Numerous studies have examined the association between positive surgical margins and treatment outcomes after radical prostatectomy. Some have suggested that positive margins do not necessarily portend a worse prognosis; others have shown that patients with positive margins are significantly more likely to experience postoperative biochemical progression, clinical progression, and death.OBJECTIVE
This study examined the relationship between positive surgical margins and treatment outcomes following radical prostatectomy.DESIGN AND INTERVENTION
This was a review of a large, international data set from consecutive patients who underwent radical prostatectomy at eight institutions between 1983 and 2000. Those who received preoperative androgen-deprivation therapy or adjuvant radiation therapy were excluded. Positive surgical margins were defined as the presence of tumor cells at the inked margin of resection; biochemical progression was defined by each individual center. Univariate and multivariate Cox regression analyses were performed to test whether PSA level, pathologic Gleason score, extracapsular tumor extension, seminal-vesicle invasion, lymph-node metastasis, or surgical-margin status predicted biochemical progression after radical prostatectomy. In order to investigate interactions between pathologic variables, a separate model was used to determine the combined effect of surgical-margin status and each pathologic variable on biochemical progression.OUTCOME MEASURES
The main endpoints were actuarial 5-year and 10-year progression-free survival rates.RESULTS
Among the 5,831 patients comprising the study population, the mean initial PSA level was 9.98 ng/ml (range 0-182 ng/ml). A positive surgical margin was found in 1,554 specimens (26.7%). Over a median follow-up period of 25 months (range 0-166 months), 955 patients (16.4%) had biochemical progression. Progression-free survival rates at 5 and 10 years were 83.8% and 70.1%, respectively, for men with negative surgical margins, compared with 53.1% and 36.1%, respectively, for men with positive surgical margins. On multivariate analysis, all investigated variables, including the presence of positive surgical margins, were independent predictors of biochemical progression (all P<0.001). In the separate multivariate analysis that included interaction terms between positive surgical margins and other adverse pathologic features, positive margins were associated with a 3.7-fold increased risk of biochemical recurrence. The combined effect of positive surgical margins and a Gleason sum of 7-10 (P=0.008), and of positive surgical margins and lymph-node metastasis (P<0.001), exceeded the sum of the risks associated with each individual variable. No such synergy was observed when the presence of positive surgical margins was combined with extracapsular extension or seminal-vesicle invasion.CONCLUSION
Positive surgical margins were independently associated with an increased risk of biochemical progression following radical prostatectomy. This risk is even greater in patients with a combination of adverse pathologic variables.