José Maria Abascal

Objective: This study compares machine learning (ML) and logistic regression (LR) algorithms in developing a predictive model for sPCa using the seven predictive variables from the Barcelona (BCN-MRI) predictive model. Method: A cohort of 5005 men suspected of having PCa who underwent MRI and targeted and/or systematic biopsies was used for training, testing, and validation. A feedforward neural network (FNN)-based SimpleNet model (GMV) and a logistic regression-based model (BCN) were developed. The models were evaluated for discrimination ability, precision–recall, net benefit, and clinical utility. Both models demonstrated strong predictive performance. Results: The GMV model achieved an area under the curve of 0.88 in training and 0.85 in test cohorts (95% CI: 0.83–0.90), while the BCN model reached 0.85 and 0.84 (95% CI: 0.82–0.87), respectively (p > 0.05). The GMV model exhibited higher recall, making it more suitable for clinical scenarios prioritizing sensitivity, whereas the BCN model demonstrated higher precision and specificity, optimizing the reduction of unnecessary biopsies. Both models provided similar clinical benefit over biopsying all men, reducing unnecessary procedures by 27.5–29% and 27–27.5% of prostate biopsies at 95% sensitivity, respectively (p > 0.05). Conclusions: Our findings suggest that both ML and LR models offer high accuracy in sPCa detection, with ML exhibiting superior recall and LR optimizing specificity. These results highlight the need for model selection based on clinical priorities.

This study compares machine learning and logistic regression algorithms in developing a predictive model for sPCa using the seven predictive variables from the Barcelona (BCN-MRI) predictive model. A cohort of 5,005 men suspected of having PCa who underwent MRI, and targeted and/or systematic biopsies was used for training, validation and testing. A Feedforward Neural Network (FNN) SimpleNet-based model (GMV) and a logistic regression-based model (BCN) were developed. The models were evaluated for discrimination ability, net benefit and clinical utility. Both models demonstrated strong predictive performance. The GMV model achieved an area under the curve of 0.88 in training and 0.85 in test cohorts (95% CI: 0.83-0.90), while the BCN model reached 0.85 and 0.84 (95% CI: 0.82-0.87), respectively (p > 0.05). The GMV model exhibited superior recall, making it more suitable for clinical scenarios prioritizing sensitivity, whereas the BCN model demonstrated higher precision and specificity, optimizing the reduction of unnecessary biopsies. Both models provided net benefit over biopsying all men, reducing unnecessary procedures by 27.5-29% and 27-27.5% of prostate biopsies at 95% sensitivity, respectively (p > 0.05). Both, machine learning and logistic regression-based models exhibited high and comparable similar clinical performance in sPCa detection using a limited dataset.

Background: A transperineal approach to prostate biopsy is now recommended to reduce the risk of infectious complications associated with the transrectal route. Our aim is to compare the efficacy of transrectal- and transperineal-guided biopsies involving the magnetic resonance imaging (MRI) of index lesions in detecting significant prostate cancer (sPCas), and to evaluate the role of systematic biopsies. Methods: In a prospective and multicenter trial conducted in an opportunistic early detection program for sPCa in Catalonia (Spain), between 2021 and 2023, 4029 men suspected of having PCa underwent multiparametric MRI followed by guided and systematic biopsies. From this cohort, we retrospectively selected 1376 men with reports of the size and localization of their index lesions. A matched group of 325 pairs of men subjected to transrectal and transperineal biopsy were chosen to account for confounding variables. We compared sPCa detection rates determined via index lesions and systematic biopsies, as well as by lesion localization. Results: Transperineal and transrectal biopsies detected sPCa in 49.5% vs. 40.6% overall (p = 0.027), 44.6% vs. 30.8% from index lesions (p = 0.001), and 24.3% vs. 35.1% from systematic biopsies (p = 0.003). SPCa detection rates were higher in transperineal biopsies across all index lesion localizations, with significant increases in the anterior zone (47.8% vs. 20.8% at the mid-base, p = 0.039, and 52.9% vs. 24.2% at the apex, p = 0.024) and central zone (33.3% vs. 5.9%, p = 0.003). With regards to SPCa detected only in systematic biopsies, 10.5% of cases were detected in transrectal biopsies and 4.9% of cases were detected in transperineal biopsies (p = 0.012). Conclusions: Targeted biopsies conducted via the transperineal route showed higher sPCa detection rates than transrectal biopsies, particularly for anterior and apical lesions, with systematic biopsies showing reduced utility.

