148436-156

A nomogram to predict axillary response to neoadjuvant chemotherapy in clinically node positive breast patients.

Subcategory: 
Category: 
Breast Cancer—Triple-Negative/Cytotoxics/Local Therapy
Session Type and Session Title: 
Poster Session, Breast Cancer—Triple-Negative/Cytotoxics/Local Therapy
Abstract Number: 

1034

Poster Board Number: 
Board #148
Citation: 
J Clin Oncol 33, 2015 (suppl; abstr 1034)
Author(s): 
Jose Vila, Roland L. Bassett, Elizabeth Ann Mittendorf, Isabelle Bedrosian, Simona Flora Shaitelman, Michael Charles Stauder, Mariana Chavez-Mac Gregor, Jennifer Keating Litton, Wei Tse Yang, Lei Huo, Henry Mark Kuerer, Kelly Hunt, Abigail Suzanne Caudle; The University of Texas MD Anderson Cancer Center, Houston, TX; NRG Oncology/NSABP, and the MD Anderson Cancer Center, Houston, TX

Abstract Disclosures

Abstract: 

Background: Many clinically node positive breast cancer patients receive neoadjuvant chemotherapy (NAC), with 40% converting to node negative. Recent trials suggest a potential role for limiting axillary surgery in these patients. This study was undertaken to develop a nomogram predicting the likelihood of axillary pathologic complete response (pCR) in biopsy-proven cN1 patients receiving NAC. Methods: Patients with cT1-4N1M0 disease who received NAC and underwent axillary lymph node dissection were identified. Patients received anthracycline- and/or taxane-based therapy with trastuzumab in HER2+ patients. Estrogen receptor (ER) was recorded as the percentage of cells staining positive. Univariate and multivariate logistic regression analyses were performed to determine factors predictive of nodal conversion. A nomogram to predict the likelihood of nodal pCR was constructed. Results: A total of 614 cN1 patients were included: 93 (15%) cT1, 367 (60%) cT2, 107 (17%) cT3, and 47 (8%) cT4. Imaging showed multifocal disease in 24% (146/614). Receptor status was ER+ in 75% (458/614), PR+ in 64% (394/612), and HER2+ in 20% (124/614). Axillary pCR was achieved in 37% (228/614). On univariate analysis, nuclear grade (OR 13.67 grade 3 vs. 1, p = < .001) and HER2+ status (OR 4.6, p < .0001) were predictive of nodal pCR. Significant negative factors included multifocality on imaging (OR 0.67, p = 0.045), lobular histology (OR 0.53, p < .0001), PR+ (OR 0.24, p < .0001), ER percentage as a continuous variable (OR 0.98, p < .0001) as well as ER+ categorized as > 1% staining (OR 0.29, p < .0001). Nomograms to predict nodal pCR were created using these variables in addition to the clinically significant factors of T stage and number of abnormal nodes on US ( < 4 vs. ≥ 4). The discrimination of the nomogram using ER positive ( > 1% staining) versus negative (AUC = 0.775) was improved using the percentage of ER staining (AUC = 0.789). Conclusions: Multifocality, histology, nuclear grade, ER, PR, and HER2 status predict the ability to achieve nodal pCR with NAC. A nomogram incorporating these factors predicts the likelihood of nodal pCR with NAC which may help guide decisions regarding surgical management of the axilla in these patients.