RESUMO
OBJECTIVE: To develop a predictive model for additional inguinal lymph node metastases (LNM) at inguinal lymph node dissection (ILND) after positive dynamic sentinel node biopsy (DSNB) using DSNB characteristics to identify a patient group in which ILND might be omitted. PATIENTS AND METHODS: We conducted a retrospective study of 407 inguinal basins with a positive DSNB in penile cancer patients who underwent subsequent ILND from seven European centres. From the histopathology reports, the number of positive and negative lymph nodes, presence of extranodal extension and size of the metastasis were recorded. Using bootstrapped logistic regression, variables were selected for the clinical prediction model based on the optimization of Akaike's information criterion. The area under the curve (AUC) of the receiver-operating characteristic curve was calculated for the resulting model. Decision curve analysis (DCA) was used to evaluate the clinical utility of the model. RESULTS: Of the positive DSNBs, 64 (16%) harboured additional LNM at ILND. Number of positive nodes at positive DSNB (odds ratio [OR] 2.19, 95% confidence interval (CI) 1.17-4.00; P = 0.01) and largest metastasis size in mm (OR 1.06, 95% CI 1.03-1.10; P = 0.001) were selected for the clinical prediction model. The AUC was 0.67 (95% CI 0.60-0.74). The DCA showed no clinical benefit of using the clinical prediction model. CONCLUSION: A small but clinically important group of basins harbour additional LNM at completion ILND after positive DSNB. While DSNB characteristics were associated with additional LNM, they did not improve the selection of basins in which ILND could be omitted. Thus, completion ILND remains necessary in all basins with a positive DSNB.