Predictive factors associated with axillary lymph node metastases in T1a and T1b breast carcinomas: Analysis in more than 900 patients

Background: Axillary lymph node metastasis (ALNM) represents the single mos t important prognostic indicator in patients diagnosed with breast cancer. The proportion of less than or equal to 1-cm (T1a, T1b) invasive breast car cinomas is increasing. The incidence and predictive factors associated with ALNM in patients with less than or equal to 1-cm tumors remains unclear an d the role of axillary lymph node dissection in these patients has been que stioned. The purpose of this study was to determine clinical and pathologic factors predictive of ALNM in patients with less than or equal to 1-cm inv asive breast carcinomas by univariate and multivariate analyses. Study Design: Review analysis from a prospective database identified patien ts with less than or equal to 1-cm invasive breast cancers treated at our i nstitution between 1990 and 1996. All patients underwent a resection of the primary tumor and axillary lymph node dissections. Routine patient and tum or characteristics evaluated included: age, race, tumor size, histologic gr ade, estrogen and progesterone receptor status, and lymphatic and vascular invasion. Univariate and multivariate analyses were performed. Adjusted odd s ratios (OR) and 35% confidence intervals (CI) are presented. Results: A total of 919 patients were identified in this study with tumors less than or equal to 1 cm. These included 199 patients (21.7%) with T1a tu mors and 720 patients (78.3%) with T1b tumors. ALNM was detected in 165 pat ients with an overall incidence of 18.0%. Of the ALNM group, 32 patients (1 9.4%) had Tla tumors and 133 patients (80.6%) had T1b tumors. Four variable s were found to be significant in univariate analysis. These included: incr easing tumor size, poor histologic grade, presence of lymphatic or vascular invasion, and younger age of the patient. An increase in tumor size was as sociated with a significant risk of ALNM (OR = 2.66, 95% CI = 1.28 to 5.75; p = 0.01). Poor tumor grade and the presence of lymphatic or vascular inva sion were also associated with an increased risk of ALNM (OR = 2.69, p = 0. 003 and OR = 5.52, p = 0.0001, respectively). Patients with ALNM were more likely to have a tumor grade of 3 (25.0% ALNM versus 12.5% node-negative, p = 0.004) and lymphatic or vascular invasion (16.9% ALNM versus 3.5% node-n egative, p < 0.0001). In multivariate analysis, an increased risk of ALNM w as demonstrated with increasing tumor size (0.1-cm increments), poor histol ogic grade, and younger age. Conclusions: This study investigated clinical and pathologic factors influe ncing ALNM in patients with T1a and T1b breast carcinomas. We have identifi ed three factors by multivariate analysis as significant independent predic tors of ALNM in this group of patients. These include increasing tumor size , poor histologic grade, and younger age. Given the significant amount of A LNM demonstrated in this study (overall 18%) and the inability to identify a subgroup of patients that had an acceptable low risk of ALNM, the complet e omission of assessing the axilla for metastatic disease in patients with small breast cancers cannot be advocated. Our recommendation for patients d iagnosed with T1a and T1b tumors is to have their axilla investigated for m etastatic disease either by traditional axillary lymph node dissections or by intraoperative lymphatic mapping and sentinel lymph node biopsy techniqu es. (J Am Coil Surg 2000;191:1-8. (C) 2000 by the American College of Surge ons).