Feasibility of Sentinel Lymph Node Biopsy in Multiple Unilateral Synchronous Breast Cancer: Results of a French Prospective Multi-Institutional Study (IGASSU 0502).

Abstract

Abstract Introduction: Controversy remains over whether to perform sentinel lymph node biopsy (SLNB) in multiple (multicentric or multifocal) unilateral synchronous breast cancer. Several small retro- or prospective studies, included pre- or post-operative (or both) multiple synchronous tumours (MST) have suggested that the test performance of SLNB is similar to that seen in unifocal disease. The purpose of this study was to evaluate the feasibility and accuracy of SLNB in preoperatively diagnosed invasive MST.Patients and Methods: The Interest of Axillary SLNB in Multiple Invasive Breast Cancer (IGASSU) study was a prospective multi-institutional study with initial breast surgery, SLNB, and systematic level I to II axillary lymph node dissection (ALND). Patients eligible for the IGASSU study had an operable invasive MST, defined as two or more physically separate invasive tumours in the same or different breast quadrant. The diagnosis of invasive MST was confirmed histologically in all patients by core needle biopsy before surgery. Detection of sentinel node was performed by using either blue patent or radiocolloide injection or both. Injection sites were subareolar.Statistical Analysis: Sensitivity (Se), negative predictive value (NPV), accuracy (A), false negative rate (FNR) and their 95%-confidence intervals (95%CI) were calculated with a classical 2x2 contingency table. A univariate analysis using odds ratio calculation was performed to identify the risk factors for false negative results.Results: Between March 1, 2006, and August 31, 2007, 216 patients were prospectively included from 16 institutions. Of these patients, 211 were evaluable. The SLNB identified rate were 93.4% (197/211). A mean number of 2.2 SLN (range, 1 to 8, ± 1.4) was successfully excised. The mean number of resected nodes in ALND was 12 (range, 1 to 39, ± 5.7). The FNR was 13.6% (14/103) [95%CI: 7- 20%], Se was 86 .4% (89/103) [95%CI: 79- 93%], NPV was 87% (94/108) [95%CI: 80-93%], A was 92.9% (183/197) [95%CI: 89- 96%]. For the 14 false-negative SN, all had ≤ 3 involved nodes in ALND. Table 1 shows patterns of management of the axilla.Table 1 ALND+ALND-Non-identified SLNB113Identified SLNB+4544Identified SLNB-1494 In a univariate analysis, tumour location (only external location vs other location) was the only clinico-pathological factor influencing the FNR (22% [11-33%] vs 7% [4-10%]), even then median aggregate histological tumour size was smaller in external tumours (17mm [range, 12-80] vs 34mm [range, 8- 90], p=0.016).Conclusion: With a FNR of 13.6% (95%CI: 7-20%), we do not recommend SLNB as a routine procedure for multiple unilateral synchronous breast cancer, even for small tumour foci. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 305.