An analysis of three different prostate cancer risk calculators applied prior to prostate biopsy: A Turkish cohort validation study.
Mehmet YıldızhanMelih BalcıUnsal EroğluErem AsilSeref CoserAli Yasin ÖzercanBurak KöseoğluOzer GüzelAhmet AsfuroğluAltuğ TuncelPublished in: Andrologia (2021)
The study aimed to investigate the best-performing of three risk calculators (RCs) for the Turkish population in predicting cancer-free status and high-risk prostate cancer (PCa) in patients undergoing transrectal ultrasound-guided prostate biopsy. The electronic medical records of 527 patients who underwent prostate biopsy for the first time due to PSA of 0.3-50 ng/dl and/or cancer suspicion at digital rectal examination (DRE) between January 2017 and December 2020 were retrieved retrospectively. The predictive power of the RCs in the biopsy and the surgical cohort was calculated by two urologists using European Randomised Study of Screening for Prostate Cancer (ERSPC) RC, the North American Prostate Cancer Prevention Trial-RC (PCPT-RC), and the Prostate Biopsy Collaborative Group (PBCG)-RC. All three RCs were successful in predicting PCa and high-risk disease at ROC analysis (p < 0.0001). Of these three nomograms, PBCG-RC outperformed PCPT-RC 2.0 and ERSPC-RH in predicting benign pathology outcomes at biopsy. A better performance of PBCG-RC was also observed in terms of prediction of high-risk disease at biopsy. Using any of the available RCs prior to biopsy is of greater assistance to prostate-specific antigen and DRE than examination alone. The study results show that PBCG-RC performed before biopsy has a higher predictive power than the other two RCs.
Keyphrases
- prostate cancer
- ultrasound guided
- fine needle aspiration
- radical prostatectomy
- patients undergoing
- end stage renal disease
- chronic kidney disease
- papillary thyroid
- newly diagnosed
- randomized controlled trial
- benign prostatic hyperplasia
- ejection fraction
- type diabetes
- prognostic factors
- skeletal muscle
- young adults
- childhood cancer
- open label