Non-sentinel Node Involvement Research Articles

Criteria in Sentinel Lymph Nodes of Melanoma Patients that Predict Involvement of Nonsentinel Lymph Nodes

Previous studies described various criteria in sentinel lymph nodes (SLN) of melanoma patients that predict the involvement of further, nonsentinel lymph nodes (NSLN). Such criteria may facilitate the selection of patients who might benefit from a completion lymph node dissection (CLND). However, it is currently unclear which parameters are most important. A total of 180 melanoma patients with positive SLNB and subsequent CLND were investigated. Histopathologic parameters in the SLN were systematically evaluated and compared with regard to NSLN positivity. Twenty-eight of these patients (16.0%) had positive NSLN. By univariate analysis several criteria with regard to tumor burden and location of melanoma cells in the SLN correlated with NSLN involvement, such as positivity by hematoxylin-eosin (H&E) staining (P < .001), largest diameter of clusters (P < .001), capsular involvement (P = .001), extranodal extension (P < .001), and tumor penetrative depth (P < .001). Multivariate analysis revealed three independent parameters: (1) positivity of the SLN by H&E staining (versus by immunohistochemistry alone), (2) relative tumor burden > 10% of total lymph node tissue, and (3) perinodal intralymphatic tumor. In 23 of 28 patients with positive NSLN the SLN was positive by H&E staining, in 15 of 28 patients the relative tumor burden was > 10%, and 13 of 28 showed perinodal intralymphatic tumor. In 5 of 28 patients with NSLN involvement, these three parameters were negative. Histopathologic examination of the SLN can identify patients at risk for NSLN positivity.

Sentinel lymph node biopsy for breast cancer using a new camera system for simultaneous capturing color and near-infrared fluorescence of indocyanine green

The sentinel node (SN) technique has gained a key role in breast cancer surgery, allowing for an accurate staging of the axillary status with a minimally invasive resection. In this study, we explored the implication of three proteins (E-cadherin, a- and b-catenins) that form the cadherin–catenin complex, a receptorial structure strictly involved in tumoral vascular invasion and embolization in this biologic event. We studied the immunohistochemical expression of the complex in patients with metastatic SN, matching the group with involved nonsentinel lymph nodes (NSNs) with that having free axillary NSNs. The simultaneous staining of the SN metastases for the three proteins has been considered an indicator of preserved function. Our data confirmed the lack of cadherin–catenin complex in tumors with lobular morphology even in SN metastasis, but statistical evaluation could not prove a significant relation between complex integrity and NSN involvement. Moreover, considering traditional histopathologic parameters, only vascular peritumoral embolization was related to an increased risk of metastatic spread to axillary NSNs.

Factors Predictive of Nonsentinel Lymph Node Involvement and Clinical Outcome in Melanoma Patients With Metastatic Sentinel Lymph Node

Factors Predictive of Nonsentinel Lymph Node Involvement and Clinical Outcome in Melanoma Patients With Metastatic Sentinel Lymph Node

The MSKCC nomogram for prediction the likelihood of non-sentinel node involvement in a German breast cancer population

To assess whether the Memorial Sloan Kettering Cancer Center (MSKCC) nomogram for prediction of NSLN metastasis is useful in a German breast cancer population and whether the characteristics of the breast tumor and the sentinel lymph node (SLN) are able to predict the likelihood of non-sentinel lymph node (NSLN) metastasis. A total of 545 patients with primary breast cancer and SLN examination were evaluated. The MSKCC nomogram was applied to 98 patients with a positive SLN who subsequently had completion axillary lymph node dissection (ALND). Predictive accuracy was assessed by calculating the area under the receiver-operator characteristic (ROC) curve. The collective was evaluated by correlating the prevalence of NSLN and SLN metastasis to pathological features. The MSKCC nomogram achieved a ROC of 0.58 indicating a bad accuracy of the nomogram. Tumor size, histology, lymphovascular infiltration, multifocality, Her-2-neu positivity, and nuclear grade correlated with the probability of SLN metastasis. Histology and primary tumor localization correlated significantly with the probability of NSLN metastasis. The MSKCC nomogram did not provide a reliable predictive model in our study population. However, the likelihood of SLN metastasis correlated with the presumed risk factors and no obvious differences between the MSKCC population and our population could be seen. In order to achieve interinstitutional reproducibility, standardization of surgical procedure and of the pathological assessment of the SLN is desirable.

