite carrier to deliver an optimal BMP retention profile is dependent on a number of parameters intrinsic to the granules. The BMP binding affinity of the CaP granules determines the rate of BMP binding and the in vitro/in vivo retention. BMP binding affinity is, in turn, dependent on the surface chemistry and the specific surface area (nanoscale roughness) of the granules. These parameters help define the degree to which BMP becomes associated with the surfaces of the granules, but are not sufficient to mimic the BMP retention and release kinetics observed in cements; the granule must also have an appropriate internal architecture. The ideal granule architecture incorporates both macropores (pores having diameters or cross-sectional dimensions of between 40 and 100 microns or so) and micropores (pores of approximately 10 microns in cross section), in a manner that (a) provides sufficient internal surface area to load the granules with BMP in quantities similar to those achievable in cements, while (b) permitting fluid infiltration of the granules to allow BMP in solution to access the internal pores surfaces of the granules and (c) does not reduce the compressive strength of the granules below the threshold required for their use in vivo. While currently available granules do not meet these criteria, in CaP granules which meet these requirements (as described below), BMP release profiles are in line with those observed in CaP cements, and release of BMP from such granules is thought (without wishing to be bound by theory) to be dependent on osteoclast resorption, as BMP associated with internal pore surfaces is not readily accessible to osteoclasts.

(35) CaP granules with specific surface area (SSA) in the range of 50-80 m.sup.2/g appear to have optimal in vitro BMP binding (FIGS. 3-6). High specific surface area likely increases binding by increasing the available surface area for binding of BMP to the CaP. High specific surface area CDHA granules loaded BMP-2 more rapidly and loaded a higher percentage of the available BMP-2 compared to the low specific surface area BCP granules (FIG. 3). In addition, more than 80% of the loaded BMP-2 remained bound to the granules over the subsequent 12-day period (FIG. 3). In contrast, both low specific surface area BCP granules rapidly released a significant amount of BMP-2 in the first 24 hours and retained less than 20% of the initially loaded BMP-2 at 13 days. Evaluating a large number of commercial granules with a wide range of specific surface areas confirmed this result.

(36) Interestingly, the addition of internal granule architecture (for instance, via the incorporation of micropores) also appears to further increase the rate of BMP-2 binding and the amount of BMP-2 bound particularly to larger CDHA granules with high specific surface area (FIGS. 4-6). The explanation for this phenomenon may be related to additional internal surfaces available for binding and the ability of granules with internal architecture to wick BMP containing fluid into the granules. This may be important if the external surfaces of granules without internal architecture become saturated with BMP or preferential BMP binding sites on external surfaces become saturated. Diffusion limitations of BMP-2 out of surfaces within the microporous structure compared to granules without internal architecture may also play a role in increasing the BMP retention profiles.

(37) Additionally, internal architecture appears to be required for optimal in vivo BMP retention in CaP granules (FIG. 5). High SSA granules with sufficient internal architecture to allow BMP to penetrate inside the granules have superior in vivo BMP retention than high SSA granules without internal architecture. The optimal in vivo BMP retention is likely due to the specific dimensions of the internal architecture of the granules to conduct BMP solution inside the granules to allow BMP to bind to the internal surfaces of the granule while preventing osteoclasts from entering. As a result, osteoclasts have to resorb the majority of the granule from the outside in to release the internally bound BMP. This phenomenon mimics the optimal osteoclast mediated gradual release of BMP from CaP cements.

(38) The internal architecture of calcium ceramic granules of the present invention is defined, in many cases, by micropores. Micropore dimensions generally vary with the size of the particles themselves, but as a general matter, the micropores in calcium ceramic granules of the invention are sized to permit fluids, to flow or migrate deep into the particle before encountering a granule surface. Thus, when these granules are incubated in BMP-containing solutions, the BMP is able to penetrate and adhere to surfaces that are deep within the interior of the granule.

(39) The design criteria for the geometry of the internal granule architecture required to allow optimal BMP retention are complex. Micropores on the order of 2-10 um are required to allow BMP in solution to penetrate into the granules without immediately contacting the CaP surfaces on the outside of the granules. This is required as a result of the high binding affinity of BMP for CaP, particularly with high specific surface area. These types of surfaces have an extremely high carrying capacity to bind BMP. Without wishing to be bound by theory, it is thought that without the appropriate microporosity, the outside surfaces of the granules would filter out all the available BMP in solution preventing BMP penetration into the internal structure of the granules. This size range of micropores is also small enough to prevent multinucleated osteoclasts from penetrating into the inside of the granules. For granules with a diameter larger than about 200 um, larger secondary macropores on the order of 20-50 um are required to conduct the BMP solution further into the interior of the granules. In preferred embodiments, macro- and micropores form a network of interconnected passageways through which fluid can penetrate into a granule. It should be noted, however, that internal architecture is not sufficient for optimal BMP retention for granules that have low specific surface area.

