Brain Tumor: Imaging Brain Tumor Treatment & Response
Tuesday 21 April 2009
Room 311 13:30-15:30


Alberto Bizzi and Meng Law

13:30  279. 2008 ISMRM Grant Recipient:  Systematic Evaluation of Stereotatic Radiosurgery Effects in Metastasis and Acoustic Neurinomas Using MRI
    Cristian Tejos1,2, M.E. Andia2,3, P. Besa4, J. Lorenzoni5, A. Vieria4, L. Meneses2,6, P. Irarrazaval1,2
Department of Electrical Engineering, Pontificia Universidad Catolica de Chile, Santiago, Chile; 2Biomedical Imaging Center, Pontificia Universidad Catolica de Chile, Santiago, Chile; 3Radiology Department, Pontificia Universidad Catolica de Chile, Santiago, Chile; 4Radiotherapy Department, Cancer Center, Pontificia Universidad Catolica de Chile, Santiago, Chile; 5Departamento de Neurocirugia, Pontificia Universidad Catolica de Chile, Santiago, Chile; 6Radiology Department, Pontificia Universidad Catolica de Chile , Santiago, Chile
13:50 280. Apparent Diffusion Coefficient Measures on MR Correlate with Survival in Glioblastoma Multiforme
    Gerard Thompson1, John Robert Cain1, Samantha Jane Mills1, Alan Jackson1
Imaging Science and Biomedical Engineering, School of Cancer and Imaging Sciences, University of Manchester, Manchester, UK
    Described in this work is a method for quantifying the change in apparent diffusion coefficient (ADC) which occurs across tissue boundaries in glioblastoma multiforme on diffusion-weighted imaging. The gradient of the change in ADC moving from peri-tumoural oedema into solid, enhancing tumour was found to correlate with length of survival (n=18; Kendall’s tau -0.401, p0.021; Cox’s Hazards Ratio 1.007 (1.001-1.014) p=0.032), whereas the ADC gradient measured from normal appearing white matter into peri-tumoural oedema did not (n=19; Kendall’s τ 0.164, p0.327; Cox’s Hazards Ratio 0.998 (0.998-1.008) p=0.695).
14:02 281.

A Fully Automated Method for Predicting Glioma Patient Outcome from DSC Imaging; A Second Reference to Histopathology?

