Presentation Authors: Mariaconsiglia Ferriero*, Rocco Simone Flammia, Rome, Italy, Guglielmo Zeccolini, Bernardino De Concilio, Bassano del Grappa, Italy, Gabriele Tuderti, Umberto Anceschi, Aldo Brassetti, Riccardo Mastroianni, Salvatore Guaglianone, Rome, Italy, Antonio Celia, Bassano del Grappa, Italy, Michele Gallucci, Giuseppe Simone, Rome, Italy
Introduction: Computer aided diagnosis (CAD) systems have shown promise in identification of prostate cancer (PCa) on mpMRI in several studies. A wrong interpretation of MRI and a not appropriate contouring of targets can compromise the detection rate of a Fusion (US/MRI guided) prostate biopsy (FPB).We investigated the potential impact of CAD system on the detection rate of PCa in a series of FPB performed in two different centres.
Methods: From March 2016 to October 2018, data on FPB were collected from prospective data set at two different centres. We compared PCa detection rate with a per patient and per target area analysis, in two groups of FPB performed either with the assistance of CAD system or using MRI only.Chi Square and Student t test were used to compare categorical and continuous variables, respectively.
Results: A total of 183 FPB were performed transperineally with Biopsee Fusion System, 89 with the assistance of CAD System (Watson Elementary). The two cohorts were homogeneous for age, baseline PSA, prostate volume, number of target areas and number of target cores, while the CAD assisted group had a significantly higher number of total cores (p < 0.001). (Table 1)On a patient based analysis the overall detection rate of any PCa and clinically significant (cs) PCa were 56.3% and 30.6%, respectively. The same outcomes were comparable between the CAD assisted and MRI only FPB groups (59.6% vs 53.2% [p=0.45] and 30.3% vs 30.9%, [p=0.99]) for any prostate cancer and csPca, respectively. On a target area based analysis, the overall detection rate of any PCa was 54.1%, with non-significant differences between groups (51.5 vs 48.5, for the CAD assisted and MRI only FPB, respectively p=0.45). Similarly, overall cs PCa detection rate was 29.5%, with comparable outcomes between groups (29.8% vs 29.2%, respectively p=0.98).After stratifying for PI-RADS score, the use of CAD did not impact significantly on diagnosis (all p>0.05). A subgroup of FBP in transitional zone lesions, detection rate of csPCa was significantly higher in the cohort of CAD assisted procedures (54.5%vs11.1%, respectively p=0.028).
Conclusions: These data showed how CAD assistance for FPB did not impact on detection rate of PCa in a per patient and per target analysis. For suspicious area located in transition zone, the use of CAD increased detection rate of csPCa.