Background: Prostate cancer (PCa) is one of the most prevalent malignancies in men worldwide. Its diagnosis primarily relies on prostate biopsy, which includes systematic biopsies and targeted biopsies directed at suspicious lesions detected by MRI. The introduction of targeted prostate biopsies improved diagnostic accuracy, but the relative efficacy of cognitive versus software-assisted techniques remains debated. Purpose: This study aimed to compare the diagnostic performance of cognitive and software-assisted targeted biopsies in detecting PCa and clinically significant prostate cancer (csPCa) in a cohort of patients undergoing transperineal prostate biopsy. A secondary aim was to investigate clinical and radiological predictors of PCa and csPCa. Patients and methods: This retrospective monocentric study included 445 patients who underwent transperineal prostate biopsy at the University Hospital of Padova from January 2023 to June 2024. Among these, 261 underwent cognitive targeted biopsy and 184 underwent software-assisted targeted biopsy. Systematic biopsies were performed alongside targeted biopsies in all patients. Detection rates for PCa and csPCa were compared between the two groups and multivariable logistic regression was used to identify significant predictors of PCa and csPCa. Results: No statistically significant differences were found between detection rates of cognitive and software-assisted techniques, either when only considering targeted biopsies (csPCa: 27.7% vs 34.1%, p = 0.3) or when combining targeted and systematic biopsies (csPCa: 44.4% vs 38.6%, p = 0.4). However, a slight trend favoring a sotware-assisted technique was observed. Multivariable analysis identified key predictors of csPCa, including age, prostate volume, lesion localisation (anterior vs posterior), PI-RADS score and lesion diameter. Conclusions: Cognitive and software-assisted targeted biopsies demonstrated similar diagnostic accuracy for PCa and csPCa detection. These findings suggest that, to date, the choice between the two techniques is influenced more by practical considerations, such as operator expertise and resource availability, rather than differences in diagnostic efficacy. Regarding predictors of PCa and csPCa, the identification of robust predictors, particularly PI-RADS score and anterior lesion localisation, highlights their critical role in improving biopsy planning.
Presupposti dello studio: Il carcinoma della prostata (PCa) è uno delle neoplasie maligne più comuni negli uomini. La diagnosi si basa principalmente sulla biopsia prostatica, che comprende biopsie sistematiche e biopsie mirate alle lesioni sospette identificate alla risonanza magnetica (RM). L’introduzione delle biopsie mirate ha migliorato l’accuratezza diagnostica, ma l’efficacia relativa delle tecniche cognitive rispetto a quelle assistite da software è ancora dibattuta. Scopo dello studio: Lo scopo di questo studio è quello di confrontare la performance diagnostica delle biopsie mirate cognitive e di quelle assistite da software nel rilevare PCa e carcinoma prostatico clinicamente significativo (csPCa) in pazienti sottoposti a biopsia prostatica transperineale. L’obiettivo secondario era investigare i predittori clinici e radiologici di PCa e csPCa. Pazienti e metodi: Questo studio retrospettivo monocentrico ha incluso 445 pazienti sottoposti a biopsia prostatica transperineale presso l’Azienda Ospedale-Università di Padova tra gennaio 2023 e giugno 2024. Di questi, 261 hanno eseguito una biopsia mirata con tecnica cognitiva e 184 con tecnica assistita da software. In tutti i pazienti sono state effettuate anche biopsie sistematiche. I tassi di rilevamento di PCa e csPCa sono stati confrontati tra i due gruppi e un’analisi di regressione logistica multivariata è stata utilizzata per identificare i predittori significativi di PCa e csPCa. Risultati: non sono state rilevate differenze statisticamente significative nei tassi di rilevamento di tra le tecniche cognitive e assistite da software, né considerando solo le biopsie mirate (csPCa: 27,7% vs 34,1%, p = 0.3) né considerando la combinazione di biopsie mirate e sistematiche (csPCa: 44.4% vs 38.6%, p = 0.4). Tuttavia, è stata osservata una tendenza verso una maggiore rilevazione da parte delle tecniche assistite da software. L’analisi multivariata ha identificato diversi predittori di csPCa, tra cui età, volume prostatico, localizzazione della lesione (anteriore vs posteriore), punteggio PI-RADS e diametro della lesione. Conclusioni: Le biopsie mirate e quelle assistite da software hanno dimostrato un’accuratezza diagnostica simile nel rilevare PCa e csPCa. Questi risultati suggeriscono che la scelta tra le due tecniche possa dipendere più da fattori pratici, quali l’esperienza, dell’operatore e le risorse disponibili, piuttosto che differenze nell’efficacia diagnostica. Riguardo i predittori di PCa e csPCa, l’identificazione di forti predittori, in particolare il punteggio PI-RADS e la localizzazione anteriore della lesione, sottolinea il loro ruolo fondamentale nel migliorare la pianificazione della biopsia.
Biopsia Transperineale MRI guidata per la Diagnosi del Carcinoma Prostatico: confronto fra fusione di immagini Cognitiva vs assistita da Software
PASTORI, MARTA
2023/2024
Abstract
Background: Prostate cancer (PCa) is one of the most prevalent malignancies in men worldwide. Its diagnosis primarily relies on prostate biopsy, which includes systematic biopsies and targeted biopsies directed at suspicious lesions detected by MRI. The introduction of targeted prostate biopsies improved diagnostic accuracy, but the relative efficacy of cognitive versus software-assisted techniques remains debated. Purpose: This study aimed to compare the diagnostic performance of cognitive and software-assisted targeted biopsies in detecting PCa and clinically significant prostate cancer (csPCa) in a cohort of patients undergoing transperineal prostate biopsy. A secondary aim was to investigate clinical and radiological predictors of PCa and csPCa. Patients and methods: This retrospective monocentric study included 445 patients who underwent transperineal prostate biopsy at the University Hospital of Padova from January 2023 to June 2024. Among these, 261 underwent cognitive targeted biopsy and 184 underwent software-assisted targeted biopsy. Systematic biopsies were performed alongside targeted biopsies in all patients. Detection rates for PCa and csPCa were compared between the two groups and multivariable logistic regression was used to identify significant predictors of PCa and csPCa. Results: No statistically significant differences were found between detection rates of cognitive and software-assisted techniques, either when only considering targeted biopsies (csPCa: 27.7% vs 34.1%, p = 0.3) or when combining targeted and systematic biopsies (csPCa: 44.4% vs 38.6%, p = 0.4). However, a slight trend favoring a sotware-assisted technique was observed. Multivariable analysis identified key predictors of csPCa, including age, prostate volume, lesion localisation (anterior vs posterior), PI-RADS score and lesion diameter. Conclusions: Cognitive and software-assisted targeted biopsies demonstrated similar diagnostic accuracy for PCa and csPCa detection. These findings suggest that, to date, the choice between the two techniques is influenced more by practical considerations, such as operator expertise and resource availability, rather than differences in diagnostic efficacy. Regarding predictors of PCa and csPCa, the identification of robust predictors, particularly PI-RADS score and anterior lesion localisation, highlights their critical role in improving biopsy planning.File | Dimensione | Formato | |
---|---|---|---|
Pastori_Marta_tesi.pdf
accesso aperto
Dimensione
987.15 kB
Formato
Adobe PDF
|
987.15 kB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/78891