Objective : To compare the efficacy of transrectal and transperineal prostate-guided biopsies to magnetic resonance imaging (MRI) index lesions in detecting clinically significant prostate cancer (csPCa), and to evaluate the role of systematic biopsies. Methods: This prospective and multicenter trial, conducted in the early detection program of csPCa of Catalonia (Spain) between 2021 and 2023, involved 4,029 men suspected of ha-ving PCa who underwent multiparametric MRI followed by guided and systematic biopsies. From this cohort, 1,376 men with reported size and localization of index lesions were selected. A matched group of 325 pairs of men subjected to transrectal and transperineal biopsy was chosen to account for confounding variables. We compared csPCa detection rates at index lesions and systematic biopsies, as well as by lesion localization. Results: Transperineal and transrectal biopsies detected csPCa in 49.5% vs. 40.6% overall (p = 0.027), 44.6% vs. 30.8% at index lesions (p = 0.001), and 24.3% vs. 35.1% at systematic biopsies (p = 0.003). CsPCa detection rates were higher in transperineal biopsies across all index lesion localizations, with significant increases in anterior zone (47.8% vs. 20.8% at mid-base, p = 0.039; 52.9% vs. 24.2% at apex, p = 0.024) and central zone (33.3% vs. 5.9%, p = 0.003). CsPCa detected only in systematic biopsies was 10.5% in transrectal biopsies and 4.9% in transperineal biopsies (p = 0.012). Conclusions: Targeted biopsies via the transperineal route showed higher csPCa detection rates than transrectal biopsies, parti-cularly for anterior and apical lesions, with systematic biopsies showing reduced utility.

The role of digital rectal examination (DRE) in the early detection of significant prostate cancer (PCa) is being questioned in the era of magnetic resonance imaging (MRI). However, some men with suspected PCa may still be identified solely through DRE, even with low serum prostate-specific antigen (PSA) levels. Additionally, most predictive models designed to improve significant PCa diagnostic pathways incorporate DRE findings. We assessed the role of DRE among 5005 men with serum PSA levels > 3.0 ng/mL and/or suspicious DRE findings, who underwent pre-biopsy MRI and targeted and/or systematic biopsies, as part of the significant PCa opportunistic screening program in Catalonia (Spain) between 2016 and 2023. Significant PCa, defined as grade group > 2, was detected in 2097 men (41.9%). Suspicion of PCa was based solely on DRE in 206 cases (4.1%) with significant PCa detected in 50 of them (2.4%). Two pathways using the Barcelona predictive models, before and after MRI, with and without DRE findings showed specificities of 52.8 and 38.7%, respectively (p < 0.001), after fixing sensitivity at 90%. Prostate biopsy was avoided in 35.1 and 26.7%, respectively (p < 0.001), while its efficacy increased from 52.8 to 58%. We conclude that DRE improved the effectiveness of an opportunistic significant PCa-screening program.

