Micrometastases in sentinel lymph node in a multicentric study: predictive factors of nonsentinel lymph node involvement--Groupe des Chirurgiens de la Federation des Centres de Lutte Contre le Cancer.

PURPOSE: To determine the rate of nonsentinel lymph node (NSN) involvement at axillary lymph node dissection (ALND) and predictive factors of this involvement following detection of micrometastasis in sentinel nodes (SN).
METHODS: We analyzed 700 observations of SN micrometastases with additional ALND with the characteristics of the patients, tumors, and SN.
RESULTS: Involvement of SN was diagnosed 388 times by serial sections (55.4%) with standard hemoxylin and eosin staining (HES) and 312 times solely on immunohistochemical analysis (IHC; 44.6%). The accurate size of the micrometastases was indicated in 488 cases: 301 larger than 0.2 mm (61.7%) and 187 < or = 0.2 mm (38.3%). Ninety-four patients (13.4%) presented an NSN involvement with only one NSN involved in 62 cases (66%). Predictive factors of NSN involvement were in univariate analysis (pT stage [P < .000], menopausal status [P = .048], T stage [P = .006], grade [P = .013], lymphovascular invasion [LVI; P = .013], histologic tumor type [P = .017], and method of micrometastasis detection, by HES or IHC [P = .015]) and in multivariate analysis (pT stage < or = or > 20 mm [odds ratio, 2.54], micrometastases detected by HES or IHC [odds ratio,1.734], presence or absence of LVI [odds ratio, 1.706]). Micrometastasis size < or = or greater than 0.2 mm was not predictive.
CONCLUSION: This study confirms the value of serial sections and the vital role played by IHC in screening for small micrometastases. Omission of additional ALND may be envisaged with minimal risk for pT1a and pT1b tumors, and pT1a-b-c tumors corresponding to tubular, colloidal, or medullar cancers.