Prognostic Impact of Subclassification of Radical Prostatectomy Positive Margins by Linear Extent and Gleason Grade

Abstract

We evaluated the relationship of progression to positive surgical margin linear length and Gleason grade at a positive surgical margin. We studied 2,150 prostatectomies done for pT2 or pT3a disease to determine grade, stage and surgical margin status. In patients with positive surgical margins we recorded the location, number, positive margin linear length and highest Gleason grade at a positive margin. The Kaplan-Meier method and log rank test were used to determine differences in progression-free probability among positive margin features. The concordance index was used to discriminate the accuracy of grouping surgical margin status as negative/positive vs positive margin linear length/highest Gleason grade. A total of 207 cases (10%) showed positive surgical margins, including 93 (45%) that were pT2+ and 114 (55%) that were pT3a. Patients with pT3a and positive margins had greater prostate specific antigen and tumor volume, and Gleason score 7 or greater than those with pT2+. A total of 45 patients with positive margins progressed. We then subcategorized positive margins. Of the patients 164 (79%) had 1 positive margin. Positive margin linear length was 1 mm or less, 1.1 to 3 and greater than 3 in 104 (50%), 55 (27%) and 48 cases (23%), respectively. Two-year progression-free probability was 95%, 91%, 83% and 47% in patients with negative margins and the 3 positive margin linear length groups, respectively (p <0.001). Gleason grade at a positive margin was 3 and 4/5 in 154 (74%) and 53 patients (26%), respectively. The latter group was significantly more likely to progress (p <0.001). The overall margin status concordance index was 0.636. It was not considerably enhanced by categorizing by positive surgical margin linear length/highest Gleason grade at positive margins. The linear extent of and highest Gleason grade at a positive surgical margin are associated with progression. However, subcategorization does not importantly add to predictive models using margin status only. More robust markers are needed in patients with positive surgical margins to warrant routine reporting and identify those at risk for biochemical recurrence.