ADC and tumor contact length with the capsule as predictors of extraprostatic extension of prostate cancer on mpMRI
Number:27ESURABS0014
Type:Scientific Electronic Presentation
Authors: Lorenzo Vassallo, Mirella Fasciano, Andrea Cerutti, Gisella Lingua
Keywords:ADC - prostate - extracapsular - MRI - cancer
SECTIONS
Objectives

Extracapsular extension (ECE) of prostate cancer is a negative prognostic factor and is associated with an increased likelihood of cancer recurrence. Thus, accurate preoperative assessment of ECE is essential for the selection and planning of appropriate therapy in patients with prostate cancer. Several studies have assessed the value of preoperative MRI in patients with known prostate cancer for evaluation of ECE. However, the accuracy of mp-MRI in determining ECE has been variable among studies. This may stem from the fact that EPE evaluation on MRI has traditionally been based on the subjective assessment of imaging findings of abutment, irregularity or prostate capsule, bulging, and neurovascular bundle thickening on T2-weighted images (T2WIs). Poor interreader agreement and dependence on the level of experience renders consistent reporting of ECE among radiologists...

...read more


Material and methods

We retrospectively analyzed 112 patients who underwent preoperative MRI and radical prostatectomy for PCa between October 2017 and December 2020.

Images were acquired under fasting conditions with a 1.5 T scanner (Optima MR450w, GE Healthcare, Milwaukee, Illinois, USA) using a 32-channel phased-array coil combined with air-inflated endorectal coil (Medrad, Indianola, Pa). Imaging protocol included three orthogonal T2w sequences, axial Diffusion Weighted Imaging (DWI) and Dynamic Contrast Enhanced (DCE) which was triggered to start simultaneously with the power injection of 0.1 mmol/kg gadobutrol(Gadovist, Bayer Schering, Berlin, Germany) through a peripheral line at 0.7 mL/s, followed by infusion of 20 cc normal saline at the same rate. Axial DWI was acquired at b-values of 0, 1000 and 2000 s/mm2. The average time to complete the whole MRI examination was 40...

...read more

Figure 1. Example of patient with frank breach of prostate capsule at axial-T2-weighted MRI (A) and at diffusion-weighted-imaging (B) excluded from the study.
Figure 2. Example of patient with overt seminal vesicle invasion at axial-T2-weighted MRI (A), at ADC map (B) and at sagittal-T2-weighted MRI (C) excluded from the study.

Results

41 patients were excluded from the study. Mean participants age was 66.8 (range years 55.2 - 78.5), median prostate-specific antigen was 7.3 ng/mL (range, 3.2-21.9 ng/mL) and median prostate volume was 52.2 ml (range, 36.5 - 80.4 mL).

In 71 men we found 87 PCa, 45 with ECE and 42 without.

The optimal cutoff for ADC in predicting ECE was 0.73, which yielded an area under the ROC curve of 71.1% (95% CI: 59.8%-82.4%), corresponding to a sensitivity of 67% and a specificity of 82% (Figure 3). The optimal cutoff for TCL was 16.5 mm, which yielded an area under the ROC curve of 82.5% (95% CI: 73.8%-91.3%), corresponding to a sensitivity of 71% and a specificity of 84% (Figure 4). Combining the two criteria improved the diagnostic performance, yielding an area under the ROC curve of 83.2% (95% CI: 74.5%-91.8%), corresponding to a sensitivity of 74% and a specificity of 87%...

...read more

Figure 3. The optimal cutoff for ADC in predicting ECE was 0.73, which yielded an area under the ROC curve of 71.1% (95% CI: 59.8%-82.4%), corresponding to a sensitivity of 67% and a specificity of 82%.
Figure 4. The optimal cutoff for TCL was 16.5 mm, which yielded an area under the ROC curve of 82.5% (95% CI: 73.8%-91.3%), corresponding to a sensitivity of 71% and a specificity of 84%.

Conclusions

Our results shows that in patients with prostate cancer and no frank signs of ECE, ADC ratio and TCL were useful predictors of ECE and the best results were obtained when both the values were used in combination. 

Our study has several limitation. First it is a retrospective study with a small number of patients. In addition mpMRI were analyzed by a single reader, who however had over 7 years of experience in interpreting prostate MRI. Inter-reader agreement was not evaluated, however previous studies showed that ADC and TCL had greater inter-reader reproducibility .

Further studies, possibly multicentric, are required to confirm our preliminary results in a larger series of patients.