Abstract 1560: Early tumor detection and characterization of a novel mouse model of Shh-driven medulloblastoma using contrast-enhanced micro-MRI.

Abstract

Abstract Introduction: An important translational application of preclinical models is the ability to draw insights into the etiology and molecular pathways altered in cancers. This is especially true for brain tumors since the early stages of disease are rarely detectable in patients and advanced-stage tumors may not accurately reflect the mutations responsible for tumor initiation. Therefore, there is a critical need for sensitive imaging protocols that allow the study of the early stages of tumor formation. In this study we optimized an in vivo high-resolution contrast-enhanced MRI protocol for the early detection and characterization of a novel mouse model of medulloblastoma (MB), the most common pediatric brain tumor which originates in the cerebellum (Cb). Method: The Patched (Ptc1) mutant mouse is one of the most studied models of MB. In our study we generated a new variant of this model (referred as Ptc1-CKO) by breeding Ptf1acre/+ mice with mice homozygous for a floxed allele of the Ptc1 gene (Ptc1fl/fl). Using this approach, Ptc1 was deleted in less than 1,000 embryonic Ptf1a-expressing granule cell progenitors (GCPs) compared to the extensive mutation of many GCPs seen in many models. Based on our results, tumors in Ptc1-CKO mice likely initiate from a single mutated GCP, and thus better models the clonal origin of sporadic human MBs. Thus, the Ptc1-CKO mouse is an ideal platform to study the early stages of MB tumorigenesis. For in vivo imaging, we used manganese-enhanced MRI (MEMRI) to obtain images 24h after intraperitoneal (IP) injection of a manganese chloride (MnCl2) solution using a 7T micro-MRI system. We acquired three-dimensional (3D) images with a 15-minute and 2-hour protocol for screening and high-resolution images, respectively. Image analysis was performed using AMIRA software. After imaging, tumors were extracted for histological and molecular analysis. Results: The contrast obtained with our MEMRI protocol showed detailed cerebellar morphology and allowed detection of MBs in Ptc1-CKO mice at different stages. Our in vivo MRI protocol was sensitive enough to detect pre-neoplastic lesions as early as 2 weeks postnatal age (N=22) and was validated with histology (N=12). Furthermore, in vivo longitudinal micro-MRI allowed the noninvasive monitoring of individual pre-neoplastic lesions and showed that individual lesions have different tumorigenic potential. Volumetric studies were performed to analyze tumor morphology and growth rates (N=25), and the advanced phenotypes were correlated with immunohistochemistry. In addition, 3D MRI images were used to guide the dissection of tumor tissue for microarray expression analysis (N=6). Our results show that Shh-driven MBs in Ptc1-CKO mice display at least two distinct imaging and molecular phenotypes, contrary to previous reports that all Shh-driven tumors converge to a common molecular endpoint. Citation Format: Giselle A. Suero-Abreu, Praveen B. Raju, Diane Pham, Edward J. Houston, Alexandra L. Joyner, Daniel H. Turnbull. Early tumor detection and characterization of a novel mouse model of Shh-driven medulloblastoma using contrast-enhanced micro-MRI. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1560. doi:10.1158/1538-7445.AM2013-1560