A preoperative marker panel for the prediction of residual tumor and the decision making for repeat transurethral resection.

OBJECTIVE

To assess the ability of a combined preoperative marker panel to identify patients with residual non-muscle-invasive bladder cancer who might benefit from repeat transurethral resection (reTUR).

METHODS

Ki67, p53, vascular endothelial growth factor-C, E-cadherin, and survivin expressions were evaluated by immunohistochemical staining of surgical specimens from 72 patients who underwent reTUR. Related clinical and molecular markers were analyzed by univariate analyses to develop a marker panel. The predictive value of the marker panel was calculated by receiver operating characteristic curves.

RESULTS

Univariate analyses identified tumor size, number of tumors, p53 expression, E-cadherin expression, and the number of altered markers as risk factors for residual tumor (P = 0.03, 0.05, 0.06, 0.024, and 0.005, respectively). After adjusting for the effects of tumor stage and grade, multivariate analyses identified the number of altered markers as a risk factor for residual tumor (P = 0.004). The addition of tumor size, E-cadherin, and the number of altered markers to the base model (based on tumor stage and tumor grade) increased its discrimination for predicting residual tumor (5%, 6%, and 10%, respectively).

CONCLUSION

Some clinical and molecular markers could improve the accuracy of residual tumor prediction at reTUR. Such a marker panel may help to identify patients with non-muscle-invasive bladder cancer who have residual tumor after first TUR and who may therefore benefit from reTUR.