A new parameter to increase the predictive value of multiparametric prostate magnetic resonance imaging for clinically significant prostate cancer in targeted biopsies: lesion density

Abstract

Aim

To investigate the predictive value of lesion length in multiparametric prostate magnetic resonance imaging with respect to prostate volume for clinically significant prostate cancer diagnosis in targeted biopsies.

Materials and methods

The data of biopsy-naïve patients in the Turkish Urooncology Association Prostate Cancer Database who underwent targeted prostate biopsies were included in this study. Lesion density is calculated as the ratio of lesion length (mm) in MR to prostate volume (cc). The biopsy results were divided into either clinically significant or insignificant cancer and benign groups. The difference in parameters between groups is evaluated by multivariable analysis to determine independent risk factors for clinically significant prostate cancer diagnosis.

Results

A total of 590 lesion biopsies were included in the study. In univariable analysis, prostate-specific antigen (PSA), PSA density, number of cores taken, lesion length, lesion density, patient age, and digital rectal examination findings were found to be different at a statistically significant level between groups (P values, respectively: 0.001, <0.001, <0.001, <0.001, <0.001, 0.012, 0.001). Subgroup analysis demonstrated that the lesion density was still significantly different between groups for all Prostate Imaging - Reporting and Data System (PI-RADS) 3, 4, and 5 subgroups (P values, respectively: 0.001, <0.001, <0.001). The multivariable analysis demonstrated that lesion density, along with the number of cores taken and the PI-RADS score of the lesion is an independent risk factor for predicting clinically significant prostate cancer, with the highest odds ratio among all parameters (OR: 27.31 [CI: 7.9–94.0]).

Conclusion

This study demonstrated that lesion size with respect to prostate volume is an important independent risk factor for the prediction of clinically significant prostate cancer in the lesion-targeted biopsy. Combined with the PI-RADS score and parameters like digital rectal examination (DRE) findings and PSA density may further increase predictive power and help clinicians decide whether to perform a biopsy in low-risk patients or perform a re-biopsy for high-risk patients subsequent to an initial negative biopsy.