Ahmed E. Fetit^1,2, Jan Novak^2,3, Simrandip K. Gill^2,3, Martin Wilson^2,3, Andrew C. Peet^2,3, and Theodoros N. Arvanitis^1,2
1Institute of Digital Healthcare, WMG, University of Warwick, Coventry, West Midlands, United Kingdom, ²Birmingham Children's Hospital NHS Foundation Trust, Birmingham, West Midlands, United Kingdom, ³University of Birmingham, Birmingham, West Midlands, United Kingdom

There has been an increasing interest in childhood brain tumour characterisation using non-invasive MR image analysis methods, such as texture analysis (TA) over the past decade. However, much of this work focused on diagnostic classification of tumour types. This raises the question: If textural features could capture powerful patterns that aid the diagnosis of tumours, can they also be used to predict patients’ survival prognosis? Following diagnosis, determination of prognosis is an important step in tumour management, with implications that determine treatment options. In this regard, the primary aim of this study was to determine whether three-dimensional TA of conventional MR images could predict the survival of paediatric medulloblastoma – the most common malignant brain tumour occurring in childhood.

Benedikt Feuerecker¹, Markus Durst², Dieter Saur³, Marion I Menzel⁴, Markus Schwaiger¹, and Franz Schilling^1
1Nuclear Medicine, Technische Universität München, Munich, Bavaria, Germany, ²GE Global Research, Munich, Germany, ³Internal Medicine, Technische Universität München, Munich, Bavaria, Germany, ⁴GE Global Research, Garching, Bavaria, Germany

Lisa J Wilmes¹, Wei-Ching Lo¹, Wen Li¹, David C Newitt¹, Suchandrima Banerjee², Evelyn Proctor¹, Emine U Saritas³, Ajit Shankaranarayanan², and Nola M Hylton^1
1University of California San Francisco, San Francisco, CA, United States, ²GE Healthcare, Menlo Park, CA, United States, ³Bilkent University, Ankara, Turkey

This work measured tumor ADC using a high resolution reduced field of view diffusion weighted imaging (HR-DWI) technique and investigated the resultant tumor apparent diffusion coefficient (ADC) metrics as predictors of pathologic complete response (pCR) in patients with locally advanced breast cancer undergoing neoadjuvant chemotherapy. For early percent change in tumor ADC a trend of increasing AUC with decreasing percentile ADC was observed. Additionally, at the early treatment time point the AUCs for the lower percentile tumor ADC were higher than for the early tumor volume change.

Jorge E Jimenez¹, Leah C Henze Bancroft¹, Roberta M Strigel^1,2, Kevin M Johnson¹, Scott B Reeder^2,3, and Walter F Block^1,3
1Department of Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, ²Department of Radiology, University of Wisconsin School of Medicine and Public health, Madison, WI, United States, ³Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States

Successful MR methods must combine capabilities for rapid acquisition, reliable fat suppression, and significant data undersampling. We present our experience in making three MR technologies mutually compatible: IDEAL signal decomposition, L1-based compressed sensing (CS) reconstruction, and a 3D radial trajectory; Vastly undersampled Isotropic Projection (VIPR). We successfully demonstrate incorporation of IDEAL and CS to 3D T1-Weighted VIPR breast MRI capable of providing high isotropic resolution. The addition of CS markedly improved image quality and SNR. Further study is necessary to determine the lower limit of the temporal footprint the method will support.

Maria A Schmidt¹, Eva Kousi¹, Araminta Ledger¹, Erica Scurr², Cheryl Richardson², Georgina Hopkinson², Elizabeth O'FLynn¹, Steven Allen², Romney Pope², Robin Wilson², and M Leach^1
1CR-UK and EPSRC Cancer Imaging Centre, Royal Marsden NHS Foundation Trust and Institute of Cancer Research, Sutton, Surrey, United Kingdom, ²Department of Radiology, Royal Marsden NHS Foundation Trust, Chelsea, London, United Kingdom

Breast Dynamic Contrast-Enhanced (DCE) examinations are usually performed with fat suppression, providing qualitative enhancement curves. In contrast, pharmacokinetic modelling often uses spoiled gradient-echo images and a proton density weighted image as a reference to calculate T1 values. It would be desirable to calculate T1 values from fat suppressed clinical DCE examinations to quantify contrast-agent uptake in longitudinal studies and to assess parenchymal enhancement. In this work we introduce a reference image of low flip angle and a calibration process, and evaluate the accuracy of T1 values thus obtained. T1 measurements are viable in fat suppressed breast DCE.

Ella F Jones¹, Natalie Hartman¹, Helen Park¹, Ania Azziz¹, David C Newitt¹, John Kornak², Catherine Kilfa¹, Bonnie N Joe¹, and Nola M Hylton^1
1Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, ²Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA, United States

This work presents a systematic quadrant-based analysis of background parenchymal enhancement (BPE) in 172 patients with BI-RADS 1 (normal) or 2 (benign). The goal is to investigate if regional variations in BPE will influence the corresponding qualitative BI-RADS BPE assessment.

Araminta EW Ledger¹, Maria A Schmidt¹, Marco Borri¹, Erica D Scurr², Julie Hughes², Alison Macdonald², Toni Wallace², Robin Wilson², and Martin O Leach^1
1CR-UK Cancer Imaging Centre, The Institute of Cancer Research and Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom, ²Radiology, The Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom

Percent-water (%Water) calculated from Dixon fat-water separation techniques can provide a volumetric measurement of breast density, an established risk factor for breast cancer. In this work, we calculate breast %Water in repeat volunteer datasets to evaluate measurement reproducibility from a high-resolution proton-density (PD) weighted two-point Dixon sequence, and assess the error arising at lower spatial resolution and with T1/T2 weighting. %Water measurements from high-resolution PD weighted sequences are found to be highly reproducible. Statistically significant differences in %Water measurement arise at lower spatial resolutions and with the introduction of T1 or T2 weighting, even with correction for fat/water signal differences.

Colleen Bailey¹, Sarah Vinnicombe², Eleftheria Panagiotaki¹, Shelley A Waugh², John H Hipwell¹, Patsy Whelehan², Sarah E Pinder³, Andrew Evans², Daniel C Alexander¹, and David J Hawkes^1
1Centre for Medical Image Computing, University College London, London, United Kingdom, ²Dundee Cancer Centre, Ninewells Hospital and Medical School, Dundee, United Kingdom, ³Breast Research Pathology, Research Oncology, King's College London and Guy's Hospital, London, United Kingdom

We have examined regions of breast tumours using diffusion MRI and fitting to compartment models that characterize diffusion in the vascular, extracellular and intracellular spaces. These were compared with monoexponential apparent diffusion coefficient (ADC), intravoxel incoherent motion (IVIM) and tensor models. In the tumour center, diffusion ADC and IVIM best explained the data, while tumour rim was best characterized by a 3-compartment model including isotropic restricted diffusion for the intracellular component. This was in agreement with the low cellularity in the tumour center and higher cellularity in the rim observed on histology.

Martin D Pickles¹, Dan W Rettmann², Kang Wang³, and Lindsay W Turnbull^1
1Centre for Magnetic Resonance Investigations, Hull York Medical School at University of Hull, Hull, East Yorkshire, United Kingdom, ²Global MR Applications and Workflow, GE Healthcare, Rochester, MN, United States, ³Global MR Applications and Workflow, GE Healthcare, Madison, WI, United States

The aim of this work is to report initial clinical experience of DIfferential Subsampling with Cartesian Ordering (DISCO) dynamic breast examinations. In DISCO k-space is divided into central and outer portions. The outer portion of k-space is segmented into a number of equal distributions regions via a pseudo-random segmentation scheme. The central portion of k-space is sampled every temporal frame while the outer edges are subsampled sequentially. Temporal resolution is minimised by implementing view-sharing. The results of this work demonstrate that by utilising DISCO both high spatial and high temporal resolution dynamic sequences are possible.

Martin D Pickles¹, Daniel Litwiller², Ersin Bayram³, Lloyd Estkowski⁴, and Lindsay W Turnbull^1
1Centre for Magnetic Resonance Investigations, Hull York Medical School at University of Hull, Hull, East Yorkshire, United Kingdom, ²Global MR Applications and Workflow, GE Healthcare, Rochester, MN, United States, ³Global MR Applications and Workflow, GE Healthcare, Waukesha, WI, United States, ⁴Global MR Applications and Workflow, GE Healthcare, Menlo Park, CA, United States

The aim of this work is to develop a T2W flip angle modulation SSFSE protocol. In breast MRI, FSE images are frequently utilised to acquire T2W images. SSFSE represents an ultrafast sequence where efficiencies in scan time could be made, however, SSFSE images are associated with blurring. Refocusing flip angle modulation has recently been utilised with SSFSE, benefits include, faster acquisitions and improved sharpness by limiting T2 decay. In this work T2W fat nulled SSFSE images were acquired in 70 seconds and provide radiologists with the same T2 information available in longer FSE based sequences.

Anup Singh¹, Prativa Sahoo², Vedant Kabra³, Indrajit Saha², Meenakshi Singhal³, and Rakesh Kumar Gupta^3
1Center for Biomedical Engineering, Indian Institute of Technology Delhi, New Delhi, Delhi, India, ²Philips India Limited, Gurgaon, Haryana, India, ³Fortis Memorial Research Institute, Gurgaon, Haryana, India

Objective of the current study was to obtain T₁ mapping of in vivo human breast using conventional 3D T₁, T₂ and PD weighted images and comparing the results of with and without fat saturation. T₁ is an important parameter and its estimation is usually required during quantitative analysis, particularly during DCE-MRI data analysis. This simple approach, based upon conventionally acquired T₁, T₂ and PD weighted images, provided absolute quantitation of T₁ in in vivo breast tissue. In the current study, we have shown that three data point (PD, T₂ and T₁ W) based approach for T₁ estimation works well for both with and without fat saturation.

José Angel Rosado-Toro¹, Tomoe Barr², Marilyn T Marron³, Jean-Phillipe Galons⁴, Patricia Thompson³, Alison Stopeck³, Jeffrey Joel Rodríguez⁵, and María I Altbach^4
1Electrical and Computer Engineering, University of Arizona, Tucson, Arizona, United States, ²Biomedical Engineering, University of Arizona, Tucson, AZ, United States, ³Arizona Cancer Center, University of Arizona, Tucson, Arizona, United States, ⁴Medical Imaging, University of Arizona, Tucson, Arizona, United States, ⁵Electrical and Computer Engineering, University of Arizona, Tucson, Arizon, United States

A fully automated breast segmentation algorithm has been developed to segment the breast anatomy using various types of imaging pulse sequences. The segmentation first finds the chest and breast pixels using a clustering technique. Next it removes the chest pixels using a dynamic programming technique on the vertical gradient. Then it removes the skin pixels using a thinning algorithm and finally it splits the two breasts using a morphological technique. The performance of the algorithm is evaluated on 202 breast imaging slices using manually traced breast outlines as reference.