(40) Solutions and Kits for Protein Loading of Granules

(41) In constructs of the present invention, BMPs are primarily carried by the ceramic granules embedded within the construct. However, as discussed above, current methods for loading granules with BMP result in BMP accumulation on the exterior surfaces of the granules, creating a protein “rim” rather than penetrating the highly porous structures of the granules that are described above. The protein rim may contribute to burst release of BMP following construct implantation, while penetration of BMP into the granules may contribute to an extended release profile insofar as BMP associated with internal surfaces of the granules are shielded from release until the implanted granules degrade, for example due to osteoclastic activity at the site of implantation. Without wishing to be bound by any theory, it is believed that, in general, BMPs are highly soluble in low pH, low-ionic strength buffers; state of the art BMP buffers with these characteristics generally have low buffering capacities. On the other hand, CaP granules are generally alkaline; this difference may contribute to the relatively limited infiltration of BMP into granules using current methods.

(42) The inventors have discovered two factors that facilitate the penetration of BMP into the interior pore structure of the granules: first, the inventors have found that pre-treatment of the granules with an acid solution (for example, 50 mM hydrochloric acid (HCl) or, more preferably, 500 mM acetic acid), also referred to as “etching” the granules, prior to incubation of the granules (either alone or incorporated into constructs) with BMP-containing solutions (referred to as the “protein loading” step) aids in the transport of BMP into the interior pore structure of the granules. An alternative strategy to achieve a similar result is to manufacture the granules in such a way that the final pH of the granules remains below pH 7, more preferably at or below pH 6. Second, the inventors have found that certain compositions, including generally those with relatively higher buffering capacities, also facilitate transport of BMP into the interior pores of the granules. Each of these factors is discussed in turn below:

(43) With respect to etching of granules prior to protein loading, the inventors have found that pre-treatment of the granules by 15 minute-4 hour incubation in an acidic solution, preferably 500 mM acetic acid, at a ratio of 10 mL per gram weight of calcium-deficient hydroxyapatite (CDHA) granules improves infiltration of the BMP into the granules. Treating granules with acid lowers the pH of the granules to pH 5.5-6.0. pH of granules is measured according to a standardized protocol, in which 0.25 grams of granules are mixed with 12.5 mL of a solution comprising 1% NaCl, then a pH measurement is made from the resulting slurry.

(44) Etching can also result, in certain cases when the buffer is of sufficiently high buffer capacity and sufficiently low pH, in increased release of BMP into solution, e.g. during washing of the granules. The skilled artisan will appreciate, additionally, that protein loading and elution results similar to those obtained by etching of granules may be obtained in some instances by utilizing granules having an inherently low (e.g. less than 7.0) pH in their as-manufactured state.

(45) With respect to new BMP buffers, the inventors have found that several buffer components and/or compositions can result in improved infiltration of BMPs into the internal pores of granules. A non-limiting listing of buffers according to the present invention (as well as the clinically used rhBMP-2 buffers for purposes of comparison) appears in Table 2:

(46) TABLE-US-00001 TABLE 2 EXEMPLARY BUFFER COMPOSITIONS Buffer Composition pH BVB-001 5 mM Glutamic Acid 4 Buffer 0.15% Glycine 1% Sucrose 0.01% Polysorbate-80 (OPTIONAL) Water for Injection BVB-005 25 mM Glutamic Acid 3.5 Buffer 0.75% Glycine 1% Sucrose 0.01% Polysorbate-80 (OPTIONAL) Water for Injection BVB-010 50 mM Glutamic Acid 3 Buffer 1.5% Glycine 1% Sucrose 0.01% Polysorbate 80 (OPTIONAL) Water for Injection pH Adjusted with HCl BVB-012 25 mM Glutamic Acid 4 Buffer 2% Glycine 1% Sucrose Water for Injection BMP2 5 mM Glutamic Acid 4.5 buffer-1 2.5% Glycine 0.5% Sucrose 5 mM NaCl 0.01% Polysorbate 80 Water for Injection BMP2 25 mM Glutamic Acid 4.5 buffer-2 2.5% Glycine 0.5% Sucrose 2 mM NaCl 0.01% Polysorbate 80 Water for Injection

