The Effect of Phenotypic and Genomic Changes in Primary IDH-Negative Glioblastoma on Progression Free Survival

Abstract

Background: GBM is a deadly brain tumor, representing 15% of all primary brain tumors and affecting more than 10,000 people annually. While a few genomic alterations have been useful in guiding the treatment course, they are not always applicable due to the molecular heterogeneity of GBM. Moreover, analyses of cellular changes in GBM have not been explored extensively. Objective: This project aims to discover new predictors of longer progression-free survival (PFS) among Glioblastoma (GBM) patients, including phenotypic changes at the cellular level and genomic copy number variations (CNV). Public Health Impact: GBM is a highly malignant brain tumor of unclear mechanism with varying molecular heterogeneity. Further studies of genomic events and phenotypic cellular changes are needed to advance the understanding of GBM survivability. Methods: The study design is a retrospective cohort study of 60 patients who were diagnosed with primary, IDH-negative GBM, treated at the Dana Farber Cancer Institute (DFCI), and had genomic data and biopsy tissue available. Clinical data were extracted from clinical charts. OncoCopy data for CNV’s were accessed through the DFCI database, while phenotypic cellular data were obtained directly from imaging of the biopsy samples. The analyzed measure of clinical outcome was progression-free survival (PFS). ANOVA, 2-sided t-tests, and Spearman correlations were used to detect differences in genomic and phenotypic states among the samples. Kaplan-Meier analyses, Log-rank tests, and Wilcoxon tests were conducted to evaluate survival-related data. P-values of <.05 were considered to be significant. Results: The data was validated through confirmation of existing knowledge of GBM, including longer PFS associated with MGMT methylation status, gross total resection, and chemotherapy (p ≤0.005 for all). Of the 33 CNV’s tested, 2 were significantly associated with shorter PFS: Monosomy 14 (p=0.010) and MET gene gain/amplification (p=0.049). However, the amount of cilia, total scarring, acute scarring, or micronuclei were not indicative of PFS (R2 ≤0.01 for all). These cellular events were also not associated with tumor size, age, MGMT status, or each other (p>0.40 for all). Conclusions: Monosomy 14 and MET gene gain/amplification were found to be associated significantly with shorter survival. Similar findings for Monosomy 14 have previously been reported in meningiomas, while the MET gene is a known growth factor, which add credibility to these findings. The significance of cellular changes in GBM remains unclear. Further analysis correlating the phenotypic changes to the genomic alterations may further shed light on their mechanisms.