Jeon-Hor Chen^1,2, Hon J Yu¹, Yifan Li¹, Yoon Jung Choi³, Po Yun Huang⁴, and Min-Ying Su^1
1Center for Functional Onco-Imaging, University of California, Irvine, CA, United States, ²Department of Radiology, Eda Hospital and I-Shou University, Kaohsiung, Taiwan, ³Department of Radiology, Kangbuk Samsung Hospital, Seoul, Korea, ⁴Department of Medical Imaging, China Medical University, Taichung, Taiwan

This study investigated the association of 3D MR-based breast density with apparent diffusion coefficient (ADC) acquired from the fibroglandular tissue of the breast. MR images from the contralateral normal breast of 38 women diagnosed with unilateral breast cancer were retrospectively analyzed. The results from our study noted a positive correlation of ADC values with MR-measured percent breast density. This study proved that women with higher breast density had higher stromal matrix, which can be assessed using ADC.

Reem Bedair¹, Mary McLean², Andrew Patterson³, Roie Manavaki¹, John Griffiths², Fiona Gilbert¹, and Martin Graves^3
1University of Cambridge, Department of Radiology, Cambridge, Cambridgeshire, United Kingdom, ²Cancer Research UK Cambridge Research Institute, Cambridge, Cambridgeshire, United Kingdom, ³Department of Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, Cambridgeshire, United Kingdom

T10 mapping acquisitions of quantitative DCE-MRI is generally performed without fat suppression. Recent studies have used 2-point Dixon (2PD) reconstruction as a time-effective method for obtaining robust water-only images. This work compares the pre-contrast T10 relaxation times obtained using a conventional non-fat suppressed multiple flip angle technique with a fat-suppressed 2PD method in a cohort of patients with locally advanced breast cancer at 3.0T. Our results show an increase in the T1 measurement of fibroglandular tissue on fat suppression. However, the mean T10 values in tumours decrease. Our study demonstrates the feasibility of rapid fat-suppression techniques for quantitative DCE-MRI analysis.

Nina L. Purvis¹, Peter Gibbs², Martin D. Pickles², and Lindsay W. Turnbull^2
1Centre for MR Investigations, Hull York Medical School, Hull, East Yorkshire, United Kingdom, ²Centre for MR Investigations, University of Hull at HYMS, Hull, East Yorkshire, United Kingdom

An investigation to find an optimised clinical b-value protocol for IVIM imaging in breast lesions and its application. B-value schemes were generated using exponential and power-law spacing then the Cramer-Rao Lower Bound of IVIM parameters was calculated. A b-value scheme was chosen based on a figure of merit to balance the relative errors of the parameters. Data was fitted using mono and bi-exponential models. The RMSEs indicated that the biexponential fit was better. The results agree well with previously reported values. The b-value scheme samples low b-values well, and allows an acceptable amount of NEX for a short scan duration.

Nithin N Vajuvalli¹, C K Dharmendra Kumar¹, Manoj G Bhosale^1,2, and Sairam Geethanath^1
1Medical Imaging Research Centre, Dayananda Sagar College of Engineering, Bangalore, Karnataka, India, ²Government College of Engineering (COEP), Pune, Maharastra, India

DCE-MRI is used to assess tumor perfusion, microvascular vessel wall permeability and extravascular–extracellular volume fraction. Tofts Model (TM) in Time Domain (TD) is computationally intensive for 3D data to obtain Pharmacokinetic (PK) map using curve fitting. The current work involves acceleration of curve fitting process for the computation of PK map using Frequency Domain (FD) based TM. Current work was demonstrated on seven breast DCE datasets. Result show reduced computational time for estimating PK map and reduced NRMSE values in FD as compared to TD tofts model with respect to randomly generated ground truth and in vivo data.

Tse Chiang Chen¹ and Philip Beatty^1
1Medical Biophysics, University of Toronto, Toronto, ON, Canada

Rozhin Yousefi¹, Xiaoyong Huang², Stanley K. Liu², and Greg J. Stanisz^1,2
1Medical Biophysics, University of Toronto, Toronto, Ontario, Canada, ²Sunnybrook Research Institute, Toronto, Ontario, Canada

In recent years, the number of available treatment modalities for cancer has increased significantly while their effectiveness remains uncertain. Finding a non-invasive imaging method to provide early assessment of tumor response to therapy will significantly enhance treatment outcome. Our proposed method is to apply quantitative MRI (qMRI) to evaluate tumor micro-structure and metabolism during treatment using two MRI techniques: chemical exchange saturation transfer (CEST) and diffusion weighted MRI (DW-MRI). In vivo experimental results showed significant differences in qMRI measurements of tumors including apparent diffusion coefficient (ADC) and amine CEST peak, before and after treatment.

Ashley M Stokes¹, Charles P Hart², and C. Chad Quarles^1
1Institute of Imaging Science, Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States, ²Threshold Pharmaceuticals, California, United States

Tumor hypoxia leads to increased tumor aggressiveness and chemotherapeutic resistance, which has led to the development of hypoxia-activated cytotoxic prodrugs, such as TH-302. As tumor hypoxia is spatially heterogeneous, and thus response to TH-302 is expected to vary spatially, there is a need for imaging-based measures of treatment response. Here, we determined tumor response to TH-302 in two rat glioma models with known differences in tumor hypoxia, and functional diffusion mapping was used to quantify treatment-induced changes in diffusion characteristics that could be indicative of response to hypoxia activated drugs such as TH-302.

Rajiv Ramasawmy^1,2, Sean Peter Johnson^1,2, Thomas Anthony Roberts¹, Daniel J Stuckey¹, Anna L David³, Rosamund Barbara Pedley², Mark Francis Lythgoe†¹, Bernard Siow†¹, and Simon Walker-Samuel†^1
1Centre for Advanced Biomedical Imaging, University College London, London, Greater London, United Kingdom, ²Cancer Institute, University College London, London, Greater London, United Kingdom, ³Institute for Women's Health, University College London, London, Greater London, United Kingdom

Orthotopic tumor models are thought to provide a more clinically-representative model of disease than traditional subcutaneous implantations, although their siting often renders them more difficult to assess. In this study, we compared 1T “benchtop” MRI and ultrasound with our gold-standard techniques of bioluminescence imaging (for cell detection) and 9.4T MRI (for tumour volume assessment) in their ability to characterise the development of liver metastases over four weeks. No significant differences were observed in the measured tumour doubling, showing that each of these techniques can be used for characterising tumour growth in deep-sited tumours, although each has characteristic advantages and disadvantages.

Jessica K.R. Boult¹, Jin Li¹, Yann Jamin¹, Maria Vinci^2,3, Sergey Popov^2,3, Karen Barker⁴, Zai Ahmad⁴, Craig Cummings¹, Suzanne A Eccles³, Jeffrey C Bamber¹, Ralph Sinkus⁵, Louis Chesler⁴, Chris Jones^2,3, and Simon P Robinson^1
1Division of Radiotherapy and Imaging, The Institute of Cancer Research, London, United Kingdom, ²Division of Molecular Pathology, The Institute of Cancer Research, London, United Kingdom, ³CR-UK Division of Cancer Therapeutics, The Institute of Cancer Research, London, United Kingdom, ⁴Division of Clinical Studies, The Institute of Cancer Research, London, United Kingdom, ⁵Division of Imaging Sciences & Biomedical Engineering, Kings College London, London, United Kingdom

Refined imaging strategies that could improve diagnosis and management of children with brain malignancies are urgently required. MR elastography (MRE) has been used to assess viscoelastic properties in the brain and brain tumors clinically and preclinically. We evaluated orthotopic D-212 MG pediatric glioblastoma xenografts and GTML/Trp53^KI/KI transgenic medulloblastomas using MRE. Both tumor types demonstrated reduced elasticity (G_d) and viscosity (G_l) relative to the surrounding brain. A bimodal distribution of G_d, not seen previously, was observed in GTML/Trp53^KI/KI tumors. These data reinforce the potential of MRE for the detection and differential diagnosis of pediatric brain malignancies based on their mechanical properties.

Zhi Zuo^1,2, Tatiana Syrovets³, Felicitas Genze³, Alireza Abaei², Genshan Ma⁴, Thomas Simmet³, and Volker Rasche^1,2
1Internal Medicine II, University Hospital Ulm, Ulm, Baden-Wurttemberg, Germany, ²Core Facility Small Animal MRI, Medical Faculty, Ulm University, Ulm, Baden-Wurttemberg, Germany,³Institute of Pharmacology of Natural Products and Clinical Pharmacology, Ulm University, Ulm, Baden-W¨¹rttemberg, Germany, ⁴Department of Cardiology, Zhongda Hospital, Medical School of Southeast University, Nanjing, Jiangsu, China

High-resolution in ovo imaging is employed for monitoring the growth of human breast cancer cells grafted on the chorioallantonic membrane of chick embryos. High-resolution imaging is achieved by using an age-adopted precooling regime for immobilization of the chick embryo. The proposed scheme is proven safe for the chick embryo showing same survival rates as in a control group. The high-resolution MRI enables quantification of the tumor xenografts starting from d4 after grafting. A good correlation between tumor volumes derived by MRI and the weight of the tumor after extraction is shown.

Rheal A. Towner¹, Patricia Coutinho De Souza¹, Krithika Balasubramanian², Charity Njoku¹, Nataliya Smith¹, David L. Gillespie³, Andrea Schwager⁴, Osama Abdullah⁵, Kar-Ming Fung⁶, Debra Saunders¹, and Randy L. Jensen^3
1Advanced Magnetic Resonance Center, Oklahoma Medical Research Foundation, Oklahoma City, OK, United States, ²Radiology & Biomedical Imaging, University of California San Francisco, CA, United States, ³Huntsman Cancer Insitute, University of Utah Health Sciences Center, UT, United States, ⁴Neurobiology & Anatomy, University of Utah Health Sciences Center, UT, United States, ⁵Small Animal Core Facility, University of Utah, UT, United States, ⁶Pathology, University of Oklahoma Health Sciences Center, OK, United States

Gliomas are the most lethal adult primary brain tumors with a poor outcome. Here, we report the effects of OKN-007 on the necrotic tumor core and non-necrotic tumor parenchyma in the F98 rat glioma model assessed by 1H-MRSI, DWI, and histological analysis. Our results that OKN-007 was able to reduce necrosis and tumor cell proliferation. There was also an increase in apoptosis following OKN-007 treatment which seemed to correlate with spectroscopic lipid peak assessments. Our results also indicated that both ADC and spectroscopic choline measures are related to glioma cell density in the F98 rat glioma model.

Henk M. De Feyter¹, Kevin L. Behar², Kevan L. Ip¹, Fahmeed Hyder¹, Lester L. Drewes³, Robin A. de Graaf¹, and Douglas L. Rothman^1
1Department of Diagnostic Radiology, Yale University, New Haven, Connecticut, United States, ²Department of Psychiatry, Yale University, CT, United States, ³Department of Biomedical Sciences, University of Minnesota, MN, United States

The ketogenic diet (KD; fat, protein, no carbohydrates) creates a plasma nutrient profile similar to starvation: increased levels of ketone bodies and reduced plasma glucose levels, and has been proposed as metabolic therapy for brain tumors. Brain tumor cells supposedly cannot oxidize ketone bodies for energy metabolism in contrast to normal brain cells, and therefore the KD would result in starving of glucose-dependent brain tumors. We investigated the capability of glioma cells to oxidize beta-hydroxybutyrate, the most abundant ketone body, using ¹³C magnetic resonance spectroscopy. 9L and RG2 glioma cells were studied both in vitro and in vivo while administering [2,4-¹³C₂]- beta-hydroxybutyrate.

