Utility of prostate multiparametric MRI (mpMRI) for recurrent prostate cancer after radiation therapy and cryotherapy.

Abstract

71 Background: We evaluated the accuracy of mpMRI for identifying locally recurrent prostate cancer after primary radiotherapy and cryotherapy. Methods: Between 2009-2015, 61 patients with evidence of rising PSA after external-beam radiotherapy (EBRT) (N = 33), brachytherapy (N = 6), and cryotherapy (N = 22) were evaluated for locally recurrent prostate cancer with mpMRI and prostate biopsy. Of these patients, 6 (10%) received androgen deprivation therapy (ADT) in combination with EBRT for a median of 24 months. Three of the cryotherapy patients received prior EBRT. Patients were identified from a prospective mpMRI database. All patients with a lesion of interest (LOI) underwent a ≥ 12-core, post-mpMRI cognitive fusion prostate biopsy. We excluded 16 patients with mpMRI who did not undergo prostate biopsy (5 positive, 11 negative). Results: Median age was 70 (IQR: 64-77). The median time from primary treatment to mpMRI was 5 years (IQR: 3-9) and the median PSA at mpMRI was 3.6 ng/mL (IQR: 2.1-5.5). Median prostate volume was 18.8 cc (IQR: 11.0-28.0 cc). mpMRI revealed lesions of interest (LOI) in 39 (64%) and 41 (67%) had biopsy-proven local recurrence. Of the 22 patients with negative mpMRI, 8 (36%) had a positive biopsy, with a median prostate volume of 19 cc, median maximum cancer length of 5 mm, median PSA of 2.5 and biopsy Gleason scores 3+3 (N = 1), 4+3 (N=2), 5+4 (N = 1), 5+5 (N = 1) and ungraded due to treatment effect (N = 3). Of the 39 patients with LOI on mpMRI, 33 (85%) had a positive biopsy. Table 1 summarizes the mpMRI and biopsy results. The sensitivity, specificity, PPV and NPV of mpMRI to predict cancer diagnosis at biopsy was 80.5%, 70.0%, 84.6% and 63.6%, respectively. On univariate analysis, gland size (p=0.367), PSA (p=0.872), biopsy Gleason score (p=0.892) and primary treatment modality (p=0.177) did not significantly predict discrepancy between mpMRI and biopsy findings. Conclusions: mpMRI reliably identifies prostate cancer recurrence after primary radiation therapy and cryoablation. [Table: see text]