To develop and validate a nomogram for preoperative prediction of lymph node metastasis in patients with progressive muscle-invasive bladder cancer.
We retrospectively recruited patients, divided them into training and validation cohorts, and gathered patient demographics, pathology data of transurethral bladder tumor resection specimens, imaging findings, and laboratory information. We performed logistic regression analyses, both single-variable and multi-variable, to investigate independent preoperative risk variables and develop a nomogram. Both internal and external validations were conducted to evaluate the predictive performance of this nomogram.
The training cohort consisted of 144 patients with advanced muscle-invasive bladder cancer, while the validation cohort included 62 individuals. The independent preoperative risk factors identified were tumor pathology grade, platelet count, tumor size on imaging, and lymph node size, which were utilized to develop the nomogram. The model demonstrated high predictive accuracy, as evidenced by the area under the receiver operating characteristic curve values of 0.898 and 0.843 for the primary and external validation cohorts, respectively. Calibration curves and decision curve analysis showed a good performance of the nomogram in both cohorts, indicating its high clinical applicability.
A nomogram for preoperative prediction of lymph node metastasis in patients with advanced muscle-invasive bladder cancer was successfully developed; its accuracy, reliability, and clinical value were demonstrated. This new tool would facilitate better clinical decisions regarding whether to perform complete lymph node dissection in cases of radical cystectomy.