Factors Predictive of Nonsentinel Lymph Node Involvement and Clinical Outcome in Melanoma Patients With Metastatic Sentinel Lymph Node

Identification of melanoma patients who need completion lymphadenectomy and adjuvant treatment after positive sentinel lymph node (SLN) biopsy would be a fundamental step forward toward personalized medicine. This study tested the hypothesis that the microscopic features of metastatic SLNs might predict not only nonsentinel lymph node (NSLN) status, but also patients' clinical outcomes. A retrospective analysis was performed on 96 consecutive melanoma patients who underwent completion lymphadenectomy after positive SLN biopsy. Patients' age and sex, primary tumor Breslow thickness, number of positive SLNs, the largest diameter and depth of invasion of metastatic deposits in the SLN, S stage, and pattern of nodal involvement were correlated with the presence of metastatic disease in NSLNs as well as with the likelihood of tumor recurrence and patient death. At pathological examination, 20 patients (20.8%) had metastatic melanoma in the NSLN. Pattern of nodal involvement, depth of invasion of SLN by metastatic disease, and S stage were statistically significantly associated with the presence of metastatic disease in NSLN. Multivariate analysis revealed that only the SLN depth of invasion was an independent predictor of NSLN status (P = .0035). This parameter was also significantly associated with disease-free and overall survival, both by univariate (P < .0001 and P = .0006, respectively) and multivariate (P < .0001 and P = .0013, respectively) survival analysis. These findings support further investigation of SLN depth of invasion as a predictive factor of potential clinical use to select patients as candidates for completion lymphadenectomy and adjuvant treatment.

O-67 Prediction of axillary lymph node metastasis by size and grade of tumour an aid for the discussion of axillary surgery in patients with operable breast cancer

Recent literature has suggested that completion axillary lymph node dissection (ALND) in breast carcinoma patients with positive SLN may not be necessary. However, a method for determining the risk of non-SLN or extranodal disease remains to be established.To determine if pathological variables from primary tumors and sentinel lymph node (SLN) metastases could predict the probability of non-sentinel lymph node (NSLN) metastases and extranodal disease in patients with breast carcinoma and SLN metastases.84 women with T1-3 breast cancer and clinically-negative axillae underwent completion ALND. Maximum diameter and width of SLN metastases were measured to calculate metastatic area. When multiple SLNs contained metastases, areas were summed to calculate the Total Metastatic Area (TMA). Multiple linear regression models were used to identify predictive factors.Her-2/neu over-expression increased the odds of NSLN metastases (OR 4.3, p = 0.01) and extranodal disease (OR 7.9, p < 0.001). Independent SLN predictors were ≥1 positive SLN (OR, 7.35), maximum diameter and area of SLN metastases (OR 2.26, 1.85 respectively) and TMA (OR, 2.12). Maximum metastatic diameter/SLN diameter (OR 3.71, p = 0.04) and the area of metastases/SLN area (OR 3.4, p = 0.04) were predictive. For every 1 mm increase in diameter of SLN metastases, the odds of NSLN extranodal disease increased by 8.5% (p = 0.02). TMA >0.40 cm2 was an independent predictor for NSLN metastases and extranodal disease.Her-2/neu over-expression and parameters assessing metastatic burden in the SLN, particularly TMA, predicted the presence of NSLN involvement and extranodal disease in patients with breast carcinoma and SLN metastases.

