Role of the combined use of clinical stage, serum PSA and Gleason score for the improvement of preoperative staging of prostate cancer

Objective. To improve prediction accuracy of non-organ confined prostate cancer before radical prostatectomy.
Material and methods. Diagnostic efficacy of clinical staging, pretreatment PSA, Gleason score in prostate biopsy and combination of aforementioned factors in prediction or capsular penetration or seminal vesicle invasion were assessed in 272 pts. undergone radical prostatectomy in RIOMR between 1996 and 2007. Sensitivity, specificity and diagnostic accuracy of these factors were compared. Receiver-OperatingCharacteristic (ROC) curves were constructed and areas under curves were calculated.
Results. Sensitivity, specificity and diagnostic accuracy for clinical staging were 39,3%, 88,1% and 68,0%; for Gleason score 7—10 — 41,1%, 80,6%—64,3%; for pretreatment PSA≥20 ng/ml — 45,5%, 80,6%—66,2% accordingly; for only 1 aforementioned factor — 78,6%, 58,1% and 66,5%; for combination of 2 factors — 39,3%, 92,5% and 70,6%; for 3 — 5,4%, 98,2% and 42,7% accordingly. Areas under ROC curves for clinical stage, Gleason score, pretreatment PSA and combination of aforementioned factors were 0,66 (95%CI 0,60 — 0,73); 0,64 (95%CI 0,57 — 0,71); 0,69 (95%CI 0,63 — 0,76) and 0,74 (95%CI 0,68 — 0,80).
Conclusion. Diagnostic accuracy for combination of clinical stage, Gleason score and pretreatment PSA was highest comparing to use of these single variables. Presence of 0 to 1 unfavorable factors may be used as an indication for radical prostatectomy according to our data.