Apparent diffusion coefficient in the evaluation of side-specific extracapsular extension in prostate cancer:...

The aim of this study is to develop a nomogram of clinical utility based on apparent diffusion coefficient (ADC) from diffusion-weighted imaging to predict extracapsular extension (ECE), and to validate externally its clinical utility.

A total of 101 men (70 for the creation and 31 for external validation of the nomogram) underwent 1.5T multiparametric magnetic resonance imaging followed by radical prostatectomy at 2 different institutions. ADC values were assessed for normal and pathological tissue. Clinical and pathological variables were investigated by univariate and multivariate logistic regression analyses on 70 patients and logistic regression coefficients were used to develop our nomogram. Receiver operating characteristic curve analysis was performed to determine the optimal ADC cut off for ECE. The nomogram was then externally validated on 31 patients at another institution.

At univariate analysis, the following variables were associated with ECE: pathological ADC and Gleason at biopsy (P<0.001) along with tumor volume and ECE at imaging (P = 0.003). At multivariate analysis, pathological ADC (P = 0.027), tumor volume (P = 0.011), and biopsy Gleason (P = 0.040) maintained their independent predictor status and were included in our nomogram together with normal ADC and ECE at imaging. Our nomogram showed a significant higher sensitivity (88%) than T2-weighted imaging (54%; P = 0.010). External validation resulted in an overall accuracy of 81%.

ADC represents a potential imaging biomarker to predict side-specific ECE in patients with prostate cancer. Our nomogram could improve the current diagnostic pathway and possibly the therapeutic approach for this disease.