Is sentinel node biopsy reliable in large breast tumors?

Purpose: The value of sentinel lymph node biopsy (SNB) in patients with larger breast tumors (diameter > 3 cm) has been questioned due to high false-negative rates reported from initial studies. The aim of this study was to analyze the safety and prognostic reliability of SNB in this group of patients. Methods: During a 6-year period (2001-2007), 84 women with mean age 51.7 +/- 11.6 years diagnosed with a breast tumor larger than 3 cm in diameter on pathological analysis were retrospectively identified from the database of Our institution. Sentinel node identification was performed after injection of blue dye subcutaneously at the subareolar area. The sentinel node specimen was sent for frozen section analysis. Regardless of the SNB results, all patients underwent completion axillary clearance. Results: Breast surgery consisted of mastectomy in 62 patients (73.8%) and partial mastectomy in 22 patients (26.2%). There were 69 invasive ductal cancers (82.1%), 14 lobular cancers (16.6%) and one case of anaplastic carcinoma (1.3%). Nine tumors (10.7%) were identified to be multifocal after the histopathological report. The mean number of sentinel nodes removed was 1.5 +/- 0.7 (range 1-4) while SNB detection was not feasible in three patients (3.6%). Of 56 positive SNBs, seven (12.5%) were not identified by routine hematoxylin and eosin staining during frozen section analysis but were detected by subsequent immunohistochemistry on the final histopathological report. All patients with multifocal tumors presented nodal metastases on pathological analysis (100%), while the rate of nodal metastatic disease in patients with unifocal tumors was 16% (12 patients), although no statistical significance was documented. The overall false-negative rate. defined as the percentage of all node-positive tumors in which the SNB was negative. was 14.3%. The false-negative rate was significantly higher for the group of patients with multifocal tumors (55.5%) compared to the group with unifocal tumors (9.3%) (p < 0.001). Conclusions: The present study indicates that sentinel node biopsy is feasible in patients with larger breast tumors (max. diameter > 3 cm), with comparable false-negative and sentinel detection rates (14.3% and 96.4%, respectively). Larger tumor size seems to be associated with increased incidence of nodal metastases while multifocality appears to be related to increased false-negative rates: hence completion axillary clearance should be initially considered for these cases.