Abstract PO1-23-01: A novel nomogram for predicting the possibility of omitting axillary lymph node dissection after neoadjuvant chemotherapy for clinical node-positive breast cancer

Abstract Purpose: Currently, there is no validated strategy established for omitting axillary lymph node dissection (ALND), when axillary node metastases are expected to disappear with neoadjuvant chemotherapy (NAC) for clinical node-positive (cN+) breast cancers. In contrast, it was reported that the axillary pathological complete response (pCR) rate to NAC for cN+ breast cancer is approximately 40%. Therefore, it is necessary to develop a more accurate method for predicting nodal pCR (ypN0) to select patients who can omit ALND after NAC for cN+ breast cancer. Methods: This is a single-centered, retrospective clinical study conducted in 128 patients with cN+ primary breast cancer who underwent ALND after NAC between January 2014 and July 2021 at Saitama Medical Center. Clinical complete response (cCR) was defined as the disappearance of the primary tumor or axillary node metastasis, as confirmed by the findings of imaging. The pCR was defined as the absence of microscopic evidence of residual invasive carcinoma in primary breast tumor or axillary nodes. Clinicopathological variables were compared using Fisher’s exact test or logistic regression, as appropriate. Statistical significance was set at p< 0.05. Results: The median age of the patients was 52 years; 48.4% were premenopausal, 38.3% had cT3 or higher, 22.7% had cN2 or higher, and 50% had stage III disease. Estrogen receptor (ER), progesterone receptor (PgR), and human epidermal growth factor receptor 2 (HER2) positivity accounted for the patients were 58.6%, 50%, and 31.3%, respectively. NAC included sequential taxanes and anthracyclines in 47.7% of the patients, and anti-HER2 therapy was used in all patients of HER2-positive breast cancer. Breast primary tumor cCR (ypT0) was 29.7% and ypN0 was 44.5%. The ypN0 rate was 19.5% higher than the ypT0 rate in all patients (p < 0.001); it was 4.8% higher in the luminal subtype (p=0.009), 25.5% higher in the triple-negative subtype (p < 0.001), and 29.3% higher in the HER2 subtype (p=0.015). Additionally, the ypN0 rate was 3.9% higher than the axillary node cCR (ycN0) rate in all patients (p < 0.001), 7.6% lower in luminal type (p < 0.001), 13.6% higher in triple-negative type (p < 0.001), and 5% higher in the HER2 subtype (p=0.006). When ycN0 was predicted to be ypN0, the negative predictive value (NPV) was 77.2% and the false negative rate (FNR) was 19.7% (Table 1). In the association between ypN0 and preoperative clinico-pathological factors in cN+ breast cancer treated with NAC, univariate analysis showed significant differences in ER negativity (ER-) [hazard ratio (HR): 0.186, 95% confidence interval (CI): 0.086–0.401, p< 0.001], PgR negativity (HR: 0.360, 95% CI: 0.176–0.736, p=0.005), HER2 positivity (HER2+) (HR: 4.394, 95% CI: 1.941–9.949, p< 0.001), ycT0 (HR: 10.667, 95% CI: 4.020–28.304, p< 0.001), and ycN0 (HR: 9.966, 95% CI: 4.399–922.576, p< 0.001). The multivariate analysis revealed that ER, ycT0, and ycN0 were independent predictive factors for ypN0 after NAC in the patients with cN+ breast cancer (Table 2). These independent factors in multivariate analysis were used to create a nomogram for ypN0 prediction. The points added were 82 if the patient was ER-, 56 if the patient was ycT0, and 100 if the patient was ycN0, and the total of these scores was used to predict whether the patient had ypN0. When the Receiver Operating Characteristic curve was used to determine the cut-off value of the total points, 119 or more points were predicted to be ypN0, the NPV was 92.9%, and the FNR was 4.5%, which demonstrated an approximately 15% improvement over the prediction of ypN0 by ycN0 alone (Table 3). Conclusion: ER, ycT0, and ycN0 are independent predictive factors of ypN0 in cN+ breast cancer, suggesting that ypN0 prediction using a nomogram may contribute to individualized axillary treatment. Table 1. Accuracy of ypN0 prediction with ycN0 after NAC for clinical node-positive breast cancer Table2. Univariate and multivariate analyses of the relationship between preoperative clinicopathological factors and pathologically axillary node-negative after neoadjuvant chemotherapy in clinical node-positive breast cancer Table 3. Accuracy of ypN0 prediction after NAC for clinical node-positive breast cancer using Nomogram Citation Format: Hirohito Seki, Yuki Ishiguro, Ken Shimizu, Akitsugu Makino, Yoshiharu Ishizaka, Ai Tsuchiya, Hirotsugu Isaka, Shigeru Imoto. A novel nomogram for predicting the possibility of omitting axillary lymph node dissection after neoadjuvant chemotherapy for clinical node-positive breast cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-23-01.