Introduction

PI-RADS ver.2 recommends that DWI should include ADC map and high b-value images (b-value≥1400 sec/mm²), and if only two b-values can be acquired, it is preferred that the lowest b-value should be set at 50-100 sec/mm² and the highest at 800-1000 sec/mm².¹ This recommendation sounds reasonable but has yet to be assured sufficiently regarding repeatability and correlation with cancer aggressiveness. The present study was investigated to evaluate what combination of b-values is appropriate for calculating ADC with respect to ADC repeatability and correlation with Gleason score.

Methods

Consecutive 20 patients with histologically confirmed prostate cancer (21 lesions) who underwent pelvic MRI using whole-body clinical MRI system (3T of Ingenia, Philips Medical Systems, Best, the Netherlands) to evaluate prostate in our hospital from July to October in 2017 were included. All patients gave informed consent to participate in the study. Two sequential free-breathing single shot diffusion-weighted spin-echo echo-planar imaging (DWI) with b-values of 0, 100, 1000 and 1500 sec/mm² and identical positioning were obtained during examination. The regions of interest (ROI) were assigned at 21 cancers on the 1st DWI and pasted them onto the 2nd DWI holding their size and location. Slight visual adjustment of the ROI on the 2nd DWI was performed when two observers considered it is necessary to cope with minimal distortions and/or movements. Each voxel ADC was calculated by fitting signal intensity of each voxel into mono-exponential curve. ADCs calculated from several combinations of b-values were expresses as follows: ADC (0, 1000), ADC (0, 1500), ADC (100, 1000), ADC (100, 1500) and ADC (1000, 1500) were calculated from b-values of 0 and 1000 sec/mm², 0 and 1500 sec/mm², 100 and 1000 sec/mm², 100 and 1500 sec/mm² and 1000 and 1500 sec/mm², respectively. The repeatability of ADC histogram metrics of 10% (ADC 10%), 25% (ADC 25%), median (ADC median), mean (ADC mean), 75% (ADC 75%), 90% (ADC 90%), skewness (ADC skewness) and kurtosis (ADC kurtosis) of voxel-wise ADC within ROI was evaluated using intraclass correlation coefficient (ICC), Bland-Altman plot and within-subject coefficient of variance. The correlation between ADC and Gleason score was also analyzed with Spearman’s rank-order correlation. In cases in which radiotherapy (IMRT) was selected and only prostate biopsy, not prostatectomy, was performed, representative and maximum Gleason scores were used separately. In cases prostatectomy was performed, Gleason score was used for both representative and maximum Gleason scores. p<0.05 was considered statistically significant.

Results

ADC (0, 1500) showed the highest repeatability in all ADC histogram metrics except ADC 10%. ADC skewness and kurtosis showed very low repeatability irrespective of b-value combinations. Therefore, Spearman’s rank-order correlation test was not performed for ADC skewness and kurtosis. Significant correlation with maximum Gleason score was obtained in ADC median and mean calculated from ADC (0, 1500) and ADC 25% calculated from ADC (0, 1000), respectively. Representative Gleason score showed no significant correlation with ADC histogram metrics. ADC histogram metrics calculated from ADC (100, 1000) recommended by PI-RADS ver. 2 showed lower repeatability compared with those calculated from ADC (0, 1500) and no significant correlation with Gleason score. A representative case was shown in Fig. 1. Details of the results were summarized in Table 1.

Conclusion

ADC calculated with b-values of 0 and 1500 sec/mm² is considered appropriate with respect to ADC repeatability and correlation with Gleason score when b-values of DWI were restricted to only two values due to clinical conditions.

Acknowledgements

No acknowledgement found.

References

1. PI-RADS Prostate Imaging - Reporting and Data System: 2015, Version 2. Weinreb JC, Barentsz JO, Choyke PL, Cornud F, Haider MA, Macura KJ, Margolis D, Schnall MD, Shtern F, Tempany CM, Thoeny HC, Verma S. Eur Urol. 2016 Jan;69(1):16-40.