MP20-05: Clinical significance of grade 5 lesions on MRI

Clinical significance of grade 5 lesions on MRI

View Poster

INTRODUCTION

Grade 5 Regions of Interest (ROI) elicit the highest level of suspicion for prostate cancer (CaP) on mpMRI. Because of the likelihood of cancer, treatment on the basis of a Grade 5 ROI alone, without biopsy confirmation, has been practiced by some. To evaluate this practice, we analyzed tissue findings in men undergoing targeted biopsy of a Grade 5 ROI.

METHODS

Between 2010-2017, 1825 men underwent MRI/US fusion biopsies. 255 men (14%) had grade 5 MRI ROIs (PI-RADS or UCLA scoring); 2/3 of all ROI were in the peripheral zone, 1/3 in the transition zone. MRI was 3T multi-parametric and biopsy was via MRI-US fusion (Cancer 122:884, 2016). At least 3 cores were obtained from each ROI, and systematic cores were also obtained. All cores were analyzed for Gleason score and maximum cancer core length. When immediate treatment was not performed, follow-up targeted biopsies were obtained.

RESULTS

Mean age was 66.7 (SD 7.2); 76% were Caucasian. Median PSA was 8.3 ng/ml (IQR 5.9, 14.7), prostate volume was 41 cc (IQR 32.0, 59.0). At first biopsy, cancer detection rate was 93% for all cancers GS≥6, N=238/255; maximum cancer core length was median 8.0 mm (IQR 4.5, 11.0). In 47% (N=120), Gleason Score was ≤3+4 (Figure); in these men a later re-biopsy was possible. Of the 56 who had repeat targeted biopsy once or more (6-42 months after initial biopsy), 48% had a Gleason upgrade (Table).

CONCLUSION

93% of Grade 5 Regions of Interest on mpMRI were found to harbor cancer, using targeted biopsy. However, nearly half the cancers found in Grade 5 ROIs were potential candidates for Active Surveillance, rather than immediate treatment. Repeat targeted biopsy of Grade 5 ROIs may identify the men who harbor a tumor of higher grade than initially diagnosed. Curative intervention based on MRI findings alone, without biopsy confirmation, would potentially result in over-treatment in up to 47% of men.

Funding: NIH (R01 158627), Jean Perkins Foundation, and UCLA CTSI (UL1TR000124)