G-4593

Can Full Protocol MRI Reduce the Reliance on Invasive Axillary Biopsies in Breast Cancer Staging? A Number Needed to Diagnose and Biopsy Approach

 Amirmohammad Azizzadeh ¹ ℗, Fahimeh Zeinalkhani ² ©, Fatemeh Mahdavi Sabet ³, Peyman Kamali Hakim ⁴, Hadise Zeinalkhani ⁵, Elahe Baban Taher ², Simin Hashemi ⁶, Afrooz Moradkhani ², Mahdiyeh Movahedi ⁷, Maryam Movahedi ²   

 
Abstract

Introduction: Accurate axillary lymph node (ALN) staging is critical for breast cancer prognosis and treatment. While sentinel lymph node biopsy (SLNB) is the current care standard, it is an invasive surgical procedure with associated morbidities, including lymphedema and paresthesia. This study evaluated the diagnostic accuracy of full protocol magnetic resonance imaging (MRI) for preoperative ALN assessment, aiming to frame its clinical utility with novel, patient-centered metrics that can better inform clinical decision-making. Methods and Materials: This retrospective cohort included 88 women with biopsy-proven, Stage I-III, non-distant-metastatic, breast cancer between 2022–2024. All imaging was performed on a 1.5T Philips system. Patients with prior neoadjuvant therapy or incomplete data were excluded. Two blinded expert radiologists independently reviewed the MRI scans. The criteria for metastatic nodes included short-axis size 10mm, cortical thickening (3mm), loss of the fatty hilum, and a round shape. SLNB or axillary lymph node dissection (ALND) served as the definitive reference standard. Analysis was conducted using R v4.4.3 and Jamovi v2.6. We calculated sensitivity, specificity, positive and negative likelihood ratios (LR+ and LR-), and clinical utility indices. The number needed to biopsy (NNB) is the number of biopsies required to detect one case of disease. The number needed to diagnose (NND) represents the average number of tests necessary to detect one person with disease. And finally, the number needed to misdiagnose (NNM) is the number of tests for every one misdiagnosed case. Results: Histopathology confirmed ALN metastases in 53 patients (60.2%). Full protocol MRI demonstrated a sensitivity of 88.7% and a specificity of 82.9%. The LR+ was 5.17 (95% CI: 2.48–10.78) and the LR- was 0.14 (95% CI: 0.06–0.29). The overall diagnostic odds ratio was 37.86 and the Youden's index was 0.71. In terms of clinical utility, the NND was 1.40 and the NNM was 7.35. Most notably, MRI guidance reduced the NNB to detect one case of metastasis from a baseline of 1.66 (without MRI) down to 1.12. Conclusion and Discussion: Full protocol MRI demonstrated high diagnostic accuracy for ALN staging of breast cancer. The LR-, approaching the 0.1 threshold for strong evidence, signifies that a negative MRI result substantially decreases the post-test probability of ALN metastasis. The NNM demonstrates that the potential for MRI misdiagnosis is relatively low and the NND of 1.40 indicates high diagnostic efficiency. The NNB suggests that MRI can help prevent some of the invasive staging procedures. This study provides a practical framework for assessing how imaging can impact the clinical pathway. These findings suggest that full protocol MRI can help guide decisions regarding the necessity of invasive procedures, potentially sparing patients from surgical morbidities.

Keywords: Breast neoplasms, Magnetic resonance imaging, Lymphatic metastasis, Sensitivity and specificity