Multiparametric whole-body 3.0-T MRI in newly diagnosed intermediate- and high-risk prostate cancer: diagnostic accuracy and interobserver agreement for nodal and metastatic staging

Abstract

Objectives: To determine the diagnostic accuracy and interobserver concordance of whole-body (WB)-MRI, vs. 99mTc bone scintigraphy (BS) and 18fluoro-ethyl-choline (18F-choline) PET/CT for the primary staging of intermediate/high-risk prostate cancer.

Methods: An institutional review board approved prospective cohort study carried out between July 2012 and November 2015, whereby 56 men prospectively underwent 3.0-T multiparametric (mp)-WB-MRI in addition to BS (all patients) ± 18F-choline PET/CT (33 patients). MRI comprised pre- and post-contrast modified Dixon (mDixon), T2-weighted (T2W) imaging, and diffusion-weighted imaging (DWI). Patients underwent follow-up mp-WB-MRI at 1 year to derive the reference standard. WB-MRIs were reviewed by two radiologists applying a 6-point scale and a locked sequential read (LSR) paradigm for the suspicion of nodal (N) and metastatic disease (M1a and M1b).

Results: The mean sensitivity/specificity of WB-MRI for N1 disease was 1.00/0.96 respectively, compared with 1.00/0.82 for 18F-choline PET/CT. The mean sensitivity and specificity of WB-MRI, 18F-choline PET/CT, and BS were 0.90/0.88, 0.80/0.92, and 0.60/1.00 for M1b disease. ROC-AUC did not show statistically significant improvement for each component of the LSR; mean ROC-AUC 0.92, 0.94, and 0.93 (p < 0.05) for mDixon + DWI, + T2WI, and + contrast respectively. WB-MRI had an interobserver concordance (κ) of 0.79, 0.68, and 0.58 for N1, M1a, and M1b diseases respectively.

Conclusions: WB-MRI provides high levels of diagnostic accuracy for both nodal and metastatic bone disease, with higher levels of sensitivity than BS for metastatic disease, and similar performance to 18F-choline PET/CT. T2 and post-contrast mDixon had no significant additive value above a protocol comprising mDixon and DWI alone.

Key points: • A whole-body MRI protocol comprising unenhanced mDixon and diffusion-weighted imaging provides high levels of diagnostic accuracy for the primary staging of intermediate- and high-risk prostate cancer. • The diagnostic accuracy of whole-body MRI is much higher than that of bone scintigraphy, as currently recommended for clinical use. • Staging using WB-MRI, rather than bone scintigraphy, could result in better patient stratification and treatment delivery than is currently provided to patients worldwide.

Keywords: Choline; Magnetic resonance imaging; Positron emission tomography; Prostate.

Conflict of interest statement

Guarantor

The scientific guarantor of this publication is Shonit Punwani.

Conflict of interest

The authors declare that they have no conflict of interest.

Statistics and biometry

No complex statistical methods were necessary for this paper.

Informed consent

Written informed consent was obtained from all subjects (patients) in this study.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• Prospective diagnostic study performed at one institution

Figures

Fig. 1

Patient recruitment flow diagram

Fig. 2

Flow diagrams of the statistical methods used in the study. a Flow diagram of intermodality and inter-reader concordance. b Flow diagram for WB-MRI LSR ROC-AUC and intermodality diagnostic accuracy analyses. c Patient-based reference standard

Fig. 2

Flow diagrams of the statistical methods used in the study. a Flow diagram of intermodality and inter-reader concordance. b Flow diagram for WB-MRI LSR ROC-AUC and intermodality diagnostic accuracy analyses. c Patient-based reference standard

Fig. 3

Example of discordant skeletal site in a 64-year-old man with a serum prostate-specific antigen level of 25.2. Top left: Negative 99mTc bone scintigram (BS), Top right: 18F-choline PET/CT showing an area of increased tracer avidity at the right inferior pubic ramus consistent with a metastasis. Bottom left: Axial T2W TSE showing the lesion is of low signal intensity. Bottom right: the lesion has increased diffusion-weighted signal on b = 1000s/mm2. The lesion was considered as an example of a true positive whole-body MRI, true positive 18F-choline PET/CT, and false negative BS

Fig. 4

Example of a discordant skeletal site in a 73-year-old man with a serum prostate-specific antigen level of 19.4. Top: Axial T2-weighted turbo spin echo with fused b = 1000s/mm2 showing a metastasis in the left iliac body, which is occult on BS (middle) and 18F-choline PET/CT (bottom). The site was considered to represent a true positive whole-body MRI, false negative 18F-choline PET/CT, and false negative 99mTc bone scintigram