O-69 Axillary recurrence in breast cancer patients after a negative sentinel lymph node biopsy

Recent literature has suggested that completion axillary lymph node dissection (ALND) in breast carcinoma patients with positive SLN may not be necessary. However, a method for determining the risk of non-SLN or extranodal disease remains to be established.To determine if pathological variables from primary tumors and sentinel lymph node (SLN) metastases could predict the probability of non-sentinel lymph node (NSLN) metastases and extranodal disease in patients with breast carcinoma and SLN metastases.84 women with T1-3 breast cancer and clinically-negative axillae underwent completion ALND. Maximum diameter and width of SLN metastases were measured to calculate metastatic area. When multiple SLNs contained metastases, areas were summed to calculate the Total Metastatic Area (TMA). Multiple linear regression models were used to identify predictive factors.Her-2/neu over-expression increased the odds of NSLN metastases (OR 4.3, p = 0.01) and extranodal disease (OR 7.9, p < 0.001). Independent SLN predictors were ≥1 positive SLN (OR, 7.35), maximum diameter and area of SLN metastases (OR 2.26, 1.85 respectively) and TMA (OR, 2.12). Maximum metastatic diameter/SLN diameter (OR 3.71, p = 0.04) and the area of metastases/SLN area (OR 3.4, p = 0.04) were predictive. For every 1 mm increase in diameter of SLN metastases, the odds of NSLN extranodal disease increased by 8.5% (p = 0.02). TMA >0.40 cm2 was an independent predictor for NSLN metastases and extranodal disease.Her-2/neu over-expression and parameters assessing metastatic burden in the SLN, particularly TMA, predicted the presence of NSLN involvement and extranodal disease in patients with breast carcinoma and SLN metastases.

Validation of clinical prediction rules for a low probability of nonsentinel and extensive lymph node involvement in breast cancer patients

Background Two recently developed clinical prediction rules aim to anticipate the lack of nonsentinel lymph node metastases and the involvement of less than 4 lymph nodes in breast cancer patients with positive sentinel lymph nodes (SLNs). Methods The University of Louisville Breast SLN Study clinical prediction rules were validated on an independent set of SLN-positive patients with tumors ≤15 mm. Results The data on 475 and 473 patients, respectively, were used for the validation. The areas under the receiver operating characteristic curves were similar to the originals for both predictive tools (.70 and .76). The lowest score of 1 identified 5 of 7 patients with disease limited to the SLNs and 161 of 165 as having less than 4 involved lymph nodes. Conclusions A subset of patients with SLN-only involvement and less than 4 metastatic lymph nodes can probably be identified by means of the Louisville clinical prediction rules, but prediction of the lack of non-SLN metastasis seems less reliable.

Prediction of Nonsentinel Lymph Node Involvement in Patients with a Positive Sentinel Lymph Node in Malignant Melanoma

Completion lymph node dissection (CLND) is routinely performed after metastatic melanoma is detected at sentinel lymph node (SLN) biopsy. Nonsentinel lymph node (NSLN) involvement is found in less than one-third of the cases. Possible predictors of NSLN involvement are examined. A retrospective review of 70 patients with a positive SLN biopsy for melanoma and drainage to one lymphatic basin was performed. The size of metastatic deposits was defined as macrometastases (>2 mm), micrometastases (< or =2 mm), a cluster of cells (10-30 grouped cells) in the subcapsular space or interfollicular zone, or isolated melanoma cells (1-20 or more individual cells) in subcapsular sinuses. Tumor stage, ulceration, SLN tumor burden, mitoses, number of positive SLNs, and total number of lymph nodes removed were examined as predictors of NSLN involvement after CLND. Two additional models based on SLN tumor burden and the number of nodes biopsied were designed. Nineteen patients (24.3%) were found to have NSLN metastases after CLND. Tumor stage, ulceration, SLN tumor burden, mitoses, number of positive SLN, and number of lymph nodes removed were not statistically significant. Residual disease at CLND stratified by SLN tumor burden was: isolated melanoma cells, 0; cluster of cells, 8 (38.1%); < or =2 mm, 5 (20.8%); and >2 mm, 6 (27.3%). A comparison of the means for the models was not predictive of NSLN involvement. None of the risk factors or models examined could predict nonsentinel lymph node involvement with melanoma. The SLN sample and minimal SLN metastatic disease when defined as isolated clusters of cells warrant further study as a potential indicator against CLND after positive SLN.

