Influence of MRIs performed in a 6-week interval on the histopathological detection of prostate cancer with different PIRADS classifications: A real-world data analysis.

Abstract

334 Background: In the early phases of MRI-guided prostate biopsies, it was often necessary to perform an additional in-house multiparametric MRI after an external MRI to accurately assess lesions for biopsy. This study aims to evaluate the consistency of lesion identification and the associated clinical and histological outcomes at a high-volume tertiary care center performing over 500 biopsies annually. Methods: This analysis included real-world data from 111 patients treated between 2018 and 2022. Each patient underwent two multiparametric MRI scans of the prostate at different institutions, with a median interval of 42 days [32–52] between scans. An MRI-fused biopsy followed 7 days [4–10] after the second MRI. Data were analyzed using the Chi-square test. Results: The PIRADS scores for index lesions on the in-house MRI were as follows: PIRADS V in 33.3% (n=37), PIRADS IV in 49.5% (n=55), PIRADS III in 12.6% (n=14), and PIRADS II in 4.5% (n=5) (Table 1). Most lesions were located in the peripheral zone of the prostate (68.8%, n=75). Cancer detection rates for randomized and/or targeted biopsies were 91.9% (n=34) for PIRADS V, 65.5% (n=36) for PIRADS IV, 21.4% (n=3) for PIRADS III, and 20% (n=1) for PIRADS II. Clinically significant prostate cancer (ISUP > 1) was detected in 91.9% (n=34) of PIRADS V, 60% (n=33) of PIRADS IV, and 21.4% (n=3) of PIRADS III lesions. Overall, malignant histology was found in 64.9% (n=72) of targeted lesions and 59.5% (n=66) of randomized biopsies. A median of 7 [5.5–8.5] biopsies were taken per lesion, with an additional 13 [11–15] randomized biopsies. In the initial external MRI, 18% (n=20) of patients were assigned a higher PIRADS category, while 10.8% (n=12) were had a lower score (p<0.001). The biopsy plan was adjusted for 57 patients (51.4%) based on the findings of the second MRI. However, given the cancer detection rates in both targeted and randomized biopsies, it is likely that all cancers would have been detected regardless of these adjustments. Conclusions: Although PIRADS classifications varied slightly between the two MRI scans performed six weeks apart, and biopsy plans were adjusted in about half of the cases, all cancers would likely have been identified. The six-week interval between MRI scans does not appear to significantly affect biopsy accuracy. PIRADS classification comparison of internal and external MRI. External MRI PIRADS II PIRADS III PIRADS IV PIRADS V All Internal MRI PIRADS II 0 40% (2) 40% (2) 20% (1) 5 PIRADS III 0 28.6% (4) 57.1% (8) 14.3% (2) 14 PIRADS IV 0 9.1% (5) 81.8% (45) 9.1% (5) 55 PIRADS V 0 2.7% (1) 16.2% (6) 81.1% (30) 37 All 0 12 60 39