Development of a new nomogram to predict insignificant prostate cancer in patients undergoing radical prostatectomy.

OBJECTIVE

The aim of this study was to develop a nomogram to predict the probability of insignificant prostate cancer.

MATERIALS AND

METHODS

A retrospective analysis was conducted of patients who underwent radical prostatectomy at a Korean hospital between January 2005 and December 2014. Patients with pathologically insignificant prostate cancer were defined as having organ-confined disease with tumor volume less than 0.5 cm³ without Gleason scores of 4 or 5. Multivariable logistic regression analysis with a stepwise selection was used to model the relationship between preoperative characteristics and insignificant prostate cancer, and a nomogram to predict the probability of insignificant prostate cancer was created. Receiver operating characteristics (ROC) analysis was performed to assess the predictive value of the model.

RESULTS

The final study population consisted of 1343 patients. Among these patients, insignificant prostate cancer was confirmed in 188 men (14.0%) at the time of prostatectomy. Six independent predictors of insignificant prostate cancer were identified: number of positive cores (p < 0.001), maximal single core tumor involvement (p < 0.001), biopsy Gleason score (p < 0.001), prostate volume (p = 0.024), patient age (p < 0.001) and prostate-specific antigen density (p < 0.001) in the multivariable model. A nomogram to predict insignificant prostate cancer was developed using these six preoperative characteristics. The area under the ROC curve for nomogram predictions was 0.87.

CONCLUSION

The nomogram developed in this paper identifies the probability of insignificant prostate cancer and gives providers more information to guide their clinical decisions.