Predicting nonsentinel node status in breast cancer preoperatively by molecular characterization of core needle biopsies with multiparametric protein analyses

21068 Background: Axillary lymph node dissection (ALND) still remains the standard treatment for breast cancer patients with sentinel lymph node (SLN) metastases. However, since only 40% to 60% of patients show additional lymph involvement, complete ALND offers no additional benefit for almost 50% of patients but carries a significant risk of axillary morbidity. In an attempt to achieve a more precise prediction for the individual patient a multivariable logistic-regression analysis of a large data set of eight histopathological variables has been published by the Memorial Sloan-Kettering Cancer Center (MSKCC) that predicts the likelihood of metastases in Non-SLNs with a ROC of 0,71. To achieve a more precise estimation of additional lymph node involvement preoperatively we analysed molecular markers of potential predictive value. Methods: Beside histopathological variables, fifty proteins of potential prognostic and predictive value were preoperatively quantified in lysates from 120 core needle biopsies with multiplex sandwich immunoassays. Biopsies, taken at the Breast Cancer Center of the University of Ulm Medical School, were classified as invasive mammary carcinomas. Variables which could be used to improve the accuracy of prediction of non-sentinel lymph node (NSLN) involvement were defined. Results: We demonstrated appropriate sensitivity, reproducibility, and robustness for this protein microarray technology to characterize proteins in core needle biopsies and generate reliable data sets. In an initial univariate data analysis (Mann Whitney test) six of the analysed parameters (TIMP2, p= 0,027; FasR, p = 0,0025; MIF, p= 0,030; FGF-2, p= 0,0020, PDGF AA/BB p= 0,027; RANTES p= 0,024) correlated with NSLN involvement beside known histopathological variables (i.e. tumor size, p= 0,006; grading, p = 0,0009). Conclusions: We defined molecular markers of diagnostic value to predict NSLN involvement. This might be an acceptable substitute to the known histopathological variables by the MSKCC nomogram to predict potential NSLN involvement preoperatively. No significant financial relationships to disclose.

Expression of the E-cadherin–catenins complex in sentinel node is related to tumor morphology but not to spread to nonsentinel nodes

The sentinel node (SN) technique has gained a key role in breast cancer surgery, allowing for an accurate staging of the axillary status with a minimally invasive resection. In this study, we explored the implication of three proteins (E-cadherin, a- and b-catenins) that form the cadherin–catenin complex, a receptorial structure strictly involved in tumoral vascular invasion and embolization in this biologic event. We studied the immunohistochemical expression of the complex in patients with metastatic SN, matching the group with involved nonsentinel lymph nodes (NSNs) with that having free axillary NSNs. The simultaneous staining of the SN metastases for the three proteins has been considered an indicator of preserved function. Our data confirmed the lack of cadherin–catenin complex in tumors with lobular morphology even in SN metastasis, but statistical evaluation could not prove a significant relation between complex integrity and NSN involvement. Moreover, considering traditional histopathologic parameters, only vascular peritumoral embolization was related to an increased risk of metastatic spread to axillary NSNs.

Size of Sentinel Node Tumor Deposits and Extent of Axillary Lymph Node Involvement: Which Breast Cancer Patients May Benefit From Less Aggressive Axillary Dissections?

In most breast cancer series, nearly 30% to 40% of all patients are sentinel node positive; however, in a large proportion of these, the disease is limited to three or fewer positive nodes. On the basis of these observations, the object of this study is to identify a subset of patients who might benefit from a less aggressive axillary dissection, without compromising staging or local disease control. We reviewed known clinicopathologic variables associated with a higher risk for axillary metastasis in 467 patients who underwent sentinel node mapping at our institution. We then compared the incidence of these variables in patients with N1a versus N2-3 stage disease. Although the presence of lymphvascular invasion in the primary tumor and extracapsular extension of tumor in the sentinel node were statistically significantly different between N1a and N2-3 patients (P < .025 and P < .01, respectively), the variable that most reliably separated N1a from N2-3 patients was the size of the tumor deposits in the sentinel node (P < .001). All patients with sentinel node tumor deposits <or=5 mm had three or fewer positive nodes; 95% were sentinel node-positive only, and 91% had single-node involvement. Patients at low risk for extensive axillary nonsentinel node involvement may benefit from a more conservative surgical approach to their axillae, perhaps limited to sentinel node biopsy only or to axillary procedures restricted to the group of axillary nodes in close proximity to those designated as sentinel nodes.

