Prediction of prostate cancer extracapsular extension with high spatial resolution dynamic contrast-enhanced 3-T MRI
- First Online:
- Cite this article as:
- Bloch, B.N., Genega, E.M., Costa, D.N. et al. Eur Radiol (2012) 22: 2201. doi:10.1007/s00330-012-2475-5
- 479 Downloads
To assess the value of dynamic contrast-enhanced (DCE) combined with T2-weighted (T2W) endorectal coil (ERC) magnetic resonance imaging (MRI) at 3 T for determining extracapsular extension (ECE) of prostate cancer.
In this IRB-approved study, ERC 3-T MRI of the prostate was performed in 108 patients before radical prostatectomy. T2W fast spin-echo and DCE 3D gradient echo images were acquired. The interpretations of readers with varied experience were analysed. MRI-based staging results were compared with radical prostatectomy histology. Descriptive statistics were generated for prediction of ECE and staging accuracies were determined by the area under the receiver-operating characteristic curve.
The overall sensitivity, specificity, positive predictive value and negative predictive value for ECE were 75 %, 92 %, 79 % and 91 %, respectively. Diagnostic accuracy for staging was 86 %, 80 % and 91 % for all readers, experienced and less experienced readers, respectively.
ERC 3-T MRI of the prostate combining DCE and T2W imaging is an accurate pretherapeutic staging tool for assessment of ECE in clinical practice across varying levels of reader experience.
• Endorectal coil (ERC) magnetic resonance imaging is widely used for imaging prostatic disease.
• ERC 3-T MRI is reasonably accurate for local prostate cancer staging.
• High diagnostic accuracy is achievable across different levels of reader experience.
• MRI facilitates therapeutic decisions in patients with prostate cancer.
KeywordsProstate cancer Magnetic resonance imaging Cancer staging Clinical interpretation Learning curve
American Joint Committee on Cancer
area under the receiver-operating characteristic curve
institutional review board
National Institute of Health
negative predictive value
positive predictive value
- 3 T