Biochemical recurrence after radical prostatectomy: multiplicative interaction between surgical margin status and pathological stage

A positive surgical margin after radical prostatectomy is considered an adverse prognostic feature. However, few groups have explored the potential interaction between surgical margin status and other cancer characteristics, specifically pathological stage. We addressed the first degree of interaction between positive surgical margins and other established adverse predictors of biochemical recurrence after radical prostatectomy.

We used univariate and multivariate analysis to test the effect of surgical margin status on biochemical recurrence in 4,490 patients treated at a single institution between 1992 and 2008. We systematically tested all first-degree interactions between surgical margin status, and pretreatment prostate specific antigen, pT and pN stage, and radical prostatectomy Gleason sum. If interactions were significant, we quantified the effect on the biochemical recurrence rate.

Overall 850 patients (18.9%) had positive surgical margins. In those with negative vs positive surgical margins the 5-year biochemical recurrence-free survival rate was 95% vs 83%, 74% vs 62% and 47% vs 29% for pT2, pT3a and pT3b disease, respectively. In multivariate models only the pT stage-surgical margin status interaction achieved independent predictor status (p = 0.003). Negative vs positive surgical margin multivariate HRs were 1 vs 2.9, 2.3 vs 4.3 and 4.1 vs 5.6 in pT2, pT3a and pT3b cases, respectively.

Compared to negative surgical margins, positive surgical margins increase the absolute biochemical recurrence 5-year rate by 12% to 18%. More importantly, positive surgical margins may substantially worsen the prognosis beyond that of the original pathological disease stage.