Value Of Sentinel Lymph Node Metastasis Status In Predicting The Presence Of Residual Disease In The Non-sentinel Lymph Node Of Breast Cancer Patients

BackgroundSentinel lymph node biopsy (SLNB) has been widely used in the treatment of breast cancer. Many large-scale clinical studies have shown that SLNB accurately predicts the status of axillary lymph node metastasis. Axillary lymph node dissection (ALND) is not performed for sentinel lymph node (SLN)-negative patients; a complete ALND is performed when SLN is positive.Some studies indicated that40%-60%of SLN-positive patients did not show non-sentinel lymph node (nSLN) involvement.4-9This potentially implies that those patients may not require ALND.Objective:To investigate the value of sentinel lymph nodes metastasis status in predicting the presence of residual disease in the non-sentinel lymph nodes and the feasibility of avoiding or reducing the scope of axillary lymph node dissection for the patients with single positive SLN. The characteristics and predictive factors for non-sentinel lymph node metastasis of patients with single positive SLN were also analyzed.MethodsThis article reported a retrospective study of2265patients with invasive breast carcinomas who were undergone sentinel lymph nodes biopsy (SLNB) in Shandong Cancer Hospital between November1999and December2011.There were1228 (54.2%) patients who satisfied the following criteria:(1) diagnosis of breast cancer by needle aspiration or excisional biopsy, and no significant palpation of axillary lymph nodes,(2)Preoperative ultrasonography, CT and/or MRI and other imaging examination for evaluation of axillary lymph node status (3) no preoperative anti-cancer therapy,(4) different levels of ALND. Divided into (-),(1/n),(1/1),(n/N),(n/n)(n>2,N>3,N>n) five groups according to the SLN metastasis,and comparative analysis the transfer conditions of non-sentinel lymph node metastasis between them. For the1/n group, the factors that influenced the nSLN metastasis were analyzed by univariate and multivariate analysis.SPSS19.0software was applied according to the criteria of the University of Louisville SLNB. Chi-squareor test was used to compare the sample rate (multiple pair comparisons with Chi-square segmentation) and Fisher exact test with (α=0.05). Single factor analysis was performed with fourfold table exact test (Fisher’s exact test). Multivariate analysis was performed with the logistic regression model.Results1. SLNB false negative rate of this group was10.69%(73/683),sensitivity was89.3%(610/683). negative predictive value was88.2%(545/618),positive predictive value was100%,accuracy rate was94.1%(1155/1228).Clinical tumor size (P=0.030),dates of surgery(P=0.001) and numbers of detected SLNs (P=0.000) were significantly related with false negative rate of SLNB by single factor analysis.Dates of surgery(P=0.034) and numbers of detected SLNs (P=0.012) were the independently predictive factors for the false negative rate of SLNB by logistic regression model.2. Then-SLN metastasis rate of the SLN(-),(1/n),(1/1),(n/N),(n/n) groups was11.8%(73/618),25.2%(65/258),49.6%(67/135),48.4%(60/124) and65.6%(61/93)respectively.Comparison of SLN (-),(1/n),(1/1),(n/N), and (n/n) groups of nSLN metastasis showed a significant difference (P=0.000) The differences of the nSLN metastasis between the SLN(-) and other groups(includel/n,1/1, n/N, n/n group) were significant (P=0.000). This difference was also significant between SLN (1/n) and other positive groups (include1/1, n/N, n/n group)(P=0.000), but not significant between SLN(1/1),(n/N) and (n/n) groups.(P=0.842,0.017,0.042respectively,Chi-square segmentation).There were no significant difference between the axillary lymph node metastasis on Level Ⅱ and Ⅲ of the SLN1/n group and SLN(-) group(P=0.012,0.570Chi-square segmentation).3. When the total SLN number was (4and with one positive case, the nSLN metastasis was not significantly different from that in the0/n group(P=0.118). In the1/n group, clinical tumor size (P=0.012), over-expression of HER-2(P=0.003), tumor grade(P=0.018) and the total number of SLN (P=0047) significantly correlated with non-SLN metastasis. Clinical tumor size (P=0.015) and the over expression of HER-2(P=0.010) were independent predictive factors for non-SLN metastasis by the logistic regression model.Conclusion1. False negative rate of SLNB was significantly related to dates of surgery and numbers of detected SLNs. The improvement of operation proficiency,increaseing numbers of detected SLNs and standard operation procedure could reduce the false negative rate of SLNB.2.The status of SLN metastasis is one of factors that influence the nSLN metastasis of the patients with invasive breast cancer. The possibility of non-sentinel lymph node involvement for the patients with single SLN metastasis is smaller than other SLN-positive patients.We can not avoid ALND only according this condition for the patients with single positive SLN (SLN1/n n>2),clinicopathological variables should be considered.3. SLN1/n metastasis patients cannot completely avoid ALND. However, under conditions of n≥4, primary tumor≤<2cm, no HER-2over-expression (-or+), and histological grade of I, ALND could be avoided.For SLN1/n metastasis patients with the above risk factors, ALND of level I maybe enough, but prospective studies with larger number and multi-center are needed.