Microscopic tumor burden in sentinel lymph nodes predicts synchronous nonsentinel lymph node involvement in patients with melanoma

Purpose We and others have demonstrated that additional positive lymph nodes (LNs) are identified in only 8% to 33% of patients with melanoma who have positive sentinel LNs (SLNs) and undergo complete therapeutic LN dissection (cTLND). We sought to determine predictors of additional regional LN involvement in patients with positive SLNs. Patients and Methods Patients with clinically node-negative melanoma who underwent SLN biopsy (1991 to 2003) and had positive SLNs were identified. Clinicopathologic factors, including extent of microscopic disease within SLNs, were evaluated as potential predictors of positive non-SLNs. Results Overall, 359 (16.3%) of the 2,203 patients identified had a positive SLN. Positive non-SLNs were identified in 48 (14.0%) of the 343 patients with positive SLNs who underwent cTLND. On univariate analysis, several measures of SLN microscopic tumor burden, one versus three or more SLNs harvested, tumor thickness more than 2 mm, age older than 50 years, and Clark level higher than III were predictive of positive non-SLNs; primary tumor ulceration and number of positive SLNs had no apparent impact. On multivariable logistic regression analysis, measures of SLN microscopic tumor burden were the most significant independent predictors of positive non-SLNs; tumor thickness more than 2 mm and number of SLNs harvested also predicted additional disease. A model was developed that stratified patients according to their risk for non-SLN involvement. Conclusion In melanoma patients with positive SLNs, SLN tumor burden, primary tumor thickness, and number of SLNs harvested may be useful in identifying a group at low risk for positive non-SLNs and be spared the potential morbidity of a cTLND.