Back to Annual Meeting Posters
Sustained Delivery of rhBMP-2 via PLGA Microspheres Significantly Reduces the Effective Dose Required for Bone Formation in Small and Large Animal models: A Potential Solution to Heterotopic Ossification Induced by Bolus Delivery of High Dose rhBMP-2
Jason D. Wink, B.A.1, Patrick Gerety, M.D.1, Rami Sherif, B.A.1, Nadya Clarke, M.D.1, Jennifer Mcgrath, M.D.2, Hyun-Duck Nah, D.M.D., Ph.D.3, Jesse A. Taylor, M.D.3.
1University of Pennsylvania, Philadelphia, PA, USA, 2Northwestern University Feinberg School of Medicine, Chicago, IL, USA, 3The Children's Hospital of Philadelphia, Philadelphia, PA, USA.
Clinical use of rhBMP-2 has increased dramatically in recent years. Delivery using a collagen sponge and supra-physiologic dose, however, has been shown to produce heterotopic ossification (HO), premature closure of cranial sutures (PCCS), and a prolonged inflammatory response, all of which limit its potential for cranial applications. Previously, our lab has demonstrated that rhBMP-2 can be encapsulated in 15% Poly (lactic-co-glycolic acid) (PLGA) microspheres (PLGA-MS), with defined release kinetics. We hypothesize that the clinical dose required for bone formation can be significantly reduced when rhBMP-2 is delivered in a sustained and controlled manner.
We compared the efficacy of free rhBMP-2 (1ug and 0.1ug/implant) versus PLGA-rhBMP2 (0.1ug/implant). Free rhBMP2 or PLGA-rhBMP2 was mixed in 250ul Matrigel with and without 250K human bone marrow-derived mesenchymal stem cells (BM-MSCs) and injected subcutaneously in the dorsum of immuno-compromised nude mice (n=5/group). Implants were harvested at 4 weeks and analyzed by micro computed tomography (microCT) and histology. Control implants contained empty PLGA-MS. Critical-sized, 10mm rabbit cranial defect model was used to further test our hypothesis. Collagen scaffolds (Duragen) loaded with either free rhBMP2 (0, 0.1, 1 or 30ug/implant) or PLGA-rhBMP2 (0.1ug/implant) in the presence and absence of allogeneic BM-MSCs were directly placed into the defects. After 6 weeks, the defects were harvested and analyzed by microCT and histology to assess newly formed bone volume (BV) and surface area (SA). All statistical analyses used a Mann-Whitney Test between experimental group and PLGA-rhBMP2 group (p<0.05).
In nude mice, mineralized bone formation using 0.1ug PLGA-rhBMP2 was robust in the presence of human BM-MSCs, which was statistically indistinct from that receiving 1ug rhBMP2 (p=0.251) (Figure 1). Alternatively, bone formation was hardly detectable in the 0.1ug rhBMP2 group. In our rabbit model, 0.1ug PLGA-rhBMP2 again outperformed 0.1 ug rhBMP2 (p<0.05) and the negative controls (no treatment and scaffold only groups) without signs of HO or PCCS (p<0.05) (Figure 2). However, cranial defects were not fully repaired at these doses. 30ug rhBMP2 showed extensive bone formation with HO and PCCS. Interestingly, the use of human BM-MSCs in our mice model significantly enhanced bone formation while the addition of allogeneic BM-MSCs negated the osteogenic effect in the rabbit model, showing no statistically significant differences between groups. PLGA-MS were present in the defect sites at the 6-week time point.
The sustained delivery of rhBMP2 may reduce the effective dose required for bone formation by 10 fold in both small and large animals. While the inclusion of human BM-MSCs resulted in enhanced bone formation in nude mice, the use of allogeneic BM-MSCs negated the osteogenic effect of rhBMP2. It is plausible that an immunological reaction to allogeneic cells may have contributed to the lack of bone formation in our rabbit model.
Back to Annual Meeting Posters