Background: Staging patients with high-risk prostate cancer (HRPCa) with conventional imaging of computed tomography (CT) and bone scintigraphy (BS) is suboptimal. Therefore, we aimed to compare the accuracy of whole-body magnetic resonance imaging (WBMRI) with conventional imaging to stage patients with HRPCa.
Methods: We prospectively enrolled patients with newly diagnosed HRPCa (prostate-specific antigen ≥20 ng/ml and/or Grade Group ≥4). Patients underwent BS, CT of the abdomen and pelvis, and WBMRI within 30 days of evaluation. The primary endpoint was the diagnostic performances of detecting metastatic disease to the lymph nodes and bone for WBMRI and conventional imaging. The reference standard was defined by histopathology or by all available clinical information at 6 months of follow-up. To compare diagnostic tests, Exact McNemar's test and area under the curve (AUC) of the receiver operating characteristics curves were utilized.
Results: Among 92 patients enrolled, 15 (16.3%) and 8 (8.7%) patients were found to have lymphatic and bone metastases, respectively. The sensitivity, specificity, and accuracy of WBMRI in detecting lymphatic metastases were 0.60 (95% confidence interval 0.32-0.84), 0.84 (0.74-0.92), and 0.80 (0.71-0.88), respectively, while CT were 0.20 (0.04-0.48), 0.92 (0.84-0.97), and 0.80 (0.71-0.88). The sensitivity, specificity, and accuracy of WBMRI to detect bone metastases were 0.25 (0.03-0.65), 0.94 (0.87-0.98), and 0.88 (0.80-0.94), respectively, while CT and BS were 0.12 (0-0.53), 0.94 (0.87-0.98), and 0.87 (0.78-0.93). For evaluating lymphatic metastases, WBMRI demonstrated a higher sensitivity (p = 0.031) and discrimination compared to CT (0.72 versus 0.56, p = 0.019).
Conclusions: For staging patients with HRPCa, WBMRI outperforms CT in the detection of lymphatic metastases and performs as well as CT and BS in the detection of bone metastases. Further studies are needed to assess the cost effectiveness of WBMRI and the utility of combined PSMA PET and WBMRI.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1038/s41391-024-00893-1 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!