Oral Presentation Annual Meetings of the Endocrine Society of Australia and Society for Reproductive Biology and Australia and New Zealand Bone and Mineral Society 2016

Testing therapeutic interventions in a novel preclinical model of neurofibromatosis type 1 related tibial pseudarthrosis (#90)

Nikita Deo 1 2 , Kathy Mikulec 2 , Lauren Peacock 2 , Tegan Cheng 2 , David Little 1 2 , Aaron Schindeler 1 2
  1. Discipline of Child and Adolescent Health, Faculty of Medicine , The University of Sydney, Sydney , New South Wales , Australia
  2. Kid's Research Institute, The Children's Hospital at Westmead, Westmead, NEW SOUTH WALES, Australia

Pseudarthrosis of the tibia is a severe orthopaedic complication frequently associated with the genetic condition neurofibromatosis type 1 (NF1). Following pathological fracture in a localised tibial dysplasia, healing is recalcitrant, culminating in a pseudarthrosis (non-union). This defect has substantial clinical impact and management is complex due to inherent deficiencies in the bone arising from the loss of the NF1 gene. These include poor osteoblast differentiation and mineralisation, excessive osteoclast-driven bone resorption, and the development of fibrous tissue.
Following some reports of double gene inactivation in patient pseudarthrosis biopsies, it was inferred that a “second hit” could underlie the focal bone pathology. To model this, we developed a mouse model of conditional Nf1-null fractures using Cre-loxP technology, effectively recapitulating the human pathology. Using this model to screen treatment strategies, we demonstrated that a signalling inhibitor that targeted a pathway (JNK) downstream of NF1 led to increased union (36%) compared to vehicle treatment (7%*, p<0.01). This was associated with decreased fibrosis rather than increased bone anabolism. We also tested generic combinations of bone anabolic (10µg local rhBMP-2/ACS) and anti-resorptive agents (5x 0.02mg/kg systemic zoledronic acid/ZA). Animals co-treated with rhBMP-2 and ZA showed the highest rate of bone union (93%) compared to vehicle (7%*), ZA (0%*), and rhBMP-2 alone (86%) (*p<0.01). Co-treatment also led to significantly increased bone volume compared to vehicle**, ZA** and rhBMP-2** (**p<0.01) and decreased pathological callus fibrous tissue compared to vehicle and rhBMP-2 groups.
These data highlight that a generalised approach not specifically targeting the deficient pathways in NF1 can nevertheless effectively promote healing and is consistent with a clinical case series where tibial pseudarthroses were treated with BMP-2 and systemic bisphosphonates. We are continuing to elaborate upon this model to examine bone lesion formation in neonatal mice to reproduce early pathobiology.