Introduction

A risk stratification of N+ prostate cancer (PCa) patients according to tumor characteristics was recently proposed (Abdollah et al. JCO 2014). The aim of this study was to assess whether the rate of other cause mortality (OCM) differs in N+ PCa patients according to these 5 risk-groups, after accounting for the risk of cancer specific mortality (CSM). This would allow to improve patient stratification and to identify candidates for additional therapies

Methods

We evaluated 1,312 N+ PCa patients treated with radical prostatectomy and pelvic lymph node dissection at two tertiary referral centers between 1988 and 2014. Patients were stratified into the 5 established risk-groups: very low-risk (≤2 N+, and Gleason score ≤6); low-risk (≤2 N+, Gleason score 7-10, pT2/pT3a and negative surgical margins); intermediate-risk (≤2 N+, Gleason score 7-10 and pT3b/pT4 or SM+); high-risk (3-4 N+); very high-risk (>4 N+). Poisson smoothed cumulative incidence methods were used to assess 8-year OCM according to risk groups, after accounting for the risk of CSM. The same analyses were performed among men who experienced BCR after RP (n= 633, 48.2%)

Results

The median follow-up after RP was 82 months (IQR 37.1-147). During the study period, 18.1% of men died from other causes and 14.6% died from PCa. The leading cause of death at 8-year was OCM in the very low and low risk groups (12.6 and 8.3% vs. 2.5 and 7.6% for CSM, respectively). Conversely, CSM was the main cause of death in the remaining groups (Fig 1a). When the same analyses were repeated in men who had BCR after RP, CSM was the leading cause of death in all risk-groups except in very low-risk patients, where the 8-year OCM and CSM were similar (5.9 and 5.8%, respectively; Fig. 1b).

Conclusions

PCa is not invariably the main cause of death in all N+ patients treated with curative intent. Patients with less aggressive disease are more likely to die from other causes. Conversely, when N+ patients recur, they will likely succumb from PCa rather than from other causes, regardless of tumor aggressiveness. These results could help physicians sparing unnecessary treatments in N+ men with lower likelihood of dying from PCa and to plan timely salvage treatments in virtually all N+ men who recur

Funding

none