Microvascular Imaging in Cancer & Response Evaluation
Tuesday 21 April 2009
Room 316A 16:00-18:00


Jeffrey L. Evelhoch and Martin O. Leach

16:00  308. Young Investigator Award Finalist:  Use of Cardiac Output to Improve Measurement of Tracer Input Function in Dynamic Contrast-Enhanced MRI
    Jeff Lei Zhang1, Henry Rusinek1, Louisa Bokacheva1, Qun Chen1, Pippa Storey1, Vivian S. Lee1
Radiology, New York University, New York, NY , USA
    We present a new method for computing AIF from MR arterial signals using a constrained conversion that takes into account the subject¡¯s cardiac output. Monte Carlo simulations showed that by using the proposed method, the reproducibility of tumor perfusion parameters and renal function parameters was significantly improved (by at least a factor of three). Dynamic MR renography was repeated for volunteers on three separate days. We obtained similar results. The proposed method may be especially useful for analyzing repeated contrast-enhanced scans such as monitoring tumor therapy or ACE-inhibitor renography.
16:20 309. Noninvasive Multimodality Imaging of the Tumor Microenvironment: Registered Dynamic 1H MRI and 18F PET Studies of a Preclinical Model of Tumor Hypoxia
    HyungJoon Cho1, Ellen Ackerstaff1, Sean Carlin1, Mihaela Lupu1, Ya Wang1, Asif Rizwan1, Joseph O'Donoghue1, Clifton Ling1, John Humm1, Pat Zanzonico1, Jason Koutcher1
Memorial Sloan Kettering Cancer Center, New York, NY, USA
    In vivo knowledge of the spatial distribution of viable, necrotic and hypoxic areas can provide prognostic information about the risk of developing metastases, the distribution of radiation sensitivity, and may possibly be used for localized dose escalation in radiation treatment. In this study, multimodality in vivo Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) imaging using stereotactic fiduciary markers in the Dunning R3327-AT prostate tumor was performed, focusing on the relationship between Dynamic Contrast-Enhanced (DCE)-MRI using Magnevist® (Gd-DTPA), and dynamic 18F-fluoromisonidazole (18F-Fmiso) PET. The non-invasive measurements were verified using tumor tissue sections stained for haematoxylin/eosin (H&E) and pimonidazole.
16:32 310. Distinct Molecular Profiles Characterize Hypoxic Breast Tumor Regions Detected by Combined MRSI, Optical Imaging, and Imaging Mass Spectrometry
    Erika R. Amstalden1, Tiffany R. Greenwood2, Zaver M. Bhujwalla2, Venu Raman2, Ronald M. A. Heeren1, Kristine Glunde2
FOM Institute for Atomic and Molecular Physics, Amsterdam, Netherlands; 2JHU ICMIC Program, Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA
    Breast tumors are characterized by spatially distinct hypoxic regions, which lead to radio- and chemoresistance and poor clinical outcome. Our goal is to identify to date unknown metabolites, small molecules, and proteins that, as a result of hypoxia-driven signaling pathways, have a high concentration in hypoxic regions. We combined 3D MRSI detection of metabolites with optical detection of hypoxic regions using hypoxia-driven fluorescence reporter and Imaging Mass Spectrometry (IMS) detection of undiscovered biomolecules in breast tumor models. Necrotic/hypoxic regions showed a markedly different molecular-metabolic profile compared to well-vascularized regions, which can provide molecular insights and differentiate these regions.
16:46 311. 3D Radial Multi Gradient Echo Dynamic MRI for Characterization of Microvasculature in Tumor Models Subjected to Respiratory Motion
    Julien Vautier1,2, Melanie Heilmann3, Christine Walczak1,2, Joël Mispelter1,2, Andreas Volk1,2
U759, INSERM, Orsay, France; 2Research Center, Institut Curie, Orsay, France; 3Computer Assisted Clinical Medicine, University Medicine, Mannheim, Germany

Preclinical dynamic MRI to characterize microvasculature in tumor models is typically performed with gradient- or spin-echo sequences and Cartesian k-space sampling which is not optimal for tumors subjected to respiratory motion. Radial k-space sampling, rather insensitive to motion, should be more appropriate. In this study, we have developed and validated a 3D radial multi gradient echo sequence. It allows for simultaneous measurement of T1 and T2* upon contrast agent administration at 1min30 time resolution. A first in vivo study comparing 3D Ktrans maps for two contrast agents of different molecular weights was performed on motion animated subcutaneous tumors in mice.

