PROJECT SUMMARY/ABSTRACT
Glioblastoma (GBM) is a uniformly fatal disease with very few clinical options. Despite modest advancements in surgical
procedures, radiation and chemotherapy, median survival from diagnosis is only around 14 months. Upon recurrence, few
effective treatment options exist. Bevacizumab, a humanized monoclonal antibody that inhibits VEGF-A, received
accelerated FDA approval in May 2009 for use in recurrent GBM and quickly became the standard of care for recurrent
GBM in the United States. Almost all patients receive bevacizumab at some point in their treatment. Because bevacizumab
plays such an important role in the management of GBM, the development of imaging biomarkers to improve risk
stratification and predict patient benefit is highly desired. Such a biomarker would be clinically useful for identifying
patients that will benefit from bevacizumab as well as scientifically useful for cohort enrichment in the next phase of
combination therapies or exploratory studies aimed at high-risk patients, where conventional therapies like bevacizumab
are likely to fail.
Extensive preliminary data (>7 trials in >400 patients) suggests diffusion MRI characteristics are a strong, independent
predictor of anti-VEGF therapeutic efficacy in recurrent GBM, with patients exhibiting a significant survival benefit if they
present with a high apparent diffusion coefficient (ADC) within contrast enhancing tumor. Data also suggests these diffusion
MR signatures may result from an elevated expression of decorin (DCN), a glycoprotein with a variety of functions. We
hypothesize that the survival advantage and imaging signatures arise from the multifaceted functions of DCN, which include
anti-angiogenic characteristics and softening of the extracellular matrix, which we theorize would result in increased
effectiveness of anti-VEGF therapies and an increase in ADC.
The current study will explore the causal, mechanistic links between DCN expression, diffusion MRI, and anti-VEGF
treatment efficacy. First, Aim 1 will involve a deep exploration into the association between diffusion MR phenotypes and
DCN expression in human GBM using image-guided biopsies and examining DCN protein expression using
immunohistochemistry and gene expression using in-situ hybridization. The relationship between diffusion MRI, DCN
expression, and corresponding genotypes using whole exome analysis, genetic subtypes using bulk RNA sequencing,
cellular states using single-cell RNA sequencing, and blood plasma levels of circulating DCN will also be performed.
Concurrently, Aim 2 will establish the causal, mechanistic links between DCN expression, diffusion MRI measurements,
and anti-VEGF treatment in GBM through ca complex, genetically modified patient-derived orthotopic xenograft (PDX)
preclinical trial. To accomplish this, a series of patient-derived cell lines will be edited to silence of overexpress DCN within
PDX models using a tetracycline-controlled gene expression system. The direct role of DCN expression in changing
diffusion MRI measurements and increasing survival following anti-VEGF therapy by turning on or off DCN expression
using doxycycline will be determined.