Sentinel lymph node biopsy and non-sentinel node involvement in special type breast carcinomas with a good prognosis

This study aimed at identifying factors related to sentinel lymph node (SLN) involvement in patients with tubular, cribriform, mucinous or papillary breast carcinoma and those related to non-SLN metastases if an SLN was positive. Multivariate analyses involved logistic and stepwise regressions. The SLNs harboured metastases in 85 of 572 cases, 78 of whom underwent axillary dissection; 19 presented non-SLN positive disease. Lack of lymphovascular invasion, a tumour size ⩽10 mm and a single SLN removed were the factors predicting an SLN metastasis rate <10%, and patients with these features could be candidates for no surgical axillary staging. A positive SLN proportion of ⩽50% and no lymphovascular invasion were associated with a <10% rate of non-SLN invasion; patients with a positive SLN and these features could be candidates for the omission of completion axillary dissection. The opposite presentation of these factors would mandate SLN biopsy and axillary dissection, respectively.

28 ORAL Validation and limitations of use of a Breast Cancer Nomogram predicting the likelihood of non-sentinel node involvement after positive sentinel node biopsy

28 ORAL Validation and limitations of use of a Breast Cancer Nomogram predicting the likelihood of non-sentinel node involvement after positive sentinel node biopsy

Evaluation of criteria in sentinel nodes of melanoma patients that predict involvement of non-sentinel nodes

Department of Dermatology, Hannover Medical School, Hannover, Germany Abstracts of the Perspectives in Melanoma X and The Third Annual International Melanoma Research Congress, 14–16 September 2006, Noordwijk, The Netherlands

Primary Breast Cancer Features Can Predict Additional Lymph Node Involvement in Patients with Sentinel Node Micrometastases

The aim of this retrospective study was to identify biological features of primary breast cancer from which to predict the presence of further axillary involvement in patients bearing micrometastases in the sentinel lymph node (SLN). From a starting group of 690 patients, we isolated patients with micrometastases in the SLN. Those patients were classified according to the presence/absence of further metastases in nonsentinel lymph nodes (NSLNs). We examined primary tumor features to identify any relevant difference. Analysis of primary tumors evaluated histology, tumor size, lymphovascular invasion, mitotic index (Mib-1), estrogen and progesterone receptor status (ER/PR status), C-erb B-2 (HER-2/neu) expression and amplification, and p53 expression. Chi square analysis for statistical significance was applied. Of the original 690 patients, 296 showed some kind of metastases in the SLN; 238 patients had gross metastases in the SLN. After axillary lymph node dissection (ALND), 102 patients (43%) had NSLNs with metastases, and 136 (57%) had negative axillary non-sentinel nodes. Another 58 patients harbored solitary micrometastases in the SLN. After ALND, 8 (14%) patients had further NSLN involvement, and 50 (86%) had negative axillary nodes. Analysis of the primary breast lesion in patients with micrometastatic SLN and metastatic NSLNs revealed the presence of lymphovascular invasion, Mib-1 index > 10%, and tumor size > 2 cm. Patients without lymphovascular invasion, Mib-1 < 10% and T size < 2 cm could avoid further ALND.

