Exploration of strategies to promote fracture healing and prevent atrophic non-union in a clinical relevant model

Abstract

Atrophic non-unions are the cause of major post traumatic morbidity. They are more common in diabetics, smokers and patients suffering high energy injuries especially those with open fractures. Treatment of atrophic non-union is usually prolonged, over years, and imposes a major burden on hospital and staff expenditures. The psychological impact on the patients is another concerning aspect as most of these patients describe their lives as being ‘’on hold’’ during the treatment course.

Atrophic non-union result from biological failure at the fracture site, in view of which there is a growing trend towards using mesenchymal stem cells in prevention and treatment of atrophic non-union. In particular, prophylactic local injection of stem cells is a promising novel therapeutic method to prevent a fracture progressing to an established non-union.

However, before they are used prophylactically for non-unions, the strategy for stem cells injection needs to be optimised. A recent clinically relevant model of atrophic non-union used an external frame to stabilise a tibial fracture. However, this kind of fixation is not commonly employed to stabilize long bone fractures, instead, intramedullary nails are widely used. Thus the aim of this thesis was to develop and evaluate (in a preclinical model), a clinically relevant strategy to stimulate fracture healing using local stem cell injection in patients stabilized with intramedullary nails.

An ex vivo experiment was conducted to compare the mechanical stability between different fixation devices. Results showed that intramedullary nails were stronger and stiffer than compression plates and external fixators.

Next, a model of tibial atrophic non-union was established using intramedullary nail fixation, in which a non-critical gap of 1 mm was created in the tibial mid-shaft. To maintain this gap, a specialised spacer was inserted with minimal effect on the surface area of healing. The periosteum was stripped and the endosteum was curetted to a distance equal to one bone diameter proximal and distal to the fracture site. This resulted in the development of a non-union with the fracture remaining un-united even after eight weeks from surgery. A range of outcome measures were employed to evaluate the model, including biomechanical, radiographic and histological modalities. In contrast, the control group in which the periosteum and endosteum were kept intact but with osteotomy and non-critical gap fixation showed a complete healing with stable union confirmed biomechanically, radiographically and histologically.

Using this novel model, a double blinded randomised controlled trial was carried out comparing the quality of bone repair between fractures injected with MSCs at day 3 with those injected at week 3 after non-union induction. Radiographic, biomechanical and histological outcome measures were used.

The results showed significant callus formation and fracture union rates after stem cells injection at both time points (day 3 and week 3) suggesting both times were suitable for cell injection. However, the day 3 injected group had better results than the week 3 injected group. Comparison groups which were fixed with intramedullary nails and injected with PBS (at day 3 and week 3) failed to unite and developed atrophic non-union after eight weeks from injection. The difference between union and non-union rates was significant at P-value < 0.05 in both time points.

The mechanism by which stem cells undergo multilineage differentiation was investigated. In vitro work by collaborators in Milan –Italy identified a non-coding circular RNA called FOXP1 which was responsible for cell fate and keeping the undifferentiated identity of stem cells. According to their results, silencing of this RNA led to impaired cell differentiation. To determine the behaviour of this kind of cell in vivo, MSCs in which circFOXP1 had been silenced were injected into the model described in this thesis and were compared with standard MSCs. The results showed a higher union rate in the standard MSCs group compared to that of the silenced circ FOXP1 MSCs injected group. The in vivo results in this thesis confirmed the in vitro findings, suggesting that the role of FOXP1 RNA is to determine the fate of MSCs.

Thus, it was concluded that FOXP1 was a controlling factor for MSCs differentiation.

Autologous stem cells have a number of potential advantages over allogeneic or xenogeneic stem cells such as fewer ethical concerns and minimal risk of disease transmission. Furthermore, subcutaneous fat is an accessible and abundant source for stem cells harvesting. Therefore, the feasibility of autologous fat-derived cells implantation was tested. The results showed a robust union in all fractures that were injected with autologous cells. The quality of repair was greater than that of allogeneic, xenogeneic and PBS control groups. The radiographic parameters (callus indexes, radiopacity, RUST score) and the histological analysis showed significant bone formation with autologous cell implantation.

It was concluded that stem cells injected early after fracture/osteotomy were as effective as stem cells injected 3 weeks after fracture and that autologous cells resulted in stronger fracture repair than allogeneic and xenogeneic cells.

Therefore, a reasonable clinical strategy would be injection of autologous fat derived stem cells within a week of injury.