Apparent diffusion coefficient predicts risk of grade reclassification in men on active surveillance for prostate cancer
Concerns about the presence of occult, higher-grade prostate cancer limit retention in active surveillance (AS) and drive intensive monitoring. Multivariable risk prediction can improve identification of men with higher-grade disease, while also minimizing repeat biopsies. Our objective was to determine if apparent diffusion coefficient (ADC) on baseline MRI (bMRI) is predictive of grade reclassification while on AS.
We retrospectively examined a cohort of 259 patients enrolled in the Johns Hopkins AS program from 2010-2017 with a bMRI showing at least one prostate lesion with ADC value followed by surveillance biopsies (systematic and/or targeted). The median follow-up prior to bMRI was 17 months (IQR: 7 mos.-52 mos.) and median ADC of the study group was 900 x 10-6 mm2/s (IQR 700 - 1070). To evaluate the clinical significance of diffusion-weighted imaging, we used the lowest quartile ADC value of 700 as a cut-off to predict the risk of eventual detection of higher-grade cancer on biopsy. Using survival analysis, we assessed the association between ADC values and upgrading to grade group (GG) ≥ 2 [Gleason Score (GS)≥3+4] and the more clinically aggressive GG ≥ 3 (GS≥4+3) cancers.
Out of the 259 men, 74 (29%) and 24 (9%) upgraded to GG≥2 and GG≥3, respectively. Compared to non-upgraders, men who upgraded to GG ≥2 had significantly higher PSAD (median, 0.10 vs. 0.08, p = 0.001), higher proportion of men with low-risk cancer (54% vs. 35%, p = 0.005), and prostate lesions with lower ADC on bMRI (median, 805 vs. 930, p = 0.01). There was no significant difference in median follow-up prior to bMRI between groups (13 mos. vs. 19 mos. p = 0.72). 1, 2 and 5-year survival free of upgrading was lower for men with ADC<700 compared to men with ADC≥700 for both GG≥2 (67%, 54%, and 42% vs. 86%, 75%, and 57%, respectively) as well as GG≥3 (86%, 80%, and 68% vs. 97%, 92%, and 85%, respectively), both p<0.05. Additionally, controlling for PSAD and risk-status, men with ADC<700 still had elevated risk of upgrading to GG≥2 (HR = 1.9, 95% CI 1.1 - 3.1, p=0.01) and GG≥3 (HR = 2.5, 95% CI 1.1 - 5.8, p=0.04).</p>
For men in AS, lower ADC scores are associated with increased risk of upgrading to GG>1 and could be a useful component of risk prediction tools. MRI evaluation with ADC scores may reduce cost, invasiveness, and time spent under a magnetic field by potentially reducing the role of dynamic contrast-enhancement MRI.