Dynamic contrast-enhanced MRI of prostate cancer at 3 T: A study of pharmacokinetic parameters

OBJECTIVE. The objectives of our study were to determine whether dynamic contrast-enhanced MRI performed at 3 T and analyzed using a pharmacokinetic model improves the diagnostic performance of MRI for the detection of prostate cancer compared with conventional T2-weighted imaging, and to determine which pharmacokinetic parameters are useful in diagnosing prostate cancer. SUBJECTS AND METHODS. This prospective study included 50 consecutive patients with biopsy-proven prostate cancer who underwent imaging of the prostate on a 3-T scanner with a combination of a sensitivity-encoding (SENSE) cardiac coil and an endorectal coil. Scans were obtained at least 5 weeks after biopsy. T2-weighted turbo spin-echo images were obtained in three planes, and dynamic contrast-enhanced images were acquired during a single-dose bolus injection of gadopentetate dimeglumine (0.1 mmol/kg). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were estimated for T2-weighted and dynamic contrast-enhanced MRI. The following pharmacokinetic modeling parameters were determined and compared for cancer, inflammation, and healthy peripheral zone: K-trans (forward volume transfer constant), k(ep) (reverse reflux rate constant between extracellular space and plasma), v, (the fractional volume of extracellular space per unit volume of tissue), and the area under the gadolinium concentration curve (AUGC) in the first 90 seconds after injection. RESULTS. Pathologically confirmed cancers in the peripheral zone of the prostate were characterized by their low signal intensity on T2-weighted scans and by their early enhancement, early washout, or both on dynamic contrast-enhanced MR images. The overall sensitivity, specificity, PPV, and NPV of T2-weighted imaging were 94%, 37%, 50%, and 89%, respectively. The sensitivity, specificity, PPV, and NPV of dynamic contrast-enhanced MRI were 73%, 88%, 75%, and 75%, respectively. K-trans, k(ep), and AUGC were significantly higher (p < 0.001) in cancer than in normal peripheral zone. The v(e) parameter was not significantly associated with prostate cancer. CONCLUSION. MRI of the prostate performed at 3 T using an endorectal coil produces high-quality T2-weighted images; however, specificity for prostate cancer is improved by also performing dynamic contrast-enhanced MRI and using pharmacokinetic parameters, particularly K-trans and k(ep), for analysis. These results are comparable to published results at 1.5 T.