A nomogram for predicting upgrading in patients with low‐ and intermediate‐grade prostate cancer in the era of extended prostate sampling

Abstract

Diagnostic (exploratory cohort). 2b. To develop a nomogram to predict the probability that the pathological Gleason sum (GS) will be higher than that indicated by the biopsy, suggesting a higher risk for the patient presumed to be at low risk, as a substantial proportion of patients with low and intermediate grade on biopsy are upgraded on interpretation of the radical prostatectomy (RP) specimens, but a similar clarification of accurate Gleason scoring is not available in patients with no surgical histology. The study included 1017 patients who had RP after biopsy showing GS 6 and 7 (3 + 4) from 2000 to 2007. Nomogram predictor variables included age, race, digital rectal examination, prostate-specific antigen (PSA) level, number of cores taken, number of positive cores, maximum percentage cancer in any core, number of previous biopsies, prostate volume, clinical stage, high-grade prostatic intraepithelial neoplasia, atypical small acinar proliferation, inflammation and perineural invasion. We calculated the nomogram-predicted probability in each patient. The area under the receiver operating characteristic curve was calculated as a measure of discrimination, and the calibration was assessed graphically. The mean age of the patients was 60 years, the mean PSA level 6.62 ng/mL; 336 patients were upgraded (33%), 623 remained the same (61.3%) and 58 were downgraded (5.7%). A nomogram for predicting the possibility of upgrading was constructed that had a concordance index of 0.68. The nomogram was well calibrated. Our nomogram for predicting upgrading provides important additional information for deciding on treatment to both the urologist and the patient with low- and intermediate-grade prostate cancer. It might prove useful when the possibility of a more aggressive Gleason variant can change the management, and is especially meaningful when management options other than surgery are selected based on the inability to recognize the true pathological actual GS.