Development of a Multiparametric Magnetic Resonance Imaging-Based Nomogram for Clinically Insignificant Prostate Cancer
10.22465/kjuo.2020.18.3.222
- Author:
Jae Woo SUNG
1
;
Dongho SHIN
;
Yong Hyun PARK
;
Hyuk Jin CHO
;
U-Syn HA
;
Sung-Hoo HONG
;
Ji Youl LEE
;
Sae Woong KIM
Author Information
1. Department of Urology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
- Publication Type:Original Article
- From:Korean Journal of Urological Oncology
2020;18(3):222-229
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
Purpose:Various predictive tools have been developed to predict insignificant prostate cancer (PCa) for active surveillance, however, these models cannot reflect all the refinements of current medicine. Thus, we aimed to develop a novel model to predict clinically insignificant PCa incorporating these factors.
Materials and Methods:We developed a novel nomogram to predict the probability of insignificant PCa (total tumor volume less than 2.5 cm3, index tumor volume less than 1.3 cm3, organ confined disease and no Gleason pattern 4 or 5) using preoperative data of 790 Korean patients who underwent radical prostatectomy. To evaluate the predictive accuracy, the area under the receiver operating characteristic curve (AUC) was calculated. Next, the predicted probability versus the actual probability was compared. This examination was performed by calibration plotting using 1,000 bootstrap resamples.
Results:Of the 790 patients, 668 (84.6%) had clinically significant PCa, and 122 (15.4%) had insignificant PCa. We developed a novel predictive model for clinically insignificant PCa using clinical stage less than T2a, biopsy Gleason sum less than 7, ratio of positive biopsy cores less than 10%, neutrophil-to-lymphocyte ratio, and multiparametric magnetic resonance imaging (mpMRI) visibility, which discriminated patients with clinically insignificant PCa from those with significant PCa with an AUC of 0.9135 (95% confidence interval, 0.9127–0.9143). The calibration plot showed a well-calibrated prediction that had little over- or underestimation.
Conclusions:We proposed a novel predictive model for insignificant PCa to more accurately select patients for active surveillance using the results from mpMRI and prebiopsy laboratory marker.
