Performance of node reporting and data system (node-RADS): a preliminary study in cervical cancer.

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• The Node Reporting and Data System (Node-RADS) is a standardized approach for evaluating lymph nodes (LNs) across various anatomical locations.
• This study focuses on assessing the diagnostic utility of Node-RADS specifically in the context of cervical cancer.

Purpose of the Study:

• To evaluate the diagnostic performance of the Node-RADS in predicting lymph node metastasis (LNM) in patients with cervical cancer.
• To determine the accuracy of Node-RADS scores in identifying LNM at both the lymph node and patient levels.

Main Methods:

• Retrospective analysis of 81 cervical cancer patients who underwent radical hysterectomy and lymph node dissection.
• Preoperative MRI scans were independently reviewed by two radiologists using the Node-RADS criteria.
• Statistical analyses, including Chi-square, Fisher's exact tests, ROC curve analysis, and AUC calculations, were performed to assess diagnostic performance.

Main Results:

• Lymph node metastasis (LNM) rates varied by anatomical region, with the highest rates in the external iliac (21.0%) and internal iliac (19.8%) regions.
• At the patient level, increasing Node-RADS scores correlated with higher LNM rates, ranging from 26.1% for score 1 to 90.9% for score 5.
• The optimal cut-off value for Node-RADS at both patient and LN levels was determined to be >3, yielding the best AUC and accuracy.

Conclusions:

• Node-RADS demonstrates effectiveness in predicting high likelihood of LNM for scores 4 and 5.
• Lower Node-RADS scores (1 and 2) showed higher-than-expected LNM rates (>25%), indicating a potential discrepancy with the system's intended probability stratification.
• Further refinement or validation may be needed for lower Node-RADS scores to improve their predictive accuracy in cervical cancer.