(47) The inventors have found that the BVB-005 (“5×”) and BVB-010 (“10×”) buffer formulations improve BMP infiltration into granules relative to the BMP-2 buffers used previously or the relatively weak buffer BVB-001 (“1×”). More generally, and without wishing to be bound by any theory, weakly acidic buffer solutions with buffering capacities sufficient to maintain low pH (e.g. less than 5.0) when incubated with granules are preferred in various embodiments of the present invention. Specifically, while the solutions set forth in Table 2 are buffered by glutamic acid and/or glycine, a variety of other buffering agents are useful in protein loading solutions of the present invention, including lactic acid, acetic acid, formic acid, malic acid, malonic acid, aspartic acid, citric acid, tartaric acid, phosphoric acid, fumaric acid and/or succinic acid. In preferred cases, the buffering agent has a pKa between about 2.3 and 4.5, while the buffer as a whole preferably has a pH between 3.5 and 4.0.

(48) In addition to buffering agents, solutions of the present invention can optionally incorporate one or more additives, including without limitation 0.01%-0.1% (w/v) polysorbate-80, 0.5%-5% (w/v) Sucrose, 0.5%-5% (w/v) Trehalose, 0.5%-5% (w/v) Sorbitol or 0.5%-5% (w/v) Mannitol.

(49) By way of example (and not limitation), one buffer solution according to the present invention has a composition of (i.e. consists essentially of) 5 mM Glutamic acid, 0.15% (w/v) Glycine, 1% Sucrose, and water with a pH of 4.0. Another buffer solution is 50 mM Glutamic acid, 1.5% (w/v) Glycine, 1% Sucrose, and water with a pH adjusted to 3.0 with hydrochloric acid, and another suitable buffer is 50 mM Glutamic acid, 0.75% (w/v) Glycine, 1% Sucrose, and water with a pH of 3.5. Yet another buffer solution is 25 mM Glutamic acid, 0.75% (w/v) Glycine, 1% Sucrose, and water with a pH adjusted to 3.5 with HCl, another is 25 mM Glutamic acid, 0.75% (w/v) Glycine, 1% Sucrose, and water with a pH of 3.7, and another buffer solution is 25 mM Glutamic acid, 2% (w/v) Glycine, 1% Sucrose, and water with a pH of 4.0. Still another suitable buffer solution is 50 mM Glutamic acid, 1.5% (w/v) Glycine, 1% Sucrose, and water with a pH of 3.7. Other buffer solutions may have similar compositions with the inclusion of additional or modified components.

(50) FIG. 8 includes several fluorescent micrographs of naïve (panels A, B, C) and etched (panels E, F, G) granules treated with buffers having various pH and buffering capacities during the protein loading step. In naïve granules, a protein rim was visible in all conditions, and in both naïve and etched granules, infiltration of the granules improved as the buffering capacity and acidity of the buffer solution increased. Importantly, infiltration of BMP into granules was improved by etching when intermediate buffering capacity buffers were used (panels B and E), but good infiltration was also observed in naïve granules loaded in buffers with high-buffering capacity. These results indicate that it is not strictly necessary to pair high-buffering capacity, low pH protein loading buffers with granule etching to achieve good BMP infiltration into the granules; accordingly, various embodiments of the present invention utilize etching and/or the improved protein loading buffer. Again, the skilled artisan will appreciate, additionally, that protein distribution results similar to those obtained by etching of granules may be obtained in some instances by utilizing granules having an inherently low (e.g. less than 7.0) pH in their as-manufactured state.

(51) The inventors have also found, however, that different combinations of granule pH (e.g. etched treatment) and buffering solutions can result different degrees of apparent binding between the granules and BMP, as evidenced by the fraction of BMP eluted in a saline wash of the BMP loaded granules. FIG. 10 illustrates the BMP fractions eluted in the saline wash (dark bars) and a subsequent guanidinium chloride protein extraction (light bars). In general, about 20% or less of the BMP elutes during the saline wash, though in etched granules loaded with BVB-010 buffer almost 40% of the BMP eluted during the wash, while etched granules loaded with BVB-005 buffer released less than 10% of their BMP during the wash. Without wishing to be bound by any theory, it is believed that these differences may correlate with differences in BMP release by implanted constructs, and the present invention encompasses constructs with BMP release kinetics that can be tuned by varying the BMP loading buffer and/or the pH (etching) of the granules.

