Magnetic Resonance Imaging for Prediction and Assessment of Treatment Response in Bevacizumab-Treated Recurrent Glioblastoma

DSpace/Manakin Repository

Magnetic Resonance Imaging for Prediction and Assessment of Treatment Response in Bevacizumab-Treated Recurrent Glioblastoma

Citable link to this page

 

 
Title: Magnetic Resonance Imaging for Prediction and Assessment of Treatment Response in Bevacizumab-Treated Recurrent Glioblastoma
Author: Rahman, Rifaquat M ORCID  0000-0002-9443-1797
Citation: Rahman, Rifaquat M. 2014. Magnetic Resonance Imaging for Prediction and Assessment of Treatment Response in Bevacizumab-Treated Recurrent Glioblastoma. Doctoral dissertation, Harvard Medical School.
Full Text & Related Files:
Abstract: Glioblastoma is the most common primary brain tumor in adults, and it is associated with a dismal prognosis with a median survival of 15 months. Despite treatment with chemotherapy, radiation therapy and surgery, patients inevitably have disease recurrence. Bevacizumab is a monoclonal humanized antibody that inhibits vascular endothelial growth factor signaling, and it has been shown to be effective in recurrent glioblastoma with respect to prolonging progression-free survival (PFS). The use of bevacizumab and other anti-angiogenic agents in recurrent glioblastoma have created novel challenges in interpreting magnetic resonance imaging (MRI) of patients. Furthermore, since only some patients appear to have a durable benefit from bevacizumab, there is a need for imaging biomarkers that can reliably identify this subgroup of patients.

Partly due to the challenges created by anti-angiogenic agents, the Response Assessment in Neuro-Oncology (RANO) was proposed to address some of the limitations with traditional response assessment criteria. In the first part of this project, we attempted to validate the RANO criteria by performing a comparative analysis of the RANO criteria vs. the Macdonald criteria using imaging from the phase II BRAIN trial. As we hypothesized, the RANO criteria yielded a significantly decreased PFS by identifying a subset of patients who had progression of nonenhancing tumor evident on T2-weighted imaging. Additionally, response and progression as defined by the RANO criteria correlated with subsequent overall survival (OS) in landmark analyses. While this supports the implementation of RANO criteria for response assessment in glioma clinical trials, future research will be necessary to further improve response assessment by incorporating advanced techniques such as volumetric anatomic assessment, perfusion-weighted MR (PWI-MR), diffusion-weighted MR (DWI-MR), MR spectroscopy (MRS) and positron emission tomography (PET).

Advanced imaging techniques are becoming increasingly recognized for their ability to provide objective, non-invasive assessment of treatment response but also to serve as predictive and prognostic biomarkers allowing for stratification of patient subgroups with better treatment outcome. In the second part of the project, we attempted to perform volumetric analysis of tumor size based on conventional MRI, as well as a histogram analysis of apparent diffusion coefficients (ADC) derived from diffusion-weighted MRI, to evaluate imaging parameters as predictors for PFS and OS in a single institution database of recurrent glioblastoma patients initiated on bevacizumab. Volumetric percentage change and absolute early post-treatment volume (3-6 weeks after initiation) of enhancing tumor can stratify survival for patients with recurrent glioblastoma receiving bevacizumab therapy. ADC histogram analysis using a multi-component curve-fitting technique within both enhancing and nonenhancing components of tumor prior to the initiation of bevacizumab can also be used to stratify OS in recurrent glioblastoma patients. While prospective studies are necessary to validate findings, future studies will increasingly incorporate multiparametric approaches to elucidate biomarkers that combine the value of conventional MRI with advanced techniques such as DWI-MR, PWI-MR, MRS and PET to obtain better predictors for PFS and OS in recurrent glioblastoma.
Terms of Use: This article is made available under the terms and conditions applicable to Other Posted Material, as set forth at http://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA
Citable link to this page: http://nrs.harvard.edu/urn-3:HUL.InstRepos:12407623
Downloads of this work:

Show full Dublin Core record

This item appears in the following Collection(s)

 
 

Search DASH


Advanced Search
 
 

Submitters