Prognostic significance of a positive surgical margin in pathologically organ-confined prostate cancer.

OBJECTIVE

To identify risk factors of a positive surgical margin (PSM) and the significance of a PSM after radical prostatectomy (RP) on biochemical recurrence (BCR) in exclusively pathologically confirmed organ-confined (OC) prostate cancer, as despite an excellent prognosis after RP, some patients with pathologically confirmed OC disease have BCR, and the prognostic significance of a PSM in these men remains unclear.

PATIENTS AND METHODS

We assessed 932 men with pathologically OC disease who were treated with RP by nine different surgeons between 1992 and 2004. The prognostic significance of clinical and pathological variables, including tumour volume (TV) and percentage of high-grade TV (%HGTV) were assessed. Logistic and Cox regression models were fitted to identify risk factors of a PSM and BCR. BCR was defined as a prostate-specific antigen (PSA) level of 0.1 ng/mL and increasing after an undetectable PSA level.

RESULTS

The total PSM rate was 12.9% (120 men); the mean TV (P < 0.001), but not %HGTV (P= 0.2) was significantly higher in patients with PSM. TV, nerve-sparing RP technique and surgical volume were independent risk factors for a PSM (P= 0.03). After a median follow-up of 35 months the overall BCR rate was 8.8% (82 men). Patients with a PSM had significantly higher BCR rates (21.7% vs 6.9%; P < 0.001). In univariable analysis, a high %HGTV (70.4%) was the most informative risk factor of BCR, followed by RP Gleason score (65.8%) and PSM (65.7%). Removal of PSM from a multivariable Cox model decreased the accuracy by 12.1% (P < 0.001).

CONCLUSIONS

Our findings show that in OC prostate cancer, the risk of a PSM depends on TV, surgical technique and surgical volume. PSM is a significant risk factor for BCR. However, only 20% men with OC disease and a PSM develop BCR; conversely, 80% of men are cured despite a PSM. Therefore, adjuvant therapy must be considered, with caution to avoid unnecessary overtreatment.