B. Nicolas Bloch MD, Presenter: Nothing to Disclose

Tania Velez MD, Abstract Co-Author: Nothing to Disclose

Robert E. Lenkinski PhD, Abstract Co-Author: Research grant, General Electric Company

Herbert Y. Kressel MD, Abstract Co-Author: Nothing to Disclose

Ivan Pedrosa MD, Abstract Co-Author: Nothing to Disclose

Martin Paul Smith MD, Abstract Co-Author: Nothing to Disclose

Mary Gabriella Hochman MD, Abstract Co-Author: Stockholder, Nomir Medical Technologies, Inc Stockholder, Abbott Laboratories Consultant, SafeMed Systems Consultant, UpToDate, Inc

Long Ngo, Abstract Co-Author: Nothing to Disclose

Elizabeth Genega MD, Abstract Co-Author: Nothing to Disclose

Neil Mitchell Rofsky MD, Abstract Co-Author: Research grant, General Electric Company Research grant, Bayer AG Consultant, EPIX Pharmaceuticals, Inc Advisory Board, Bayer AG Advisory Board, General Electric Company

et al, Abstract Co-Author: Nothing to Disclose

To determine the sensitivity and specificity of high spatial resolution dynamic contrast enhanced (DCE-) combined with T2 weighted (T2-W) endorectal (ER) coil magnetic resonance imaging (MRI) at 3 Tesla for assessment of extracapsular extension (ECE) and staging in prostate cancer patients in routine clinical practice, using histopathology as the reference standard.

In this IRB approved study, 3T MRI was performed with combined surface and ER coils in 106 patients (mean age: 58.5; range 47-72; mean total PSA: 10.6 ng/ml; range: 2-117; mean Gleason score: 6; range: 6-9) prior to radical prostatectomy. T2-W fast spin-echo and high spatial resolution DCE 3D gradient echo images were acquired using Gd- DTPA. DCE-images were analyzed with a computer generated color-coded scheme. One very experienced, two experienced, and two less experienced readers independently assessed ECE, and tumor stage in a routine clinical setting. MRI based staging results (as prospectively described) were compared with detailed histopathologic results. For the prediction of ECE sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) along with the corresponding 95% confidence intervals were calculated. Staging accuracy was determined by the area under the receiver-operating-characteristic (AUC) using the Wilcoxon-Mann-Whitney index of diagnostic accuracy.

Excluding 7 patients with microscopic ECE the mean sensitivity, specificity, PPV, and NPV for ECE for all readers was 92%, 95%, 86% and 97%, respectively. Diagnostic accuracy of staging was 92% for experienced readers (47/51) and 92% for inexperienced readers (44/48). Overall diagnostic accuracy (AUC) of staging was 93%; AUC for experienced readers 93% (47/51) and 85% for inexperienced readers (44/48). Including the patients with microscopic ECE the results for all readers were 74%, 91%, 77% and 89%, respectively. The interreader variability was 0.85 without (0.65 with) microscopic ECE data.

With an overall staging accuracy of 92%, this study confirms the pretherapeutic utility of 3T MRI , yielding improved assessment of ECE in routine clinical practice. The main challenge for accurate assessment of ECE was microscopic disease.

High resolution ER coil 3T MRI routinely provides successful staging results in clinical practice, particularly when irrelevant microscopic disease is excluded.

Bloch, B, Velez, T, Lenkinski, R, Kressel, H, Pedrosa, I, Smith, M, Hochman, M, Ngo, L, Genega, E, Rofsky, N, et al, , Accurate Determination of Extracapsular Extension and Staging of Prostate Cancer with High Resolution 3 Tesla MR in a Routine Clinical Setting.  Radiological Society of North America 2008 Scientific Assembly and Annual Meeting, February 18 - February 20, 2008 ,Chicago IL. http://archive.rsna.org/2008/6022385.html