“Regeneration of a Pediatric Alveolar Cleft Model using 3D Printed Bioceramic Scaffolds and Osteogenic Agents: Comparison of Dipyridamole and rhBMP-2.”

Abstract

Purpose:

Alveolar clefts are traditionally treated with secondary bone grafting, but this is associated with morbidity and graft resorption. Although rhBMP-2 is under investigation for alveolar cleft repair, safety concerns remain (e.g. growing suture pathology). Dipyridamole (DIPY) is an adenosine receptor indirect agonist with known osteogenic potential. This study compared DIPY to rhBMP-2 at alveolar cleft defects delivered via 3D-printed bioceramic (3DBC) scaffolds.

Methods:

Skeletally immature New Zealand White rabbits underwent unilateral, 3.5mm x3.5mm alveolar resection adjacent to the growing suture. Five served as negative controls. The remaining defects were reconstructed with 3DBC scaffolds coated with 1000μm-DIPY (n=6), 10,000μm-DIPY (n=7), and 0.2 mg/mL-rhBMP-2 (n=5). At t=8 weeks, new bone was quantified using Amira 6.1 software. Non-decalcified histology was performed, and new bone was mechanically evaluated. Statistical analysis was performed using a generalized linear mixed model and Wilcoxon rank sum test.

Results:

Negative controls did not heal while new bone formation bridged all 3DBC treatment groups. 1,000μm-DIPY scaffolds regenerated 28.03±7.38%, 10,000μm-DIPY scaffolds regenerated 36.18±6.83% (p=0.104 1,000μm vs. 10,000μm DIPY), and rhBMP-2 coated scaffolds regenerated 37.17±16.69% bone (p=0.124 vs. 1,000μm-DIPY and p=0.938 vs. 10,000μm-DIPY). On histology/electron microscopy, no changes in suture biology were evident for DIPY, while rhBMP-2 demonstrated early signs of suture fusion. Healing was highly cellular and vascularized across all groups. No statistical differences in mechanical properties were observed between either DIPY or rhBMP-2 when compared to native bone.

Conclusion:

Dipyridamole generates new bone without osteolysis and early suture fusion associated with rhBMP-2 in skeletally immature bone defects.

Background

Congenital defects of the alveolus affect approximately three out of four patients with cleft lip and palate.¹ Structural consequences of an alveolar cleft include maxillary arch instability, inability to support and erupt dentition, and facial asymmetry. Secondary alveolar bone grafting is the reconstructive modality used by most centers, and iliac cancellous bone graft (ICBG) remains a commonly used donor site due to factors such as straightforward operative approach and considerable donor bone supply.^1–5 However, autologous grafts are limited by donor site morbidity, delayed healing, and resorption that may necessitate revision surgery.^3,6,7 These challenges provide the impetus for tissue engineering based alternatives capable of generating new, functional bone in the immature skeleton.

A key tenet of tissue engineering is the use of a structural matrix or scaffold to support cellular regeneration.^8–10 To date, clinical applications of scaffolds for alveolar clefts have focused on their capacity for local delivery of osteogenic agents or cells.^11–16 However, it was not until recently that by using 3D printing, deliberate geometric design of scaffold lattice dimensions were able to facilitate regeneration of vascularized, lamellar bone across critical-sized segmental mandibular defects in a translational rabbit model without bioactive agents.¹⁷ These scaffolds were made with calcium-phosphate based bioceramics, which have well-established osseoconductive^18,19 and safety characteristics.^20,21 The initial success of this scaffold inspired the investigation of combining this scaffold with an effective osteogenic agent while prioritizing patient safety.

rhBMP-2 is among the most investigated bone regenerative agents;^22–28 several groups have reported their experience with off-label rhBMP-2 use for alveolar cleft repair.^{12,16,27,29–33} Although promising regenerative outcomes have been reported, BMPs remain contraindicated in the pediatric population.³⁴ FDA concerns are based on reports of exuberant bone growth, inhibited bone healing, and edema.³⁴ Life-threatening events have been associated with BMPs,³⁵ and the biologic importance of BMP signaling on osteosarcoma has been described.³⁶ In the craniomaxillofacial skeleton, premature suture fusion has also been reported.^37–39 These concerns have provided the impetus for investigation of alternative, potentially safer osteogenic agents, such as those that activate the recently described adenosine receptor pathway.