Lara Leoni¹, Martin Andrews², Chin-Tu Chen³, Barry Lai⁴, and Brian B. Roman^5
1University of Chicago, Chicago, Il, United States, ²University of Chicago, IL, United States, ³Radiology, University of Chicago, IL, United States, ⁴Argonne National Laboratory, IL, United States, ⁵radiology, university of chicago, Chicago, Illinois, United States

We have developed a MnMRI approach to detecting pancreatic cancer in a murine model of human disease. We utilized bioluminesence, MRI and XFM imaging to determine the uptake of Mn by developing tumors. T1 maps of the lesions indicated a decrease due to Mn accumulation.

Youping Xiao¹, Yunbin Chen¹, Jianji Pan², Ying Chen¹, Yiqi Yao¹, Xiang Zheng¹, Xiangyi Liu¹, Dechun Zheng¹, and Weibo Chen^3
1Radiology, Fujian Provincial Cancer Hospital, Fuzhou, Fujian, China, ²Radiation Oncology, Fujian Provincial Cancer Hospital, Fuzhou, Fujian, China, ³Philips Healthcare, Shanghai, China

The results of this present study shows that intravoxel incoherent montion diffusion weighted imaging(IVIM-DWI) derived parameters present a good producibility and feasibility on xenografts of human nasopharyngeal carcinoma(NPC) cell line CNE-2 which have a significant higher D(pure diffusion coefficient) and a lower f(fraction of perfusion) parameter. It is indicated that IVIM-DWI would be valuable in helping evaluate the micro-environment characteristic of diffusion and perfusion in NPC's xenograft, especially during the process of chemoradiotherapy.

Jing Guo¹, Simon Bayerl², Jürgen Braun³, Peter Vajkoczy², and Ingolf Sack^4
1Radiology, Charité - Universitätsmedizin Berlin, Berlin, Berlin, Germany, ²Department of Neurosurgery, Charité - Universitätsmedizin Berlin, Berlin, Germany, ³Department of Medical Informatics, Charité - Universitätsmedizin Berlin, Berlin, Germany, ⁴Radiology, Charité - Universitätsmedizin Berlin, Berlin, Germany

MR elastography (MRE) was performed to study the viscoelasticity of glioblastoma (GB) in the GL261 mouse model. We found that the magnitude of the complex shear modulus |G*| of the tumor was significantly reduced while the phase angle showed a large heterogeneity of values compared to healthy brain tissue. Our results corroborate clinical studies of MRE in GB and raise the prospect of assessing tumor malignancy by tissue mechanical constants.

Marie-France Penet¹, Balaji Krishnamachary¹, Flonné Wildes¹, Yelena Mironchik¹, Chien-Fu Hung², TC Wu², and Zaver M Bhujwalla^1
1JHU ICMIC Program, Division of Cancer Imaging Research, The Russell H Morgan Department of Radiology, The Johns Hopkins University School of Medicine, Baltimore, MD, United States,²Department of Pathology, The Johns Hopkins University School of Medicine, Baltimore, MD, United States

Epithelial ovarian cancer remains the leading cause of death from gynecologic malignancy among women in developed countries. Identifying mechanisms that drive the aggressiveness of ovarian cancers and its associated pathologies, such as the formation of metastases and the build-up of ascitic fluid, is urgently needed to provide new targets in effective control and treatment. Noninvasive magnetic resonance imaging (MRI) and magnetic resonance spectroscopic imaging (MRSI) provide opportunities to characterize the tumor microenvironment and to assess its relationship with ascites and metastases. We applied MRI and MRSI to better characterize ascites formation in a syngeneic orthotopic experimental model of ovarian cancer.

Devkumar Mustafi¹, Erica Markiewicz¹, Marta Zamora¹, Xiaobing Fan¹, Jeffrey Mueller², Suzanne D Conzen³, and Gregory S Karczmar^1
1Radiology, The University of Chicago, Chicago, IL, United States, ²Pathology, The University of Chicago, Chicago, IL, United States, ³Medicine, Section of Hematology and Oncology, The University of Chicago, Chicago, IL, United States

Precise MRI-histopathology correlation demonstrates that MRI accurately identifies mammary cancer at various stages of development in the widely used C3(1)SV40Tag mouse model, and provides a tool for the development of image-based markers that differentiate indolent from aggressive cancer. 96% of in situ and 100% of invasive cancers identified on in vivo MRI agreed with histology. Methods described here will allow investigators to develop better MRI-based markers for tumor progression, improve understanding of cancer initiation and progression, evaluate response to therapy in murine models of breast cancer, and provide valuable insights regarding clinical management of patients with early breast cancers.

Carlos J Perez-Torres¹, Liya Yuan², Robert E Schmidt³, Keith M Rich², Robert E Drzymala⁴, Joseph JH Ackerman^1,5, and Joel R Garbow^1
1Radiology, Washington University, Saint Louis, MO, United States, ²Neurosurgey, Washington University, Saint Louis, MO, United States, ³Neuropathology, Washington University, Saint Louis, MO, United States, ⁴Radiation Oncology, Washington University, Saint Louis, MO, United States, ⁵Chemistry, Washington University, Saint Louis, MO, United States

The use of anti-VEGF antibodies represents a new treatment approach for radiation necrosis. We present a timeline of VEGF expression and test the effectiveness and specificity of anti-VEGF antibody treatment in our mouse model of radiation necrosis. While the anti-VEGF antibody treatment has an effect on the extent of injury, it does not reduce VEGF expression. The persistence of VEGF could lead to the recurrence of injury once treatment is stopped.

Anna G Sorace^1,2, Stephanie L Barnes^1,2, Jennifer G Whisenant^1,2, Mary E Loveless¹, and Thomas E Yankeelov^1,2
1Radiology and Radiological Sciences, Vanderbilt University, Nashville, Tennessee, United States, ²Vanderbilt University Institute of Imaging Science, Vanderbilt University, Nashville, Tennessee, United States

This study evaluated the relationship between the apparent diffusion coefficient (ADC), the extravascular extracellular volume fraction (v_e), and quantitative histology measurements using two preclinical breast cancer models, BT474 and MDA-MB-231. Quantitative imaging biomarkers can reveal treatment response, however better validation of the techniques needs to occur prior to translation. While ADC reveals significant correlations with the extracellular space in both preclinical models, this data adds to the growing body of literature which suggests that v_e is not a reliable biomarker of extracellular space.

Marie-France Penet¹, Balaji Krishnamachary¹, Tariq Shah¹, Yelena Mironchik¹, Anirban Maitra², and Zaver M Bhujwalla^1
1JHU ICMIC Program, Division of Cancer Imaging Research, The Russell H Morgan Department of Radiology, The Johns Hopkins University School of Medicine, Baltimore, MD, United States,²MD Anderson Cancer Center, The University of Texas, TX, United States

There is a critical need for identifying and developing new noninvasive biomarkers and therapeutic targets in pancreatic cancer. We recently observed aberrant choline metabolism in subcutaneous and orthotopically implanted human pancreatic cancer xenografts using 1H MR spectroscopic imaging. However, high-resolution proton spectra of tumors and pancreatic tissue extracts normalized to the water signal assuming similar water content did not reflect the significantly increased total choline observed in vivo. Our purpose here was to determine the differences in water content between pancreatic tumors and the pancreas to accurately quantify differences in metabolism when using the water signal for normalization.

Daisuke Kokuryo¹, Eiji Yuba², Kenji Kono², Tsuneo Saga¹, and Ichio Aoki^1
1Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Chiba, Japan, ²Graduate School of Engineering, Osaka Prefecture University, Sakai, Osaka, Japan

A combination treatment using carbon-ion-beam irradiation and multimodal thermo-sensitive polymer-modified liposomes (MTPLs) containing contrast agents and anti-cancer drug was developed. The accumulation and treatment effects of the MTPLs were evaluated in in vivo experiments on subcutaneously xenografted tumor model mice. MTPLs accumulated in the tumor region regardless of the influence of carbon-beam irradiation and the efficacy of the combination treatment increased after heat-triggering of drug release. It was concluded that the proposed strategy provided an effective anti-cancer treatment.

Gaurav Sharma¹, Somenath Ghatak¹, Arun Kumar Verma², Thirumurthy Velpandian³, and Rama Jayasundar^1
1NMR, All India Institute of Medical Sciences, New Delhi, Delhi, India, ²Biotechnology, All India Institute of Medical Sciences, New Delhi, Delhi, India, ³Pharmacology, All India Institute of Medical Sciences, New Delhi, Delhi, India

Ivan Jambor¹, Marko Pesola¹, Harri Merisaari², Pekka Taimen³, Peter J Boström⁴, Timo Liimatainen⁵, and Hannu J Aronen^1
1Department of Diagnostic Radiology, University of Turku, Turku, Finland, ²Turku PET Centre, University of Turku, Turku, Finland, ³Department of Pathology, Turku University Hospital, Turku, Finland, ⁴Department of Urology, Turku University Hospital, Turku, Finland, ⁵Department of Biotechnology and Molecular Medicine, A.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio, Finland

Fifth-one patients with histologically confirmed PCa underwent 3T MRI consisting of relaxation along fictitious field (RAFF), diffusion weighted imaging (DWI), b valeus of 0, 100, 300, and 500 s/mm², and T₂ mapping. Using whole mount prostatectomy sections and anatomical T₂-weighted images as reference, one ROI was placed in the center of PCa area and the same sized ROI in the peripheral zone, and central gland not containing PCa. High spearman correlation coefficient value of -0.68 was found for correlation of RAFF relaxation values with Gleason score groups, outperforming DWI (ADC_m) and T₂ mapping.

Kamila U Szulc¹, Ade Oyefiade², Lily Riggs^1,2, Eric Bouffet^3,4, Suzanne Laughlin⁵, Brian W Timmons⁶, Jason P Lerch⁷, Cynthia B de Medeiros², Jovanka Skocic¹, and Donald J Mabbott^1,2
1Neurosciences and Mental Health, Hospital for Sick Children, Toronto, Ontario, Canada, ²Department of Psychology, Hospital for Sick Children, Toronto, Ontario, Canada, ³Division of Haematology/Oncology, Hospital for Sick Children, Toronto, Ontario, Canada, ⁴Department of Pediatrics, Hospital for Sick Children, Toronto, Ontario, Canada, ⁵Diagnostic Imaging, Hospital for Sick Children, Toronto, Ontario, Canada, ⁶Department of Pediatrics, McMaster University, Hamilton, Ontario, Canada, ⁷Mouse Imaging Centre, Hospital for Sick Children, Toronto, Ontario, Canada

Cranial radiation is a standard form of treatment for malignant brain tumors. While radiation increases survival rates, it also leads to long-term cognitive impairments and neurodegeneration. Unfortunately, there is no cure or standard of care for these treatment-related effects. Recently, there have been a growing number of studies showing the benefits of physical activity for the brains of healthy children, but its potential as a rehabilitative technique remains unknown. We conducted a 12-week program to examine whether aerobic exercise can stimulate brain repair processes in pediatric brain tumor survivors treated with cranial radiation. Specifically, we examined the effects of exercise on cortical thickness.

Hai-Ling Margaret Cheng^1,2, Tameshwar Ganesh², Reza Bayat Mokhtari³, Mosa Alhamami², and Herman Yeger^3
1Institute of Biomaterials & Biomedical Engineering, University of Toronto, Toronto, Ontario, Canada, ²Physiology & Experimental Medicine, Hospital for Sick Children, Toronto, Ontario, Canada, ³Developmental & Stem Cell Biology, Hospital for Sick Children, Toronto, Ontario, Canada

Gopal Varma¹, Xiaoen Wang¹, Han Xie², Gerburg Wulf³, Pankaj Seth², David C Alsop¹, Aaron K Grant¹, and Vikas P Sukhatme^2
1Radiology, Division of MR Research, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, United States, ²Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, United States, ³Division of Hematology and Oncology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, United States

Proton pump inhibitors (PPIs) hold potential for treatment of cancer treatment by disrupting the transport of excess hydrogen ions out of cancer cells that maintains a normal intracellular pH and forms an acidic extracellular environment. The effect of a PPI combination on preclinical models of breast and non-small cell lung cancer was studied using slice-selective 31P spectroscopy through the tumor up to 40 minutes post drug injection. Intracellular pH decreased significantly following treatment. In addition, a more significant increase in Pi/γATP ratio from pre- to post-injection periods was observed.

Jetse van Gorp¹, Peter Seevinck¹, Anna Andreychenko², Alexander Raaijmakers², Peter Luijten³, Miriam Koopman⁴, Vincent Boer³, and Dennis Klomp^3
1Image Sciences Institute, University Medical Center Utrecht, Utrecht, Utrecht, Netherlands, ²Department of Radiotherapy, University Medical Center Utrecht, Utrecht, Utrecht, Netherlands,³Department of Radiology, University Medical Center Utrecht, Utrecht, Utrecht, Netherlands, ⁴Department of Medical Oncology, University Medical Center Utrecht, Utrecht, Utrecht, Netherlands

In this work, the feasibility to detect orally administered chemotherapy (capecitabine) in the liver was investigated at a clinical 7T MR system. The system was equipped with a broadband radiative antenna to acquire both 1H and 19F signal, and dual-frequency excitation pulses were implemented to overcome excitation issues. Pulse acquire and 3D spectroscopic imaging 19F spectra were successfully acquired in two patients at 1 and 10 hours after drug intake. The results show that it is feasible to monitor chemotherapy metabolism at 7T in the human body.

Sarah Ann Mason¹, Nina Tunariu¹, Dow-Mu Koh¹, David J Collins¹, Martin O Leach¹, and Matthew D Blackledge^1
1Institute of Cancer Research and Royal Marsden Hospital, Sutton, Surrey, United Kingdom

No single MR sequence can fully represent the underlying biology in bone metastases, which necessitates that clinicians employ complementary image data for disease diagnoses, response assessments, and treatment decisions. The sheer volume of data can make image interpretation complex and overwhelming. We introduce a method for consolidating information by identifying like regions in the bone (e.g. active disease) based on a mean-shift analysis of fat fraction (FF), apparent diffusion coefficient (ADC), and spatial location. This non-parametric method provides superb data visualization, makes no assumptions about the underlying data distributions, and can track changes in the region of interest over time.

Eugene Kim¹, Astrid Jullumstrø Feuerherm^2,3, Berit Johansen^2,3, Olav Engebraaten⁴, Gunhild Mari Mælandsmo⁴, Tone Frost Bathen¹, and Siver Andreas Moestue^1
1MR Cancer Group, Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway, ²Department of Biology, Norwegian University of Science and Technology, Trondheim, Norway, ³Avexxin AS, Trondheim, Norway, ⁴Department of Tumor Biology, Institute for Cancer Research, Oslo University Hospital, Oslo, Norway

DCE-MRI and ex vivo micro-CT angiography were used to investigate the relationship between tumor vascular function and structure, and the effect of a cytosolic phospholipase A2 (cPLA2) inhibitor (AVX235, Avexxin AS) on this relationship, in a patient-derived breast cancer model. In control tumors (n=7), there were good correlations between Ktrans and vascular surface area (SA) (r=0.67) and between vp and SA (r=0.7). In AVX235-treated tumors (n=9), these correlations were weaker (r=0.25 and 0.33, respectively). This suggests that cPLA2 inhibition modulated the link between vascular structure and function and had a spatially heterogeneous effect on blood flow and/or vessel permeability.

Derek A White^1,2, Zhang Zhang³, Heling Zhou¹, Debu Saha³, Peter Peschke⁴, Zhongwei Zhang¹, and Ralph P Mason^5
1Radiology, University of Texas Southwestern, Dallas, Texas, United States, ²Bioengineering, University of Texas at Arlington, Texas, United States, ³Radiation Oncology, University of Texas Southwestern, Dallas, Texas, United States, ⁴Clinical Cooperation Unit Molecular Radiooncology, German Cancer Center, Heidelberg, Germany, ⁵Radiology, University of Texas Southwestern, Dallas, TX, United States

Non-invasive prognostic biomarkers promise new insights into tumor pathophysiology potentially allowing therapy to be optimized. Notably hypoxia influences radiation responses and Oxygen sensitive MRI (BOLD and TOLD) are sensitive to tissue oxygenation. This study further explores relationships between R1, R2* of rat prostate tumors with respect to oxygen breathing challenge and the tumor growth delay induced by a split dose radiation regimen. Oxygen breathing was found to enhance tumor growth delay and correlations were found with R1 and R2* assessed before the first dose of radiation.

Jacob Antunes¹, Satish Viswanath¹, Mirabela Rusu¹, Laia Valls², Norbert Avril², Christopher Hoimes², and Anant Madabhushi^1
1Center for Computational Imaging and Personalized Diagnostics, Case Western Reserve University, Cleveland, OH, United States, ²University Hospitals Case Medical Center, Cleveland, OH, United States

We present a framework for quantitatively evaluating early treatment response of renal cell carcinoma (RCC). A single RCC patient was imaged using an integrated T2W/PET and DWI sequence before and during cytostatic drug treatment. Sequences within an acquisition protocol were spatially aligned and co-registered between acquisition time points. SUV, ADC-map, and T2W textural feature intensities were extracted on a per-voxel basis. A weighted difference map combination of multiple PET/MRI parameters was computed to optimize for expected changes within annotated RCC and normal tissue regions. The integrated multi-parametric PET/MRI map demonstrated high specificity in identifying early treatment response in metastatic RCC.

Xiaoyu Jiang¹, Hua Li¹, Ping Zhao¹, H. Charles Manning¹, Junzhong Xu¹, and John C. Gore^1
1Institute of Imaging Science, vanderbilt university, nashville, Tennessee, United States

Amaresha Shridhar Konar¹, Nithin N Vajuvalli¹, Rashmi R Rao¹, Divya Jain¹, Dharmendra CK Kumar¹, and Sairam Geethanath^1
1Medical Imaging Research Center, Dayananda Sagar College of Engineering, Bangalore, Karnataka, India

Current work demonstrates a technique, Region Of Interest Compressed Sensing (ROICS) on DCE-MRI data where ROI was selected around the tumor, restricting the reconstruction to the ROI. Compressed Sensing (CS) and ROICS methods were applied on seven breast DCE-MRI data sets at chosen acceleration factors from 2x to 20x. The reconstructed images were used to obtain quantitative Pharmacokinetic maps (PK) and the error in reconstruction was quantified by Normalized Root Mean Square Error (NRMSE) value. A significant increase in NRMSE value and artifacts in PK maps are observed from 5x acceleration onwards for CS as compared to ROICS.

Christopher Hensley¹, Eunsook Jin^2,3, Naama Lev-Cohain⁴, Qing Yuan⁴, Kemp Kernstine⁵, Craig Malloy^6,7, Robert Lenkinski^6,7, and Ralph Deberardinis^8,9
1Children's Research Institute, University of Texas Southwetern Medical Center, Dallas, Texas, United States, ²Advanced Imaging Research Center, University of Texas Southwetern Medical Center, Texas, United States, ³Internal Medicine, University of Texas Southwetern Medical Center, Texas, United States, ⁴Radiology, University of Texas Southwetern Medical Center, Texas, United States, ⁵Cardiovascular and Thoracic Surgery, University of Texas Southwetern Medical Center, Texas, United States, ⁶Advanced Imaging Research Center, University of Texas Southwetern Medical Center at Dallas, Texas, United States, ⁷Radiology, University of Texas Southwetern Medical Center at Dallas, Texas, United States, ⁸Children's Research Institute, University of Texas Southwetern Medical Center at Dallas, Texas, United States, ⁹Pediatrics, University of Texas Southwetern Medical Center at Dallas, Texas, United States

Significant inter- and intratumoral heterogeneity in in vivo glucose metabolism as assayed by [U-13C] glucose infusions exist in non-small cell lung cancer (NSCLC) patient primary tumors. We believe microenvironment-based advanced imaging methods can be used as pre-operative markers for intraoperative tumor fragment sampling to begin to study the magnitude of non-cell autonomous regulation of lung tumor metabolism. We demonstrate this proof of concept with DCE-MRI to assay the effects of heterogeneity in tumoral perfusion on both oxidative and non-oxidative NSCLC patient tumor glucose metabolism.

Michel Sarraf^1,2, Flavien Caraguel¹, François Berger¹, Boudewijn Van Der Sanden¹, and Hana Lahrech^1
1CEA-CLINATEC, Grenoble, Isère, Rhône-Alpes, France, ²Saint Joseph University, Beyrouth, Lebanon

Seung-Cheol Lee¹, Jeffrey Roman¹, Kavindra Nath¹, David Nelson¹, Kevin Muriuki¹, Alexander Shestov¹, and Jerry Glickson^1
1Department of Radiology, University of Pennsylvania, Philadelphia, PA, United States

Time course 13C NMR study was performed in the perfused lymphoma and melanoma cells as well as in vivo xenografts for detailed metabolic flux calculation. mTOR signaling inhibitor rapamycin was administered to lymphoma cells and xenografts. Well resolved time course 13C NMR spectra were obtained from both perfused cells and in vivo tumors. mTOR signaling inhibition decreased fluxes to lactate, glutamate as well as glycogen. Melanoma xenografts exhibited higher TCA cycle flux than lymphoma xenografts. Quantitative flux calculation is under process.

Alexander A. Shestov¹, Anthony Mancuso², Pierre Gilles Henry³, Dennis B. Leeper⁴, and Jerry David Glickson^5
1Radiology, University of Pennsylvania, Philadelphia, PA, United States, ²Radiology, University of Pennsylvania, PA, United States, ³University of Minnesota, MN, United States, ⁴Radiation Oncology, Thomas Jefferson University, PA, United States, ⁵Radiology, Univesity of Pennsylvania, PA, United States

The bioreactor techniques combined with 13C MRS are an important tool to study cancer cell metabolism. Modeling of intracellular MRS isotopomer data obtained during perfusion with 13C labeled substrates allows quantitative determination of transport and metabolic parameters in vivo/in situ. In this work, we apply a novel 13C metabolic flux analysis technique to elucidate cancer metabolism bionetwork and calculate fluxes through important cancer metabolic pathways.

Dragana Savic¹, Youngho Seo¹, Randall Hawkins¹, Soonmee Cha¹, Miguel Pampaloni¹, Sharmila Majumdar¹, and Ramon Barajas^1
1Radiology and Biomedical Imaging, University of California, San Francisco (UCSF), San Francisco, California, United States

We present a non-invasive image-based quantification method for estimating the activity in patients using an investigational simultaneous TOF PET/MRI scanner. Patients were injected with 18F-fluormisonidazole, and the average activity was calculated from the whole blood samples, and compared to the activity from the PET/MRI scanner. The average MRI blood-to-blood ratios were 1.16 and 1.18 respectively for the carotid arteries and the jugular veins. These results suggest that we can do image-derived blood activity concentration calculation reproducibility, and potentially avoid invasive blood sample procedures, by the use of our image derived quantification method using a hybrid simultaneous TOF PET/MRI scanner.

Emily Decelle¹, Taylor Fuss¹, Shulin Wu¹, Adam Feldman², Douglas Dahl², Aria Olumi², W Scott McDougal², Chin-Lee Wu¹, and Leo L Cheng^3
1Pathology, Massachusetts General Hospital, Boston, MA, United States, ²Urology, Massachusetts General Hospital, Boston, MA, United States, ³Pathology and Radiology, Massachusetts General Hospital, Boston, MA, United States

Prostate cancer (PCa) is the second leading cause of cancer death and the most frequently diagnosed malignancy in men worldwide. Previous observations show that PCa metabolic information can delocalize from PCa glands and into histologically-benign tissue creating “field effects” resulting in “metabolomic lesions” that are larger than histology lesions. In this study, we evaluate these effects with location defined prostate biopsy samples from patients suspected of harboring PCa using high-resolution magic angle spinning MR spectroscopy at 14.1T with constructions of training and testing cohorts created through quantitative histopathology of the biopsy cores.

Noriko Mori¹, Flonné Wildes¹, Tomoyo Takagi¹, Kristine Glunde^1,2, and Zaver M. Bhujwalla^1,2
1The Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, United States, ²The Sidney Kimmel Comprehensive Cancer Center, The Johns Hopkins University School of Medicine, MD, United States

Abnormal phospholipid and lipid metabolism are characteristic features of cancer. Cancer cells in culture do not completely mirror observations made in vivo because of the strong influence of the tumor microenvironment. In our ongoing studies we are comparing phosphocholine/glycerophosphocholine and lipid levels in prostate and breast cancer cells and tumors using 1H MR spectroscopy, to further understand metabolic processes. We have compared protein levels of the related enzymes to the metabolisms in these cells and tumors. Significantly different protein levels observed between cells in culture and tumors demonstrate the importance of the tumor microenvironment in phospholipid and lipid metabolism.

Tom Peeters¹, Vincent Breukels¹, Corina van den Heuvel², Anna Navis², Sanne van Lith², Jack van Asten¹, Remco Molenaar³, William Leenders², and Arend Heerschap^1
1Department of Radiology and Nuclear Medicine, Radboudumc, Nijmegen, Netherlands, ²Department of Pathology, Radboudumc, Nijmegen, Netherlands, ³Department of Cell Biology and Histology, Academic Medical Center, Amsterdam, Netherlands

Mutational changes in cytosolic isocitrate dehydrogenase 1 (IDH1) result in production of NADP⁺ and the oncometabolite D-2HG at the expense of αKG and NADPH. Replenishment of αKG from glutamine is one of the compensatory anaplerotic mechanisms that allow tumor cells to survive the induced metabolic stress. We investigated the effect of epigallocatechin gallate (EGCG), a known inhibitor of glutamate dehydrogenase 1 (GDH1), on the fate of ¹³C-glutamine using ¹³C MRS in IDH1^wt/R132H and IDH1^wt/wt cancer cells. EGCG significantly inhibits proliferation of IDH1^wt/R132H cells. EGCG also prohibits the conversion of glutamine into D-2HG and changes intracellular glutamate and glutamine pool sizes.

Ingolf Sack¹, Anatol Fritsch², Steve Pawlizak², Martin Reiss-Zimmermann³, Karl-Titus Hoffmann³, Felix Arlt⁴, Wolf Müller⁵, Jing Guo¹, Jürgen Braun⁶, and Josef Käs^2
1Radiology, Charité - Universitätsmedizin Berlin, Berlin, Berlin, Germany, ²Physics and Earth Sciences, University of Leipzig, Saxony, Germany, ³Department of Neuroradiology, University Hospital, University of Leipzig, Saxony, Germany, ⁴Department of Neurosurgery, University Hospital, University of Leipzig, Saxony, Germany, ⁵Department of Neuropathology, University Hospital, University of Leipzig, Saxony, Germany, ⁶Department of Medical Informatics, Charité - Universitätsmedizin Berlin, Berlin, Germany

This study aims to bridge the gap between micro and macro scales of viscoelastic tissue properties in tumors. High resolution MRE was applied within clinical routine exams for measuring in-vivo viscoelastic properties of 7 cerebral tumors of different entity. Single cell viscoelastic properties of the same tumors were measured after surgery by the optical stretcher. We observed a significant correlation between in-vivo MRE and single cell elasticity with marked softening in high-WHO grade tumors. Thus, cellular mechanics is an important marker for in-vivo tissue viscoelasticity by which tumors can be staged in the future using MRE.

Xiao-Yong Zhang¹, Jingping Xie¹, Hua Li¹, Junzhong Xu¹, John C. Gore¹, and Zhongliang Zu^1
1Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States

Maarten W. Barentsz¹, Valentina Taviani², Jung M. Chang³, Debra M. Ikeda², Kanae K. Myiake⁴, Suchandrima Banerjee⁵, Maurice A.A.J. van den Bosch¹, Brian A. Hargreaves², and Bruce L. Daniel^2
1Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ²Radiology, Stanford University, Stanford, CA, United States, ³Radiology, Seul National University Hospital, Seul, Korea,⁴Diagnostic Imaging and Nuclear Medicine, Kyoto University Hospital, Kyoto, Japan, ⁵Global Applied Science Laboratory, GE Healthcare, Menlo Park, CA, United States

The diagnostic value of reduced FOV (r-FOV) high resolution diffusion-weighted imaging (DWI) in discriminating between benign and malignant breast lesions was assessed with respect to conventional, bilateral (bil) DWI and contrast-enhanced (CE) MRI. The discriminatory ability of r-FOV based on ADC was similar to bil-DWI. The area under the receiver operating characteristics for the predictive value of lesion shape and BI-RADS classification was higher when r-FOV DWI was used instead of bil-DWI. CE-MRI alone had a higher predictive value for malignancies than both r-FOV and bil-DWI. High-resolution DWI of targeted lesions allowed more accurate evaluation of tumor morphology than bil-DWI.

Catherine J Moran¹, Jung Min Chang², Marcus T Alley¹, Kanae Kawai Miyake¹, Debra M Ikeda¹, Brain A Hargreaves¹, Kristin L Granlund¹, and Bruce L Daniel^1
1Radiology, Stanford University, Stanford, CA, United States, ²Seoul National University Hospital, Seoul, Korea

The DW-PSIF sequence may provide an alternative to EPI-DWI acquisitions for diffusion weighted MRI in the breast as it provides diffusion weighted images with higher resolution and none of the characteristic distortion of EPI-DWI. We performed a pilot study of DW-PSIF in the breast which included analysis of effect on diagnostic accuracy and assessment of diffusion weighting, sharpness, SNR, distortion and level of artifact.

Yongming Dai¹, Junxiang Zhang², and Dongmei Wu^3
1Philips Healthcare, Shanghai, China, ²Department of Radiology, The First Affiliated Hospital of Bengbu Medical College, Anhui, China, ³Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China

In this abstract we investigated and evaluated the role of magnetic resonance (MR) diffusion kurtosis imaging (DKI) in characterizing breast lesions, and demonstrate the potential utility of DKI for the characterization of breast lesions.

Roberta M Strigel^1,2, Courtney K Morrison², Leah C Henze Bancroft¹, James H Holmes³, Kang Wang³, Wendy B DeMartini¹, Alejandro Munoz del Rio^1,2, and Frank R Korosec^1,2
1Radiology, University of Wisconsin, Madison, WI, United States, ²Medical Physics, University of Wisconsin, Madison, WI, United States, ³Global MR Applications and Workflow, GE Healthcare, Madison, WI, United States

High spatiotemporal resolution (STR) breast MRI offers the potential to improve diagnostic accuracy and characterization of cancer. We utilized a novel approach to sampling k-space and reconstructing images with view-sharing to provide high STR MRI with a temporal resolution of 27 seconds, six-times faster than the clinical standard-of-care, while maintaining high spatial resolution. We performed an intra-patient study to compare image quality between exams. Image quality of the high STR MRI was affected by the higher parallel imaging factor; however, differences were small and diagnostic image quality was maintained. This technique is promising for allowing advanced analysis of perfusion.

Evanthia Kousi¹, Maria A. Schmidt¹, Marco Borri¹, Cheryl Richardson², Georgina Hopkinson², Elizabeth A.M. O'Flynn¹, Robin M. Wilson², Steven Allen², Romney J.E. Pope², and Martin O. Leach^1
1CR-UK and EPSRC Imaging Centre, Royal Marsden NHS Foundation Trust and Institute of Cancer Reasearch, Sutton, Surrey, United Kingdom, ²Department of Radiology, Royal Marsden NHS Foundation Trust, Chelsea, London, United Kingdom

T2* relaxation time has been proposed as an imaging biomarker to evaluate hypoxia associated with tumour grade and therapeutic response. In this study, we explore the relationship between T2*, ADC and contrast agent uptake. Substantial variations for all functional parameters detected within tumours. Considerable inter-subject variability observed in the association between T2* and ADC for breast tumours. Less variability observed in the association between T2* and contrast uptake. Considering all patients, T2* and ADC correlated weakly and no correlation observed between T2* and contrast uptake. Our results suggest that T2* is an independent parameter and may provide new clinical information.

Andrew J Patterson¹, Mary M McLean², Reem Bedair¹, Andrew N Priest¹, John R Griffiths², Martin J Graves¹, and Fiona J Gilbert^1
1Department of Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, England, United Kingdom, ²Cancer Research UK Cambridge Insititute, Li Ka Shing Cambridge, Cambridge, England, United Kingdom

Magnetization Transfer Ratio (MTR) quantifies the interactions of water protons with different macromolecular environments. This study reports on changes in MTR in tissue proximal to malignancy and investigates the differences in MTR between tumor type and grade. Thirty patients with biopsy confirmed breast cancer were recruited. MTR maps are computed at 3T following a B1 field correction. This study noted a statistically significant difference between tumor type with the invasive mucinous type having lower MTR. This study finds that parenchyma proximal to malignancy was lower than that of contralateral parenchyma which may be due to desmoplastic reactions.

Melanie Freed^1,2, Pippa Storey^1,2, Alana Amarosa Lewin¹, Melanie Moccaldi¹, Linda Moy¹, and Sungheon G. Kim^1,2
1Center for Biomedical Imaging, Department of Radiology, NYU School of Medicine, New York, NY, United States, ²Center for Advanced Imaging Innovation and Research (CAI2R), Dept. Radiology, NYU School of Medicine, New York, NY, United States

Obesity is a known risk factor for developing breast cancer. So far, there has not been a comprehensive study comparing the lipid composition in the breast for patients with benign tissue and cancer. Part of the reason this remains a challenge is the lack of methods to non-invasively investigate lipid composition in breast tissue. We use a multiple gradient echo acquisition to rapidly acquire lipid spectral maps in the breast. These data suggest that for post-menopausal women, lower monounsaturated and higher saturated fatty acids may be related to breast cancer development.

Elisabeth Weiland¹, Sandra Peter², Dominik Nickel¹, Rolf Janka², Michael Uder², and Evelyn Wenkel^2
1MR Application Developement, Siemens Healthcare, Erlangen, Germany, ²Radiology, University of Erlangen, Germany

We evaluated the potential of Dixon fat-water separation using an ultrafast view sharing DCE protocol to differentiate between benign and malignant lesions of very early contrast phases. Analysis of maximum slope (MS) and time-to-enhancement (TTE) of the kinetic curves were performed on water-only images for 37 mass lesions. Best differentiation was achieved with MS, derived from the water-only images (ROC analysis with area under curve of 0.914). Due to its short acquisition time and its robust image quality ultrafast dual-echo sequences with Dixon fat-water separation may be beneficial for MR-based breast screening.

Jessica Buck¹, Saadallah Ramadan¹, Leah Best², Judith Silcock³, Jameen Arm², Scott Quadrelli¹, Gorane Santamaria¹, Kin Men Leong², Peter Lau², Peter Malycha¹, David Clark^1,3, and Carolyn Mountford^1,4
1Centre for MR in Health, University of Newcastle, Newcastle, NSW, Australia, ²Calvary Mater Hospital, Newcastle, NSW, Australia, ³The Breast and Endocrine Centre, Gateshead, NSW, Australia, ⁴Centre for Clinical Spectroscopy, Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States

In vivo 2D L-COSY identifies premalignant changes in women at high risk of developing breast cancer that are not seen by routine imaging, and allows women to be identified as MR spectroscopy Low Risk or MR spectroscopy High Risk according to changes recorded. Changes in the MR spectroscopy High Risk group include deregulation of lipid pathways and increased levels of metabolites. If these changes are confirmed in larger populations, it is possible that this information will allow women at increased clinical risk for breast cancer an objective means to monitor changes that may be taking place in their breast tissue.

Alana Amarosa Lewin¹, Sungheon Kim¹, James S Babb¹, Amy N Melsaether¹, Jason McKellop¹, Melanie Moccaldi², Ana Paula Klautau Leite³, and Linda Moy^1
1Radiology, New York University School of Medicine, New York, New York, United States, ²Radiology, New York University Cancer Institute, New York, New York, United States, ³Radiology, Hospital das Clínicas, School of Medicine, University of São Paulo, Brazil

This study investigates whether quantitative kinetic analysis of benign lesions and background parenchyma (BP) in breast MRI can elucidate differences between BRCA carriers and sporadic controls with high risk for breast cancer. We identified 49 BRCA mutation carriers and 49 control cases with benign lesions for comparison of their initial and delayed enhancement ratios (IER and DER) of BP. The control group showed significant differences in IER and DER between pre and post-menopausal women, but the BRCA group did not. Our results also indicate that the BRCA gene mutation has multifactorial and complex clinical and biologic implications.

Reem Bedair¹, Andrew Patterson², Mary McLean³, Roie Manavaki¹, Scott Reid⁴, John Griffiths³, Martin Graves², and Fiona Gilbert^1
1University of Cambridge, Department of Radiology, Cambridge, Cambridgeshire, United Kingdom, ²Department of Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, Cambridgeshire, United Kingdom, ³Cancer Research UK Cambridge Research Institute, Cambridge, Cambridgeshire, United Kingdom, ⁴GE Healthcare, Diagnostic Imaging, Buckingham, Buckinghamshire, United Kingdom

Quantitative DCE-MRI together with pharmacokinetic modelling enables the assessment of tumour-vessel permeability and leakage space in vivo. This work compares the early changes in the PK-derived parameter (Ktrans ) using both a voxel-wise histogram analysis across the entire tumour and the average Ktrans from the largest slice in cohort of breast cancer patients undergoing neoadjuvant chemotherapy. Although pixel-based analyses can demonstrate significant differences in the distribution of Ktrans, this can be a time-consuming process. Our results show that the average Ktrans calculated from the single largest slice provides statistically similar results that can be easily incorporated into routine clinical practice.

Reem Bedair¹, Martin Graves², Mary McLean³, Scott Reid⁴, Roie Manavaki¹, John Griffiths³, Andrew Patterson², and Fiona Gilbert^1
1University of Cambridge, Department of Radiology, Cambridge, Cambridgeshire, United Kingdom, ²Department of Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, Cambridgeshire, United Kingdom, ³Cancer Research UK Cambridge Research Institute, Cambridge, Cambridgeshire, United Kingdom, ⁴GE Healthcare, Diagnostic Imaging, Buckingham, Buckinghamshire, United Kingdom

DCE-MRI has proven a promising non-invasive modality for characterizing the pathophysiological microenvironment of tumours. Pharmacokinetic modelling can yield results of tumour-vessel permeability, perfusion and extracellular-extravascular volume fraction. This work exploits the improved spatiotemporal resolution achievable at 3.0T to investigate the relationship between the modelled vascular parameters and their histopathological profile within a cohort of breast cancer patients. Hotspot Ktrans and ve were found to be higher for the more common malignant types. A significant difference was found between hotspot Ktrans and ve in grades 1 and 3 tumours. Our results indicate that Ktrans and ve provide important and independent information concerning tumor biology and microvascular structure that supports the use of these more complex analysis protocols.

Xia Li¹, Vandana G Abramson¹, Lori R. Arlinghaus¹, Hakmook Kang¹, Jason M Williams¹, Richard G Abramson¹, A. Bapsi Chakravarthy¹, Praveen Pendyala¹, and Thomas E Yankeelov^1
1Vanderbilt University, Nashville, Tennessee, United States

This study determines if classifying breast cancer patients by subtype improves the ability of integrated DCE-MRI and DW-MRI to predict eventual response after the first cycle of NAC. The patients were divided into three groups according to receptor status: 1) ER-/PR-/HER2-, 2) HER2+, and 3) HR+/HER2-. These preliminary results demonstrate that DCE- and DW-MRI may be able to better predict treatment response for patients with particular receptor status. This observation should be confirmed in a large cohort in the future.

Michael Fox¹, Peter Gibbs¹, Martin Pickles¹, and Lindsay W. Turnbull^1
1Centre for MR Investigations, HYMS at University of Hull, Hull, East Yorkshire, United Kingdom

The performance of GLCM texture analysis is dependent on a well-drawn ROI segmenting a tumour from “healthy” surrounding tissue. There may be changes to the surrounding tissue in breast cancer patients which can be an indicator of tumour status, or response prediction. Software was developed to incrementally expand the ROI for 100 patients and compare the performance of GLCM texture analysis against previously recorded values. An increase in number of differences between TNEG patient sub-groups was found when peritumoral tissue was included, and performance was comparable for other patient sub-groups, suggesting surrounding tissue should be included in analysis.

Jeon-Hor Chen^1,2, Yifan Li¹, Yoon Jung Choi³, Chen-Pin Chou⁴, Tsung-Lung Yang⁴, and Min-Ying Su^1
1Center for Functional Onco-Imaging, University of California, Irvine, CA, United States, ²Department of Radiology, Eda Hospital and I-Shou University, Kaohsiung, Taiwan, ³Department of Radiology, Kangbuk Samsung Hospital, Seoul, Korea, ⁴Department of Radiology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan

Characterizing the association between breast density and risk of tumor subtypes may enhance our understanding of how breast cancer subtypes differ in etiology. The goal of this study was to use 3D MR-based density method to investigate the association of breast density with the development of different subtypes of breast cancer. In total 114 women diagnosed with unilateral breast cancer were analyzed. A negative correlation between age and percent breast density was noted, with older women tended to have lower density. ER negative and triple negative cancer had higher PD than other tumor subtypes, but no statistical difference was identified.

Michael Fox¹, Peter Gibbs¹, Martin Pickles¹, and Lindsay W. Turnbull^1
1Centre for MR Investigations, HYMS at University of Hull, Hull, East Yorkshire, United Kingdom

Minkowski Functinals may provide a new method to observe or diagnose breast cancer tumours in MRI, as they have done in CT. Software was created in-house, using 101 threshold levels, to create binary images of segmented breast lesions, from T1W MR images. Three 5th order polynomial fits were created for each of the 100 patients to describe their change in MF values as the threshold was raised. Two-thirds of the measured parameters were able to distinguish between TNEG patient sub-groups, with differences between biopsy grades being found also. This work supports the introduction of Minkowski Functionals into MRI texture analysis.

Khushbu Agarwal¹, Uma Sharma¹, Smriti Hari², Vurthaluru Seenu³, Rajinder Parshad³, and Naranamangalam R Jagannathan^1
1Department of NMR & MRI Facility, All India Institute of Medical Sciences, New Delhi, Delhi, India, ²Department of Radiodiagnosis, All India Institute of Medical Sciences, New Delhi, Delhi, India, ³Department of Surgical Disciplines, All India Institute of Medical Sciences, New Delhi, Delhi, India

Isabel Dregely¹, Daniel AJ Margolis², Kyung Sung¹, Novena Rangwala¹, Steve Raman³, and Holden H Wu^1
1Radiological Sciences, University of California Los Angeles, Los Angeles, CA, United States, ²University of California Los Angeles, Los Angeles, CA, United States, ³University of California Los Angeles, CA, United States

The central component of prostate cancer MRI is a high-resolution T2-weighted (T2w) acquisition. Quantitative T2-mapping can provide added value, however current methods are limited by compromises between spatial resolution, 3D coverage, and scan time. In this work we show initial results in a pilot patient study of rapid, quantitative T2 MRI in ~1 min total scan time using a 3D Dual Echo Steady State (DESS) acquisition.

Colleen Bailey¹, Eleftheria Panagiotaki¹, Nina Tunariu², Matthew R Orton³, Veronica A Morgan³, Thorsten Feiweier⁴, David J Hawkes¹, Martin O Leach³, David J Collins³, and Daniel C Alexander^1
1Centre for Medical Image Computing, University College London, London, United Kingdom, ²Radiology, Royal Marsden NHS Foundation Trust and Institute of Cancer Research, Sutton, United Kingdom, ³CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden NHS Foundation Trust, London, United Kingdom, ⁴Healthcare Sector, Siemens AG, Erlangen, Germany

Bone metastases in men with advanced prostate cancer were examined by diffusion MRI with varying diffusion times and gradient strengths. Data were fitted to 3-compartment models of diffusion that included a perfusion compartment, an extracellular compartment and an intracellular spherically-restricted compartment. Data showed variation for points with the same b-value but different diffusion time and were better explained by a model incorporating restriction than either conventional ADC or 2- or 3-compartment models with free diffusion. Model parameters indicated a low perfusion fraction (<10%), intracellular volume fraction 0.22-0.56 and a cell radius between 4.7-9.9 µm.

Seyed Reza Mousavi¹, Seyyed Mohammad Hesabgar², Timothy Scholl^2,3, and Abbas Samani^2,3
1Clinical Neurological Sciences, University of Western Ontario, London, Ontario, Canada, ²Medical Biophysics, University of Western Ontario, London, Ontario, Canada, ³Robarts Research Institute, London, Ontario, Canada

T2-weighted MR imaging is an evolving modality in prostate cancer diagnosis. Although T2-weighted MRI has a good sensitivity but it has low specificity. The specificity of the T2-weighted MRI, however, may be improved by calculating prostate tissue elasticity distribution. Tissue elasticity properties are known to be prostate cancer biomarkers. In this research project, an MR elastography technique is proposed to reconstruct tissue Young’s modulus with the aim of improving T2-weighted MR imaging specificity. The proposed technique only utilizes T2-weighted MR imaging information and does not require any additional peripheral devices. It also does not appreciably disrupt the clinical flow.

Lucy E Kershaw^1,2, Andrew J McPartlin^2,3, and Ananya Choudhury^2,3
1CMPE, The Christie NHSFT, Manchester, United Kingdom, ²Institute of Cancer Sciences, The University of Manchester, Manchester, United Kingdom, ³Oncology, The Christie NHSFT, Manchester, United Kingdom

SYNOPSIS

Hugh Harvey¹, Veronica Morgan², David Dearnaley³, Sharon Giles², Alison Macdonald², Julia Murray³, and Nandita deSouza^1
1CRUK Cancer Imaging Centre, The Institute of Cancer Research, Sutton, Surrey, United Kingdom, ²The Royal Marsden NHS Foundation Trust, Surrey, United Kingdom, ³Radiotherapy & Imaging, The Institute of Cancer Research, London, United Kingdom

Delineation of the tumor region-of interest is crucial when planning boost doses of radiation therapy to a dominant intraprostatic tumor nodule. Comparison of prostatic tumor volume measurements defined on T2W images, Apparent Diffusion Coefficient maps and Dynamic Contrast Enhanced images in 16 patients due for radiation dose-boosting to tumor revealed a significant difference in measured volumes between the sequences, with T2W-derived volumes being up to 48.7% larger than DCE-derived volumes and therefore most useful in the context of radiation therapy planning.

Veronica Clavijo Jordan^1,2, Christian Preihs¹, Shiuhwei Chen³, Shanrong Zhang¹, Wen-hong Li³, Neil Rofsky², and Dean Sherry^1,4
1Advanced Imaging Research Center, UT Southwestern Medical Center, Dallas, Texas, United States, ²Department of Radiology, UT Southwestern Medical Center, Dallas, Texas, United States, ³Departments of Cell Biology and of Biochemistry, UT Southwestern Medical Center, Dallas, Texas, United States, ⁴Department of Chemistry, UT Dallas, Texas, United States

Due to the lack of accuracy in current prostate cancer testing methods, there is an increasing need for more reliable tests to differentiate prostate cancer from benign prostatic hyperplasia (BPH), prostatic intraepithelial neoplasia, and/or inflammation. The reported loss of zinc in prostate cancer offers a potential mechanism to distinguish malignancy from the other conditions. Here we introduce a gadolinium-based sensor that can report the available zinc(II) present in the prostate sensitively. We also report the role of glucose in secreting zinc(II) in prostate secretory cells.

Thiele Kobus^1,2, Andriy Fedorov¹, Clare Tempany¹, Robert Mulkern³, Ruth Dunne¹, and Stephan E. Maier^1
1Radiology, Brigham and Women's Hospital, Boston, Massachusetts, United States, ²Radiology, Radboud UMC, Nijmegen, Netherlands, ³Radiology, Children's Hospital, Boston, Massachusetts, United States

We investigated the performance of the bi-exponential fit for extended b-factor DWI (b=0 to 3500 s/mm²) in discriminating between different tissues in prostate imaging. The means of the fitted parameters, ADC_fast, ADC_slow and f_slow (fraction of ADC_slow component), were significantly different between normal tissue (PZ and TZ) and tumor tissue in both PZ and TZ. There was also a difference for f_slow between normal PZ and TZ (p<0.001). No differences in the parameters were found between tumors in the PZ and TZ. The increase of f_slow in tumor tissue might represent the occupation of the luminal space by tumor cells.

Haibo Wang¹, Satish viswanath², Asha Singanamalli³, and Anant Madabhushi^4
1Case Western Reserve University, Cleveland Heights, OHIO, United States, ²Biomedical Engineering, Case Western Reserve University, Cleveland Heights, OHIO, United States, ³Case Western Reserve University, OHIO, United States, ⁴Biomedical Engineering, Case Western Reserve University, OHIO, United States

This paper presents a computerized approach to assist radiologist for prostate cancer diagnosis on T2w MRI. It detects prostate cancer more faster and more accurate than the conventional computerized solutions. Novel image processing and machine learning techniques are successfully used in the approach.

Holden H Wu¹, Alan Priester^2,3, Shyam Natarajan^2,3, Kyunghyun Sung¹, Daniel Margolis¹, Warren Grundfest^2,3, Leonard Marks^3,4, and Steven Raman^1
1Radiological Sciences, University of California Los Angeles, Los Angeles, CA, United States, ²Biomedical Engineering, University of California Los Angeles, CA, United States, ³Center for Advanced Surgical and Interventional Technology (CASIT), University of California Los Angeles, CA, United States, ⁴Urology, University of California Los Angeles, CA, United States

to add ...

Olga Starobinets^1,2, John Kornak³, John Kurhanewicz^1,2, and Susan M Noworolski^1,2
1Radiology and Biomedical Imaging, UCSF, San Francisco, CA, United States, ²Graduate Group in Bioengineering, UC Berkeley, Berkeley, CA, United States, ³Epidemiology and Biostatistics, UCSF, San Francisco, CA, United States

Majority of imaging parameters obtained during a prostate mpMRI scan demonstrate a better separation between low-grade cancerous and benign regions for prostate tissues exposed to anti-androgen therapy. Pretreatment with dutasteride or finasteride may facilitate a better detection of low-grade prostate cancer.

Martin Thomas Freitag¹, Jan Radtke^1,2, Boris Hadaschik², Uwe Haberkorn³, Heinz-Peter Schlemmer¹, Matthias Roethke¹, and Ali Afshar-Oromieh^3
1Department of Radiology, German Cancer Research Center, Heidelberg, Baden-Wuerttemberg, Germany, ²Department of Urology, University hospital of Heidelberg, Heidelberg, Baden-Wuerttemberg, Germany, ³Department of Nuclear Medicine, University hospital of Heidelberg, Heidelberg, Baden-Wuerttemberg, Germany

The recently introduced 68Gallium-PSMA-ligand is regarded as a significant step forward in the diagnosis of prostate cancer and its metastases for PET examinations. Multiparametric PET/MR could be advantageous over PET/CT to better delineate PET-positive findings due to higher soft tissue contrast. Here, we systematically compare the diagnostic performance of multiparametric PET/MRI versus PET/CT with focus on bone and lymph node metastases of prostate cancer. PET-positive results were significantly correlating between both methods indicating a strong reliability of PET-measurements using 68Ga-PSMA-PET/MRI. For the delineation of lymph nodes, PET/MRI was superior to PET/CT whereas the conspicuity of bone metastases was comparable.

Mariska P Luttje¹, Catalina S Arteaga de Castro², Peter R Luijten¹, Marco van Vulpen¹, Uulke A van der Heide², and Dennis WJ Klomp^1
1Imaging Division, University Medical Center Utrecht, Utrecht, Netherlands, ²Department of Radiotherapy, the Netherlands Cancer Institute - Antoni van Leeuwenhoek hospital, Amsterdam, Netherlands

In contrast to 1H MRSI at lower field strength, where polyamine signals overlap with choline signals, our results at 7 tesla demonstrate the ability to fit each of the prostate metabolite resonances distinctively. This facilitated investigation of the role of polyamines in the results of 1H MRSI in detecting prostate cancer. The conventional metabolite ratio (tCho+PA+Cr)/Cit significantly discriminates between tumor and healthy tissue areas at 7 tesla, whereas the new ratio tCho/(Cit+PA+Cr) performs less in discriminating tumor from healthy tissue in the prostate in contrast to reports obtained from tissue samples.

Xinran Zhong^1,2, Novena Rangwala¹, Steven Raman¹, Daniel Margolis¹, Holden Wu^1,2, and Kyunghyun Sung^1,2
1Radiological Sciences, University of California Los Angeles, Los Angeles, CA, United States, ²Biomedical Physics Interdepartmental Program, University of California Los Angeles, Los Angeles, CA, United States

B1+ variations in the pelvis is measured using the reference region VFA method and differences on the T1 relaxation values in prostate and pelvic muscles with and without compensating for B1+ variation were analyzed in a total of 108 prostate patients at 3T. Significant improvement on T1 value could be seen after RF correction through comparison between different tissues and different systems, indicating the need for B1 correction in prostate T1 imaging.

Victoria Sherwood¹, Donald MacDonald², James Harding³, Nicholas Hedley³, Kieran Jefferson², Chris Koller¹, and Charles Hutchinson^3
1Department of Radiology Physics, University Hospitals Coventry & Warwickshire NHS Trust, Coventry, Warwickshire, United Kingdom, ²Department of Urology, University Hospitals Coventry & Warwickshire NHS Trust, Coventry, Warwickshire, United Kingdom, ³Department of Radiology, University Hospitals Coventry & Warwickshire NHS Trust, Coventry, Warwickshire, United Kingdom

A clinical development study was designed to validate targeted MR-ultrasound fusion-guided prostate biopsies against transperineal template saturation biopsies, using Real-time Virtual Sonography (RVS). Twenty-seven patients underwent fusion and template biopsies. Results were compared and related back to multiparametric MRI. Eleven of 19 patients with Gleason 6 disease or above, as diagnosed by template biopsy, tested positive on fusion biopsy. Almost all the remaining patients had Gleason 6 disease, which may not be detected on MRI. RVS has the potential to reduce patient complications and lighten the burden on healthcare services by significantly limiting the number of necessary template saturation biopsies.

