Development of a high-grade glioma preclinical surgery model using an inducible KRAS/TP53 Oncopig
Caleb Y Kwon1, Ryan B Duke1, Augustino V Scorzo1
1Thayer School of Engineering, Dartmouth College, Hanover, NH, United States.
Introduction:
Surgery remains the primary treatment modality for high-grade glioma; yet, Q10 median overall survival has seen only modest improvement over the past few decades. Lack of reliable preclinical models that recapitulate the volumes and infiltrative nature of human tumors is a persistent challenge for development and evaluation of new surgical techniques. In this context, the Oncopig model has been shown to produce realistic tumors in other organs. It is a transgenic model that encodes for the oncogenic P53 and KRAS mutations which can be switched on through delivery of Cre-recombinase, usually using a viral vector, driving endogenous tumor induction. The objective of this study was to explore the potential of Oncopigs as preclinical high-grade glioma model.
Methods:
A cocktail of adeno-associated viruses (AAV) encoding Cre-recombinase was implanted in the brains of ten Oncopigs. Animals were then monitored for tumor growth using MRI. Once tumors were detected, an additional MRI was acquired several days later to track growth and animals then either underwent surgical resection or tissue harvest. Tumor specimens were collected for histopathological analysis.
Results:
Tumors were observed on MRI scans for six of the ten pigs. The median time to observed tumor onset was 24 days and ranged from 13 to 104 days. Tumor volumes prior to euthanasia in these animals ranged from 250 to 2000 mm3. Pathological analysis of tumor specimens revealed infiltrative tumors in six subjects, with features consistent with high-grade glioma. Fluorescence-guided surgical resection with ALA-PpIX in one animal revealed high levels of PpIX fluorescence, also consistent with high-grade gliomas in humans.
Discussion:
These results indicate that the Oncopig can be used to induce brain tumors that exhibit pathological features consistent with high-grade glioma in humans. A realistic high-grade glioma model such as this should support better preclinical development and evaluation of new surgical and other treatment strategies.
