AUTHOR=Ren Lei , Chen Yanling , Liu Zixiong , Huang Guankai , Wang Weifeng , Yang Xu , Bai Baohua , Guo Yan , Ling Jian , Mao Xiaopeng 

TITLE=Integration of PSAd and multiparametric MRI to forecast biopsy outcomes in biopsy-naïve patients with PSA 4~20 ng/ml

JOURNAL=Frontiers in Oncology

VOLUME=Volume 14 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2024.1413953

DOI=10.3389/fonc.2024.1413953

ISSN=2234-943X

ABSTRACT=Abstract
Introduction: To investigate whether the transrectal ultrasound-guided combined biopsy (CB) improves the detection rates of prostate cancer (PCa) and clinically significant PCa (csPCa) in biopsy-naïve patients. We also aimed to compare the Prostate Imaging Reporting and Data System (PI-RADS v2.1) score, ADC values, and PSA density (PSAd) in predicting csPCa by the combined prostate biopsy. 
Methods: This retrospective and single-center study included 389 biopsy-naïve patients with PSA level 4 ~ 20 ng/ml, of whom 197 patients underwent prebiopsy mpMRI of the prostate. The mpMRI-based scores (PI-RADS v2.1 scores and ADC values) and clinical parameters were collected and evaluated by logistic regression analyses. Multivariable models based on the mpMRI-based scores and clinical parameters were developed by the logistic regression analyses to forecast  biopsy outcomes of CB in biopsy-naïve patients. The ROC curves measured by the AUC values, calibration plots, and DCA were performed to assess multivariable models. 
Results: The CB can detect more csPCa compared with TRUSB (32.0% vs 53%). The Spearman correlation revealed that Gleason scores of the prostate biopsy significantly correlated with PI-RADS scores and ADC values. The multivariate logistic regression confirmed that PI-RADS score 4, 5 and prostate volume were important predictors of csPCa. The PI-RADS+ADC+PSAd (PAP) model had the highest AUCs of 0.913 for predicting csPCa in biopsy-naïve patients with PSA level 4 ~ 20 ng/ml. When the biopsy risk threshold of the PAP model was greater than or equal to 0.10, 51% patients could avoid unnecessary biopsy, and only 5% patients of csPCa were missed.
Conclusion: The pre-biopsy mpMRI and the combined prostate biopsy has a high CDR of csPCa in biopsy-naïve patients. A multivariable model based on the mpMRI-based scores and PSAd could provide a reference for clinicians in forecasting biopsy outcomes in biopsy-naïve patients with PSA 4 ~ 20 ng/ml and make a more comprehensive assessment during the decision-making of the prostate biopsy.