Adenosine is a purine nucleoside with effects at almost every organ system,⁴⁰ and activation of its receptors has demonstrated bone regeneration equal to BMP-2 without concerning side effects in a murine model.⁴¹ There are four adenosine receptors: A1, A2A, A2B, and A3. A2A receptor (A2AR) ligation attenuates osteoclast activity and numbers,^42–44 and augments osteoblast differentiation.⁴⁵ These effects are achieved with either direct or indirect A2AR activation, as demonstrated with Dipyridamole (DIPY), an inhibitor of Type 1 equilibrative nucleoside transporter (ENT1).⁴⁵ Dipyridamole has a well-established history of safe use spanning decades in both adult and pediatric patients as an antithrombotic and vasodilatory medication.^46–48 The long-standing safe human application combined with, Dipyridamole’s more recently discovered capacity for bone regeneration in skeletally mature calvarial models⁴⁹ makes it an osteogenic agent of interest for creation of safe bone tissue engineering constructs (Figure 1).

Figure 1:

Schematic illustrating adenosine receptor activation & pathways affected in osteoblast & osteoclast activity

This study compares the effects of locally delivering Dipyridamole and rhBMP-2 on alveolar clefts in a skeletally immature translational rabbit model. The objective of this experiment was to assess the capacity of 3D-printed bioceramic (3DBC) scaffolds coated with these therapeutic agents to effectively and safely regenerate bone in an alveolar cleft defect. We hypothesized that 3DBC scaffolds coated with DIPY would achieve bone regeneration comparable to rhBMP-2 without concerning side effects.

Materials & Methods

3D Ink Formulation & Scaffold Design

Colloidal Gel formulation

Ceramic powders were calcined and milled to synthesize colloidal gel formulation.⁵⁰ β-TCP phosphate gel formulation with a solid volume fraction (ϕ_ceramics) of ~46% was made by blending ceramic powder, ammonium polyacrylate (Darvan 821A; RT Vanderbilt, Norwalk, CT, USA), deionized water (DI)-H₂O, hydroxypropyl methylcellulose (Methocel F4M; Dow Chemical Company, Midland, MI, USA) and polyethylenimine (Sigma-Aldrich, St. Louis, MO, USA).^50,51

Scaffold Design & 3D Printing

Scaffolds were designed via computer-aided design (CAD) (RoboCAD 4.3; 3D Inks LLC, Tulsa, OK, USA) (Figure 2). They were 3D printed via additive manufacturing (Aerotech Inc., Pittsburgh, PA, USA) and made of 100% beta tricalcium phosphate, ~3.5mm length & width & thickness, and were printed with 250μm struts & 330 μm pore spacing (Figure 3). The colloidal gel formulation, or ink, was loaded into a syringe (Nordson Corp., Westlake, OH, USA) with a 250 μm-diameter extrusion nozzle (Nordson Corp.). Scaffolds were printed layer-by-layer in a paraffin oil tray to prevent the construct from drying. After printing, scaffolds were sintered at 400°C, 900°C, and 1100°C to densify constructs and burn out impurities.⁵¹

Figure 2:

CAD/CAM design of scaffold used in alveolar cleft defect. A) top view B) front view C) first layer iso view D) layer 9 iso view and E) final scaffold iso view

Figure 3:

3D Printed B-TCP scaffolds with osseoconductive geometries (A) gross figure (B) High magnification scanning electron microscopy illustrating strut and pore print.

