Anne-Marie Bleau, Brian M. Howard, Lauren A. Taylor, Demirkan Gursel, Jeffrey P. Greenfield, H. Y. Lim Tung, Eric C. Holland and John A. Boockvar
Brain tumor stem cells (TSCs) hypothetically drive the malignant phenotype of glioblastoma multiforme (GBM), and evidence suggests that a better understanding of these TSCs will have profound implications for treating gliomas. When grown in vitro, putative TSCs grow as a solid sphere, making their subsequent characterization, particularly the cells within the center of the sphere, difficult. Therefore, the purpose of this study was to develop a new method to better understand the proteomic profile of the entire population of cells within a sphere.
Tumor specimens from patients with confirmed GBM and glioma models in mice were mechanically and enzymatically dissociated and grown in traditional stem cell medium to generate neurospheres. The neurospheres were then embedded in freezing medium, cryosectioned, and analyzed with immunofluorescence.
By sectioning neurospheres as thinly as 5 μm, the authors overcame many of the problems associated with immunolabeling whole neurospheres, such as antibody penetration into the core of the sphere and intense background fluorescence that obscures the specificity of immunoreactivity. Moreover, the small quantity of material required and the speed with which this cryosectioning and immunolabeling technique can be performed make it an attractive tool for the rapid assessment of TSC character.
This study is the first to show that cryosectioning of neurospheres derived from glioma models in mice and GBM in humans is a feasible method of better defining the stem cell profile of a glioma.
Sacit Bulent Omay, Yu-Ning Chen, Joao Paulo Almeida, Armando Saul Ruiz-Treviño, John A. Boockvar, Philip E. Stieg, Jeffrey P. Greenfield, Mark M. Souweidane, Ashutosh Kacker, David J. Pisapia, Vijay K. Anand and Theodore H. Schwartz
Exome sequencing studies have recently demonstrated that papillary craniopharyngiomas (PCPs) and adamantinomatous craniopharyngiomas (ACPs) have distinct genetic origins, each primarily driven by mutually exclusive alterations: either BRAF (V600E), observed in 95% of PCPs, or CTNNB1, observed in 75%–96% of ACPs. How the presence of these molecular signatures, or their absence, correlates with clinical, radiographic, and outcome variables is unknown.
The pathology records for patients who underwent surgery for craniopharyngiomas between May 2000 and March 2015 at Weill Cornell Medical College were reviewed. Craniopharyngiomas were identified and classified as PCP or ACP. Patients were placed into 1 of 3 groups based on their genomic mutations: BRAF mutation only, CTNNB1 mutation only, and tumors with neither of these mutations detected (not detected [ND]). Demographic, radiological, and clinical variables were collected, and their correlation with each genomic group was tested.
Histology correlated strongly with mutation group. All BRAF tumors with mutations were PCPs, and all CTNNB1 with mutations and ND tumors were ACPs. Preoperative and postoperative clinical symptoms and radiographic features did not correlate with any mutation group. There was a statistically significant relationship (p = 0.0323) between the age group (pediatric vs adult) and the mutation groups. The ND group tumors were more likely to involve the sella (p = 0.0065).
The mutation signature in craniopharyngioma is highly predictive of histology. The subgroup of tumors in which these 2 mutations are not detected is more likely to occur in children, be located in the sella, and be of ACP histology.