Lesion size on prostate magnetic resonance imaging predicts adverse radical prostatectomy pathology

Abstract

Objectives: To investigate the value of the maximal lesion diameter on preoperative multiparametric/bi-parametric magnetic resonance imaging for estimating the risk of adverse radical prostatectomy pathology.

Patients and methods: Consecutive patients (n = 162) with prostate multiparametric or biparametric magnetic resonance images acquired before prostatectomy were retrospectively stratified into two groups: 65 patients with normal MRI (n = 18) or a suspicious lesion <15 mm in diameter (n = 47), and 97 patients with a lesion diameter ≥15 mm. The presence of extraprostatic extension, margin positivity, seminal vesicle invasion, and lymph node metastasis was examined in these groups using logistic regression analysis, including preoperative clinical parameters (prostate-specific antigen concentration, biopsy Gleason grade group, clinical T-stage, and D’Amico risk group).

Results: The prevalence of extraprostatic extension, margin positivity, and seminal vesicle invasion was 53.1% (86/162), 22.8% (37/162), and 17.9% (29/162), respectively. Lymphadenectomy was performed in 64 men, of whom 14 had lymph node metastasis. Lesion diameter ≥15 mm strongly predicted extraprostatic extension (Odds ratio: 7.94, 95% confidence interval: 3.87−16.28, p < 0.001), margin positivity (Odds ratio: 7.86, 95% confidence interval 2.63−23.51, p < 0.001), and seminal vesicle invasion (Odds ratio: 7.57, 95% confidence interval 2.18−26.22, p = 0.001). Lesion diameter ≥15 mm was an independent risk factor for adverse prostatectomy pathology. Lesion diameter ≥20 mm, but not ≥15 mm, was a significant risk factor for lymph node metastasis.

Conclusion: Magnetic resonance imaging lesion diameter ≥15 mm is an independent risk factor for extraprostatic extension, margin positivity and seminal vesicle invasion.

Keywords:

• Magnetic resonance imaging
• pathology
• prostate cancer
• radical prostatectomy

Acknowledgements

We are grateful to Pasi Ohtonen, M.Sc. for assistance with statistical analyses. This work was supported in part by the Finnish Urological Association to Panu Tonttila.

Disclosure statement

The authors have no conflicts of interest to disclose.