To design slow resorption patient-specific bone graft whose properties of bone regeneration are increased by its geometry and composition and to assess it in in-vitro and in-vivo models.

A graft composed by hydroxyapatite (HA) and β-TCP was designed as a cylinder with 3D gyroid porosities and 7 mm medullary space based on swine's anatomy. It was produced using a stereolithography 3D-printing machine (V6000, Prodways).

Sterile bone grafts impregnated with or without a 10µg/mL porcine BMP-2 (pBMP-2) solution were implanted into porcine femurs in a bone loss model. Bone defect was bi-weekly evaluated by X-ray during 3 months. After sacrifice, microscanner and non-decalcified histology analysis were conducted on biopsies.

Finally, osteoblasts were cultured inside the bone graft or in monolayer underneath the bone graft. Cell viability, proliferation, and gene expression were assessed after 7 and 14 days of cell culture (n=3 patients).

3D scaffolds were successfully manufactured with a composition of 80% HA and 20% β-TCP ±5% with indentation compressive strength of 4.14 MPa and bending strength of 11.8MPa.

In vivo study showed that bone regeneration was highly improved in presence of pBMP-2. Micro-CT shows a filling of the gyroid sinuses of the implant (Figure 1).

In vitro, the presence of BMP2 did not influence the viability of the osteoblasts and the mortality remained below 3%. After 7 days, the presence of BMP2 in the scaffold significantly increased by 85 and 65% the COL1A1 expression and by 8 and 33-fold the TNAP expression by osteoblasts in the monolayer or in the scaffold, respectively. This BMP2 effect was transient in monolayer and did not modify gene expression at day 14.

BMP2-impregnated bone graft is a promising patient-personalized 3D-printed solution for bone defect regeneration, by promoting neighboring host cells recruitment and solid new bone formation.

For any figures and tables, please contact the authors directly.

Decellularization techniques have advanced to reduce the risk of immune rejection in transplantation. Validation of these protocols typically relies on Crapo's criteria¹, which include the absence of visible nuclei and low DNA content. In our study, five decellularization protocols were compared to determine the optimal approach for human fascia lata (HFL) samples. However, our findings raised questions as to why recipients can still develop immunity despite meeting validation criteria.

HFL samples were decellularized using four protocols with SDS-Triton X100-DNase (D1 to D4-HFL) and one protocol using solvent-detergent-based baths (D5-HFL). The decellularized samples (D-HFL) were compared to native samples (N-HFL) using histology, and DNA content was measured. The human leukocyte antigen (HLA) content within the matrix was assessed using western blot analysis. Both D-HFL and N-HFL samples, along with negative control patches, were implanted in the backs of 28 Wistar rats. Anti-human IgG serum levels were evaluated after one month.

H&E and Hoechst staining revealed the absence of residual cells in all decellularization protocols. DNA content was consistently below the critical threshold (p<0.05). All implanted D-HFL samples resulted in significantly lower anti-human IgG levels compared to N-HFL (p<0.01). However, 2.5 out of 4 rats developed immunity after being implanted with D1 to D4-HFL, with varying levels of anti-human IgG. Only rats implanted with D5-HFL showed undetectable levels of IgG and were considered non-immunized. Western blot analysis indicated that only D5-HFL had a residual HLA content below 1%.

The literature on decellularization has primarily relied on Crapo's criteria, which do not consider the role of HLA mismatch in acute immune rejection. Our results suggest that a residual HLA content below 1% should also be considered to prevent immunization, even if other validation criteria are met. Further research is needed to evaluate the impact of residual HLA levels on human allotransplantation outcomes.

Periosteal mesenchymal stem cells (PMSC) are an emerging niche of stem cells to enhance bone healing by tissue engineering process. They have to be differentiated into osteoprogenitors in order to synthesize new bone matrix. In vitro differentiation with specific differentiation medium (DM) is not exactly representative of what occurs in vivo. The interaction between PMSC and growth factors (GF) present in biological matrix is somewhat less understood. The goal of this study is to explore the possibility of spontaneous PMSC differentiation in contact with different biological matrices without DM.

500.000 porcine PMSC were seeded on 6-well plates and cultured with proliferation medium (PM). When reaching 80% confluence, biological samples (n=3) of demineralized bone matrix (DBM), decellularized porcine bone allograft (AOp), human bone allograft (AOh), human periosteum (HP) and human fascia lata (HFL) were added. Negative and positive control wells included cells with only PM or DM, respectively. The differentiation progress was assessed by Alizarin Red staining at days 7, 14 and 21. Bone morphogenetic protein content (BMP 2, 4, 5, 6, 7, 8, 9 and 11) of each sample was also investigated by western blot.

Alizarin red highlighted bone nodules neoformation on wells containing AOp, AOh and DBM, like positive controls. HP and HFL wells did not show any nodules. These results are correlated to a global higher BMP expression profile in AOp than in HP and HFL but not statistically significant (p=0.38 and p>.99, respectively). The highest expression in each tissue was that of BMP2 and BMP7, which play an important role in osteoinduction.

PMSC are well known to participate to bone formation but, despite BMP presence in HP and HFL, they did not permit to achieve osteogenesis alone. The bone contact seems to be essential to induce in vitro differentiation into osteoprogenitors.

The purpose of this study is to enhance massive bone allografts osseointegration used to reconstruct large bone defects. These allografts show >50% complication rate requiring surgical revision in 20% cases. A new protocol for total bone decellularisation exploiting the vasculature can offer a reduction of postoperative complication by annihilating immune response and improving cellular colonization/ osseointegration.

The nutrient artery of 18 porcine bones - humerus/femur/radius/ulna - was cannulated. The decellularization process involved immersion and sequential perfusion with specific solvents over a course of one week. Perfusion was realized by a peristaltic pump (mean flow rate: 6ml/min). The benefit of arterial perfusion was compared to a control group kept in immersion baths without perfusion.

Bone samples were processed for histology (HE, Masson's trichrome and DAPI for cell detection), immunohistochemistry (IHC : Collagen IV/elastin for intraosseous vascular system evaluation, Swine Leukocyte Antigen – SLA for immunogenicity in addition to cellular clearance) and DNA quantification. Sterility and solvent residues in the graft were also evaluated with thioglycolate test and pH test respectively.

Compared to native bones, no cells could be detected and residual DNA was <50ng/mg dry weight. Intramedullary spaces were completely cleaned. IHC showed the preservation of intracortical vasculature with channels bounded by Collagen IV and elastin within Haversian systems. IHC also showed a significant decrease in SLA signaling. All grafts were sterile at the last decellularization step and showed no solvent residue. The control group kept in immersion baths, paired with 6 perfused radii/ulnae, showed that the perfusion is mandatory to ensure complete decellularisation.

Our results prove the effectiveness of a new concept of total bone decellularisation by perfusion. These promising results could lead to a new technique of Vascularized Composite Allograft transposable to pre-clinical and clinical models.