Correlation of apparent diffusion coefficients (ADCs) on diffusion-weighted (DW) magnetic resonance (MR) imaging with lesion size and prostate cancer risk.

Abstract

48 Background: A negative correlation between ADC values and Gleason risk of prostate cancer (CaP) has been reported by the NIH. We correlate ADC values on DW MR imaging with lesion size and risk stratification of CaP. Methods: Men with abnormal DRE or PSA underwent a 3T MP-MRI (T2, DWI, and DCE) with an endorectal coil. ADC maps were calculated using b values of 0, 500, 1,000 and 1,500; additional b 2000 images were obtained separately. Three radiologists (EB, RV, AR) reviewed and graded all lesions using the 5-point Likert scale and PI-RADS score. A total of 268 men with suspicious lesion (s) on MRI were prospectively enrolled. The UroNav system (Invivo, Florida) was used to perform MR/TRUS fusion-guided prostate biopsies, obtaining one biopsy core in axial and sagittal planes from each lesion. Our institution’s pathologist reviewed all biopsies. Lowest ADC values of suspicious lesions were correlated against lesion size and Gleason risk stratification. Results: The median age and PSA were 65.0 years and 7.4 ng/mL respectively. An average of 1.5 (378/256) suspicious lesions per patient was identified on the MP-MRI. Median volume of MRI target lesions was 0.36 cm3. The cancer detection rate of fusion biopsy on lesion level analysis was 45.2 % (171/378). The mean ± SD ADC values (x 10-6 mm2/sec) for lesions negative and positive for CaP were 1022 ± 260 and 724 ± 223 respectively (p <0.0001). The mean ± SD ADC values (x 10-6 mm2/sec) for lesion size < 0.2 cm3, 0.2 – 0.5 cm3, 0.5 – 1.0 cm3, and > 1 cm3 were 771 ± 246, 677 ± 206, 629 ± 119, and 500 ± 86respectively. Higher risk CaP was associated with lower ADC values with AUC of 0.77 (p=0.0001). The mean ± SD ADC values (x 10-6 mm2/sec) for low (≤6), intermediate (7), and high (8-10) Gleason grade CaP were 845 ± 285, 687± 195, and 628 ± 145 respectively. Conclusions: The ADC value of suspicious lesions on MRI has an inverse correlation with lesion size and prostate cancer risk. It may be a useful in predicting CaP and its clinical significance. Further studies are needed to determine its value in stratifying patients for prostate biopsy, active surveillance and/or treatment.