Purpose: We evaluated whether a positive surgical margin at the apex confers a different likelihood of biochemical recurrence than at other sites.
Material and methods: A total of 3,087 men underwent radical prostatectomy between January 2000 and June 2008. Patients with prior treatment, positive seminal vesicles, lymph node involvement or less than 6 months of followup were excluded from analysis. The remaining 1,667 men were grouped by margin status, including negative surgical margins, a solitary positive apical margin, a solitary nonapical positive margin and multiple positive margins. Kaplan-Meier analysis was used to compare biochemical recurrence across groups. Cox proportional hazards models were constructed to determine whether a solitary positive apical margin is an independent risk factor for biochemical recurrence.
Results: Median followup was 21.1 months. Of the cases 1,295 (77.7%) had negative surgical margins, 96 (5.8%) had a solitary positive apical margin, 82 (4.9%) had a solitary positive nonapical margin and 194 (11.6%) had multiple positive margins. The likelihood of biochemical recurrence in solitary positive apical margin cases was between that of negative surgical margins and a solitary positive nonapical margin with multiple positive margins showing the highest likelihood of biochemical recurrence (p <0.001). Three-year likelihood of freedom from biochemical recurrence was 94.7% (95% CI 92.7-96.2), 87.0% (95% CI 74.1-93.7), 81.4% (95% CI 67.2-89.9) and 73.0% (95% CI 63.9-80.2) for negative surgical margins, a solitary positive apical margin, a solitary positive nonapical margin and multiple positive margins, respectively. In the multivariate model a solitary positive nonapical margin (2.17, 95% CI 1.17-4.03, p = 0.01) and multiple positive margins (3.03, 95% CI 1.99-4.61, p <0.001) were independent predictors of biochemical recurrence but a solitary positive apical margin was not (1.34, 95% CI 0.65-2.75, p = 0.43).
Conclusions: A solitary positive apical margin was associated with worse biochemical recurrence but on multivariate analysis it was not an independent predictor of recurrence. Models to predict biochemical recurrence after radical prostatectomy should account for differences in the prognostic significance of different positive margin sites.