Correlation of margin status and extraprostatic extension with progression of prostate carcinoma.

BACKGROUND: The correlation of surgical margins and extraprostatic extension (EPE) with progression is uncertain with regard to prostate carcinoma patients treated by radical prostatectomy. The objective of this study was to define factors predictive of cancer progression; emphasis was placed on surgical margins and their relation to extraprostatic extension. METHODS: The study group consisted of 377 patients who were treated by radical retropubic prostatectomy and bilateral pelvic lymphadenectomy at the Mayo Clinic between 1986 and 1993. All specimens were totally embedded and whole-mounted. Patients ranged in age from 41 to 79 years (mean, 65 years). Those with seminal vesicle invasion or lymph node metastasis and those treated preoperatively with radiation or androgen deprivation were excluded. Final pathologic T classifications were pT2a (41 patients), pT2b (237), and pT3a (99). Progression was defined as biochemical failure (prostate specific antigen [PSA] >0.2 ng/mL), clinical or biopsy-proven local recurrence, or distant metastasis. The mean follow-up was 5.8 years (range, 0.2-11.4 years). Seventy-nine patients who received adjuvant treatment within 3 months after surgery were excluded from survival analysis. RESULTS: The overall margin positivity rate was 29%. Seventy-two patients (19%) had only positive surgical margins without evidence of EPE ("surgical incision"), 53 (14%) had only EPE, 37 (10%) had both, and 215 (57%) had neither. Positive margins were correlated with the finding of EPE (P = 0.003). Progression free survival rates at 5 and 10 years were 88% and 67%, respectively. In univariate analysis, preoperative PSA concentration, positive surgical margins, Gleason grade, cancer volume, and DNA ploidy were significant in predicting progression (P values, <0.001, <0.001, 0.01, 0.007, and <0.001, respectively). In multivariate analysis, margin status and DNA ploidy were independent predictors of progression (relative risk for margin status, 1.9; 95% confidence interval [CI], 1.1-3.4; P = 0.03; relative risk for DNA ploidy, 5.1; 95% CI, 2.4-10.9; P<0.001). Among patients with positive margins, 5-year progression free survival was 78% for those with negative EPE and 55% for those with positive EPE. CONCLUSIONS: Surgical margin status and DNA ploidy were independent predictors of progression after radical prostatectomy. To improve cancer control, adjuvant therapy may be considered for patients with positive surgical margins or nondiploid cancer.