Assessment of Tumor Volume by Frozen Section Examination of the Radical Prostatectomy Specimen: Correlation with Long-Term Follow-Up.
Rafael E Jimenez, John C Cheville, R Jeffrey Karnes, Jodi L Hilderbrand, Christine M Lohse, Thomas J Sebo. Mayo Clinic, Rochester, MN
Background: While tumor volume (TV) in radical prostatectomy (RP) specimens (RPS) is considered a significant prognostic factor in prostate cancer (PCa), few studies have included enough follow-up (f/u) to evaluate its long-term impact on prognosis. We describe our experience in assessing this parameter on frozen section (FS) examination on a cohort of PCa patients (pts) followed for over a decade.
Design: 451 consecutive RPS from 1995-98 with no neoadjuvant therapy were evaluated at time of surgery by a standardized, limited sampling protocol using FS technique, followed by reevaluation next day with permanent sections (sct). Average number of sct evaluated per RPS was 14 (range 13-63). Crude estimates of TV were calculated after correlation of microscopic exam results with gross locations where sct were taken. F/u data was obtained during routine clinical f/u and/or via questionnaires to pts no longer followed at our institution.
Results: Mean time from RP to last f/u was 11.3 years (y) (median 12.5; range 0.1-15.1). Mean and median TV was 6.69 and 2.6 cc, respectively (range 0.0005-131.25). 49% of tumors had a Gleason score (GS) of <6, 77% were organ confined, and 59% had negative surgical margins (mgn). 90.4% of cases were signed-out the same of surgery. 7.7% were amended/addended to correct/supplement information rendered at time of FS. 135 pts experienced cancer progression at a mean of 4.3 y after RP. Progression-free survival (PFS) rates at 5 and 10 y were 80% and 69%, respectively. 18 of 440 pts died from PCa at a mean of 6.9 y after RP. Cancer-specific survival (CSS) rates at 5 and 10 y after RP were 99% and 97%. GS, mgn status and pathologic stage correlated with PFS, while all of them but mgn status correlated with CSS. TV correlated with PFS and CSS.
C indexes for TV were 0.675 and 0.802 for PFS and CSS, respectively, compared to those of GS (0.698 and 0.711) and stage (0.639 and 0.775).
Conclusions: TV, as estimated by a combination of careful gross assessment and microscopic examination at time of FS, is a strong predictor of both PFS and CSS and, in our study actually outperforms pathologic staging categories. FS examination of the RPS is highly efficient, as it provides accurate diagnostic and prognostic information with excellent turnaround time.
Category: Genitourinary (including renal tumors)
Monday, February 28, 2011 1:00 PM
Poster Session II # 114, Monday Afternoon