Sentinel lymph node metastases detected by immunohistochemistry only do not mandate complete axillary lymph node dissection in breast cancer

Background: The significance of breast cancer sentinel lymph node (SLN) metastases detected only by immunohistochemistry staining (IHC) remains poorly understood. This study attempted to quantify the risk of non-SLN metastases. Methods: A prospectively collected database of 750 consecutive SLN biopsy procedures in breast cancer patients was reviewed. Medical records were reviewed to supplement the database. Results: SLNs were identified in 738 (98.4%) of these procedures in 723 patients. Of these, 151 patients (20.5%) had metastases detected by hematoxylin and eosin staining (H&E), and 33 (4.6%) of the 718 with known IHC staining results had metastases detected by IHC only. Twenty-eight (84.8%) of 33 patients with IHC-detected metastases underwent complete axillary lymph node dissection (CALND). The median primary tumor size was 2.0 cm among those undergoing CALND and 0.9 cm among the five patients treated without CALND (P = .10). Two of the 28 patients (7.1%) had additional metastases detected with CALND. These patients had a T3 or T4 invasive lobular primary tumor. Of 24 patients with T1 or T2 primary tumors and IHC-detected metastases who underwent CALND, none had additional metastases detected. Median follow-up was 14.5 months. All patients with IHC-detected SLN metastases were treated with adjuvant systemic therapy. None of the five patients with IHC-detected metastases not undergoing CALND has subsequently manifested clinical axillary disease. Conclusions: CALND could have been or was safely omitted in 29 of 29 patients with T1 or T2 primary tumors and metastases detected by IHC. Such patients should be counseled about this low risk before CALND is recommended.