Evaluating the clinical utility of Node-RADS 1.0 for axillary lymph node assessment in breast cancer: comparison with conventional MRI diagnosis.

[OBJECTIVES] To evaluate the clinical utility of Node-Reporting and Data System (Node-RADS) 1.0 in assessing axillary lymph node (ALN) on breast MRI.

[METHODS] This retrospective study included women with clinical T1-T2 stage breast cancer who underwent surgery between April 2014 and March 2023. MRI-ALN status was first evaluated using routine clinical diagnostic criteria. Node-RADS scores were subsequently assigned based on preoperative MRI after a 2-week interval. The optimal cut-off was determined using the Youden index. Diagnostic performance was compared for assessing ALN metastasis (ALNM) and pathological ALN (pALN) burden. In the secondary analysis, differences in imaging features were further evaluated in patients with Node-RADS < 3.

[RESULTS] The optimal cut-off for diagnosing ALNM was Node-RADS >1. Compared with MRI-ALN status, Node-RADS score demonstrated lower sensitivity (56% vs. 71%, P < .01), but comparable specificity (92% vs. 85%, P = .16). For assessing pALN burden, Node-RADS > 2 was the optimal cut-off with lower sensitivity (55% vs. 77%, P < .01) but higher specificity (82% vs. 62%, P < .01). In the secondary analysis, patients with ALNM showed significant differences in cortical morphology and the suspicious breast-side lymph node sign.

[CONCLUSIONS] Node-RADS 1.0 showed lower sensitivity than conventional MRI diagnosis in assessing ALNM but demonstrated good specificity in evaluating ALNM and pALN burden.

[ADVANCES IN KNOWLEDGE] Node-RADS 1.0 showed lower sensitivity than conventional MRI diagnosis in assessing ALNM. The potential values of cortical morphology and the suspicious breast-side lymph node sign in improving the sensitivity of Node-RADS 1.0 warrant further investigation.