Surgical Injury Model

Following approval of Institutional Animal Care and Use Committee (IACUC), all surgeries were performed under sterile conditions. 24 skeletally immature New Zealand White rabbits underwent unilateral, ~3.5mm × ~3.5mm alveolar cleft defect injury. This defect size was established as critical-sized in preliminary experiment and demonstrated to be non-healing in this work. A ~13 mm skin incision was made on the right aspect of the midface to visualize the maxilla. The alveolar ridge, including the suture at the interface of the pre-maxilla and maxilla, was visualized. Using a scaffold as template, a ~3.5mm length and width defect was made with an oral surgery burr and, after proper visualization and elevation of the maxillary sinus membrane, the bone defect was extended vertically ~3.5mm to match the volume of the scaffold (Figure 4). The scaffolds were inset in the defect by a fit and fill technique, paying careful attention to two aspects: i) obtaining primary stability of the scaffold, and II) avoiding violation of the maxillary sinus membrane/nasal cavity fibromucosa. The location of this defect was within 2mm of the alveolar growth suture. Scaffolds were immersed in in 2% bovine collagen solution (Collagen I, Bovine; Corning Inc., Corning, NY, USA) to carry the experimental molecules investigated. After being coated with collagen, scaffolds were coated with either one of two different concentrations of Dipyridamole (DIPY) (1,000 uM & 10,000 uM) or an rhBMP-2 dose consistent with other studies (0.2 mg/mL) (INFUSE^®, Bone Morphogenetic Protein, Medtronic, Memphis, TN, USA).^38,52,53 Our group has previously reported that 100μm DIPY dosing can augment bone regeneration in skeletally mature models;⁴⁹ the two higher concentrations used in this experiment were chosen to assess if these effects are dose dependent and whether or not supra-therapeutic dosage causes any harm to growing skeletal sutures. Alveolar defects were reconstructed with 3DBC scaffolds coated with 1000μm DIPY (n=6), 10,000μm DIPY(n=8), or 0.02 mg/mL rhBMP-2 (n=5). Five defects which received no intervention served as negative controls. Positive controls were not included because previous work has already established the regenerative benefits of Dipyridamole when delivered via 3D-printed ceramic scaffolds.⁵⁴ Furthermore, the aim of this work was to compare the effects of Dipyridamole to rhBMP-2 as a potential therapeutic agent in pediatric skeletal defects. Animals were given food ad libitum postoperatively without activity restrictions and were euthanized via anesthetic overdose at 8-weeks.

Figure 4:

Intraoperative placement of 3D printed scaffold at skeletally immature rabbit alveolus. (A) Exposure of alveolar ridge (B) After surgical resection of alveolar ridge/creation of cleft defect (C) Defect replaced with 3D printed scaffold in fit-and-fill manner.

Sample analysis

The rabbit midface was removed en bloc and excess soft tissue and bone were dissected. The maxillae were dehydrated in a series of ethanol solutions (70–100% EtOH) and embedded in a methyl methacrylate resin. After embedding, rabbit maxillae were scanned using micro-computed tomography (μCT 40, Scanco Medical, Basserdorf, Germany) with an 18 μm slice resolution. Data were exported in DICOM format, imported into Amira 6.3 software (Visage Imaging GmbH, Berlin, Germany), and analyzed for bone regeneration, scaffold degradation/resorption, and gross changes in bone morphology.

Areas of new bone were distinguished from scaffold structure by image thresholding⁵⁴. Regions of bone, scaffold, and scaffold interstices (empty space/soft tissue infiltrate) were isolated and cumulatively added to a total volume of 100% (Figure 5). A single, blinded investigator completed all microCT analysis.

Figure 5:

MicroCT slices imported into AMIRA software to distinguish between scaffold and quantification of bone (A) Scaffold (yellow arrow) easily distinguished from bone (green arrow) based on density. (B) Scaffold highlighted in red for volume-editing and isolation of scaffold from alveolus. (C) Bone selected without scaffold to demonstrate selection of bone without scaffold.

Histomorphology & Scanning Electron Microscopy

After microCT imaging, plasticized blocks were subjected to serial sectioning with a diamond saw (Isomet 2000, Buehler Ltd., Lake Bluff, IL, USA). Using a grinding machine (Metaserv 3000, Buehler, Lake Bluff, IL, USA) with water irrigation, samples were treated with a series of SiC abrasive paper until approximately 130 μm thick. Polished samples were subsequently stained in Stevenel’s blue and Van Geison fuchsine to differentiate soft, connective, and bone tissue.^55–57 Histomorphometric analysis was not conducted because of its limitations in measuring 3-dimensional structures. Backscatter electron microscopy was performed to qualitatively assess suture patency.

Nanoindentation & Assessment of Mechanical Properties

Nanoindentation was performed with a triboindenter (TI 950, Hysitron, Minneapolis, MN) to assess for mechanical properties of newly formed bone, as previously described.^58–61 In brief, nanoindentation is a well-described mechanical test utilized for investigating mechanical properties such as elastic modulus and hardness of bone at localized regions and at small scales.^58–61 An indenter tip is used to penetrate a material to apply a pre-determined load, and the corresponding displacement of the material is recorded, quantified as the resulting elastic and plastic deformation that occurs in response to this pre-determined load. For this experiment, the nanoindenter was loaded with a Berkovich diamond pyramidal-shaped probe/tip. Water was added to each sample surface to assess tip and probe calibration. A loading profile with a peak load of 300 μN at a rate of 60 μN/s was applied, with a holding period of ten seconds and unloading period of two seconds. For each specimen, indentation (n=30 indent points) was performed within scaffold interstices at sites of newly regenerated bone, as well as outside of the scaffold (n=30 indent points) at an uninjured, native bone site to serve as an internal control. Regions of bone were chosen based with a light microscope (Hysitron). Each loading profile generated a force–displacement curve via Hysitron TriboScan software with slopes that yielded the reduced modulus Er and hardness H of bone tissue in giga-pascals (GPa) using the following formulae:

