Andreas Lemke¹, Anja Weidner², Jörg Döpfert¹, Dietmar Dinter², and Lothar Rudi Schad^1
1Computer Assisted Clinical Medicine, Heidelberg University, Mannheim, Germany, ²Department of Clinical Radiology and Nuclear Medicine, Heidelberg University, Mannheim, Germany

Diffusion kurtosis parameter maps of six patients with biopsy confirmed prostate cancer (PCa) were obtained using a 3 Tesla Scanner with an endorectal coil and 6 b-values (b=0,100,500,1000,1500,2000 s/mm²). All patients showed an increased value of the kurtosis in the tumor compared to healthy tissue and five of the patients a decreased ADC- and D-value. The mean parameter values in the tumor and the contralateral healthy tissue were: K=1.02±0.25 vs. 0.65±0.06, D=1.33±0.29 µm²/ms vs. 1.78±0.17 µm²/ms, ADC=0.82±0.2 µm²/ms vs. 1.15±0.08 µm²/ms.

Amita Shukla-Dave¹, Hedvig Hricak¹, Oguz Akin¹, Changhong Yu², Kristen L. Zakian¹, Victor Reuter¹, Peter T. Scardino¹, James Eastham¹, and Michael W. Kattan^2
1Memorial Sloan-Kettering Cancer Center, New York, NY, United States, ²Cleveland Clinic, Cleveland, OH, United States

Many more prostate cancers (PCa) are detected each year. Better methods are needed to predict which patients have insignificant PCa. The aims of the present study were to 1) validate the previously published MR-based nomograms; and 2) design new nomograms incorporating MR Imaging (MRI), MR Spectroscopic Imaging (MRSI) and clinical data without detailed biopsy data for predicting the probability of insignificant PCa. MRI/ MRSI data was examined prospectively in 181 patients studied prior to surgery. We have successfully validated the previously published MR models. The new BaseMRI and BaseMRI/MRSI models after validation may obviate the need for repeat biopsy.

Baris Turkbey¹, Haresh Mani², Vijay Shah^3,4, Marcelino Bernardo^3,4, Ardeshir Rastinehad⁵, Thomas Pohida⁶, Yuxi Pang⁷, Dagane Daar³, Compton Benjamin⁵, Yolanda McKinney¹, Joanna Shih⁸, Maria J Merino², Peter A. Pinto⁵, and Peter L. Choyke^1
1Molecular Imaging Program, NCI, NIH, Bethesda, MD, United States, ²Laboratory of Pathology, NCI, NIH, United States, ³Molecular Imaging Program, NCI, NIH, United States, ⁴Imaging Physics, SAIC Frederick, Inc., NCI-Frederick, ⁵Urologic Oncology Branch, NCI, NIH, United States, ⁶Division of Computational Bioscience, Center for Information Technology, NIH, United States, ⁷Philips Healthcare, United States, ⁸Biometric Research Branch, NCI, NIH, United States

Multiparametric MRI has better sensitivity for detecting larger (> 5 mm in diameter) and more aggressive (Gleason score of > 7) tumors. The use of the mold enables more exact correlation between each MR parameter and the histopathologic specimen. The customized mold provided tissues blocks that had a one to one correspondence with the in vivo MR.

Veronica A Morgan¹, Sharon L Giles¹, Sophie F Riches¹, David Dearnaley², and Nandita M deSouza^1
1CRUK & EPSRC Cancer Imaging Centre, Institute of Cancer Research &Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom, ²Academic Urology, Institute of Cancer Research & Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom

This study evaluates the accuracy of T2W plus DW-MRI for detecting recurrent tumor within the prostate following External Beam Radiotherapy (EBRT). Mean ADC measurements of tumor and irradiated non-tumor tissue, confirmed by TRUS biopsy, were documented in 26 patients with evidence of biochemical failure. ADCs of MRI-defined tumor, histology-defined tumor and non-tumor irradiated tissue were compared. Mean ADC for both MRI and histology defined tumor were significantly lower than for non-tumor irradiated tissue. A cut-off ADC of 1337 (x10-6mm2/s) predicted tumor with a 92.9% sensitivity and 100% specificity. DWI MRI can be used for identifying tumor recurrence following EBRT.

