Risk Stratification of pN+ Prostate Cancer after Radical Prostatectomy from a Large Single Institutional Series with Long-Term Followup.

PURPOSE

Lymph node positive (pN+) prostate cancer after radical prostatectomy has wide variability in long-term oncologic outcomes. We present a large institutional series with extended followup to create an oncologic risk stratification system that clarifies the prognostic heterogeneity for patients with pN+ disease after radical prostatectomy.

MATERIALS AND METHODS

Men with pN+ prostate cancer after radical prostatectomy during 1987 to 2012 were included in the study. Regression models were created to identify significant predictors of biochemical recurrence, metastasis, cancer specific mortality and overall mortality. A cancer specific mortality risk score was then created and internally validated to stratify patients in terms of risk of cancer specific mortality.

RESULTS

For our cohort of 1,011 men with a median followup of 17.6 years the 20-year rate of cancer specific mortality was 31%. On multivariate Cox regression modeling 3 or more positive nodes (HR 1.75, p=0.003), pathological Gleason score 7 vs 6 (HR 1.74, p=0.04) and 8-10 vs 6 (HR 2.63, p=0.001), and positive surgical margins (HR 1.96, p=0.001) were significantly associated with increased cancer specific mortality, while adjuvant radiotherapy (HR 0.40, p=0.008) was associated with decreased cancer specific mortality. A cancer specific mortality risk score was then created using these 4 variables to stratify patients with markedly different prognoses, yielding 20-year cancer specific mortality rates of 19.1% vs 34% vs 46% (p <0.001) for low, intermediate and high risk categories, respectively.

CONCLUSIONS

The prognosis of patients with pN+ prostate cancer varied significantly after radical prostatectomy. A risk score created using the number of positive nodes, pathological Gleason score, margin status and adjuvant radiotherapy status successfully separated patients into low, intermediate and high risk groups.