$$E_r=\frac{\sqrt{\pi }}{2\sqrt{\text{A}\left(h_c\right)}} \times s;\text{\hspace{0.17em}H}=\frac{{\text{P}}_{max}}{\text{A}\left(h_c\right)}$$

with A(h_c) representing contact area at peak load (represented by P_max) and s representing stiffness. Elastic modulus is defined as the stress (force/surface area of the material) experienced by the material for a given elastic strain (deformation). Reduced elastic modulus takes into account the contact area of the nano-indenter tip. Hardness was defined as the maximum force applied by the indenter tip divided by peak or maximum load.

Statistical analysis

MicroCT quantification was analyzed for normality using the Shapiro-Wilk test (p>0.05) prior to analysis (IBM SPSS v23, IBM Corp., Armonk, NY, USA). Quantified bone regeneration data are presented as mean values with corresponding 95% confidence intervals. Amira 3D software quantification of scaffold treatment groups were compared using one-way ANOVA analysis.

For nanoindentation, newly formed bone was compared to internal control bone by using a generalized linear mixed model (GLMM). In brief, GLLMs takes into consideration random effects within subjects for determining treatment effect of experimental groups. In the case of nanoindentation, data is clustered due to repeated indentation measures on the same samples.

Results

MicroCT

Bone was quantified as a function of regeneration/growth within scaffold interstices. Negative controls remained unhealed with limited bone growth at defect margins observed. 1,000μm-DIPY scaffolds regenerated 28.03±7.38% bone and 10,000μm-DIPY scaffolds regenerated 36.18±6.83% bone (p=0.104, highest power between all group comparisons at β=0.68). rhBMP-2 coated scaffolds regenerated 37.17±16.69% bone (p=0.124 vs. 1,000μm-DIPY and p=0.938 vs. 10,000μm-DIPY). MicroCT also revealed rhBMP-2 to be associated with alveolar ridge bone remodeling consistent with resorption and osteolysis (Figure 6A). None of these changes were observed for either DIPY concentration (Figure 6B) compared to native bone that was not subjected to surgical injury (Figure 6C).

Figure 6:

(A) MicroCT slice of rhBMP-2 treated scaffold with signs of ridge resorption depicted by blue arrow, as well as signs of shortening of the suture length. (B) MicroCT slice of Dipyridamole treated scaffold with no signs of ridge resorption depicted by blue arrow and no changes in suture length. (C) MicroCT slice of alveolus without surgical injury or intervention for comparison of baseline alveolar ride and suture length.

Histology & Scanning Electron Microscopy

Non-decalcified histologic sections depicted negative controls to remain critical-sized, and no inflammatory response was observed on histology (Figure 7A). Osseoconductive bone regeneration was observed spanning the full length of the alveolar defect with a highly cellular and vascularized structure in scaffolds coated with DIPY irrespective of concentration (Figure 7B), as well as rhBMP-2 (Figure 7C). O. New bone bridged the entire scaffold porosity while the suture at the interface of the premaxilla and maxilla remained patent when treated with Dipyridamole at both 1,000 uM and 10,000 uM concentrations. On high magnification histologic imaging, rhBMP-2 treated scaffolds demonstrated early signs of suture fusion (Figure 8A). Early suture fusion was confirmed with scanning electron microscopy (Figure 8 B and C).

