A National Cancer Database-based nomogram to predict lymph node metastasis in penile cancer.

Abstract

To evaluate the predictive nature of several clinicopathological variables by developing a nomogram predictive for lymph node-positive disease using the National Cancer Database cohort of patients with squamous cell carcinoma of the penis. Stepwise logistic regression was used to find the best-fit model; remaining clinical variables were used to create a nomogram to predict the probability of lymph node-positive disease. On multivariate analysis, high pathological grade (3-4 vs 1: odds ratio [OR] 3.27, 95% confidence interval [CI] 1.70-6.29; 2 vs 1: OR 2.58, 95% CI 1.39-4.79 [P = 0.002]), lymphovascular invasion (OR 2.49, 95% CI 1.61-3.84 [P < 0.001]), and positive clinical lymph node status (N1 vs N0: OR 20.0, 95% CI 11.4-35.7; N2 vs N0: OR 27.8, 95% CI 14.1-55.6; N3 vs N0: OR 49.2, 95% CI 14.8-162.8 [P < 0.001]) were predictors of lymph node metastasis in penile cancer. The bootstrap-corrected concordance index of this nomogram was 0.880. Using tumour grade, tumour lymphovascular invasion and clinical lymph node status, we developed a nomogram highly predictive of pathologial lymph node metastasis that, after further external validation, could be helpful in the surgical decision-making process.