Prospective study found that peripheral lymph node sampling reduced the false-negative rate of sentinel lymph node biopsy for breast cancer.

Background: Although sentinel lymph node biopsy (SLNB) can accurately predict the status of axillary lymph node (ALN) metastasis, the high false-negative rate (FNR) of SLNB is still the main obstacle for the treatment of patients who receive SLNB instead of ALN dissection (ALND). The purpose of this study was to evaluate the clinical significance of SLNB combined with peripheral lymph node (PLN) sampling for reducing the FNR for breast cancer and to discuss the effect of "skip metastasis" on the FNR of SLNB.

Methods: At Shandong Cancer Hospital Affiliated to Shandong University between March 1, 2012 and June 30, 2015, the sentinel lymph nodes (SLNs) of 596 patients with breast cancer were examined using radiocolloids with blue dye tracer. First, the SLNs were removed; then, the area surrounding the original SLNs was selected, and the visible lymph nodes in a field of 3-5 cm in diameter around the center (i.e., PLNs) were removed, avoiding damage to the structure of the breast. Finally, ALND was performed. The SLNs, PLNs, and remaining ALNs underwent pathologic examination, and the relationship between them was analyzed.

Results: The identification rate of SLNs in the 596 patients was 95.1% (567/596); the metastasis rate of ALNs was 33.7% (191/567); the FNR of pure SLNB was 9.9% (19/191); and after the SLNs and PLNs were eliminated, the FNR was 4.2% (8/191), which was significantly decreased compared with the FNR before removal of PLNs (P = 0.028). According to the detected number (N) of SLNs, the patients were divided into four groups of N = 1, 2, 3, and ≥4; the FNR in these groups was 19.6, 9.8, 7.3, and 2.3%, respectively. For the patients with ≤2 or ≤3 detected SLNs, the FNR after removal of PLNs was significantly decreased compared with that before removal of PLNs (N ≤ 2: 14.0% vs. 4.7%, P = 0.019; N ≤ 3: 12.2% vs. 4.7%, P = 0.021), whereas for patients with ≥4 detected SLNs, the decrease in FNR was not statistically significant (P = 1.000). In the entire cohorts, the "skip metastasis" rate was 2.5% (15/596); the FNR caused by "skip metastasis" was 2.1% (4/191).

Conclusions: The FNR of SLNB was associated with the number of SLNs. For patients with ≤3 detected SLNs, PLN sampling can reduce the FNR of SLNB to an acceptable level of less than 5%. Because of the existence of the "skip metastasis" and distinct metastasis patterns, the FNR of SLNB cannot be completely eliminated.