Figure 7:

Non decalcified histologic sections of rabbit alveolus defect replaced by 3D printed scaffold or negative control in sagittal slice orientation at t=8 weeks. (A) Negative control defect (B) Dipyridamole-induced bone regeneration (C) rhBMP-2 induced bone regeneration

Figure 8:

(A) High magnification of rhBMP-2 group non-decalcified histology with signs of suture fusion (yellow arrows). (B & C) Scanning electron microscopy confirming signs of early suture fusion (yellow arrows) at sutures adjacent to rhBMP-2 treated scaffolds

Nanoindentation

Reduced elastic modulus and hardness values were ranked and are depicted on Figure 9. Both 1,000 uM DIPY and 10,000 uM DIPY-treated scaffolds as well as rhBMP-2-treated scaffolds regenerated bone with a reduced elastic modulus and that was not statistically different from native bone/internal control (p=0.46, p=0.53, & p=0.28 respectively). Native bone reduced elastic modulus was quantified at 5.53±1.98. 1,000 uM DIPY, 10,000 uM DIPY, and rhBMP-2 reduced elastic moduli were quantified at 4.52±2.86, 4.69±2.86, and 7.02±2.86, respectively (highest power at β=0.25 between native bone and rh-BMP-2). For Hardness, there was again no statistical difference between any treatment groups compared to native bone (p=0.30, p=0.84 & p=0.95 for 1,000 uM DIPY, 10,000 uM DIPY, and rhBMP-2, respectively). Native bone hardness was quantified at 0.22±0.06. 1,000 uM DIPY, 10,000 uM DIPY, and rhBMP-2 hardness were quantified at 0.17±0.09, 0.23±0.09, and 0.22±0.09, respectively (highest power at β=0.24 between native bone and 1,000 uM DIPY).

Figure 9:

(A) Reduced elastic modulus with native bone comparison as an internal control. (B) Hardness with native bone comparison as an internal control.

Discussion

A safe, tissue-engineered approach to alveolar cleft reconstruction in the growing child remains elusive. This experiment investigated if bone could be regenerated safely and robustly in the rabbit alveolus via localized A2AR activation or rhBMP-2 delivery via 3D printed bioceramic scaffolds.

Although the regenerative equivalence between rhBMP-2 and Dipyridamole has previously been reported in smaller, skeletally mature models,^41,62 this is the first comparative report in a skeletally immature, translational alveolar cleft model with assessment of mechanical properties. At t=8 weeks, the indirect A2AR agonist Dipyridamole had a dose-dependent bone regenerative response that was quantitatively equal to rhBMP-2, but unlike rhBMP-2 did not results in premature suture fusion or osteolysis. Irrespective of bone forming agent, mechanical testing indicated that the bone formed within the scaffold presents with hardness and elastic modulus consistent with native bone. These data suggest that both Dipyridamole and rhMBP-2, when delivered with 3D printed bioceramic scaffolds, may be able to regenerate vascularized bone that behaves similarly to native skeletally immature bone subjected to mechanical loading.

Dipyridamole’s osteogenic potential combined with its well-established use in pediatric and adult patients, makes it an agent of interest for clinical investigation in bone tissue engineering. For decades, Dipyridamole has been used as an antithrombotic with therapeutic ranges up to 5 mg/kg every 8 hours in pediatric patients.^63–65 In 2011, Costa and colleagues reported the novel role of adenosine receptors in osteogenic cell differentiation,⁶⁶ and since then Dipyridamole’s effects have been elucidated to also include upregulation of osteoblast differentiation⁶⁷ and attenuation of osteoclast differentiation.^41,68,69 More recently, Dipyridamole has been shown to achieve bone healing comparable to BMP-2 in murine calvarial models^41,62,69 without concerning side effects attributed to BMPs.^38,70,71 This study is the first comparison between Dipyridamole and rhBMP-2 in a skeletally immature, load-bearing translational model of the rabbit alveolus. No exuberant bone formation or changes in the maxillary suture were observed at t=8 weeks for groups treated with Dipyridamole-coated scaffolds, even at supratherapeutic doses as high at 10,000 um. This lack of adverse effects highlights the potential role of Dipyridamole as regenerative therapeutic in pediatric bone tissue engineering.

In contrast, early changes in maxillary suture biology were noted for rhBMP-2. Histology and electron microscopy both depicted evidence of early suture fusion in rhBMP-2 treated scaffolds, an adverse effect that has been reported by several groups.^37–39 Of note, the translational literature has frequently reported suture fusion, but clinical reports have been less consistent. For example, premature cranial suture fusion with rhBMP-2 has been reported,⁷² and cells from fused sutures in single-suture craniosynostosis patients highly express BMPs while under expressing BMP inhibitor Noggin.⁷³ Conversely, adverse effects of rhBMP-2 for alveolar repair have not been widely reported. Many studies assessing rhBMP-2 in pediatric patients report short-term outcomes without many adverse events^74–76 including Hammoudeh and colleagues, who have reported application of rhBMP-2 in the alveolus of ~500 patients without adverse sequelae at almost three years post-operatively.¹⁶ Long term analysis of the clinical application of rhBMP-2 are still pending.

