Extraprostatic Extension in Multiparametric MRI; Is Presurgical Detection Possible?

Abstract Background To evaluate the efficacy of 1.5T MRI using lesion length (LL) and tumor-capsule contact length (TCL) in detecting extraprostatic extension (EPE) in prostate cancer (PCa). Methods A total of 110 patients who underwent radical prostatectomy due to PCa were enrolled. Preoperative MR images were evaluated retrospectively by two independent observers who did not know the histopathological results. The observers evaluated LL and TCL. The radiological findings, including lesion location, were verified using histopathological mapping. Results Multiparametric MRI examination of the prostate demonstrated low sensitivity (Observer 1; 40.4% and Observer 2; 40.4%) but high specificity (Observer 1; 96.6% and Observer 2; 84.5%), with significant differences for detecting EPE (Observer 1, p < 0.0001; Observer 2, p = 0.003). The increased PI-RADS score correlated positively with the increased EPE rate (p < 0.0001 for both observers). The mean LL and TCL values were statistically significantly higher in patients with EPE than patients without EPE. The TCL was a significant parameter for EPE, with high sensitivity and low for both observers. For both observes the cutoff value of LL for EPE was 14.5 mm, and the cutoff value of TCL for EPE was 9.5 mm. Histopathological LL value (28 ± 12,3 mm) was higher than radiological LLs (Observer 1; 22,14 ± 10,15 mm and Observer 2; 19,06 ± 8,61). Conclusion The results revealed that 1.5T MRI demonstrated low sensitivity and high specificity in detecting EPE. The LL and TCL may be indirectly beneficial in detecting EPE. Considering the radiological underestimation of LL may be helpful before PCa therapies.