Since the optimal scheme for targeted biopsies of magnetic resonance imaging (MRI) suspicious lesions remains unclear, we compare the efficacy of two schemes for these index lesions. A prospective trial was conducted in 1161 men with Prostate Imaging Reporting and Data System v 2.1 3–5 undergoing targeted and 12-core systematic biopsy in four centers between 2021 and 2023. Two- to four-core MRI-transrectal ultrasound fusion-targeted biopsies via the transperineal route were conducted in 900 men in three centers, while a mapping per 0.5 mm core method (saturated scheme) was employed in 261 men biopsied in another center. A propensity-matched 261 paired cases were selected for avoiding confounders other than the targeted biopsy scheme. CsPCa (grade group ≥ 2) was identified in 125 index lesions (41.1%) when the two- to four-core scheme was employed, while in 187 (71.9%) when the saturated biopsy (p < 0.001) was used. Insignificant PCa (iPCa) was detected in 18 and 11.1%, respectively (p = 0.019). Rates of csPCa and iPCa remained similar in systematic biopsies. CsPCa detected only in systematic biopsies were 5 and 1.5%, respectively (p = 0.035) in each group. The saturated scheme for targeted biopsies detected more csPCa and less iPCa than did the two- to four-core scheme in the index lesions. The rate of csPCa detected only in the systematic biopsies decreased when the saturated scheme was employed.

Concerns exist regarding the effects of 5-alpha reductase inhibitors (5-ARIs) on multipa-rametric magnetic resonance imaging (mpMRI) and clinically significant prostate cancer (csPCa) detection. Our objective is to analyze the effect of 5-ARI on the prostate imaging–reporting and data system (PI-RADS) distribution and csPCa and insignificant PCa (iPCa) detection. Among 2212 men with serum prostate-specific antigen levels of >3.0 ng/mL and/or suspicious digital rectal examinations who underwent mpMRI and targeted and/or systematic biopsies, 120 individuals exposed to 5-ARI treatment for over a year were identified. CsPCa was defined when the grade group (GG) was >2. The overall csPCa and iPCa detection rates were 44.6% and 18.8%, respectively. Since logistic regression revealed independent predictors of PCa, a randomized matched group of 236 individuals was selected for analysis. The PI-RADS distribution was comparable with 5-ARI exposure (p 0.685). The CsPCa detection rates in 5-ARI-naïve men and 5-ARI-exposed men were 52.6% and 47.4%, respectively (p 0.596). IPCa was detected in 37.6 and 62.5%, respectively (p 0.089). The tumor GG distribution based on 5-ARI exposure was similar (p 0.149) to the rates of csPCa and iPCa across the PI-RADS categories. We conclude that exposure to 5-ARI in suspected PCa men did not change the PI-RADS distribution and the csPCa and iPCa detection rates.

Morote J, et al. World J Mens Health. 2024 Jan;42:e3. https://doi.org/10.5534/wjmh.230028

The primary objective of this study was to analyse the current accuracy of targeted and systematic prostate biopsies in detecting csPCa. A secondary objective was to determine whether there are factors predicting the finding of csPCa in targeted biopsies and, if so, to explore the utility of a predictive model for csPCa detection only in targeted biopsies. We analysed 2122 men with suspected PCa, serum PSA > 3 ng/mL, and/or a suspicious digital rectal examination (DRE), who underwent targeted and systematic biopsies between 2021 and 2022. CsPCa (grade group 2 or higher) was detected in 1026 men (48.4%). Discrepancies in csPCa detection in targeted and systematic biopsies were observed in 49.6%, with 13.9% of csPCa cases being detected only in systematic biopsies and 35.7% only in targeted biopsies. A predictive model for csPCa detection only in targeted biopsies was developed from the independent predictors age (years), prostate volume (mL), PI-RADS score (3 to 5), mpMRI Tesla (1.5 vs. 3.0), TRUS-MRI fusion image technique (cognitive vs. software), and prostate biopsy route (transrectal vs. transperineal). The csPCa discrimination ability of targeted biopsies showed an AUC of 0.741 (95% CI 0.721–0.762). The avoidance rate of systematic prostate biopsies went from 0.5% without missing csPCa to 18.3% missing 4.6% of csPCa cases. We conclude that the csPCa diagnostic accuracy of targeted biopsies is higher than that of systematic biopsies. However, a significant rate of csPCa remains detected only in systematic biopsies. A predictive model for the partial omission of systematic biopsies was developed.