Radiogenomic Analysis of Clinically Relevant MRI Features in Glioblastoma Multiforme Open Access

Abstract

Background: Glioblastoma Multiforme (GBM) is a highly malignant form of brain cancer with one of the worst median survival times of all cancers. GBM tumors are characterized by several types of heterogeneity which ultimately lead to the failure of even the most intensive treatment regimens. Current research has uncovered diversity at genetic, epigenetic, and transcription levels of tumor cells, suggesting that an optimal standard of care should be tailored to individual patients harboring tumor cells with specific genomic aberrations.

Methods: In Part One of this study, we validate a novel semi-automated in silico volumetric image feature segmentation method and explore the potential prognostic power of several imaging features. In Part Two, we use both bottom-up and top-down approaches to correlate MR imaging features to several genomic aberrations of GBM.

Results: We find that our segmentation method agrees more with other volumetric techniques than with radiologists' scorings from qualitative standards and that several imaging features, notably percent necrosis (HR=1.862, P=0.01), are strongly correlated with survival. Bottom-up imaging feature / genomic correlations suggest MGMT promoter methylation status, but not EGFR or TP53 mutations or molecular subtype, is associated with certain imaging features. Topdown analyses using microarray data combined with bioinformatic software correlates antiapoptosis, growth and proliferative, and cell death pathways with the percent necrosis imaging feature.

Conclusion: We have developed and verified a robust image feature measurement methodology for GBM tumors and show that it has statistical power to both predict survival as well as to implicate various molecular pathways with certain imaging features. Magnetic resonance imaging features have the potential to serve as non-invasive biomarkers for several clinically relevant molecular pathways.

Table of Contents

I. Introduction...1-9

II. Materials and Methods...9-15
III. Results...15-20
IV. Discussion...20-26
V. Limitations and Future Directions...27
VI. Conclusion...28
VII. References...29-40
VIII. Tables and Figures

a. Table 1...41
b. Table 2...42
c. Table 3...43
d. Table 4...44
e. Table 5...45
f. Table 6...46
g. Table 7...47
h. Table 8...48
i. Figure 1...49
j. Figure 2...50
k. Figure 3...51
l. Figure 4...52
m. Figure 5...52
n. Figure 6...53
o. Figure 7...53
p. Figure 8...54
q. Figure 9...55
r. Figure 10...56
s. Figure 11...57
t. Figure 12...57
u. Figure 13...58
v. Figure 14...58
w. Figure 15...59
x. Figure 16...60
y. Figure 17...61
z. Appendix Figure 1...62
aa. Appendix Figure 2...62
bb. Appendix Figure 3...63
cc. Appendix Figure 4...63

About this Honors Thesis

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Emory College
Department	Neuroscience and Behavioral Biology
Degree	BS
Submission	Honors Thesis
Language	English
Research Field	Biology, Neuroscience Health Sciences, Radiology
Keyword	MRI magnetic resonance imaging glioblastoma multiforme radiogenomics brain tumor biomarker
Committee Chair / Thesis Advisor	Gutman, David Andrew, Emory University
Committee Members	Marsteller, Pat, Emory University Easterling, Keith, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Radiogenomic Analysis of Clinically Relevant MRI Features in Glioblastoma Multiforme ()	2018-08-28 12:45:47 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions