Antonio C Westphalen¹, Galen D Reed¹, Phillip P Vinh¹, Christopher K Sotto¹, Daniel B Vigneron¹, and John Kurhanewicz^1
1Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States

The results of this study suggest that MR imaging techniques (T2-weighted MR imaging, MR spectroscopic imaging, and diffusion-weighted MR imaging) should likely be combined for the assessment of patients with suspected locally recurrent prostate cancer following radiation therapy.

Kent Yip¹, N. Jane Taylor², J James Stirling², Ian C Simcock², Uma Patel¹, Nihal Shah¹, Peter Ostler¹, James A d'Arcy³, David J Collins³, Peter J Hoskin¹, Anwar R Padhani², and Roberto Alonzi^1
1Marie Curie Research Wing, Mount Vernon Cancer Centre, Northwood, Middlesex HA6 2RN, United Kingdom, ²Paul Strickland Scanner Centre, Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, United Kingdom, ³CRUK-EPSRC Cancer Imaging Centre, Institute of Cancer Research & Royal Marsden Hospital, Sutton, Surrey SM2 5PT, United Kingdom

The use of doses much higher than the conventional 2Gy/fraction during stereotoactic radiotherapy (SBRT) has allowed radiotherapy courses to be significantly shortened from weeks to days. It is increasingly being used in the treatment of localised prostate cancer. From a radiobiology perspective, the lack of time for re-oxygenation to take place during the much shortened duration of treatment may compromise its effectiveness. In this study, based on data obtained from functional MRI imaging, we have shown that there is a significant degree of re-oxygenation in the irradiated prostate during SBRT, and that this is accompanied by transient prostate gland hypervascularity.

Daniel Jason Aaron Margolis¹, Nishant Gandhi¹, Gregory Shaw¹, Shyam Natarajan², Naira Muradyan³, and Leonard Marks^4
1Department of Radiology, UCLA David Geffen School of Medicine, Los Angeles, CA, United States, ²Department of Bioengineering, UCLA David Geffen School of Medicine, Los Angeles, CA, United States, ³iCAD, Inc., ⁴Department of Urology, UCLA David Geffen School of Medicine

The relative utility of diffusion and perfusion imaging for detection and grading of prostate cancer is promising. In 89 patients who had 175 MRI-ultrasound fusion targeted biopsies, where regions of interest were copied from T2-weighted images onto other series, statistical significance was achieved for ADC, Ktrans, and Kep to discriminate between benign and cancerous targets, and between targets containing high-grade (Gleason pattern 4) disease from a combination of low-grade and benign disease. However, a manually-chosen “hot-spot” ADC was not significant for either. This suggests that the entire lesion should be evaluated to determine level of suspicion.

Russell N. Low^1,2, Donald B Fuller³, and Naira Muradyan^4
1Sharp and Children's MRI Center, San Diego, California, United States, ²San Diego Imaging, San Diego, CA, United States, ³CyberKnife Centers of San Diego, ⁴iCAD, Inc

Dynamic contrast-enhanced MRI (DCE) is used to quantitatively assess prostate cancer before and after CyberKnife radiation therapy. The MRI driven parametric values Ktrans, Kep, EFV, and iAUGC can be followed to assess changes in tumor permeability occurring on a cellular level after CyberKnife SBRT. Both quantitative and qualitative display of these parametric values can be used to monitor tumor response to treatment.

Yuxi Pang^1,2, Baris Turkbey², Marcelino Bernardo^2,3, Jochen Kruecker⁴, Samuel Kadoury⁴, Maria J Merino⁵, Bradford Wood⁶, Peter A Pinto⁷, and Peter L Choyke^2
1Philips Healthcare, Cleveland, OH, United States, ²Molecular Imaging Program, National Cancer Institute, Bethesda, Maryland, United States, ³SAIC-Frederick, Bethesda, Maryland, United States, ⁴Philips Research North America, Briarcliff Manor, NY, United States, ⁵Laboratory of Pathology, National Cancer Institute, Bethesda, Maryland, United States, ⁶Diagnostic Radiology Department-Clinical Center, National Institutes of Health, Bethesda, Maryland, United States, ⁷Urologic Oncology Branch, National Cancer Institute, Bethesda, Maryland, United States

This work is to evaluate the performance of intravoxel incoherent motion (IVIM) and dynamic contrast-enhanced (DCE) MRI for prostate cancer detection. Thirty three patients underwent diffusion weighted (DW) imaging with five b-values, T2W and DCE-MRI. Diffusion coefficients (D) and perfusion fraction (f) were derived from an asymptotic curve-fitting based on the IVIM model. The results showed that D differentiated tumors from normal tissues in the prostate using all possible combinations of non-zero b-values; however, f demonstrated large variations depending on the choice of b-values. Exclusion of the highest b-value of 750 (s/mm2) led to better correlations of f with DCE-MRI.

Durgesh Kumar Dwivedi¹, Rajeev Kumar², Sanjay Thulkar³, Sanjay Sharma³, Amit K. Dinda⁴, and Naranamangalam R. Jagannathan^1
1Department of NMR & MRI Facility, All India Institute of Medical Sciences, New Delhi, Delhi, India, ²Department of Urology, All India Institute of Medical Sciences, New Delhi, Delhi, India, ³Department of Radio-diagnosis, All India Institute of Medical Sciences, New Delhi, Delhi, India, ⁴Department of Pathology, All India Institute of Medical Sciences, New Delhi, Delhi, India

Gleason grading is predictor of aggressiveness and disease outcome in prostate cancer (PCa) patients and its reproducibility might depend upon clinical factors and pathologist experience. For accurate treatment selection for PCa it is essential to know the Gleason score (GS) accurately on TRUS biopsy or final GS on radical prostatectomy specimens. In order to improve pretreatment prediction of PCa aggressiveness, we performed a prospective study to evaluate PCa aggressiveness based upon apparent diffusion coefficient (ADC) values derived from diffusion weighted MRI. Our study indicated that ADC values could be used as a noninvasive method in the prediction of PCa aggressiveness.

Michael C. Yang¹, Yi Sui², Lei Tang^2,3, Albert Y. Yang⁴, Eric J. Zhang², Guanzhong Liu², Virgilia Macias⁵, Andre Balla⁵, Leslie Deane⁶, Xiaohong Joe Zhou^1,2, and Karen L. Xie^1
1Department of Radiology, University of Illinois at Chicago, Chicago, Illinois, United States, ²Center for MR Research, University of Illinois at Chicago,³Department of Radiology, Peking University Cancer Hospital, Beijing, China, ⁴College of Medicine, University of Illinois at Chicago, ⁵Department of Pathology, University of Illinois at Chicago, ⁶Department of Urology, University of Illinois at Chicago

Diffusion imaging using multiple b-values has great potential for revealing tissue structural information beyond what an apparent diffusion coefficient (ADC) can provide. In this study, in addition to conventional images, multi-b-value diffusion imaging using a novel stretched-exponential model was applied to patients with biopsy-proven prostate cancer to evaluate its clinical utility for cancer detection. The series of diffusion images at multi-b-values was analyzed to yield quantitative values of distributed diffusion coefficient (DDC) and stretch exponent ƒ8 5. The maps based on these new parameters demonstrate significantly higher contrast and detectibility of prostate cancer than conventional ADC maps.

Meer Basharat¹, Maysam Jafar¹, Nandita deSouza¹, and Geoffrey Payne^1
1CRUK & EPSRC Cancer Imaging Centre, Institute of Cancer Research, Sutton, Surrey, United Kingdom

2D-CSI using short-TE (TE=32ms) PRESS was performed on 13 prostate cancer patients to assess improved visualisation of metabolite signals with reduced T2-decay compared to the standard TE=100ms implementation and to correlate these metabolite concentrations to ADC. Citrate T2 was 88±34ms in the PZ (n=34) and 90±50ms (n=86) in the CG, whilst choline T2 was 61±18ms (n=13) across both zones, suggesting that short-TE is more favourable for detecting prostate metabolites. Indeed, more metabolites with improved signal were observed using TE=32ms. Citrate-ADC correlation was r2=0.146 which may be due to increase ductal volume in the secretory glandular tissues.

Ben Babourina-Brooks¹, Deming Wang¹, Glen Wood², and Gary Cowin^1
1Centre for Advanced Imaging, University of Queensland, Brisbane, QLD, Australia, ²Brisbane Urology Clinic, Brisbane, QLD, Australia

We aimed to compare two Diffusion Weighted Imaging (DWI) sequences for prostate tumour detection, Echo Planar Imaging (EPI) and Half Fourier Acquisition Single shot Turbo spin Echo (HASTE). The HASTE sequence should have reduced magnetic susceptibility and chemical shift artifacts than EPI. We quantitatively assessed the Apparent Diffusion Coefficient (ADC) of tumours compared to healthy tissue and Receiver Operator Coefficient (ROC) curves in both sequences, using whole mount histology as the gold standard. ADC values of HASTE and EPI were different, which resulted in unique ADC thresholds for each sequence. Both sequences were able to accurately distinguish the PZ from Pca as reported by the high >0.8 area under the curve values from the PZ only curves.

Alan Wright¹, Kobus Thiele¹, Kirsten M Selnæs², Ingrid S Gribbestad², Elisabeth Weiland³, Tom W. J. Scheenen¹, and Arend Heerschap^1
1Radiology, Radboud University Nijmegen Medical Centre, Nijmegen, Gelderland, Netherlands, ²Department of Circulation and Medical Imaging, NTNU, Trondheim, Norway, ³Siemens Healthcare, Erlangen, Germany

An automated quality control algorithm has been developed for 3D ¹H MRSI data sets from prostate cancer patients. This will provide an important post processing tool in automation of the analysis of MRSI data of these patients. The method is based on feature extraction using Independent Component Analysis and uses a Gold Standard of consensus decisions of spectral quality as judged by four experts. The algorithm can separate a Test Set of acceptable from unacceptable data with 95% sensitivity and 95% specificity.

Rajakumar Nagarajan¹, Daniel Margolis¹, Steven S Raman¹, Manoj K Sarma¹, Ke Sheng², Robert E Reiter³, and M.Albert Thomas^1
1Radiological Sciences, University of California Los Angeles (UCLA), Los Angeles, CA, United States, ²Radiation Oncology, University of California Los Angeles (UCLA), Los Angeles, CA, United States, ³Urology, University of California Los Angeles (UCLA), Los Angeles, CA, United States

Prostate cancer (PCa) is the most common cancer in men and is the second leading cause of cancer death in American men, behind only lung cancer. Diffusion weighted imaging (DWI) and MR spectroscopy offers improved sensitivity and specificity for PCa detection. We have investigated the functional changes in prostate cancer patients with three pathologically proven different Gleason scores (3+3, 3+4 and 4+3) using DWI and magnetic resonance spectroscopic imaging (MRSI). The combined MRSI and DWI can help in the discrimination of intermediate Gleason grade tumors from low Gleason grade tumors with histopathology as a standard reference. This information can guide subsequent definitive management, and help to optimize active surveillance programs.

Alan Wright¹, May-Britt Tessem², Helena Bertilsson^3,4, Maria T Grinde^2,5, Guro F Giskeødegård², Jostein Halgunset^3,6, Anders Angelsen⁴, Arend Heerschap¹, and Ingrid S Gribbestad^2
1Radiology, Radboud University Nijmegen Medical Centre, Nijmegen, Gelderland, Netherlands, ²Department of Circulation and Medical Imaging, NTNU, Trondheim, Norway, ³Department of Laboratory Medicine and Children’s and Women’s Health, NTNU, Trondheim, Norway, ⁴Department of Urology, St Olavs Hospital, Trondheim University Hospital, Trondheim, Norway, ⁵St Olavs Hospital, Trondheim University Hospital, Trondheim, Norway, ⁶Department of Pathology and Medical Genetics, St Olavs Hospital, Trondheim University Hospital, Trondheim, Norway

¹H HR-MAS MRS was performed on 131 cores taken from surgically resected prostates from patients with a cancer diagnosis. The cores were characterized with histopathology for the tumour and non-tumour content. LCModel was used with a novel basis set of model metabolites made from careful assignments of a subset of spectra and a formate standard solution for robust and automated quantification procedure. Glutamate, choline and glycerophosphocholine were all found to be positive markers for the presence of tumour tissue, while glucose was significantly reduced in samples containing mostly tumour. Citrate and spermine were found to negatively correlate with the percentage of stroma in benign tissue.

Charles S Springer^1,2, Ian J Tagge¹, Xin Li¹, Luminita A Tudorica^1,3, Sunitha B Thakur^4,5, Karen Y Oh³, Elizabeth A Morris⁵, Nicole Roy³, Mark D Kettler³, William D Rooney^1,6, Jason A Koutcher^4,5, and Wei Huang^1,2
1Advanced Imaging Research Center, Oregon Health & Science University, Portland, Oregon, United States, ²Knight Cancer Institute, Oregon Health & Science University, Portland, Oregon, United States, ³Diagnostic Radiology, Oregon Health & Science University, Portland, Oregon, United States, ⁴Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York, United States, ⁵Radiology, Memorial Sloan Kettering Cancer Center, New York, New York, United States, ⁶Biomedical Engineering, Oregon Health & Science University, Portland, Oregon, United States

A meta-population of 163 human breast tumors in vivo was studied by shutter-speed DCE-MRI. This spans three institutions, two magnetic field strengths, three instrument vendors, two initial screening modalities, three Gd contrast agents, and 21 lesion types. When the value of ÄKtrans is plotted vs. Äkep for all, a "threshold" behavior is manifest. The 120 benign lesion points are very tightly clustered near the origin, while the 43 malignant tumor values are strongly correlated, and some reach very large values. It is possible this is a manifestation of the metabolic "angiogenic switch" preceding breast tumor progression.

Karen Y Oh¹, Luminita A Tudorica¹, Nicole Roy¹, Mark D Kettler¹, Xin Li¹, Yiyi Chen¹, Aneela Afzal¹, John W Grinstead², Gerhard Laub³, and Wei Huang^1
1Oregon Health & Science University, Portland, Oregon, United States, ²Siemens Healthcare, Portland, Oregon, United States, ³Siemens Healthcare, San Francisco, California, United States

Thirty one patients with 36 suspicious breast lesions underwent DCE-MRI prior to biopsies using a commercially available GRE-based TWIST sequence, a k-space undersampling and data sharing acquisition strategy. 18 or 20 s temporal resolution was obtained along with 1x1x1.4 mm3 spatial resolution. Single-frame DCE images were acquired immediately following TWIST DCE-MRI using a conventional full-k-space-sampling GRE sequence. Excellent intra- and inter-radiologist reader agreement in tumor morphology evaluation was achieved when comparing the last TWIST DCE image set with the conventional GRE images. Shutter-Speed model pharmacokinetic analyses of TWIST DCE-MRI data improved diagnostic accuracy compared to that obtained with ACR BIRADS Lexicon.

Jason McKellop¹, Alana Amarosa¹, Tess Clendenen², Melanie Moccaldi¹, Anne Zeleniuch-Jacquotte², Linda Moy¹, and Sungheon Kim^1
1Radiology, NYU School of Medicine, New York, NY, United States, ²Environmental Medicine, NYU School of Medicine, New York, NY, United States

It has been shown that background parenchymal enhancement (BPE) is strongly associated with the menstrual cycle as well as breast cancer. However, it has not been shown whether the kinetics of dynamic contrast enhancement is also associated with the menstrual cycle. In this study, BPE was measured in premenopausal patients with benign breast lesions and compared across menstrual phases. Statistically significant differences were found between the 4th week and early phases of menstruation in terms of early enhancement slope (p=.02) and delayed enhancement (p=.01).

Shelley Waugh¹, Richard Lerski¹, Luc Bidaut², and Alastair Thompson^3
1Medical Physics, Ninewells Hospital, Dundee, Angus, United Kingdom, ²University of Dundee, Dundee, United Kingdom, ³Department of Surgery, Ninewells Hospital, Dundee, United Kingdom

This study considers the use of texture analysis in a clinical setting for the identification and classification of breast cancer using MRI. Consideration was made of malignant vs. normal tissue and ductal vs. lobular carcinoma in a group of patients. Texture analysis was able to accurately differentiate between normal and malignant tissue in all cases. Differentiation between lobular and ductal cancer resulted in a 96% accuracy in 83 lesions. Both the co-occurrence (COM) and run-length matrix models were found to best describe invasive breast cancers, with the COM model resulting in the best differentiation between lobular and ductal cancers.

Luke Bonello¹, Giuseppe Petralia², Paul Summers², Roberto Di Filippi², Sara Raimondi², Giuseppe Renne², Giuseppe Curigliano², and Massimo Bellomi^1,2
1School of Radiology, University of Milan, Milan, Lombardia, Italy, ²European Institute of Oncology, Milan, Italy

We evaluated the correlation of the apparent diffusion coefficient (ADC) obtained from diffusion-weighted magnetic resonance imaging (DW-MRI) with clinico-pathological and molecular prognostic factors of breast cancer in 88 patients. Mean ADC for T4 tumors was higher than for T2 tumors (p<0.05); mean ADC for lymph node positive tumors (N1/N2) was lower than for lymph node negative tumors (p≤0.02). No significant differences in ADC between the 4 different gene profiling subgroups of breast cancer (Luminal A, Luminal B, HER-2, Triple Negative) were identified. Our results suggest that ADC can potentially serve as an additional non-invasive clinical prognostic factor in breast cancer.

Ryan Brown¹, Pippa Storey¹, Kelly Anne McGorty¹, Jose Raya¹, Daniel K Sodickson¹, Graham C Wiggins¹, and Linda Moy^1
1Radiology, New York University School of Medicine, New York, NY, United States

7T breast imaging promises substantial SNR gains over 3T, although significant image quality hurdles must be overcome to approach that required for diagnosis. 7T breast imaging has been hindered by uncharted sequence parameters and inconsistent fat suppression. We focused on customizing the essential clinical sequence, T1-weighted 3D FLASH with and without SPAIR fat suppression for high image quality and fibroglandular-fat contrast. This preliminary study demonstrated high resolution (0.2mm³) breast MRI at 7T with reliable SPAIR fat suppression throughout the entire breast volume despite B0 and B1+ inhomogeneity encountered at high field.

Wolfgang Bogner¹, Katja Pinker², Hubert Bickel², Marek Chmelik¹, Michael Weber², Thomas H Helbich², Siegfried Trattnig¹, and Stephan Gruber^1
1MR Center of Excellence, Department of Radiology, Medical University Vienna, Vienna, Vienna, Austria, ²Department of Radiology, Medical University Vienna, Vienna, Vienna, Austria

This work compares the diagnostic value of diffusion-weighted imaging based on single-shot echo planar imaging (ssEPI) and readout-segmented echo planar imaging (rsEPI) for 49 histopathologically proven breast lesions at 3T. Two independent readers visually assessed image quality and lesion conspicuity. Image properties and apparent diffusion coefficient (ADC) were quantified. The diagnostic accuracy was calculated based on an ADC threshold of 1.25×10-3mm²/s. RsEPI provided significantly higher image quality and lesion conspicuity than ssEPI by reducing geometric distortions, image blurring, and artifact level. Thereby, rsEPI reached a higher diagnostic accuracy of 96% for the differentiation of benign and malignant breast lesions.

Gene Young Cho^1,2, Linda Moy³, Scott DeGregorio³, Sungheon Kim¹, Melanie Moccaldi³, Jane Kwon¹, Steven Baete¹, Daniel K Sodickson¹, and Eric E Sigmund^1
1Radiology, NYU School of Medicine - Schwartz Center for Biomedical Imaging, New York, NY, United States, ²Sackler Institute of Graduate Biomedical Sciences, NYU School of Medicine, New York, NY, United States, ³Radiology, NYU School of Medicine, New York, NY, United States

Diffusion-weighted imaging (DWI) can play an important role in cancer management as it is sensitive to both vascular and cellular components of tumors via intravoxel incoherent motion (IVIM) analysis. Data processing algorithms for IVIM analysis can be optimized by filtered voxelwise analysis of the IVIM parametric maps. In this study, we used highly sampled DWI data to perform voxel based analysis on biexponential IVIM parameters in a cohort of breast cancer patients at 3T and compared results with integrated ROI analysis, histological cancer subtype, and hormone receptor status.

Rani G Sah¹, Uma Sharma¹, Rajinder Parshad², Vurthaluru Seenu², and Naranamangalam Raghunathan Jagannathan^1
1Department of NMR & MRI Facility, All India Institute of Medical Sciences, New Delhi, Delhi, India, ²Department of Surgery, All India Institute of Medical Sciences, New Delhi, Delhi, India

Apparent diffusion coefficient (ADC) was determined in breast tissue of 203 women including 141 patients with infiltrating ductal carcinoma, 34 benign and 28 normal volunteers using diffusion MRI at 1.5 T. The mean ADC of malignant lesion was significantly lower (1.03 ± 0.18) compared to benign (1.63 ± 0.28) and normal (1.80 ± 0.12) breast tissues. The high cell proliferation in malignancy increases the cellular density, which decreases the extra-cellular volume water fraction and thereby decreases the ADC in malignancy. Cut-off values using ROC was also determined to differentiate malignant, benign and normal breast tissues.

Liuquan Cheng¹, Xiru Li², and Mei Liu^3
1Radiology, Chinese PLA General Hospital, Haidian, Beijing, China, ²Surgery, Chinese PLA General Hospital, Haidian, Beijing, China, ³Pathology, Chinese PLA General Hospital, Haidian, Beijing, China

A comprehensive diagnostic protocol integrating DWI-ADC value and dynamic contrast enhancement (DCE) MR mammography was developed to evaluate the breast diseases. It improved both the sensitivity and specificity of the prediction when compared to DCE only. The ACR BI-RADS categories based on the comprehensive diagnostic protocol were well correlated with the pathological grades.

Elizabeth AM O'Flynn¹, Veronica A Morgan², Sharon L Giles³, Erica Scurr³, Nina Tunariu³, and Nandita M deSouza^3
1Cancer Imaging Centre, Institute of Cancer Research and the Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom, ²Cancer Imaging Centre, Institute of Cancer Research and the Royal Marsden NHS Foundation Trust, United Kingdom, ³Cancer Imaging Centre, Institute of Cancer Research and the Royal Marsden NHS Foundation Trust

ADC values in the normal fibroglandular breast tissue for women at high risk of developing breast cancer have not yet been documented. 24 high risk women and 12 volunteers were scanned during the proliferative phase of the menstrual cycle. Pixel-by-pixel ADChigh (omitting b=0) was computed from a monoexponential fit. There was a significant difference between the mean ADChigh value (x10-6mm2/s) in high risk women (1893+/-251) compared to normal risk women (1689+/-237)(p=0.036). The ADChigh centiles in high risk women were also significantly higher than volunteers in the proliferative phase, although skewness was similar for the two groups (p=0.165).

Jun Chen¹, Roger Grimm¹, Kevin Glaser¹, Kugel Jennifer¹, Kay Pelletier¹, and Richard Ehman^1
1Radiology Department, Mayo Clinic, Rochester, MN, United States

Women in the US have a 1 in 8 chance of developing breast cancer. Contrast-enhanced MRI has a high sensitivity for breast cancer, but low specificity. Breast MRE is a promising tool for detecting malignancies, but the requirement for direct contact between the driver and the breast has several disadvantages. We have developed a noncontact driver design to avoid these problems and a 3D GRE MRE acquisition to image both breasts simultaneously. The results from 6 healthy female volunteers demonstrate the feasibility of these methods and motivate future studies of MRE for breast cancer diagnosis.

El-Sayed H Ibrahim¹, Derek Hamlin¹, Jean Shaffer¹, Romanie C Nichols², and Nancy Mendenhall^2
1Department of Radiology, University of Florida, Jacksonville, FL, United States, ²University of Florida Proton Therapy Institute, Jacksonville, FL, United States

Digital rectal examination, serum prostate specific antigen (PSA), and biopsy-derived Gleason score (GS) do not provide detailed information about tumor tissue characterization. In this work, a multi-parametric MRI exam, that includes T2-weighted, diffusion-weighted imaging, and MR spectroscopy, was developed, optimized for 3.0T, and tested on intermediate-risk patients, who show wide range of treatment outcomes. The results showed good (but not exact) agreements between different MRI/MRS measurements; small inter- and intra-observer variabilities; and weak correlations between MRI/MRS measurements and PSA/GS results. The detailed information by the proposed exam is expected to result in better treatment outcomes by optimizing individualized treatment plans.

Marnix C. Maas¹, Jurgen J. Fütterer¹, and Tom W.J. Scheenen^1
1Radiology, Radboud University Nijmegen Medical Centre, Nijmegen, GLD, Netherlands

Quality assurance measurements were performed for a prospective study involving 10 institutions, aimed at proving the diagnostic accuracy of 3T multi-modality MR imaging including high-resolution T2-weighted imaging, DWI, DCE and 3D MRSI in distinguishing carcinoma from other prostate tissue. ADCs, T₁s and (Choline + Spermine + Creatine) / Citrate ratios were compared in 5 institutions in this preliminary study. Modest but statistically significant differences between institutions were found in all 3 categories. Whether this warrants corrections of the quantitative data for the prospective study will be subject to further investigation.

Tsutomu Tamada¹, HIroki Higashi¹, Teruki Sone¹, Yoshimasa Jo², Atsushi Higaki¹, Akihiko Kanki¹, Akira Yamamoto¹, and Katsuyoshi Ito^1
1Radiology, Kawasaki Medical School, Kurashiki City, Okayama, Japan, ²Urology, Kawasaki Medical School, Kurashiki City, Okayama, Japan

We evaluated the utility of T2WI, DCE-MRI and DWI for detecting prostate cancer with total serum PSA levels of 4-10 ng/ml (gray zone). Combined T2WI, DWI, and DCE-MRI findings would be potentially useful for detecting and managing prostate cancer, even if performed for patients with gray-zone PSA levels.

Daniel Jason Aaron Margolis¹, Karim Chamie², Shyam Natarajan³, Nelly Tan¹, Jiaoti Huang⁴, and Robert Reiter^2
1Department of Radiology, UCLA David Geffen School of Medicine, Los Angeles, CA, United States, ²Department of Urology, UCLA David Geffen School of Medicine, ³Department of Bioengineering, UCLA David Geffen School of Medicine, Los Angeles, CA, United States, ⁴Department of Pathology, UCLA David Geffen School of Medicine

Determination of eligibility for active surveillance (AS) is based on serum and biopsy findings. We reviewed prostatectomy specimens of 104 men who had presurgical endorectal coil MRI to determine who would have been appropriate for AS based on clinical parameters alone and after adding imaging information, using surgical pathology as the standard of reference. The sensitivity and negative predictive value increased markedly by the addition of the apparent diffusion coefficient of the index tumor using a cut-off of 850 square microns/sec to the clinical parameters alone. This improves diagnostic confidence for men seeking AS.

Kemal Tuncali¹, Andriy Fedorov¹, Fiona M Fennessy¹, Sandeep N Gupta², Junichi Tokuda¹, Sam Song¹, Nobuhiko Hata¹, Robert V Mulkern^1,3, and Clare M Tempany^1
1Radiology, Brigham and Women's Hospital, Boston, MA, United States, ²GE Global Research Center, Niskayuna, NY, United States, ³Children's Hospital, Boston, MA, United States

Twelve patients enrolled in our study had 3T multiparametric MRI (mpMRI) of the prostate. T2WI, ADC maps from DWI, and pharmacokinetic maps from DCE were evaluated by three radiologists who selected targets and rated them for degree of suspicion of cancer. 67 targets were selected and biopsied using transperineal approach and 3T MRI guidance. Biopsy outcome of cancer was correlated with mpMRI ratings. Calculated ROC curves and test metrics suggested that at least two parameters, ADCb500 and v_e, should be combined to select targets for imaging-guided prostate biopsy in order to increase sensitivity and PPV for detecting prostate cancer.

Xin Li¹, Ryan A Priest^2,3, William J Woodward¹, Ian J Tagge¹, Faisal Siddiqui^2,3, Wei Huang¹, William D Rooney¹, Tomasz M Beer^4,5, Mark G Garzotto^6,7, and Charles S Springer^1,5
1Advanced Imaging Research Center, Oregon Health & Science University, Portland, Oregon, United States, ²School of Medicine, Oregon Health & Science University, Portland, Oregon, United States, ³Radiology, Oregon Health & Science University, Portland, Oregon, United States, ⁴Hematology/Oncology, Oregon Health & Science University, Portland, Oregon, United States, ⁵Knight Cancer Institute, Oregon Health & Science University, Portland, Oregon, United States, ⁶Urology, Oregon Health & Science University, Portland, Oregon, United States, ⁷Portland VA Medical Center, Portland, Oregon, United States

Standard Dynamic-Contrast-Enhanced MRI (DCE-MRI) pharmacokinetic modeling assumes inter-compartmental water exchange kinetics to be infinitely fast. Though it is physically impossible that such processes be truly infinitely rapid, in many tissue regions they seem so as far as DCE MRI water signals are concerned: the exchange MR systems remain in their fast-exchange-limit [FXL] conditions. However, there are tissue loci where these systems transiently depart their FXLs during DCE-MRI contrast reagent (CR) bolus passage. Using an exchange-sensitized DCE-MRI acquisition and analytical “shutter-speed” model, we demonstrate the feasibility of prostate cancer detection with water exchange allowed DCE-MRI pharmacokinetic modeling.

Xin Li¹, Ryan A Priest^2,3, William J Woodward¹, Faisal Siddiqui^2,3, Tomasz M Beer^4,5, Mark G Garzotto^6,7, William D Rooney¹, and Charles S Springer^1,5
1Advanced Imaging Research Center, Oregon Health & Science University, Portland, Oregon, United States, ²School of Medicine, Oregon Health & Science University, Portland, Oregon, United States, ³Radiology, Oregon Health & Science University, Portland, Oregon, United States, ⁴Hematology/Oncology, Oregon Health & Science University, Portland, Oregon, United States, ⁵Knight Cancer Institute, Oregon Health & Science University, Portland, Oregon, United States, ⁶Urology, Oregon Health & Science University, Portland, Oregon, United States, ⁷Portland VA Medical Center, Portland, Oregon, United States

For prostate Dynamic-Contrast-Enhanced MRI (DCE-MRI), water exchange effects become more influential with high interstitial contrast reagent concentration. Using real data and simulations based on region-of-interest prostate DCE-MRI, pharmacokinetic parameters extracted from the fast-exchange-limit-constrained (FXL-c) Standard pharmacokinetic modeling (SM) and versions of “shutter-speed” models (SSM) were investigated. The first generation SSM (SSM1) was found sensitive to inter-compartmental water exchange effects. With inclusion of the mean intracellular water lifetime biomarker, the effect on K^trans (a CR extravasation rate measure) can be quantified using the difference of its magnitudes returned by sequential SSM1 and SM analyses of a given DCE-MRI time-course.

Greetje Groenendaal^1,2, Alie Borren², Maaike Moman², Evelyn Monninkhof³, Paul van Diest⁴, Mariëlle Philippens², Marco van Vulpen², and Uulke van der Heide^1,2
1Radiotherapy, NKI-AVL, Amsterdam, Netherlands, ²Radiotherapy, UMC Utrecht, Utrecht, Netherlands, ³Julius Centre, UMC Utrecht, Utrecht, Netherlands,⁴Pathology, UMC Utrecht, Utrecht, Netherlands

For the purpose of focal boost radiotherapy a voxel-wise tumor prediction model was developed for the prostate peripheral zone (PZ) using diffusion weighted imaging and dynamic contrast-enhanced MRI. This model showed an area under the receiver operating curve of 0.89 on a voxel level when validated with whole-mount section histopathology. The voxel size used was 2.5x2.5x2.5 mm3 The model output can be used to define different risk levels for tumor presence, which in turn could serve as an input for dose planning. In this way the robustness of tumor delineations for focal boost therapy can be greatly improved.

Chaitanya Kalavagunta¹, Christopher A Warlick², Xiangmin Zhou¹, Xiufeng Li¹, Joseph S Koopmeiners³, Jonathan C Henriksen⁴, Andrew D Johnson⁴, Stephen Schmechel⁴, and Gregory J Metzger^1
1Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, United States, ²Department of Urologic Surgery, University of Minnesota, Minneapolis, Minnesota, United States, ³Department of Biostatistics, University of Minnesota, Minneapolis, Minnesota, United States,⁴Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, Minnesota, United States

Multi parametric maps of anatomic, vascular and metabolic data of prostate cancer (PCa) and benign tissue acquired using multiple imaging modalities can yield improved discrimination of the extent and aggressiveness of PCa. An important step in this direction is the registration of in-vivo MR images with histopathological sections from prostatectomy. In this study we use this histological ground truth tumor localization information along with quantitative histological data to find correlation between tumor nuclear densities with T2 and ADC measurements made using registered peripheral zone tumor regions.

Chaitanya Kalavagunta¹, Xiangmin Zhou¹, Jonathan C Henriksen², Stephen Schmechel², and Gregory J Metzger^1
1Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, United States, ²Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, Minnesota, United States

In this study we register in-vivo prostate MR with Pseudo Whole Mount histological sections by a non-rigid registration method using local affine transformation guided by internal structures.

Matthias C Roethke¹, Timur Kuru², Patrik Zamecnik¹, Boris Hadaschik², and Heinz-Peter Schlemmer^1
1Radiology, German Cancer Research Center, Heidelberg, Germany, ²Urology, University Hospital, Heidelberg, Germany

Commonly, prostate cancer (PCa) is diagnosed using a random sample biopsy of the prostate under transrectal ultrasound guidance (TRUS). Multiparametric MRI increased detection rate, especially of smaller lesions. Since these lesions are difficult to target with conventional TRUS biopsy, MR-guided biopsy procedures have been established. Recent technical developments integrated MRI images with ultrasound with live fusion. Most of these techniques use a transrectal approach for the biopsy. In this study, we evaluated a MR/TRUS live fusion system with a transperineal biopsy approach. We found an excellent accuracy of targeting smaller lesions <0.5cc in a phantom model.

Mariska P. Luttje¹, Catalina S. Arteaga de Castro¹, Marco van Vulpen¹, Alie Borren¹, Jan J.W. Lagendijk¹, Peter R. Luijten¹, Dennis W.J. Klomp¹, and Uulke A. van der Heide^2
1Imaging Division, University Medical Center, Utrecht, Utrecht, Netherlands, ²The Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands

Hypoxia is an important marker for the resistance of cancerous tissue to both radiotherapy and chemotherapy. By using MRI, R2* (1/T2*) can be measured, correlated with the pO₂ in prostate tissue. We investigated the use of R2* mapping at 3T. In addition, feasibility of R2* mapping has been investigated at 7T, while taking care of significant susceptibility effects. In all patients tumor regions had lower R2* values than the surrounding healthy prostate tissue. Higher fields strengths with susceptibility matching are expected to bring the R2* values closer to the intrinsic tissue R2*, which may improve correlation to pO₂.

He Zhu^1,2, Kelly Myers¹, Michael Schar^1,3, and Peter B. Barker^1,2
1Radiology, Johns Hopkins University, Baltimore, MD, United States, ²FM Kirby Research Center, Kennedy Krieger Institute, Baltimore, MD, United States,³Philips Healthcare, Cleveland, OH, United States

A 3D PRESS MRSI sequence was developed for tCho detection in the breast. This novel sequence consisted of a dualband pre-saturation and dualband dephasing before echo formation (BASING). Both methods, either method and unsuppressed scans were tested in 2D MRSI to evaluate their performances by normalizing residual signals to unsuppressed signals (water and lipid). The combination achieved reduction factors of less than 1% for both water and fat. As 3D MRSI was implemented with the combination, tCho signal was detected in healthy breast tissues.

Rani G Sah¹, Uma Sharma¹, Rajinder Parshad², Vurthaluru Seenu², and Naranamangalam Raghunathan Jagannathan^1
1Department of NMR & MRI Facility, All India Institute of Medical Sciences, New Delhi, Delhi, India, ²Department of Surgery, All India Institute of Medical Sciences, New Delhi, Delhi, India

High choline-containing metabolite (tCho) differentiates malignant breast tissues from benign lesions. However, tCho is also seen in normal and in lactating breast tissues and hence there is need for determination of absolute tCho concentration. Early breast cancer patients had significantly higher tCho concentration compared to locally advanced breast cancer patients. The cut-off values of tCho concentration obtained gave clear differentiation between early, locally advanced breast cancer, benign lesions and normal breast tissues. The results help in the progress of breast proton MRS from the stage of investigational research to the role that could be efficaciously used in routine clinical practice.

Rani G Sah¹, Uma Sharma¹, Rajinder Parshad², Vurthaluru Seenu², Sandeep Mathur³, and Naranamangalam Raghunathan Jagannathan^1
1Department of NMR & MRI Facility, All India Institute of Medical Sciences, New Delhi, Delhi, India, ²Department of Surgery, All India Institute of Medical Sciences, New Delhi, Delhi, India, ³Department of Pathology, All India Institute of Medical Sciences, New Delhi, Delhi, India

tCho concentration and tumor volume were determined using in-vivo MRS and MRI in triple negative (TN), non-triple negative (nTN) and triple positive (TP) breast cancer patients. TN patients were of younger age (28 to 39 yrs) compared to nTN and TP patients. tCho concentration was significantly lower in TN (3.8 ± 1.0 mmol/kg) compared to nTN (4.7 ± 3.1 mmol/kg) and TP (5.4 ± 2.5 mmol/kg) patients while tumor volume was similar in all patients. Reduced choline kinase expression and high degradation of phosphatidylcholine might have led to lower tCho in TN patients; however, this warrants further investigations.

Naranamangalam Raghunathan Jagannathan¹, Rani G Sah¹, Uma Sharma¹, Vurthaluru Seenu², Sandeep Mathur³, and Rajinder Parshad^2
1Department of NMR & MRI Facility, All India Institute of Medical Sciences, New Delhi, Delhi, India, ²Department of Surgery, All India Institute of Medical Sciences, New Delhi, Delhi, India, ³Department of Pathology, All India Institute of Medical Sciences, New Delhi, Delhi, India

The total choline (tCho) concentration and tumor volume were determined using in-vivo MRS and MRI in 73 invasive ductal carcinoma patients at 1.5T with known HER2/neu, ER and PR status. The tCho concentration in HER2+ (3.8±1.2), ER+ (5.0±2.7) and PR+ (5.0±2.8) patients was not significantly different as compared to HER2- (4.4 ±2.9), ER-(4.3±2.8) and PR- (4.7±2.9) patients. Tumor volume in ER+ patients was significantly lower compared to patients with ER-. The increase in the volume may be attributed to higher microvessels density in ER- patients. Also no significant difference in tCho concentration and tumor volume was observed with PR status.

Yuan Le¹, Randall Kroeker², Hal D Kipfer³, Shadie S Majidi³, Stephanie Holz³, Brian Dale², and Chen Lin^1
1Radiology, Indiana University, Indianapolis, Indiana, United States, ²Siemens Health Care, Erlangen, Germany, ³Radiology and Imaging Science, Indiana University, Indanapolis, Indiana, United States

A novel pulse sequence (TWIST DIXON)which combines k-space data sharing strategy used in Time-resolved angiography With Stochastic Trajectories (TWIST) with fat and water separation by dual-echo DIXON was evaluated for breast DCE MRI application. In this study, we compared the image quality of TWIST DIXON sequence with the conventional method of fat suppressed 3D spoiled gradient echo sequence in a group of clinical patients. The results demonstrate that TWIST DIXON technique provides significantly higher perceived SNR, more accurate fat suppression and overall better image quality.

Yuan Le¹, Christian Geppert², Randall Kroeker², Brian Dale², and Chen Lin^1
1Radiology, Indiana University, Indianapolis, Indiana, United States, ²Siemens Healthcare, Erlangen, Germany

Computer simulations have been performed to study the error caused by TWIST view sharing on the measured breast dynamic contrast enhancement. A digital breast phantom was generated with spherical lesions of different sizes and three types of enhancement curves: persistent, plateau and wash-out. The results suggest that for lesion larger than 5mm, TWIST view sharing with greater 33% central k-space region and more than 50% peripheral sampling density would preserve the curve type information. Such imaging parameters are suitable for clinical breast DCE MRI.

Ileana Hancu¹, Robert Lenkinski², and Seung-Kyun Lee^1
1GE Global Research Center, Niskayuna, NY, United States, ²UT Southwestern, Dallas, TX, United States

A population study is presented, analyzing the performance of thirteen shimming approaches for 3T breast MRI exams. Best shim results were obtained by excluding the back and the heart of the subjects. The most homogeneous breast B0 was obtained while shimming over the breast area using 3D B0 maps; this approach, however, may prove impractical, due to the long time needed for the acquisition of B0 maps. Among the multi-plane approaches considered, a rectangular shim ROI (relying on the acquisition of only 2-3 planes of B0 data) provided the best compromise between short acquisition times and homogeneous B0.

Thomas D O’Sullivan¹, Anais Leproux¹, Jeon-Hor Chen², Orhan Nalcioglu², Bruce J Tromberg¹, and Min-Ying Lydia Su^2
1Laser Microbeam and Medical Program, Beckman Laser Institute and Medical Clinic, University of California, Irvine, CA, United States, ²Center for Functional Onco-Imaging, Department of Radiological Sciences, University of California, Irvine, CA, United States

The breast density from the contralateral normal breast of patients receiving neoadjuvant chemotherapy was measured using two imaging modalities, MRI and broadband Diffuse Optical Spectroscopic Imaging (DOSI). The fibroglandular tissue volume measured on MRI was compared to the concentration of water, lipid, oxyhemoglobin, deoxyhemoglobin, and oxygen saturation measured by DOSI. The strongest MRI-DOSI correlation was found with water and deoxyhemoglobin concentration. Chemotherapy is known to decrease density due to suppressed ovarian function. The change of density may provide a prognostic marker of this process. Compared to MRI, DOSI may provide a less expensive bed side imaging device for this purpose.

Shanshan Xu^1,2, Ming-Jung Kim^3,4, Jeon-Hor Chen^3,5, Orhan Nalcioglu³, Bruce J. Tromberg^1,2, Enrico Gratton^1,2, and Min-Ying L. Su^3
1Biomedical Engineering Department, University of California, Irvine, California, United States, ²Beckman Laser Institute, Irvine, California, United States,³Center for Functional Onco-Imaging, Department of Radiological Sciences, University of California, Irvine, California, United States, ⁴Department of Radiology, Yonsei University College of Medicine, Seoul, Korea, ⁵Department of Radiology, China Medical University Hospital, Taichung, Taiwan

A quantitative correlation study was performed between optical parameters measured by Diffuse Optical Spectroscopy Imaging (DOSI) and DCE-MRI kinetic parameters in breast cancer patients undergoing neoadjuvant chemotherapy. In patients who achieved pathologic complete response, DOSI parameters (oxyhemoglobin, deoxyhemoglobin and oxygen saturation) were strongly correlated with Ktrans and kep. This relationship was changed or became weaker after chemotherapy. In patients who did not achieve pCR, there was no significant correlation between any DOSI and DCE-MRI parameters in both baseline and follow-up studies. Different findings were seen between these two groups, suggesting that these relationships may be explored for predicting therapeutic response.

S. L. Bowen¹, R. T. Seethamraju², B. L. Niell³, and C. Catana^1
1Radiology, Massachusetts General Hospital, Charlestown, MA, United States, ²Siemens Healthcare, United States, ³Radiology, Massachusetts General Hospital, Boston, MA, United States

The specificity of MR in the detection of breast cancer has been shown to be significantly increased when combined with spatially fused F-18-fluorodeoxyglucose PET images. Several factors, however, limit the efficacy of spatially registering images acquired from standalone MR and PET (i.e. limited registration accuracy). Combined MR-PET scanning is expected to significantly improve spatial registration between the separate modalities and as such may improve patient care. Since no breast coil has been developed for the Biograph mMR MR-PET (Siemens Healthcare), we aimed to optimize image quality for breast imaging for MR and PET components using a conventional breast coil.

Katja Pinker-Domenig¹, Hubert Bickel², Wolfgang Bogner³, Stephan Gruber³, Benedikt Brück², Heinrich Magometschnigg², and Thomas H Helbich^1
1Dept. of Radiology, Division of Molecular and Gender Imaging, Medical University Vienna, Vienna, Vienna, Austria, ²Dept. of Radiology, Medical University Vienna, Vienna, Vienna, Austria, ³Dept. of Radiology, Medical University Vienna, MR Centre of Excellence, Vienna, Vienna, Austria

Combined use of 3D-1H-MRSI, DWI, CE-MRI and PET in the assessment of breast lesions enabled an accurate breast cancer diagnosis with improved sensitivity, specificity and diagnostic accuracy.

Frederick Kelcz¹, Leah Henze², Alisa K Johnson¹, and Walter F. Block^2
1Radiology, University of Wisconsin, Madison, WI, United States, ²Medical Physics, University of Wisconsin, Madison, WI, United States

In an ongoing study, for 18 patients with breast lesions we tested whether advanced MRI methods (DWI, MRS, BOLD, fast temporal resolution) added to confidence in diagnosis when compared to a conventional complete breast MRI. Thus far we have found that the most common outcome is that the advanced methods are either concordant with, but do not add to the confidence in conventional diagnosis; or provide potentially confounding results that may lower confidence in the conventional findings.

Mangala Srinivas¹, Fernando Bonetto², Erik Aarntzen¹, Cornelius JA Punt³, Otto C Boerman⁴, Wim J Oyen⁴, Pauline Verdijk⁵, Carl Figdor¹, Arend Heerschap⁶, and Jolanda de Vries^1
1Tumor Immunology, RUNMC, Nijmegen, Gelderland, Netherlands, ²INTEC-CONICET, Argentina, ³Oncology, RUNMC, Netherlands, ⁴Nuclear Medicine, RUNMC, Netherlands, ⁵Vaccinology, RIVM, Netherlands, ⁶Radiology, RUNMC, Netherlands

Dendritic cell (DC) vaccination is an emerging therapy for cancer patients. Injected DCs must migrate to lymph nodes for immunostimulation. However, migration is extremely poor (<4% upon intradermal injection). Expense, logistic and sensitivity issues hinder clinical optimization. Therefore, we developed a quantitative, in vitro 19F MR-based migration assay using 3D collagen scaffolds and tissue samples, and validate our results using clinical data acquired using scintigraphy on 111In-labeled DCs. We found a strong relationship between migration rates and total cell numbers. We also studied the effect of different components such as cytokines on migration both in vitro and in the clinic.

Mathilde Wagner^1,2, Léon Maggiori^2,3, Maxime Ronot^1,2, Valérie Vilgrain^1,2, Yves Panis^2,3, and Bernard Edgar Van Beers^1,2
1Radiology, Beaujon hospital, Clichy, France, ²INSERM U773, Clichy, France, ³Colorectal surgery, Beaujon hospital, Clichy, France

The purpose of the study was to compare the sensitivity of T2-weighted and diffusion-weighted imaging (DWI) in detecting colorectal liver metastases in a rat model with precise histopathological correlations. DWI detection rate was higher than T2-weighted imaging detection rate (p = 0.001). After stratification according to metastasis average diameter, DWI detection rate was higher than T2-weighted imaging detection rate for metastases with a diameter between 0.3 and 1.2 mm, but not for metastases smaller than 0.3 mm or larger than 1.2 mm. This study shows the better sensitivity of DWI relative to T2-weighted imaging for detecting liver metastases.

Daisuke Suzuki^1,2, Masayuki Yamaguchi¹, Toshihiro Furuta^1,3, Ryutaro Nakagami^1,4, Yasuo Okuyama², Kohki Yoshikawa², and Hirofumi Fujii^1
1Functional Imaging Division, National Cancer Center Hospital East, Kashiwa, Chiba, Japan, ²Graduate Division of Health Sciences, Komazawa University, Setagaya, Tokyo, Japan, ³Department of Radiology, The University of Tokyo, Tokyo, Japan, ⁴Research Fellow of the Japan Society for the Promotion of Science, Japan

Superparamagnetic iron oxide (SPIO)-enhanced MRI can differentiate sentinel lymph node (SLN) metastasis from inflammation by the presence and absence of high signals in lymph nodes according to previous reports. In this study, we revealed that high-signal areas mimicking metastatic foci could appear even in inflamed lymph nodes. We also investigated the pathogenesis of residual high signals in analyzing the relationship between SPIO doses and high signal areas using an animal model, and found that expanded paracortices not containing macrophages and inhomogeneous distribution of macrophages may cause these high signals in inflamed lymph nodes.

Amr Morsi^1,2, Avital Gaziel³, Hana Baig¹, John G. Golfinos⁴, Eva Hernando-Monge³, and Youssef Zaim Wadghiri^5
1Radiology, NYU Langone Medical Center, New York, NY, United States, ²Neurosurgery, NYU Langone Medical Center, NY, NY, United States, ³Pathology, NYU Langone Medical Center, New York, NY, United States, ⁴Neurosurgery, NYU Langone Medical Center, New York, NY, United States, ⁵Radiology, NYU School of Medicine, New York, NY, United States

95% of patients with melanoma brain metastases (B-Mets) succumb to their disease within six months of diagnosis. The scarcity of physiologically relevant in vivo models and an accurate non-invasive modality to monitor tumorigenesis hinders the investigation of melanoma brain tropism, and response to therapies. We used Ex vivo MRI to characterize the tumor growth in large cohorts of B16F10 and 5B1 melanoma (B-Mets) mouse models. Ex vivo imaging proved to be a high throughput imaging modality. Data reveals significant changes in growth pattern between both models and suggests that the 5B1 model is more reliable and relevant for preclinical studies.

Nai-Wei Yao^1,2, Chiao-Chi V. Chen¹, Yi-Hua Hsu¹, Hsiu-Ting Lin¹, Jeou-Yuan Chen¹, and Chen Chang^1
1Institute of Biomedical Sciences, Academic Sinica, Taipei, Taiwan, ²Department of Zoology, National Taiwan University, Taipei, Taiwan

Osteopontin (OPN) expression in glioblastoma has been correlated with diverse function of tumor development. The major goal of the present study is to elucidate the involvement of OPN in tumor angiogenesis, concerning both functional and structural characteristics of the tumor vasculature. The role of OPN in tumor angiogenesis was studied using a rat C6 glioma model with OPN knockdown (KD). The functional and structural aspects of the tumor vasculature were investigated by blood oxygenation level dependant (BOLD) MRI with gas challenge and steady-state contrast-enhanced (SSCE) MRI, respectively. The study directly provided in vivo evidence on the role of OPN in the functionality of tumor vasculature.

Eugene Kim¹, Paul T. Winnard, Jr.^2,3, Venu Raman^2,3, Zaver M. Bhujwalla^2,3, and Arvind P. Pathak^2,3
1The Whitaker Biomedical Engineering Institute, Johns Hopkins University, Baltimore, MD, United States, ²The Johns Hopkins University In Vivo Cellular and Molecular Imaging Center Program, ³The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University

Hypoxia profoundly affects the tumor microenvironment, aggressiveness and invasiveness, and response to therapy. Here, we investigated how carbogen (95% O2/5% CO2) breathing effects oxygenation in hypoxic tumor regions by employing a unique combination of BOLD MRI and molecular imaging of a hypoxia-inducible fluorescent human breast cancer model. Our results showed heterogeneous MR-measured ΔR₂* responses to carbogen in hypoxic regions identified by a fluorescent HIF-1 reporter in orthotopic tumors. The response was significantly greater in one tumor than the other, suggesting that carbogen breathing may alleviate hypoxia but with high intra- and inter-tumor variability.

Ning Jin^1,2, Yang Guo², Rachel Klein², and Andrew C Larson^2,3
1Siemens Healthcare, Columbus, OH, United States, ²Department of Radiology, Northwestern University, Chicago, IL, United States, ³Robert H. Lurie Comprehensive Cancer Center, Chicago, IL, United States

Hepatocellular carcinoma (HCC) are typically hypervascular tumors. Angiogenesis, the process by which new vessels develop from pre-existing microvessels, plays an important role in HCC progression. The aim of our study was to evaluate MR vessel imaging techniques for tumor angiogenesis quantification in rat models of HCC. Two different rodent HCC models were chosen: McA-RH7777 Morris hepatoma model and N1-S1 hepatoma model.

Wenlian Zhu¹, Yoshinori Kato¹, and Dmitri Artemov^1
1JHU ICMIC Program, Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States

 

Isabelle Iltis¹, Jeunghwan Choi², Emily Colonna¹, Manda Vollmers¹, Mithun Shenoi³, Joel Slaton⁴, John Bischof², and Gregory J Metzger^1
1Radiology, University of Minnesota - CMRR, Minneapolis, MN, United States, ²Mechanical Engineering, University of Minnesota - Bioheat and Mass Transfer Laboratory, Minneapolis, MN, United States, ³Biomedical Engineering, University of Minnesota - Bioheat and Mass Transfer Laboratory, Minneapolis, MN, United States, ⁴Department of Urologic Surgery, University of Minnesota, Minneapolis, MN, United States

Cryosurgery for the treatment of tumors is greatly improved by using TNF-α-based nanoparticles injected prior to the procedure. The effect of such molecules on the tumor is called “preconditioning” and largely remains to be characterized. In this work, we assess the feasibility of DCE-MRI to detect “preconditioning” of the tumor in vivo in a mouse model of prostate cancer. A significant effect of preconditioning on the tumor, and to a lesser extent the muscle, could be detected, warranting further studies that will allow us to monitor, but also to improve the understanding of the underlying mechanisms of preconditioning in vivo.

Hans-Juergen Raatschen¹, Christian Olaf Schoenfeld¹, Birgit Hotz², and Hubert G Hotz^2
1Diagnostic and Interventional Radiology, Hannover Medical School, Hannover, Germany, ²General, Vascular and Thoracic Surgery, Charité Universitätsmedizin Berlin, Berlin, Germany

Pharmacokinetic analysis of dynamic, albumin-(Gd-DTPA)-enhanced MRI appears to be a sensitive method to quantify angiogenesis-inhibiting effects of bevacizumab and suramin well before changes in tumor volumes can be detected and thus might be a useful addition to morphological MRI in the follow-up of antiangiogenic cancer treatment.

Feng Chen¹, Kaier Zheng², Frederik De Keyzer¹, Raymond Oyen¹, and Yicheng Ni^1
1Radiology, University of Leuven, Leuven, Brabant, Belgium, ²Radiology, BenQ Medical Center, Nanjing Medical University, Nanjing, China

Our randomized, controlled study demonstrated that the combination of a vascular disrupting agent with an antiangiogenic (ZdTha) enhanced overall antitumor efficacy due to synergistic effects. ZdTha induced cumulative tumor necrosis, and thus, significantly decreased tumor volume and reduced the viable tumor rim; ZdTha also reduced tumor vessel permeability and improved tumor hemodynamic indexes via a transient normalization of tumor vasculature induced by Tha. Finally, the ADC value at 12 d was shown to be an independent predictor of changes in tumor volume. These therapeutic effects were successfully tracked and evaluated with in vivo multiparametric MRI.

Albert P Chen¹, Yi-Ping Gu², Michelle Ladouceur-Wodzak², William Chu³, and Charles H Cunningham^2,4
1GE Healthcare, Toronto, ON, Canada, ²Imaging Research, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, ³Radiation Oncology, University of Toronto, Toronto, ON, Canada, ⁴Medical Biophysics, University of Toronto, Toronto, ON, Canada

Changes in the [1-¹³C]lactate signal observed in vivo following injection hyperpolarized [1-¹³C]pyruvate have been shown to be a marker for early treatment response. With the emergence of hypofractionated radiation therapy, this tool for detecting early response to treatments non-invasively may be very important for management of certain cancers. In this study, early and significant change in hyperpolarized [1-¹³C]lactate signal in tumors was observed after a single, large dose of radiation therapy in a human breast cancer model.

Eline Vos¹, Mark van Uden¹, Jurgen Fütterer¹, and Tom Scheenen^1
1Radiology, Radboud University Nijmegen Medical Center, Nijmegen, Netherlands

The SNR of a new investigational two-channel receive endorectal coil for prostate imaging was compared to a current clinically available single-loop receive coil. Four patients underwent a staging protocol with both coils at 1.5T and four patients at 3T. The investigational coil showed higher signal to noise ratio, up to a distance of 57 mm and 40 mm from the coil for 1.5 and 3T respectively. Thus, for staging of the dorsal part of the prostate the investigational coil has a benefit over the currently available coil.

Tariq Shah¹, Flonne Wildes¹, and Zaver M Bhujwalla^1
1Radiology, Johns Hopkins University, Baltimore, Maryland, United States

Lymph node metastasis is associated with poor prognosis in prostate cancer. Since lymphatic vessels serve as the primary route for this spread it is important to determine the role of lymphatic endothelial- prostate cancer cell interactions in the invasion of lymphatic vessels. Here we investigated the role of this interaction in invasion and degradation of extracellular matrix (ECM) in our MR compatible cell perfusion assay and determined associated metabolic changes. We observed significantly increased invasion by PC-3 human prostate cancer cells when in the proximity of lymphatic endothelial cells suggesting that this interaction plays a critical role in lymphatic metastasis.

Meiyun Wang¹, Dapeng Shi¹, Zilun Liu¹, Yan Bai¹, and Yongming Dai^2
1Department of Radiology, Henan Provincial People's Hospital, Zhengzhou, Henan, China, ²Siemens Healthcare

This study was to investigate the value of HR-SWI in prostatic diseases and the results showed that SWI is not only helpful in the differential diagnosis between prostatic cancer and benign prostatic hyperplasia and prostatitis but also sensitive in detecting calcification within prostate.

John P. Mugler, III¹, Martin Requardt², Karl Engelhard³, Talissa A. Altes¹, Dominik Paul², and Berthold Kiefer^2
1Radiology & Medical Imaging, University of Virginia, Charlottesville, VA, United States, ²Siemens Healthcare, Erlangen, Germany, ³Martha Maria Medical Center, Nuremberg, Germany

An inner-volume version of 3D TSE (SPACE) was implemented to permit high-resolution 3D imaging of the prostate in conjunction with relatively long repetition times for improved image contrast. The performance of this method was compared to that for conventional 2D TSE in four healthy subjects and three patients with prostate disease. Inner-volume SPACE permitted the repetition time to be substantially extended compared to standard slab-selective SPACE, while maintaining an acceptable acquisition time and the ability to obtain high-quality multi-planar reconstructions. Promising results were obtained compared to conventional 2D TSE in three patients with prostate disease.

Marcelino Bernardo^1,2, Dagane Daar^1,2, Baris Turkbey¹, Maria J Merino³, Peter Pinto⁴, and Peter L Choyke^1
1Molecular Imaging Program, National Cancer Institute, Bethesda, MD, United States, ²SAIC-Frederick, Frederick, MD, United States, ³Laboratory of Pathology, National Cancer Institute, Bethesda, MD, United States, ⁴Urology Oncology Branch, National Cancer Institute, Bethesda, MD, United States

The correlation of prostate MRI to whole-mount sections is essential in order to validate MRI findings. Tissue blocks with the correct location, orientation and shape can be obtained by cutting the deformable prostatectomy specimen inside a mold designed from the patients MRI. In this work, we describe and demonstrate an automated method for designing these molds and creating them using open source software and a low cost 3-D printer. It should be practical in large population studies necessary to compile a training set for developing a decision support system for detecting and localizing prostate cancer.

Peter T. Fwu¹, Jeon-Hor. Chen^1,2, Siwa Chan³, Dah-Cherng Yeh⁴, Chin-Kai Chang², Julian Kao¹, Muqing Lin¹, Orhan Nalcioglu¹, and Min-Ying L. Su^1
1Center for Functional Onco-Imaging, Department of Radiological Sciences, University of California, Irvne, California, United States, ²Department of Radiology, China Medical University Hospital, Taichung, Taiwan, ³Department of Radiology, Taichung Veterans General Hospital, Taichung, Taiwan,⁴Department of Surgery, Taichung Veterans General Hospital, Taichung, Taiwan

We applied quantitative methods to investigate the symmetry of bilateral breasts. The volume, as well as the 3D morphology (Circularity, Convexity, Irregularity) and texture properties (gray-level co-occurrence matrix) of the segmented fibroglandular tissues on MRI were analyzed. A high correlation between the volume measured from the left and right breasts is observed. The correlation for morphology parameters are still high, and worse for texture parameters. The analysis of symmetry is particularly important for detection of non-mass-like lesions on breast MRI. The result may provide a reference dataset for developing intellectual computer-aided diagnostic systems for detecting abnormal lesions on MRI.

David J Manton¹, Martin D Pickles¹, Martin Lowry¹, and Lindsay W Turnbull^1
1YCR Centre for MR Investigations, The University of Hull, Hull, East Yorkshire, United Kingdom

A Tofts-Kermode-Kety (TKK) pharmacokinetic model with significant blood plasma signal contribution was applied to moderate (~30s) temporal resolution DCE-MRI breast tumour data demonstrating a range of curve types (from Type 1 shallow continuous enhancement to Type 3b instantaneous enhancement with marked wash-out). In Type 3b cases, boot-strap error estimates indicated that the model was ill-conditioned with relative standard deviations for volume parameters exceeding 100%, therefore the separate tissue and plasma curves were analysed empirically using maximum gradient and areas under the curve. When compared to empirical parameters derived from the raw data, TKK empirical parameters demonstrated better discriminatory power.

Arfan Ahmed¹, Peter Gibbs¹, Martin D Pickles¹, and Lindsay W Turnbull^1
1CMRI, University of Hull, Hull, England, United Kingdom

We compared texture analysis of DCE Breast imaging using Haralick cooccurence method. Tests were performed on single slice vs multi slice and results showed significant differences when using multi slice and no significant differences using single slice, the study showed that important tumour information is missed when using single slice for textural analysis

Geon-Ho Jahng¹, Jeong Hyun Kim¹, Jung Kyu Ryu¹, Min-Ji Kim¹, Hyug-Gi Kim¹, Dal Mo Yang¹, Dong Wook Sung², and Woo-Suk Choi^2
1Radiology, Kyung Hee University Hospital at Gangdong, Kyung Hee University, Seoul, Seoul, Korea, ²Radiology, Kyung Hee University Hospital, Kyung Hee University, Seoul, Seoul, Korea

An application of a double inversion recovery (DIR) sequence in breast MR imaging may improve lesion detections by nullifying signals from fat and fibroglandular tissue without using any contrast agent. The purpose of present work was to investigate the effectiveness of the DIR MRI sequence in detections of breast cancers. Firstly, we compared lesion contrasts among several imaging sequences, including contrast-enhanced T1-weighted image (CE-T1WI), DIR image, pre-contrast-enhanced T1WI and T2-weighted image (T2WI) from twenty-four patients. Secondly, we evaluated whether DIR image was useful to detect breast cancers without injecting contrast agent from additional thirty-two patients.

Shotaro Kanao¹, Masako Kataoka¹, Mami Iima¹, Rena Sakaguchi¹, Natsuko Onishi¹, Tomohisa Okada¹, David Andrew Porter², Thorsten Feiweier², and Kaori Togashi^1
1Diagnositic Imaging and Nulcear Medicine, Kyoto University Graduate School of Medicine, Kyoto, Kyoto, Japan, ²Neurology Applications, Siemens AG

Diffusion weighted MR imaging (DW-MRI) is expected to increase diagnostic accuracy of detecting breast cancer but compared to dynamic contrast MRI (DC-MRI), image quality is poor due to low spatial resolution and image distortion. We evaluated the feasibility of high resolution DW-MRI (special resolution of 1.0x1.0x2.5mm) of the breast, using newly-developed readout segmented echo planner imaging (RS-EPI) and single shot echo planner imaging (SS-EPI). High resolution DW-MRI of the breast was feasible in both RS-EPI and SS-EPI. DW-MRI using RS-EPI has advantage of less distortion.

Yuan Le¹, Hal D Kipfer², Shadie S Majidi², Stephanie Holz², and Chen Lin^1
1Radiology, Indiana University, Indianapolis, Indiana, United States, ²Radiology and Imaging Science, Indiana University, Indanapolis, Indiana, United States

This study compares the artifacts in fat suppressed breast MRI caused by biopsy markers from three different techniques: dual echo Dixon, conventional or Quick FatSat (QFS) and SPectrally selective Adiabatic Inversion Recovery (SPAIR) at 3T. The metal artifact caused by biopsy markers appear as a void in Dixon images versus interleaved bright and dark rings in QFS and SPAIR images. SPAIR fat suppression has the largest volume of artifacts due to biopsy markers.

Xi Liang^1,2, Kotagiri Ramamohanarao¹, Qing Yang³, Alexander pitman⁴, and Marius Staring^5
1University of Melbourne, Carlton, Victoria, Australia, ²National ICT Australia, Carlton, Victoria, Australia, ³Apollo Medical Image Technology Pty. Ltd., ⁴St. Vincent's Hospital, ⁵Leiden University Medical Center

DCE-MRI is sensitive in the detection of tumors. However the motion in between the image acquisitions can complicate the analysis. Nonrigid registration is used to achieve alignment between images. It can be formulated as an optimization problem to minimize the image dissimilarity. However it not only reduces the occurred motion but also may change the volume of enhanced regions, e.g. cancer tissue. A spatial-variant rigidity regularization term can be used to preserve the rigidity of tissues. This study proposes a practical method to compute the coefficients of rigidity terms. The evaluation result shows it can replace the manual segmentation to compute the coefficients used to reduce undesirable deformations. We propose a framework to generate regularization coefficients for nonrigid registration in DCE-MRI, where tumor locations are to be transformed in a rigid fashion. The coefficients are obtained by applying a sigmoid function on subtraction images from a pre-registration. All parameters in the function are automatically determined using k-means clustering. In the evaluation test, the proposed method can replace the binary coefficients requiring manual tumor segmentation.

Lei Tang¹, Xiao-Peng Zhang¹, Ying-Shi Sun¹, Zi-Yu Li², Jia-Fu Ji², Xiao-Ting Li¹, and Yi-Qiang Liu^3
1Radiology, Peking University Cancer Hospital, Beijing, China, ²Gastroenterology Surgery, Peking University Cancer Hospital, ³Pathology, Peking University Cancer Hospital

Borrmann type 4 (BT-4) gastric cancer often infiltrates deep mucosa, which may lead to false negative results on endoscopic biopsy. In this study, we performed correlation of DWI with histopathology and comparison with conventional sequences, to see if it can provide additional information for the diagnosis of BT-4 gastric cancer. We found the cancer-to-stomach CNR on DWI was better than T1WI and T2WI. A three-layer sandwich sign that demonstrated high signal intensity in the inner and outer layer, and low signal intensity in the intermediate layer of BT-4 gastric cancer was observed on DWI. The mean ADC value of BT-4 gastric cancer was significantly lower than the poorly distended normal stomach wall. We concluded that DWI can provide useful information for the diagnosis of BT-4 gastric cancer.

Vicky J Goh¹, N. Jane Taylor¹, Aftab Khan², J James Stirling¹, Ian C Simcock¹, Robert Kozarski³, Robert Glynne-Jones², James A d'Arcy⁴, David J Collins⁴, and Anwar R Padhani^1
1Paul Strickland Scanner Centre, Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, United Kingdom, ²Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, United Kingdom, ³University of Hertfordshire, Hatfield, Hertfordshire, United Kingdom, ⁴CRUK-EPSRC Cancer Imaging Centre, Institute of Cancer Research & Royal Marsden Hospital, Sutton, Surrey SM2 5PT, United Kingdom

The aim of this study was to assess the reproducibility of R₂* and sequential changes following chemoradiation therapy (CRT), in relation to perfusion changes shown by dynamic contrast enhanced MRI (DCE-MRI). 14 patients underwent ISW-MRI and 3D DCE-MRI at 1.5T, pre- and up to 11 weeks post- chemoradiation. Cohort R₂* increases post-CRT, corresponding to decreases in IAUGC₆₀ and K^trans. This, accompanied by loss of the baseline correlation between R₂* and K^trans suggests that colorectal tumors are made hypoxic by chemoradiotherapy. These results have implications with regard to the optimal timing of surgery following the completion of therapy.

Wael Shabana¹, Dilkash Kajal¹, and Rebecca E Thornhill^1
1Medical Imaging, The Ottawa Hospital, Ottawa, Ontario, Canada

The clinical MRI assessment of metastatic mesorectal lymph nodes plays an important role in predicting prognosis in rectal cancer. However, radiologic assessment offers limited sensitivity to predict the malignant potential of lymph nodes, particularly when conventional diameter criteria are used. We identified several diameter-independent shape and textural features from contrast enhanced T1-weighted MR images of mesorectal lymph nodes that were capable of delineating nodes with high malignant potential with 85% sensitivity, 78% specificity and 85% accuracy. With further optimization, these advanced shape and textural features promise to provide improved sensitivity compared to conventional radiologic methods.

Philippe Garteiser¹, Arnaud Dieudonne^1,2, Rachida Lebtahi², Mohamed Abdel-Rehim³, Valerie Vilgrain³, Ralph Sinkus¹, and Bernard E Van Beers^1,3
1CRB3 U773 Université Paris Diderot, Sorbonne Paris Cite, INSERM, Paris, 75018, France, ²Service of Nuclear Medicine, Hopital Beaujon, Clichy, France,³Service of Radiology, Hopital Beaujon, Clichy, France

The localization of yttrium loaded radioembolization microspheres and the detection of their effects on structural and functional integrity of the tumors remain a challenge. In the present work, the mechanical properties of the glass substrate of the microspheres was exploited in applying MR elastography sequence on a phantom and on a hepatocellular carcinoma patient.

Susanne Bonekamp¹, Zhen Li², Vivek Gowdra Halappa¹, Celia Pamela Corona-Villalobos¹, Jean-Francois H. Geschwind¹, and Ihab R Kamel^1
1Radiology, Johns Hopkins University, Baltimore, Maryland, United States, ²Radiology, Tongji Medical College, Wuhan, China

Patient survival is the gold standard for the assessment of treatment response. The purpose of this study was to evaluate if early multiparametric assessment of treatment response is associate with patient survival in patients with hepatocellular carcinoma (HCC). To this end, HCC index lesions were analyzed using proprietary software and response was assessed based on change in tumor size as well as changes in volumetric ADC, hepatic arterial enhancement and portal venous enhancement. The combination of early change in portal venous enhancement and ADC more consistently predicted survival times than a single measurement or change in tumor size.

Wieland H Sommer¹, Felix Ceelen¹, Philipp M Paprottka¹, Maximilian F Reiser¹, and Daniel Theisen^1
1Department of Clinical Radiology, University of Munich, Großhadern Hospital, Munich, Bavaria, Germany

The aim of this study was to define the role of MRI in the pretherapeutic prediction of treatment response of radioembolization in neuroendocrine liver metastases. By analyzing different combinations of predictors at baseline and correlation them with the survival, we defined vascularization, histologic markers as the most valuable predictors. Progression free survival is independent of the tumor load in the liver.

Leif Jensen¹, Mark L. Schiebler¹, Scott K. Nagle^1,2, Scott B. Reeder^1,3, Jeffrey P. Kanne¹, Cristopher Meyer¹, Tracey Weigel⁴, and Christopher P. François^1
1Radiology, UW-Madison, Madison, WI, United States, ²Medical Physics, UW-Madison, Madison, WI, United States, ³Biomedical Engineering, UW-Madison, Madison, WI, United States, ⁴Thoracic Surgery, UW-Madison, Madison, WI, United States

Non small cell lung cancer can be cured with surgical resection of disease limited to a hemi thorax. MRI has the advantage over CT and PET imaging in the diagnosis of vascular and cardiac invasion and for determining the feasibility of complex vertebral body and great vessel involvement by malignant masses. Precise delineation of tumor extent is critical in treatment planning. The use of cardiac gating bSSFP, tagged images, MRA, Diffusion Weighting, and Fat separation after contrast administration are all conventionally available pulse sequence methods that can be employed to help determine surgical resectabilty.

Shuji Nagata¹, Koji Hiraoka², Tetsuya Hamada², Takanori Shoda², Hiroshi Nishimura³, Masayoshi Kage⁴, Kimberly K Amrami⁵, and Naofumi Hayabuchi^1
1Radiology, Kurume University Hospital, Kurume, Fukuoka, Japan, ²Orthopedic Surgery, Kurume University Hospital, ³Radiology, Saiseikai Futsukaichi Hospital, ⁴Pathology, Kurume University Hospital, ⁵Radiology, Mayo Clinic

 

Harald Braun¹, Susanne Ziegler¹, and Harald H Quick^1
1Institute of Medical Physics, University of Erlangen-Nürnberg, Erlangen, Bavaria, Germany

Simultaneous whole-body hybrid PET/MR imaging has become clinical reality. As PET and MRI have traditionally been separate fields, physicians and scientists usually specialized in either method. With regards to PET/MR hybrid imaging, it seems of mutual scientific benefit to bring those two fields closer together. This presentation introduces the basics and physical principles of PET to a broad audience of MRI specialists. This includes different tracers and isotopes used in PET, physical processes involved in signal generation, some basics about PET instrumentation in the context of combined PET/MR systems and the necessary post processing and reconstruction steps for PET data.

Kathryn Jane Fowler¹, Costa Raptis², Agus Priatna³, Jonathan McConathy⁴, Rex Parker⁴, Vamsi Narra⁵, and Pamela Woodard^6
1Body MRI, Mallinckrodt Institute of Radiology, St. Louis, Mo, United States, ²Cardiothoracic imaging, Mallinckrodt Institute of Radiology, St. Louis, Mo, United States, ³Siemens, ⁴Mallinckrodt Institute of Radiology, ⁵Abdominal Imaging, Mallinckrodt Institute of Radiology, St. Louis, Mo, United States,⁶Cardiothoracic Imaging, Mallinckrodt Institute of Radiology, St. Louis, Mo, United States

Initial experience with Magnetic Resonance PET imaging for body applications with emphasis on development of optimal focused and whole-body protocols. Comparison of intraindividual PET-CT and MR-PET.

Yoshiharu Ohno^1,2, Mizuho Nishio¹, Hisanobu Koyama^1,2, Takeshi Yoshikawa¹, Sumiaki Matsumoto¹, Daisuke Takenaka¹, Katsusuke Kyotani², Nobukazu Aoyama², Saori Satou³, Hideaki Kawamitsu², Satoru Takahashi¹, and Kazuro Sugimura^1
1Radiology, Kobe University Graduate School of Medicine, Kobe, Hyogo, Japan, ²Radiology, Kobe University Hospital, Kobe, Hyogo, Japan, ³Toshiba Medical Systems, Ohtawara, Tochigi, Japan

Distant metastasis and/or recurrence are very important for management of patients with non-small cell lung cancer (NSCLC). We hypothesized that newly utilized Quick 3D and enhanced fat free (EFF) techniques could improve diagnostic performance of whole-body MRI at 3T system (whole-body 3T MRI) as compared with previously reported protocol, and might be at least as valuable as FDG-PET/CT in NSCLC patients. The purpose of this study was to directly and prospectively compare diagnostic capability for distant metastasis and/or recurrence assessment among whole-body 3T MRI with conventional protocol, that with Quick 3D and EFF and integrated FDG-PET/CT in NSCLC patients.

Miyuki Takasu¹, Chihiro Tani¹, Miho Ishikawa¹, Daisuke Komoto¹, Keizo Tanitame¹, Yuji Akiyama¹, Yoshiaki Kuroda², Akira Sakai³, and Kazuo Awai^1
1Department of Diagnostic Radiology, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan, ²Department of Hematology and Oncology, Research Institute for Radiation Biology and Medicine, Hiroshima University, Hiroshima, Japan, ³Department of Hematology, Fukushima Medical University Hospital, Fukushima, Japan

The present study was performed to prospectively evaluate the effectiveness of IDEAL MRI for discrimination of tumor stage in patients with multiple myeloma without visible focal lesions. The fat signal fractions among the four groups were compared with post hoc test. The fat signal fraction and ƒÀ2m-to-albumin ratio were entered into discriminant analysis. Among the four groups, patients with symptomatic myeloma showed the lowest fat signal fraction. With the discriminant analysis, 22 of the 24 patients (92%) were correctly classified in each category. Fat quantification using the IDEAL sequence was significantly different when comparing patients with symptomatic and asymptomatic myeloma.

Firas Moosvi^1,2, Ahmed El-Kaffas^1,2, Melissa Yin¹, and Greg Stanisz^1,2
1Imaging Research, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada, ²Medical Biophysics, University of Toronto, Toronto, Ontario, Canada

Assessing the efficacy of novel cancer therapies non-invasively remains an important goal in imaging research. DCE-MRI provides some useful information about the tumour microenvironment including vessel leakiness and extravascular extracellular volume, but falls short when determining parameters such as blood volume and flow. One approach is to combine DCE-MRI with a modality that is sensitive to blood flow and blood volume such as DCE-Ultrasound. This abstract describes a pilot pre-clinical study using a novel imaging protocol combining DCE-MRI and DCE-US to assess the tumour microenvironment after therapy.

Ellen Ackerstaff¹, Xin Li², Sohyun Han³, DongKyu Lee³, HyungJoon Cho³, Sean Carlin¹, Jason A Koutcher¹, and Wei Huang^2
1Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ²Oregon Health & Science University, Portland, OR, United States, ³Ulsan National Institute of Science and Technology (UNIST), EonYang-eup, Ulju-gun, Ulsan, Korea

The non-invasive detection and quantification of hypoxic areas in tumors is of tremendous interest for the purpose of predicting and monitoring cancer treatment response. We analyze DCE-MRI data in a preclinical cancer model with the Shutter-Speed Model to investigate its ability to distinguish tumor areas that are well perfused (oxygenated), hypoxic (viable), or necrotic. Our preliminary results indicate that a combination of K^trans and τ_i can separate areas that are predominantly viable/well-perfused or viable/hypoxic or necrotic. The successful implementation of Shutter-Speed DCE-MRI assessment of the tumor microenvironment may potentially obviate the need for supplementary imaging studies, such as ¹⁸F-Fmiso PET.

Fabian Kording¹, Claudia Weidensteiner¹, Andreas Lingnau², Stefan Zwick³, and Wilfried Reichardt^1
1Department of  Radiology Medical Physics, University Medical Center Freiburg, Freiburg, Baden-Württemberg, Germany, ²Tumor Biology Center Freiburg, ProQinase GmbH, Freiburg, Baden-Württemberg, Germany, ³R&D Engineer MRI CryoProbe Systems, Bruker BioSpin AG, Fällanden, Switzerland

Anti-angiogenic therapy in tumors can be monitored by dynamic contrast enhanced MRI (DCE-MRI) or dynamic susceptibility contrast imaging (DSC-MRI) and vessel size imaging (VSI) to detect changes in tumor microvasculature. To acquire these biomarkers it is necessary to use multiple contrast agents until now. In this work, which is supported by the 2010 ISMRM Seed-Grant, these biomarkers are acquired simultaneously in vivo using a single shot gradient echo spin echo EPI sequence with only one Gd-CA injection. We tested the sequence in an in vivo brain tumor model and were able to acquire multiple datasets that were quantified in different tissue compartments after segmentation to account for tumor heterogeneity.

Renuka Sriram¹, Kayvan R Keshari¹, Bertram Koelsch¹, Mark Van Criekinge¹, John Kurhanewicz¹, and Zhen Jane Wang^1
1Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, California, United States

Cancer cell metabolism is characterized by increased glucose uptake and lactate production eunder aerobic conditions. This phenomenon has been exploited to differentiate between tumor grades, using hyperpolarized (HP) [1-¹³C] pyruvate. The flux of pyruvate to lactate through the lactate dehydrogenase enzyme (LDH) is a function of the monocarboxylate transporters, cofators and enzyme expression and activity. In an effort to understand these factors that contribute to the HP Pyruvate/Lactate kinetics, aggressive human RCCs UMRC6 (localized) and UOK262 (metastatic) are characterized using HP ¹³C MR, thermal labeling with [3-¹³C] pyruvate and subsequent comparison to the expression of relevant transporters and LDH.

Rui Vasco Simoes¹, Ellen Ackerstaff¹, Natalia Kruchevsky¹, Yang Pu^1,2, George Sukenick¹, and Jason A Koutcher^1
1Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ²The City College of New York, New York, NY, United States

Distinguishing non-invasively metastatic from non-metastatic clinical breast cancer, would significantly improve evaluation of patients’ prognoses. Two isogenic murine breast cancer lines, metastatic 4T1 and non-metastatic 67NR, were studied in an NMR-compatible cell perfusion system to investigate their dynamic metabolic responses to (a) oxygen availability vs. hypoxia, and (b) glutamine supply vs. deprivation. When exposed to these stress conditions, each cell line adopted different metabolic strategies. Non-metastatic 67NR cells revealed overall higher glycolytic activity than metastatic 4T1 cells. In the more aggressive cell line, glycolysis is glutamine-dependent during hypoxia, and decreases when oxygen becomes available to favor TCA cycle activity.

Niki M Zacharias^1,2, Napapon Sailasuta¹, Henry R Chan¹, Maja C Cassidy³, Cameron Henneberg¹, Ashraf Imam¹, Brian D Ross¹, and Pratip Bhattacharya^1
1Enhanced MR Laboratory, Huntington Medical Research Institutes, Pasadena, CA, United States, ²Division of Chemistry, California Institute of Technology, Pasadena, CA, United States, ³Department of Physics, Harvard University, Cambridge, MA, United States

Parahydrogen Induced Polarization (PHIP) can offer 50,000 fold increase in MR signal under the right conditions. Diethyl 1-13C 2,3-d2 succinate is generated through the hydrogenation of diethyl 1-13C 2,3-d2 fumarate and hyperpolarization increases the 13C signal by 5000 fold. We have previously employed hyperpolarized diethyl succinate to image and observe real time metabolism in normal mice. In the work described here, we have applied hyperpolarized diethyl succinate for imaging cancer in two different subcutaneous tumor models in mice and compared their fingerprinting of the downstream metabolites in the TCA cycle in real time in vivo, thereby providing an exciting application of metabolic imaging employing hyperpolarization.

Minhua Zhu¹, Asha Balakrishnan², Simon HU¹, Pedar E.Z. Larson¹, Sarah J. Nelson¹, John Kurhanewicz¹, Andrei Goga¹, and Daniel B. Vigneron^1
1radiology and biomedical imaging, ucsf, San Francisco, CA, United States, ²Dept. of Medicine, Division of Hematology/Oncology, University of California, San Francisco, CA

Hyperpolarized technology using dynamic nuclear polarization has been developed and used in detecting signals of 13C metabolites in vivo at very high SNR [1]. In this work, hyperpolarized 13C 3D-MRSI using a 3D compressed sensing sequence [2] was used to measure liver metabolism in mice which developed abnormal size of the liver after expression of the RAS oncogene was switched on in the liver. After the liver developed into an abnormal state at a relatively late stage, the regression of the later disease stages before and after turning off the oncogene were studied, and significant differences in hyperpolarized lactate and alanine levels were detected.

Basetti Madhu¹, Santiago Uribe Lewis², Adele Murrell², and John R Griffiths^1
1Molecular Imaging, Cancer Research UK Cambridge Research Institute, Cambridge, England, United Kingdom, ²Epigenetics and Imprinting Laboratory, Cancer Research UK Cambridge Research Institute, Cambridge, England, United Kingdom

The Apc ^Min/+ mouse is a genetically engineered cancer mouse model that spontaneously develops tumours in the GI tract. Here we have investigated the metabolic characteristics of the GI tumours of Apc ^Min/+ mice using HRMAS ¹H NMR spectroscopy. We found that amino acid metabolism was significantly affected in tumours relative to the normal adjacent tissues. The total choline content was also increased (though not statistically significantly) whereas total creatine remained unchanged. The data show that HRMAS ¹H NMR spectroscopy of Apc ^Min/+ mouse tumour tissues recapitulates some aspects of the metabolic states observed in human tumours.

Alessia Lodi¹, Hubert J Stoppler², Matthew Firpo³, Sean J Mulvihill³, Margaret A Tempero⁴, and Sabrina M Ronen^1
1Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, California, United States, ²Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, California, United States, ³Department of Surgery, University of Utah, Salt Lake City, Utah, United States, ⁴Department of Medicine, University of California San Francisco, San Francisco, California, United States

Pancreatic cancer is the fourth leading cause of cancer-related death in the United States. Because over 80% of pancreatic cancer cases are diagnosed at an advanced stage, survival rates are poor for all stages of the disease. Survival rates improve if the tumor is localized at diagnosis and can be surgically removed. Novel, non-invasive methods for early diagnosis of pancreatic cancer are needed. Here we report a very significant decrease of the high intensity lipid signals in the cancerous specimens compared to matched uninvolved pancreatic biopsies which may represent a candidate biomarker for the early detection of pancreatic cancer.

Sanjay Annarao¹, and Sunitha Thakur^1,2
1Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ²Department of Radiology, Memorial Sloan-Kettering Cancer Centre, New York, NY, United States

We measured Lac with improved lipid and water suppression using 1331 binomial composite pulses using SS1-SelMQC sequence. Using this sequence Lac levels were quantified in orthotopic breast tumors with different growth rates. MCF-7 and BT-474 breast tumors were used to represent fast growing tumors compared with slow growing MDA-MB-231 and MDA-MB-435 tumors. The [Lac] was found to be higher in lower tumor volume, as tumor volume becomes more, [Lac] declined. The statistical analysis shows the relationship between tumor volume and Lac level in slice. In all these cell lines, the tumor volume is negatively correlated with [Lac].

Sanjay Annarao¹, Thomas Ku², Kishore Nagavara³, Jason Koutcher^1,4, and Sunitha Thakur^1,3
1Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ²Department of Molecular Pharmacology & Chemistry, Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ³Department of Radiology, Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ⁴Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, United States

SS1-SelMQC pulse sequence is proposed to detect Lac with enhanced signal intensity by effectively suppressing fat and water using 1331 binomial composite Spectral-Selective Pulses in Selective MQ Coherence (SS1-SelMQC). Lac signal enhancement was achieved by using short binomial frequency selective pulses, which reduces signal loss in the evolution period due to the effects of scalar coupling and molecular diffusion. T1 and T2 of Lac incorporating T1 and T2 variables in SS1-SelMQC. T1-SS1-SelMQC and T2- SS1-SelMQC is standardized in phantom and demonstrated in in-vivo breast tumors. The T2 of Lac was statistically different in different types of breast tumors.

Asif Rizwan¹, Xiaohui Ni¹, Rachael O'Connor², Samuel Singer², Sean Carlin³, Jason Koutcher¹, and Kristen L. Zakian^1
1Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ²Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ³Radiology, Memorial Sloan-Kettering Cancer Center, New York, NY, United States

The lack of effective chemotherapy in de-differentiated liposarcoma leaves non-surgical candidates with few options. A non-invasive marker reflecting the effect of drugs could be quite valuable in pre-clinical drug evaluations. Furthermore, in the clinic, an early marker of response/non-response could permit the physician to discontinue ineffective treatment. Lactate, an end-product of glycolysis has the potential to be a biomarker of prognosis and treatment effect. The goal of the current study was to assess the change in lactate levels in a human DDLS tumor xenograft implanted in mice in response to chemotherapy.