International Journal of Hematology and Oncology 2022, Vol 32, Num 4 Page(s): 075-081
The Accuracy of Two Different Calculators which Calculate the Probability of Prostate Cancer Before Prostate Biopsy


Dışkapı Yıldırım Beyazıt Eğitim ve Araştırma Hastanesi, 1. 2. ve 4. Üroloji Klinikleri, Ankara, TÜRKİYE

Keywords: Prostate cancer, Risk, Risk factor, Prostate biopsy
Clinical and pathological data of 1104 patients who had undergone prostate biopsy for the first time using transrectal ultrasound were evaluated to determine the efficacy of two different prostate cancer (PC) risk calculators before prostate biopsy. Patients‘ data, such as age, race, abnormal digital rectal examination, prostate specific antigen (PSA) level, family history for PC, prior negative prostate biopsy and finasteride usage were submitted to American-type PC risk calculator (type-1). Other data of the patients such as having hipoechoic nodules on ultrasound, abnormal digital rectal examination, prostate volume and PSA levels were submitted to European-type prostate risk calculator (type- 2). The capacity to predict PC of both calculators were compared in 10%, 25%, 50%, 75% and over each of these probability levels. Data collected from 829 patients were suitable for both calculators and PC was detected in 64 patients. For both calculators, capacity to predict PC was similar in ≥10% and ≥75% probabilities (p=1.000 and p= 0.607, respectively), however, at ≥25% and ≥50% probability levels, type-1 calculator was favored (p= 0.031 and p<0.001, respectively). Type-1 calculator had 7.8% and 14% positive predictive rates and 100% and 98.8% negative predictive rates at ≥25% and ≥50% probabilities; for type-2 calculators, positive predictive rates were 11.7% and19.9%, and negative predictive rates were 98.2% and 96.6%, respectively. In conclusion, we determined that the utilization of American-type calculator was more advantageous on predicting patients with PC before biopsy, but neither of the calculators were suitable because unnecessary biopsies could be performed due to low positive predictive rates.