(52) In use, protein-loaded granules generated through the use of the compositions and methods of the present invention form one part of a multipart construct for use in treating patients. As is described above, these constructs generally include an osteoinductive protein that is preferably associated with and elutes from granules with complex and interconnected networks of micropores of varying size, which granules are in turn embedded, inserted, or otherwise in contact with a polymer matrix that has a macroporous structure to facilitate cellular and vascular infiltration, and which is characterized by a residence time on the order of several weeks (advantageously permitting extended delivery of osteoinductive protein as more specifically described above) and a stiffness and compression resistance sufficient to enable the construct to remain intact and provide structural support for new bone growth when implanted. These constructs will typically include a porous polymer matrix preferably comprising collagen but, optionally comprising other naturally-occurring or synthetic polymers.

(53) Granules (alone or integrated into constructs) can be bundled into kits that can include one or more of a BMP loading solution, an applicator for applying the loading solution to the granules or a granule-containing construct or composition, and/or for placing the construct in the body of a patient, and instructional materials which describe use of the kit, or its components, to perform the methods of the invention. Although exemplary kits are described herein, the contents of other useful kits will be apparent to the skilled artisan in light of the present disclosure. Each of these kits is included within the invention.

(54) The invention includes a kit for treatment to prevent loss of and/or increase bone mass in a patient in need thereof. The kit includes a composition comprising granules as described above, along with an osteoinductive protein, which can be in lyophilized or other dry form or, more preferably in solution in a buffer described above. If the protein is lyophilized or otherwise not in solution, the kit also preferably include a diluent or loading buffer as described above. The kit also includes an applicator, including, but not limited to, a syringe for adding fluid to the protein vessel and/or wetting the granule-containing composition, or a delivery apparatus for placing the composition, including granules loaded with the osteoinductive protein, into the body of a patient. Further, the kit can optionally include an instructional material setting forth the pertinent information for the use of the kit to treat or prevent bone loss, promote union or knitting of a fracture, and/or otherwise increase bone mass or treat a bone condition in the patient.

CONCLUSION

(55) Throughout this application, reference is made to “macropores,” “micropores” and macro- and microporosity. In general, macropores have a cross-sectional dimension greater than 100 microns, while micropores are between 100 nm and 100 microns. Pores less than 100 nm are referred to as nanopores.

(56) The phrase “and/or,” as used herein should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Other elements may optionally be present other than the elements specifically identified by the “and/or” clause, whether related or unrelated to those elements specifically identified unless clearly indicated to the contrary. Thus, as a non-limiting example, a reference to “A and/or B,” when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A without B (optionally including elements other than B); in another embodiment, to B without A (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

(57) The term “consists essentially of” means excluding other materials that contribute to function, unless otherwise defined herein. Nonetheless, such other materials may be present, collectively or individually, in trace amounts of up to 2% or more in some instances.

(58) As used in this specification, the term “substantially” or “approximately” means plus or minus 10% (e.g., by weight or by volume), and in some embodiments, plus or minus 5%. Reference throughout this specification to “one example,” “an example,” “one embodiment,” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the example is included in at least one example of the present technology. Thus, the occurrences of the phrases “in one example,” “in an example,” “one embodiment,” or “an embodiment” in various places throughout this specification are not necessarily all referring to the same example. Furthermore, the particular features, structures, routines, steps, or characteristics may be combined in any suitable manner in one or more examples of the technology. The headings provided herein are for convenience only and are not intended to limit or interpret the scope or meaning of the claimed technology.

(59) Certain embodiments of the present invention have described above. It is, however, expressly noted that the present invention is not limited to those embodiments, but rather the intention is that additions and modifications to what was expressly described herein are also included within the scope of the invention. Moreover, it is to be understood that the features of the various embodiments described herein were not mutually exclusive and can exist in various combinations and permutations, even if such combinations or permutations were not made express herein, without departing from the spirit and scope of the invention. In fact, variations, modifications, and other implementations of what was described herein will occur to those of ordinary skill in the art without departing from the spirit and the scope of the invention. As such, the invention is not to be defined only by the preceding illustrative description.