Multicentric Breast Cancer: A New Indication for Sentinel Node Biopsy—A Multi-Institutional Validation Study

Multicentric breast cancer has been considered to be a contraindication for sentinel node (SN) biopsy (SNB). In this prospective multi-institutional trial, SNB-feasibility and accuracy was evaluated in 142 patients with multicentric cancer from the Austrian Sentinel Node Study Group (ASNSG) and compared with data from 3,216 patients with unicentric cancer. Between 1996 and 2004, 3,730 patients underwent SNB at 15 ASNSG-affiliated hospitals. Patient data were entered in a multicenter database. One hundred forty-two patients presented with multicentric invasive breast cancer and underwent SNB. Intraoperatively, a mean number of 1.67 SNs were excised (identification-rate, 91.5%). The incidence of SN metastases was 60.8% (79 of 130). This was confirmed by axillary lymph node dissection (ALND) in 125 patients. Of patients with positive SNs, 60.8% (48 of 79) showed involvement of nonsentinel nodes (NSNs), as did three patients with negative SNs (false-negative rate, 4.0). Sensitivity, negative predictive value, and overall accuracy were 96.0%, 93.3%, and 97.3%, respectively. Ninety-one percent of the patients underwent mastectomy, and 9% were treated with breast conserving surgery. None of the patients have shown axillary recurrence so far (mean follow-up, 28.8 months). Compared with 3,216 patients with unicentric cancer, there was a significantly higher rate of SN metastases as well as in NSNs, whereas there was no difference in detection and false-negative rates. Multicentric breast cancer is a new indication for SNB without routine ALND in controlled trials. Given adequate quality control and an interdisciplinary teamwork of surgical, nuclear medicine, and pathology units, SNB is both feasible and accurate in this disease entity.

Microscopic tumor burden in sentinel lymph nodes (SLNs) best predicts nonsentinel lymph node (NSLN) involvement in patients with melanoma

8004 Background: Therapeutic lymph node dissection (TLND) has long been the standard of care for melanoma metastatic to the regional LNs and is commonly used for patients (pts) with microscopic disease identified by SLN biopsy. We and others have demonstrated that additional histologically positive (pos) nodes are identified in 8%-33% of completion LN dissection (CLND) specimens. The purpose of this study was to determine predictors of additional LN involvement in pts. with pos SLNs. Methods: Pts with clinically node-negative melanoma who underwent SLN biopsy at our institution between 1991 and 2003 and had pos SLNs were identified. Patient and primary tumor factors, number of pos SLNs, and extent of microscopic disease within the SLN were evaluated as potential predictors of pos NSLNs. Results: Overall, 359 (16%) of 2,203 pts had a pos SLN. Pos NSLNs were identified in 48 (14%) of the 343 pts with pos SLNs who underwent CLND. On univariate analysis, several markers of SLN microscopic tumor burden, tumor thickness &gt;2mm, age &gt;50 years, and Clark level &gt;III were predictive of additional pos nodes in the TLND specimen. Primary tumor ulceration was not predictive. On multivariable logistic regression analysis, markers of microscopic tumor burden (large SLN met focus, large SLN tumor square area, presence of SLN extracapsular extension, and synchronous subcapsular and intramedullary SLN deposits) were the most powerful independent predictors of pos NSLNs; tumor thickness &gt;2mm also predicted additional disease. A model was established using tumor thickness and SLN tumor burden to identify pts with low, intermediate, or high risk for NSLN involvement (Table). Conclusions: In melanoma pts with ≥1 pos SLN, SLN tumor burden and primary tumor thickness may be useful in identifying a group at low risk for pos NSLNs. This information is valuable in designing clinical trials in which pts may be randomly assigned to TLND versus observation alone after removal of a pos SLN. [Table: see text] No significant financial relationships to disclose.

Microanatomic characteristics of melanoma metastases in the sentinel node(s) (SN) predict survival and question the need for complete lymph node dissection (cLND)