    Kyrre E. Emblem1,2, Frank G. Zoellner3, Atle Bjornerud1,4
Department of Medical Physics, Rikshospitalet University Hospital, Oslo, Norway; 2The Interventional Centre, Rikshospitalet University Hospital, Oslo, Norway; 3Department of Assisted Clinical Medicine, University of Heidelberg, Mannheim, Germany; 4Department of Physics, University of Oslo, Oslo, Norway
    We have assessed whether a fully automated, multi-parametric model for predicting outcome in glioma patients from dynamic susceptibility contrast MR imaging can be used as a second reference to pathologic findings. Based on automatically segmented tumor regions, 3D scatter diagrams of cerebral blood volume as a function of Ktrans were derived for each patient. A predictive model based on support vector machines was used to predict outcome in each patient using scatter diagrams and survival status of the remaining patients. Our results suggest that the proposed approach provides similar diagnostic accuracy values to histopathology when predicting patient outcome.
14:14 282. Acute Effects of Bevacizumab on Glioblastoma Vascularity Assessed with DCE-MRI and Relation to Patient Survival
    Weiting Zhang1, Teri N. Kreisl1, Jeff Solomon2, Richard C. Reynolds3, Danial R. Glen3, Robert W. Cox3, Howard A. Fine1, John A. Butman4
Neuro-Oncology Branch, National Cancer Institute, NIH, Bethesda, MD, USA; 2Medical Numerics, Inc., Germantown, MD, USA; 3National Institute of Mental Health, NIH, USA; 4Radiology and Imaging Sciences, Clinial Center, NIH, USA
    DCE-MRI was used to monitor the acute effects of Bevacizumab on physiologic measures of tumor vascularity, such as blood brain barrier permeability, represented as Ktrans. In addition, we relate these to progression free survival (PFS) and to overall survival (OS). Bevacizumab dramatically reduces Ktrans (46%) and enhancing tumor volume (39%) in recurrent GBMs within 96-hour of a single dose. However, tumors in which Bevacizumab resulted in larger decreases in Ktrans did not demonstrate an improved survival over those with smaller reductions in Ktrans. GBMs with larger baseline enhancing tumor volume and greater baseline Ktrans both predicted poorer PFS and OS.
14:26 283. A Composite Model of the Parametric Response Map Predicts Survival Independent of Radiographic Response in Patients with High Grade Glioma
    Craig J. Galban1, Thomas L. Chenevert1, Daniel A. Hamstra2, Charles R. Meyer1, Pia Sundgren1, Christina Tsien2, Theodore S. Lawrence2, Alnawaz Rehemtulla2, Timothy D. Johnson3, Brian D. Ross1
Radiology, University of Michigan, Ann Arbor, MI, USA; 2Radiation Oncology, University of Michigan, Ann Arbor, MI, USA; 3Biostatistics, University of Michigan, Ann Arbor, MI, USA
    A parametric response map composite model of the apparent diffusion coefficient (PRMADC) and relative cerebral blood flow (PRMrCBF) is predictive of treatment response in glioma patients independent of radiographic response (RR). Perfusion and diffusion MRI were performed on 44 patients pre and post-treatment. Survival analysis was employed on the population by stratifying based on RR, PRMADC, PRMrCBF and a composite of PRMADC and PRMrCBF (PRMADC-rCBF). Individually, PRM was closely associated to 10 week RR. A multivariate analysis showed a stronger dependence on PRM, most notably PRMADC-rCBF, than RR. This study provides a more accurate treatment response metric for cancer patients.
14:38 284. R2* Response of Brain Tumors to Hyperoxic and Hypercapnic Respiratory Challenges at 3 Tesla
    Andreas Müller1, Stefanie Remmele2, Ingo Wenningmann3, Frank Träber1, Roy König1, Hans Clusmann4, Hannes Dahnke5, Juergen Gieseke5, Sebastian Flacke1, Winfried A. Willinek1, Hans H. Schild1, Petra Mürtz1
Department of Radiology, University Hospital Bonn, Bonn, Germany; 2Philips Research Europe, Hamburg, Germany; 3Department of Anesthesiology, University Hospital Bonn, Bonn, Germany; 4Department of Neurosurgery, University Hospital Bonn, Bonn, Germany; 5Philips Medical Systems, Hamburg, Germany
    The R2* response to hyperoxic and hypercapnic respiratory challenges is of major diagnostic interest in oncology as it gives insight into tissue oxygenation and vasoreactivity. We determined the R2* response to elevated levels of O2 and CO2 in cerebral tumors at 3 Tesla. Different tumor regions show characteristically altered vessel function and reactivity. We present a robust and sensitive method to assess oxygenation changes and vasoreactivity in brain tumors. This additional information may support the selection and dosage of tumor therapies and may allow the non-invasive contrast-agent-free monitoring of treatment response.
14:50 285. Functional Diffusion Maps Applied to FLAIR Abnormal Areas Are Valuable for the Clinical Monitoring of Recurrent Brain Tumors
    Benjamin M. Ellingson1,2, Mark G. Malkin1,3, Scott D. Rand1,2, Devyani P. Bedekar1,2, Kathleen M. Schmainda1,2
Translational Brain Tumor Program, Medical College of Wisconsin, Milwaukee, WI, USA; 2Department of Radiology, Medical College of Wisconsin, Milwaukee, WI, USA; 3Departments of Neurology and Neurosurgery, Medical College of Wisconsin, Milwaukee, WI, USA
    Functional diffusion maps (fDMs) involve calculating voxel-by-voxel changes in diffusivity in order to spatially localize and quantify changes in tumor cellularity. Traditionally, fDMs are applied to contrast-enhanced tumor regions exclusively. In the current study we have expanded the application of fDMs to FLAIR abnormal regions. Results suggest this technique provides additional insight into the growth and treatment response of both enhancing and non-enhancing brain tumors, along with reflecting the neurological status of the patient.
15:02 286. Assessment of Anti-Angiogenic Treatment in Glioblastoma Using Arterial Spin-Labeling and Dynamic Susceptibility Contrast Perfusion MRI in a Phase II Trial
    Mei-Yun Wang1, Wei-Ting Zhang1, Poe-zhou chen1, Thomas Benner1, Divya S. Bolar1, Tracy T. Batchelor2, Rakesh K. Jain3, A. Gregory Sorensen1
Radiology, Massachusetts General Hospital, Boston, MA, USA; 2Neurology, Massachusetts General Hospital, Boston, MA, USA; 3Radiation Oncology, Massachusetts General Hospital, Boston, MA, USA
    Angiogenesis is an essential step in the growth and spread of solid tumor, and advances in MRI now permit detection of the hemodynamic changes of glioblastoma after treatment. In the current study, arterial spin-labeling (ASL) and first-pass dynamic susceptibility contrast (DSC) perfusion MRI were used to assess the treatment effect of an anti-angiogenic agent AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, in a phase-II clinical trial. Our results suggest ASL is almost as sensitive as DSC in assessing rCBF changes of tumor and the tumor vascular normalization and reversal after anti-angiogenesis treatment.
15:14 287. Intravoxel Water Diffusion Heterogeneity of Human High-Grade Gliomas
    Thomas Kwee1, Craig Galban1, Marko Ivancevic1,2, Pia Sundgren1, Christina Tsien3, Larry Junck4, Benjamin Hoff1, Charles Meyer1, Brian Ross1, Thomas Chenevert1
Department of Radiology, University of Michigan, Ann Arbor, MI, USA; 2Philips Healthcare, Cleveland, OH, USA; 3Department of Radiation Oncology, University of Michigan, Ann Arbor, MI, USA; 4Department of Neurology, University of Michigan, Ann Arbor, MI, USA
    Diffusion-weighted signal decay of brain tissue is multi-exponential due to the presence of multiple intravoxel proton pools (IPPs) with different diffusion coefficients. This study investigated the intravoxel water diffusion heterogeneity (IDWH) of human high-grade gliomas (N=20), using the stretched-exponential model. IDWH was significantly higher in high-grade gliomas than in normal brain tissue, which potentially offers a new method for assessing tumor extent and evaluating therapeutic response. Correlation between tumor IDWH and overall tumor diffusion coefficient was strongly negative, suggesting that highly cellular tumors contain a lower number of distinct IPPs, while cystic/necrotic tumors contain a higher number of distinct IPPs.