It is possible that the pathologic effects of rhBMP-2 can be affected by carrier agent. The Food and Drug Administration (FDA) contraindication for rhBMP-2 in patients under eighteen years of age is based on absorbable collagen sponge (ACS) delivery,⁷⁷ which facilitates more prominent BMP-mediated effects compared to tri-calcium phosphate carriers.⁷⁸ This is likely consequent to prolonged BMP retention from collagen-based carriers. Since Hammoudeh et al’s promising outcomes used a demineralized bone matrix as a scaffold carrier, it is possible that using scaffold carriers with lesser retention capacity than collagen scaffolds attenuate BMP-mediated responses through a curtailed drug delivery timeframe. Our findings support this: while early signs of suture fusion are seen in our rhBMP-2 coated scaffolds, the degree of fusion is not as marked as studies with collagen-based delivery systems.³⁸ While longer time frame studies are needed to assess these effects, it may be possible that by using 3D-printed B-TCP scaffolds with geometries that osseoconduct, safer yet robust bone regeneration mediated by rhBMP-2 can be achieved than previously reported. However, pathologic effects to the growing suture were still observed in this immature large animal craniofacial model.

The pathologic changes seen in BMP-induced osteogenesis were absent in both Dipyridamole treatment groups. Additionally, the reduced elastic modulus of Dipyridamole-treated new bone was closer to normal than rhBMP-2 treated new bone, though all were not statistically different from internal control native bone values. The fine nanoindenter tip used was confirmed under scanning electron microscopy to have only indented regions within scaffold pores made up exclusively of bone, ensuring that scaffold material did not influence outcomes. Osteogenic agent restoration of bone mechanical properties to normal under nanoindentation-induced load is highly suggestive of Dipyridamole’s and rhBMP-2’s ability to restore vascularized bone that can behave like native bone when subjected to mechanical stress.

The biomaterial choice and its 3D-printed geometric design were essential contributors to the large amounts of bone regenerated for both rhBMP-2 and Dipyridamole treated scaffolds. Scaffold design was based on well-established principles from other osseoconductive biomaterials, that report how early and late bone healing can be accelerated via geometric alterations in metallic implant design to accelerate osseointegration rate.^79,80 An additional benefit of these B-TCP constructs is the overt difference in density between B-TCP and newly formed bone within scaffold porosity. Since the scaffolds used were made of 100% B-TCP, scaffold demineralization and potential mischaracterization as bone was not a concern. Our groups has reported validated this using 3D printed bioceramic scaffolds at large surgical defect sites in translational models.^17,49

There were several limitations in this study. From a technical standpoint, this alveolar defect model cannot match the intrinsic complexity of a primary cleft palate in terms of size and tri-dimensional defect volume, presence of oro-nasal fistula, interface of different tissues such as bone, fibrosis, teeth, and mucosa. Furthermore, the intraoral approach used for correction of a human cleft, unlike our extraoral approach, increases the risks of wound dehiscence, graft/scaffold exposure and contamination. A scientific limitation was experimental group size. It is possible that greater differences in results, particularly mechanical properties, may have been observed with larger group sizes. Additionally, a comparison between B-TCP scaffold and a collagen-based delivery system will likely elucidate the effect of carriers on rhBMP-2-influenced bone pathology. Finally, both shorter and longer time points to assess healing changes influenced by both Dipyridamole and rhBMP-2 are warranted. Intramembranous-like healing was observed on histology; although some regions of new bone formation resemble the lamellar organization of native bone adjacent to the surgical injury site, distinct immature woven bone is also observed—this can be attributed to the early end point of this experiment. Longer time points are being investigated, and if these longer time points consistently report mechanical properties that are equivalent to native bone, as well as intact suture patency, the adenosine receptor pathway may prove to be a valuable osteogenic agent for pediatric bone regeneration.

Conclusion

3DBC scaffolds are capable of effectively delivering both Dipyridamole and rhBMP-2 to regenerate vascularized bone in skeletally immature defects of the rabbit alveolus. Dipyridamole generates new bone without osteolysis and early suture fusion associated with rhBMP-2.