Comparison of biparametric and multiparametric prostate MRI to assess extraprostatic tumor extension.

OBJECTIVES

Accurate assessment of extraprostatic extension (EPE) in prostate cancer is essential for surgical planning. This study aimed to compare the diagnostic performance of biparametric (bpMRI) and multiparametric MRI (mpMRI) for detecting EPE using histopathology as the reference standard.

MATERIALS AND

METHODS

Biopsy-naïve men that underwent mpMRI followed by radical prostatectomy were included retrospectively. Four radiologists independently evaluated bpMRI (T2-weighted, diffusion-weighted) and mpMRI (addition of DCE). PI-RADS, pseudocapsular (PC) bulging, PC irregularity, MRI-visible PC breach, and PC tumor contact length were assessed. Histopathological prostatectomy specimens served as reference. Sensitivities and specificities were calculated using generalized estimating equations. A scoring system, integrating these morphological characteristics with MRI-visible PC breach and tumor contact length weighted double, was evaluated by calculating the area-under-the-curve (AUC) and Youden-Index.

RESULTS

90 men (mean age: 64 ± 6) were included. 86 (95.6%) patients had clinically significant prostate cancer (ISUP ≥ 2). bpMRI and mpMRI had no significant difference regarding morphological characteristics, for instance, MRI-visible PC breach (sensitivity: 0.22 [95% CI: 0.15, 0.31] versus 0.23 [95% CI: 0.16, 0.31], p = 0.87; specificity: 0.99 [95% CI: 0.96, 0.99] versus 0.99 [95% CI: 0.96, 0.99], p = 1.00). The AUC for the scoring system was similar between bpMRI and mpMRI (mean AUC: 0.80 [95% CI: 0.69, 0.90] versus 0.79 [95% CI: 0.69, 0.90], p = 0.77) with an optimal score of 2/6 points for both protocols.

CONCLUSION

Diagnostic accuracy for detecting EPE did not differ between bpMRI and mpMRI, thus supporting bpMRI as an alternative for local prostate cancer staging.