Predicting sentinel lymph node metastasis in a Chinese breast cancer population: assessment of an existing nomogram and a new predictive nomogram

摘要

We assessed the MSKCC nomogram performance in predicting SLN metastases in a Chinese breast cancer population. A new model (the SCH nomogram) was developed with clinically relevant variables and possible advantages. Data were collected from 1,545 patients who had a successful SLN biopsy between March 2005 and November 2011. We validated the MSKCC nomogram in the modeling and validation group. Clinical and pathologic features of SLN biopsy in modeling group of 1,000 patients were assessed with multivariable logistic regression to predict the presence of SLN metastasis in breast cancer. The SCH nomogram was created from the logistic regression model and subsequently applied to 545 consecutive SLN biopsies. By multivariate analysis, age, tumor size, tumor location, tumor type, and lymphovascular invasion were identified as independent predictors of SLN metastasis. The SCH nomogram was then developed using the five variables. The new model was accurate and discriminating (with an AUC of 0.7649 in the modeling group) compared to the MSKCC nomogram (with an AUC of 0.7105 in the modeling group). The area under the ROC curve for the SCH nomogram in the validation population is 0.7587. The actual probability trends for the various deciles were comparable to the predicted probabilities. The false-negative rates of the SCH nomogram were 1.67, 3.54, and 8.20 % for the predicted probability cut-off points of 5, 10, and 15 %, respectively. Compared with the MSKCC nomogram, the SCH nomogram has a better AUC with fewer variables and has lower false-negative rates for the low-probability subgroups. The SCH nomogram could serve as a more acceptable clinical tool in preoperative discussions with patients, especially very-low-risk patients. When applied to these patients, the SCH nomogram could be used to safely avoid a SLN procedure. The nomogram should be validated in various patient populations to demonstrate its reproducibility.