Extranodal extension of axillary metastasis of invasive breast carcinoma as a possible predictor for the total number of positive lymph nodes

Introduction: By the implementation of the sentinel node procedure in the treatment of breast carcinoma routine axillary lymph node, dissection can be abandoned in patients with a tumour-negative sentinel node. When the sentinel node is positive there are two options; an axillary dissection or radiotherapy of the axilla. In the latter case one is not informed about the total number of positive lymph nodes which can be of importance for the choice of adjuvant chemotherapy. In this paper we analyse whether it is possible to use histological parameters of a lymph-node metastasis to predict the number of tumour-cell-containing nodes. Methods: Four hundred and ninety-eight patients treated for invasive breast cancer at our department from 1991 to 1996 were investigated to see whether extranodal extension of axillary metastases had a significant predictive value for the number of positive lymph nodes. Extranodal extension was scored as: no extranodal extension (NEE) and extranodal extension (EE); the latter was subdivided in minimal extranodal extension (MEE) or extensive extranodal extension (EEE). Results: Of 498 patients, 230 patients had axillary involvement. NEE was seen in 83 (36.1%) patients and 147 (63.9%) had EE. Subdivision of this latter group revealed 77 patients with MEE (52%), 65 patients with EEE (45%) and five patients not further specified (3%). The number of positive nodes for the EE-group (6.9±0.5) was significant higher compared with the NEE-group (2.1±0.2) (P<0.001). The number of positive nodes was also significantly higher for the EEE-group compared to the MEE-group, 10.6±0.8 vs 4.0±0.4 (P<0.001). The predictive value for ≥4 positive axillary lymph nodes was 84.6% for EEE, 58.5% for EE and only 14.5% for NEE. These differences were significant (P<0.001). Conclusions: The presence of extranodal extension in axillary lymph-node metastases can be a good predictor for the expected number of positive nodes. This histological parameter could be of value to determine the kind of adjuvant chemotherapy after a positive sentinel-node biopsy with only radiotherapy of the axilla and no further axillary lymph-node dissection.