8005 Introduction: The SN procedure has become a widely used staging procedure for stage I/II melanoma patients. We previously reported on the high SN+ identification rate with the use of the EORTC MG group pathology protocol (van Akkooi et al., Eur J Cancer: 2006, in press). Because only approximately 20% of SN positive patients have additional non-SN lymph node involvement in the cLND specimen, we have proposed to identify a SN+ patient group, which can be spared a cLND. Therefore further microanatomic analyses of the metastatic SNs has been performed to identify patient/tumor or SN factors which predict DFS and OS as well as additional non-SN nodal positivity. Methods: 262 stage I/II pts were included into the SN database between 10/97 and 5/04. 77 were SN+ (29%). 74 pts had SN material for re-evaluation in this study. Microanatomic analysis categorized a.o. the location of tumor cells; subcapsular, parenchymal (or combined), multifocal or extensive; the Starz classification and the measurement of the amount of tumor load (in mm). Tumor load was quantified by size: &lt;0.1mm (single melanoma cells), 0.1 - 1.0mm, &gt;1.0mm. DFS, OS and additional non-SN positivity was calculated for all the microanatomic analyses. Results: Mean/Median Breslow was 3.5/3.0 (0.8 - 12.0) mm, mean/median FU was 35/30 (6 - 81) months. There were no signif. differences in OS or additional non-SN positivity for the different loc. of involvement. The estimated 5yr OS rates were 100%, 63% and 35% for the different groups of tumor load, &lt;0.1mm, 0.1 - 1.0mm, &gt;1.0mm respectively (Figure)(P=0.03). There were no patients with add. non-SN positivity in the group with involvement &lt;0.1mm. On multivariate analysis the tumor load was the most important prognostic factor for DFS (P=0.005) and OS (P=0.03). DFS, OS and add. non-SN positivity of all other analyses will be presented at the meeting. Conclusions: It is questionable whether SN with onlysingle cells involvement (&lt;0.1mm) should be considered positive. Dist. mets are exceedingly rare (1/16 = 6.3%) and was identical to the SN- patient pop., and non-sentinel node involvement was not observed at all. These patients should probably not undergo add. LND and may not be considered stage III patients. No significant financial relationships to disclose.

Predicting non-sentinel lymph node involvement in breast cancer patients

531 Background: Current convention is to perform a completion axillary lymph node dissection (ALND) for invasive breast cancer (BC) patients (pts) with positive sentinel lymph node(s) (+SLN), even though &lt;50% will have non-sentinel lymph node metastasis (+NSLN). Our goal was to develop and compare predictive models of NSLN status among +SLN pts to identify for whom completion ALND may be omitted. Methods: We constructed 3 models by recursive partitioning with receiver-operating-characteristic curves (RP-ROC), boosted classification and regression trees (CART), and a multivariate logistic regression (MLR) informed by CART. Models were developed using a multi-institutional database of 1,040 BC pts who underwent SLN biopsy and completion ALND at academic or community hospitals as part of a prospective, consented study. Accuracies were compared to the Memorial Sloan-Kettering Breast Cancer Nomogram (Nomogram). Results: 976 BC pts had evaluable SLNs and 285 had +SLNs; 101(35.4%) +SLN pts had +NSLNs. Tumor size, lymphovascular invasion (LVI), and SLN metastasis size were the best predictors of risk (multivariate P-values&lt;0.001) of +NSLN. 39 of 156 +SLN pts (25%) with T1 tumors had +NSLNs vs. 62 of 129 (48%) with T2/T3 tumors. 73.7% with LVI vs 19.5% without LVI had +NSLNs. 4.7% of pts with isolated tumor cells, 42% with micrometastasis and 71% with macrometastasis in SLN had +NSLNs. MLR informed by CART identified 2 highly predictive variables, the LVI × size of SLN metastasis [OR 4.73(3.11–7.20, 95%CI), P&lt;0.001] and tumor size × size of SLN metastasis [OR 1.18(1.10–1.26, 95%CI), P&lt;0.001]. While RP-ROC and boosted CART stratified pts into low-risk (4.3%-9.9%), moderate-risk (33.3%-42.9%), and high-risk (62.2%-93%) groups, MLR predicted NSLN status with accuracy superior to RP-ROC, boosted CART, and the Nomogram: 83.3%, 76.7%, 67.7%, and 76.7%, respectively, after 10-fold cross validation. The Nomogram’s sensitivity was significantly inferior to those of RP-ROC, boosted CART, and MLR: 53.8%, 78.8%, 78.2%, and 78.0%, respectively. Conclusions: In pts with +SLNs, interactions between clinicopathologic characteristics are highly informative in predicting risk of +NSLN. However, neither our methods nor the Nomogram achieved sufficient accuracy to recommend a change in current clinical practice. No significant financial relationships to disclose.