Gregor Thörmer¹, Josephin Otto¹, Martin Reiss-Zimmermann¹, Matthias Seiwerts¹, Nikita Garnov¹, Michael Moche¹, Thomas Kahn¹, and Harald Busse^1
1Department of Diagnostic and Interventional Radiology, Leipzig University Hospital, Leipzig, Saxony, Germany

The differentiation between high (Gleason Score > 6) and low (¡Ü6) grade prostate cancer (PCa) is considered important for therapeutic decision. For that purpose, we investigated the diagnostic potential of the apparent diffusion coefficient (ADC). The influence of the choice of b-values (four methods) was retrospectively studied on 40 patients with known PCa. Although absolute ADC values in PCa and surrounding healthy tissue (SHT) significantly (p<0.001) depended on the b-values, the ADC ratios (PCa/SHT) were nearly independent of the method. Image appearance of ADC maps using b=50 and 800 [mm/s²] was rated most suitable for PCa detection by three readers.

Susan M Noworolski^1,2, Galen D Reed^1,2, and John Kurhanewicz^1,2
1Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, ²Graduate Group in Bioengineering, University of California, San Francisco and Berkeley, San Francisco and Berkeley, CA, United States

It has been suggested that Gd-DTPA does not reach intact prostatic ducts. A novel pharmacokinetic model of the prostate incorporating Gd-inaccessible luminal water is presented. A feature of this Luminal Water model is that it allows for a slower washout, quite common in healthy and benign, glandular prostate tissues. Simulations and an example in vivo case demonstrate that the standard extended Tofts-Kermode model would greatly underestimate K^trans and Vees in glandular tissues (14% - 44% in cases shown). Incorporation of the Gd-inaccessible luminal water may yield parameters more accurately reflecting underlying tissue and has promise for aiding prostate tissue characterization.

Johannes Georg Korporaal¹, Cornelis AT van den Berg¹, Matthias JP van Osch², Greetje Groenendaal¹, Marco van Vulpen¹, and Uulke A van der Heide^1
1Dept. of Radiotherapy, University Medical Center Utrecht, Utrecht, Netherlands, ²Dept. of Radiology, Leiden University Medical Center, Leiden, Netherlands

Arterial input function (AIF) measurements in the iliac arteries from the DCE-MRI phase signal (AIF_PHASE) were validated with DCE-CT (AIF_CT) in twelve prostate cancer patients. AIF_PHASE and AIF_CT were normalized to the injected dose of contrast agent and compared. For all patients, the shape of AIF_PHASE is very similar to the shape of AIF_CT. No significant differences were found between AIF_PHASE and AIF_CT for the area under the curve of the first pass peak and the signal 180 seconds after the peak. In conclusion, AIF_PHASE shows good agreement in both shape and amplitude with AIF_CT.

Daniel Jason Aaron Margolis¹, Shyam Natarajan², Dinesh Kumar³, Ram Narayanan³, and Leonard Marks^4
1Dept. of Radiology, UCLA David Geffen School of Medicine, Los Angeles, CA, United States, ²Center for Advanced Surgical and Interventional Technology, UCLA David Geffen School of Medicine, ³Eigen, ⁴Dept. of Urology, UCLA David Geffen School of Medicine, Los Angeles, CA, United States

MRI with diffusion and perfusion imaging was performed in 54 consecutive men referred for prostate biopsy. Target lesions based on diffusion and perfusion parameters were ranked based on level of suspicion and contoured on T2-weighted imaging. Subsequent biopsies were performed with real-time image coregistration with transrectal ultrasound. The number of positive cores was significantly higher for targeted biopsies than for systematic 14-point sextant biopsies, and in some cases, cancer was only found with targeted biopsies.

Kemal Tuncali¹, Junichi Tokuda¹, Andriy Fedorov¹, Iulian Iordachita², Sam Song¹, Sota Oguro¹, Andras Lasso³, Fiona M Fennessy¹, Yi Tang¹, Nobuhiko Hata¹, and Clare M Tempany^1
1Radiology, Brigham and Women's Hospital, Boston, MA, United States, ²The Johns Hopkins University, Baltimore, MD, United States, ³School of Computing, Queen’s University, ON, Canada

Nine patients with clinically suspected prostate cancer underwent 3T MRI-guided transperineal targeted prostate biopsy successfully in the lithotomy position without significant compications using local anesthesia and intravenous procedural sedation. Multiparametric diagnostic MRI data were registered to intraprocedural T2 weighted turbo spin echo images to facilitate targeting of index lesions. Needle location was confirmed using real-time 2D FISP. Of eight biopsy results available, two were proved cancer and six showed no malignancy. We have shown this approach to be feasible and safe. It can yield clinically useful results in patients with index lesion on diagnostic MRI.