Do prostate MRI reports consider focal therapy when describing diagnostic findings? A descriptive review and analysis of concordance among prostate MRI reports and radical prostatectomy pathology reports.

Abstract

41 Background: Focal therapy (FT) for prostate cancer (PCa) is becoming an attractive option for men seeking treatments that minimize the morbidity of whole gland therapy. Prostate MRI reports should include lesion location, presence of bilateral extension and presence of multifocality as these radiographic features are essential in selecting appropriate candidacy. We aim to evaluate concordance of disease between MRI and final pathology reports. Methods: We conducted a retrospective review of pathology reports from radical prostatectomies (RP) performed at our institution between 2016-2019, in patients whom also underwent pre-operative prostate MRI’s. The number and locations of PI-RADS ≥ 3 lesions described in MRI reports were compared against the RP findings and assessed for disease concordance in terms of lesion location, extent (unilateral vs. bilateral extent) and/or the presence of tumor multifocality. Fischer’s exact test was used for statistical analysis. Results: A total of 1593 patients underwent RP, among which 416 (26.1%) had preoperative prostate MRI’s. A majority (65.3%) (n = 273/416) of MRI’s reported a single PI-RADS > 3 lesion, 51/416 (12.3%) had negative MRI’s and 22.1% (92/416) described >2 PI-RADS ≥ 3 lesions. A total of n = 388/416 (93.2%) had detailed descriptions of their MRI and path reports for comparative analysis. MRI and final path tumor concordance with respect to unifocal extent or multifocality was low at 37.2%. When a single PI-RADS > 3 lesion was described, concordance with final path was 37.6% as compared to 55.6% when >2 lesions were described (p = 0.006). Discordant findings at final path included the presence of tumor multifocality (80.7%) or the extent of a unifocal lesion (19.3%). Lesions crossing the midline were uncommonly reported on MRI (16%). When prostate midline assessment was included in the evaluation of lesion extent, the degree of concordance with final pathology improved from 33.3% to 64.2% (p = 0.038). Conclusions: Prostate MRI reports have a large discordance with RP reports. Tumor multifocality and/or unifocal disease crossing the midline may not hold significant meaning if the disease is ultimately organ confined and the treatment is aimed at the whole gland. However, in the context of focal therapy, these findings are of germane importance. Radiologists are encouraged to consider focal therapy as a treatment strategy when describing MRI findings and acknowledge the importance of tumor multifocality/bilaterality in this context.