Steven M Shea¹, Joseph M Yacoub¹, Gopal N Gupta², Grace Yoon³, and Ari Goldberg^1
1Radiology, Loyola University Chicago, Maywood, IL, United States, ²Urology, Loyola University Chicago, Maywood, IL, United States, ³Stritch School of Medicine, Loyola University Chicago, Maywood, IL, United States

The rise of MR/US fusion for prostate cancer biopsies necessitates a true 3D imaging technique. We compared a T2-weighted 3D variable flip angle turbo spin echo acquisition versus the standard 2D T2-weighted acquisition in patients undergoing MR/US fusion biopsy at 3T. PIRADS scores, image quality scores, and relative contrast were all equivalent or better in T2w 3D as compared to T2w 2D. Given the advantages in scan time and overall voxel size, T2w 3D imaging is a viable alternative to T2w 2D imaging.

Justin Charles Peterson¹, Adam Farag², Trevor Szekeres², Eli Gibson^2,3, Aaron D Ward^2,3, Joseph Chin⁴, Stephen Pautler⁵, Glenn Bauman⁴, Cesare Romagnoli⁴, Robert Bartha^1,2, and Timothy J Scholl^1,2
1Medical Biophysics, Western University, London, Ontario, Canada, ²Robarts Research Institute, Ontario, Canada, ³Biomedical Engineering, Western University, Ontario, Canada, ⁴London Health Sciences Centre, Ontario, Canada, ⁵St. Joseph's Health Care, Ontario, Canada

Prostate cancer (PCa) is the most common malignancy in men. Currently, a combination of multi-parametric MR including T₂-weighted, diffusion-weighted, and dynamic contrast enhanced imaging, is used clinically but often provides insufficient information to determine the malignancy of a lesion. Previous studies have shown increased MRI measured tissue sodium concentration (TSC) in brain and breast cancer. In this report we demonstrate in vivo ²³Na MRI in patients with PCa. Using this proposed method TSC data was registered to histopathology and analyzed. These preliminary data show a positive correlation between tumor grade and TSC within the prostate.

Paul Summers¹, Daniel Chong², Valentina Elli³, Daniele Giardiello⁴, Mehran Vaziri¹, Giuseppe Petralia¹, and Massimo Bellomi^1,3
1European Institute of Oncology, Milan, Italy, ²Stillpig Software, Sarawak, Malaysia, ³University of Milan, Milan, Italy, ⁴University of Milan - Bicocca, Milan, Italy

Tissues with long T2s like the prostatic peripheral zone (PZ) can mimic lesions in diffusion weighted MRI due to “T2 shine-through”. We have evaluated the impact on prostate tumor contrast of Relax DWI, where four or more echo time and diffusion weighting combinations are used to isolate T2 and ADC effects. Relax DWI showed better contrast between tumor and both prostate and peripheral zone when high b-value images were calculated for TE=0 (eliminating T2 effects) than was observed in the acquired images using our scanner’s minimum TE for the same b-value, without significant change in ADC map appearance.

Olga Starobinets^1,2, Jeffry Simko^3,4, Kyle Kuchinsky³, Sonam Machingal¹, John Kurhanewicz^1,2, Peter R Carroll⁴, Kirsten L Greene⁴, and Susan M Noworolski^1,2
1Radiology and Biomedical Imaging, UCSF, San Francisco, CA, United States, ²Graduate Group in Bioengineering, UC Berkeley, Berkeley, CA, United States, ³Pathology, UCSF, San Francisco, CA, United States, ⁴Urology, UCSF, San Francisco, CA, United States

The purpose of this study was to use semi-quantitative parameters derived from dynamic contrast-enhanced (DCE) MRI, 1hydrogen-MR spectroscopic imaging (MRSI), diffusion MR and MRI to differentiate benign and malignant, as well as high-risk from low-risk prostate cancers in the anterior aspect of the prostate using whole-mount pathology as the standard of reference. A logistic regression combination of parameters provided ROC AUC=0.983 for discriminating benign from malignant tissues and AUC=0.824 for distinguishing between low-risk and high-risk cancer.

Petra J van Houdt¹, Harsh K Agarwal^2,3, Laurens B van Buuren¹, Marko Ivancevic⁴, Søren Haack⁵, Jesper Folsted Kallehauge⁶, Peter L Choyke³, and Uulke A van der Heide^1
1Radiation Oncology, the Netherlands Cancer Institute, Amsterdam, Netherlands, ²Philips Research NA, Briarcliff Manor, MD, United States, ³National Cancer Institute, National Institutes of Health, Bethesda, NY, United States, ⁴Philips Healthcare, Best, Netherlands, ⁵Clinical Engineering, Aarhus University Hospital, Aarhus, Denmark, ⁶Medical Physics, Aarhus University Hospital, Aarhus, Denmark

T2 is proposed as a biomarker for response monitoring of patients with prostate cancer treated with radiotherapy. A key requirement in multi-center longitudinal studies is good reproducibility over multiple visits. The purpose of this study was to establish the inter- and intra-center reproducibility of T2 maps. Phantom measurements were performed on three platforms, showing that the differences in T2 values within and between platforms were smaller than 5%. Average difference in T2 values of volunteers between two visits was -1.5 +/- 9.5%. This work demonstrates that T2 mapping is suitable for use in multi-center, longitudinal trials in prostate cancer patients.

Satish Easwar Viswanath¹, Chun Yeung Yim², Nicolas Bloch³, Mark Rosen⁴, John Kurhanewicz⁵, and Anant Madabhushi^6
1Case Western Reserve University, Cleveland, Ohio, United States, ²Rutgers University, New Brunswick, New Jersey, United States, ³Boston University, MA, United States, ⁴University of Pennsylvania, PA, United States, ⁵University of California San Francisco, CA, United States, ⁶Case Western Reserve University, Ohio, United States

We present the first results of a multi-site study involving post-processing T2w MRI acquisitions from multiple institutions and scanners, to calculate a new pseudo-quantitative T2w parameter with tissue-specific meaning, to more accurately and reproducibly identify prostate cancer on MRI.

Stephan Polanec¹, Katja Pinker², Martin Suasani², Peter Brader², Dietmar Georg², Thomas Helbich², and Pascal Baltzer^2
1General Hospital of the Medical University of Vienna!, Vienna, Vienna, Austria, ²General Hospital of the Medical University of Vienna!, Vienna, Austria

To compare the diagnostic performance of the ESUR PI-RADS scoring system for multiparametric magnetic resonance imaging (MP MRI) of the prostate using either three (T2w, DCE, DWI) or four MRI parameters (T2w, DCE, DWI, 3D 1H-MRSI) for the detection and grading of prostate cancer (PCa). The results demonstrate that diagnostic accuracy of MP MRI of the prostate using three MRI parameters is as good as MP MRI of the prostate using four MRI parameters. Adding 3D 1H-MRSI as a fourth parameter doesn’t increase diagnostic accuracy for differentiation of benign and malignant lesions or of high-grade and low-grade PCa.

Aytekin Oto¹, David VanderWeele², Yulei Jiang¹, Stephanie Maria McCann¹, Xiaobing Fan¹, Jianing Wang¹, and Tatjana Antic^3
1Radiology, The University of Chicago Medicine, Chicago, IL, United States, ²Internal Medicine, The University of Chicago Medicine, Chicago, IL, United States, ³Pathology, The University of Chicago Medicine, Chicago, IL, United States

Multi-parametric MRI (Mp-MRI) is playing an increasing role in the detection, staging and localization of Pca. Determination of the cancer aggressiveness by Mp-MRI has been addressed in several studies and significant correlations between quantitative MRI parameters (such as apparent diffusion coefficient [ADC] and ktrans) and Gleason score have been reported. However, to our knowledge, no study thus far has investigated associations between quantitative Mp-MRI parameters and genomic markers in Pca.

Adam T. Froemming¹, Eric A. Borisch², Joshua D. Trzasko², Roger C. Grimm², Armando Manduca², Phillip Young³, Stephen J. Riederer³, and Akira Kawashima^3
1Radiology, Mayo Clinic, Rochester, MN, United States, ²Physiology and Biomedical Engineering, Mayo Clinic, MN, United States, ³Radiology, Mayo Clinic, MN, United States

Fengguang Zhang¹, Jinrong Qu¹, Hui Liu², Xiang Li¹, Hongkai Zhang¹, Hailiang Li¹, Grimm Robert³, Kiefer Berthold³, and Xuejun Chen^1
1Radiology, Henan Tumor Hospital, Zhengzhou, Henan, China, ²NEA MR Collaboration, Siemens Ltd., China, Shanghai, China, ³Healthcare, Siemens AG, Erlangen, Germany

Accurate staging esophageal carcinoma is quite important for treatment decision. un to now, there is still no available MRI technique to well address this need. With the development of self-gating radial VIBE, we are able to evaluate this new technqiue in a large esophageal carcinoma population. The prelimary results shows confident in T1 and T2 staging.

JEONG-WON JEONG^1,2, Csaba Juhász^1,3, Sandeep Mittal^3,4, Edit Bosnyák¹, and Diane C Chugani^1,2
1Pediatrics and Neurology, Wayne State University, Detroit, MI, United States, ²Children's Hospital of Michigan, Detroit, MI, United States, ³Karmanos Cancer Institute, Detroit, MI, United States, ⁴Neurosurgery and Oncology, Wayne State University, Detroit, MI, United States

An independent component analysis with ball-stick model (ICA+BSM) was proposed to solve an intra-voxel crossing fiber problem in clinical diffusion tensor imaging (DTI). This study investigates whether the ICA+BSM analysis can be combined with isotropic diffusion spectrum imaging (IDSI) technique to assess the degree of cellularity in tumor-infiltrated white matter tissues in clinical DTI. Compared with apprarent diffusion coefficient image, which provided a moderate accuracy of 0.457 to detect tumor cells in the active tumor region delinated by á[11C]methyl-L-tryptophan (AMT)-positron emission tomography, IDSI-derived cell ratio yielded a much higher accuracy of 0.969 in conventional receiver operating characteristic curve analysis.

Hongliang Sun¹, Yanyan Xu², Aiping Song³, and Wu Wang^4
1Radiology, China-Japan Friendship Hospital, Beijing, Beijing, China, ²Radiology, China-Japan Friendship Hospital, Beijing, China, ³China-Japan-Friendship Hospital, Beijing, China, ⁴China-Japan Friendship Hospital, Beijing, China

We found the IVIM parameters demonstrated well correlation with tumor stage and grade, which could reflect clinicopathological features of rectal cancer.

Arash Latifoltojar¹, Margaret Hall-Craggs², Alan Bainbridge², Charles House², Kannan Rajesparan², Stuart Taylor³, Kwee Yong³, Neil Rabin², and Shonit Punwani^3
1University College London, London, London, United Kingdom, ²University College London Hospital, London, United Kingdom, ³University College London, London, United Kingdom