Utility of node-RADS in preoperatively detecting pelvic lymph node metastasis in bladder cancer.

BACKGROUND

An accurate preoperative nodal staging is crucial for muscle-invasive bladder cancer in risk stratification and treatment decision-making. However, conventional cross-sectional imaging cannot replace pelvic lymph node dissection (PLND) due to accuracy concerns. Recently, Node Reporting and Data System (Node-RADS), a novel structured reporting system for oncological nodal evaluation, has been recently launched. The study is designed to explore the diagnostic utility of Node-RADS in clinical management compared with PLND in bladder cancer.

METHODS

The study retrospectively collected patients diagnosed with bladder cancer who had received a preoperative computed tomography scan and subsequent radical cystectomy. Two radiologists independently reviewed the scans according to Node-RADS 1.0 criteria. Nodal staging was histopathologically confirmed by surgical specimens from PLND. The diagnostic performance of Node-RADS was compared both at patient and regional levels. Cohen's kappa was used to exhibit interreader agreement.

RESULTS

Lymph node metastasis was found in 22/85 patients by standard PLND. Both Node-RADS score>3 and 4 were suitable diagnostic cutoffs. Node-RADS scores highly correlated with the results of PLND. It exhibited promising diagnostic performance with moderate sensitivity (55.00%), high specificity (85.48%), accuracy (78.05%), and excellent interreader agreement (kappa 0.73-0.95). Whereas, Node-RADS was not observed to have obviously superior diagnostic performance to nodal size criteria in our cohort.

CONCLUSION

The study supported the utility of Node-RADS in preoperative pelvic nodal staging, which may assist urologists in selecting appropriate nodal templates. A high positive predictive value enabled the early identification of suitable patients for multimodal treatment to achieve optimal oncological outcomes.