16:58 312. High Resolution MR Angiography of Tumors Using Iron-Oxide Contrast Agent
    Dmitri Artemov1, Yoshinori Kato1
JHU ICMIC Program, Dept. of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, USA
    Many forms of anticancer therapy target tumor vasculature and monitoring their effect with noninvasive imaging is one of the most important problems in cancer imaging. MR angiography (MRA) allows direct visualization of the tumor vasculature however its spatial resolution is limited by the method and only relatively large mature blood vessels can be imaged using current techniques. We propose to use SPIO-based blood pool contrast agents for susceptibility contrast enhanced MRA. These agents remain intravascular in the leaky tumor vasculature and provide improved visualization of small blood vessels due to the ’blooming effect’ in T2* weighted MR images.
17:10 313. Early DCE-MRI Findings Predict Tumor Volume Changes
    Edward Ashton1
R&D, VirtualScopics, Inc., Webster, NY, USA
    This study examines the relationship between early DCE-MRI findings and later changes in tumor burden as measured by structural CT or MRI for cancer patients undergoing anti-vascular or anti-angiogenic therapy. In particular, it addresses the question of whether reductions in blood flow and vascular permeability observed within the first two weeks of treatment are predictive of changes in tumor burden after 8 weeks. The data set includes 164 patients drawn from 13 clinical trials using a largely uniform data acquisition and analysis protocol. Results show a probability of tumor reduction of 70% in DCE-MRI responders vs. 26% for DCE-MRI non-responders.
17:22 314. Serial Assessment of Perfusion Parameters in Patients with GBM Following Anti-Angiogenic Therapy
    Emma Essock-Burns1,2, Janine M. Lupo1, Susan M. Chang3, Soonmee Cha1,3, Sarah J. Nelson1,2
Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, USA; 2UCSF/UCB Joint Graduate Group in Bioengineering, University of California, San Francisco, San Francisco, CA, USA; 3Department of Neurological Surgery, University of California, San Francisco, San Francisco, CA, USA
    Perfusion parameters, derived from T1-weighted dynamic contrast enhanced imaging and arterial spin labeling, of tumor and normal brain tissue in patients with GBM were tracked for 6 months after receiving anti-angiogenic therapy. Kps, fBV, and CBF levels in the tumor region were seen to decrease significantly 6 months after treatment with noticeable changes within individual patients by 2 months after treatment. The results from this study suggest that perfusion imaging is a useful tool for assessing the effects of this therapy and possibly assisting early prediction of progression for patients with GBM.
17:34 315. Early Detection of Response to Antiangiogenic Therapy in Metastatic Clear-Cell Renal Cell Carcinoma with ASL MRI
    Ivan Pedrosa1, Philip Robson1, Rupal Bhatt2, David McDermott2, Michael B. Atkins2, David Alsop1
Department of Radiology, Beth Israel Deaconess Medical Center, Boston, MA, USA; 2Department of Medicine, Division Hematology-Oncology, Beth Israel Deaconess Medical Center, Boston, MA, USA
    Six patients with metastatic clear cell renal cell carcinoma (RCC) underwent arterial spin labeling (ASL) magnetic resonance imaging (MRI) before and 1 week after initiation of therapy with sorafenib and bevacizumab. Mean tumor perfusion at 1 week into therapy (80 ± 64 ml/100 g/min) was significantly lower than the tumor perfusion at baseline (176 ± 107 ml/100 g/min)(p=0.02, paired t-test) whereas there was no significant decrease in measured tumor size at 1 week (p=0.06, paired t-test). ASL MRI detects changes in the vascularity of RCC metastasis as soon as 1 week into therapy with sorafenib and bevacizumab.
17:46 316. DCE-MRI Summary and Heterogeneity Statistics Predict Response to Combined Chemo- And Anti-VEGF Therapy
    Chris James Rose1, James Patrick O'Connor1, Yvonne Watson1, Caleb Roberts1, Giovannni A. Buonaccorsi1, Susan Cheung1, Brandon Whitcher2, Gordon Jayson3, Alan Jackson1, Geoffrey J. Parker1
Imaging Science and Biomedical Engineering, School of Cancer and Imaging Sciences, The University of Manchester, Manchester, UK; 2MRI Modelling, Clinical Imaging Centre, GlaxoSmithKline, London, UK; 3Cancer Research UK Dept. of Medical Oncology, The Christie, Manchester, UK
    In the context of clinical trials of anti-cancer therapies, there is limited evidence that biomarkers derived from DCE-MRI and tracer kinetic modelling have predictive value. It has been speculated that intratumoural spatial heterogeneity—neglected by conventional analyses—may carry useful information. We applied spatial heterogeneity analysis to a clinical trial of combined chemo- and anti-VEGF therapy. Multiple regression and correlation analyses revealed a statistically significant (p ≤ 0.001) correlation between median Ktrans and spatial heterogeneity of ve maps—measured before treatment at baseline—and reduction in tumour size measured after five two week cycles, indicating that treatment response can be predicted at baseline.