Introduction

An extended pelvic lymph node dissection (ePLND) should be considered in prostate cancer (PCa) patients undergoing radical prostatectomy (RP) at increased risk of lymph node invasion (LNI). However, whether the removal of presacral and common iliac nodes should be included is still a matter of debate.

Methods

4,790 PCa patients treated with RP and an anatomically defined ePLND between 2003 and 2016 at a single center were identified. The risk of LNI was calculated according to the Briganti nomogram. The number of positive nodes and the proportion of patients with positive nodes in the common iliac and presacral stations was plotted over the calculated risk of LNI according to the Briganti nomogram using lowess-smoothed fit curve. Patients were stratified according to the risk of LNI in 3 categories: <10%, 10-30%, and >30%. The number of positive nodes and the proportion of patients with positive nodes in the common iliac and presacral stations were compared. Multivariable analyses tested the association between the calculated risk of LNI and involvement of common iliac and presacral nodes after adjusting for the number of nodes removed and pathologic Gleason score.

Results

The risk of LNI according to the Briganti nomogram was <10, 10-30, and >30% in 4,569 (95.4%), 115 (2.4%), and 106 (2.2%) patients. The median number of nodes removed was 16 and 487 (10.2%) patients had LNI. A total of 483 (99.2%) patients had LNI located to the internal iliac, external iliac, or obturator nodes, while 40 (8.2%) and 22 (4.5%) patients had positive nodes in the common iliac and presacral stations. None of the patients with positive common iliac nodes had negative pelvic nodes. The number of positive nodes increased according to the risk of LNI (P<0.001). The median number of positive nodes was 1 vs. 3 vs. 5 in patients with a risk <10, 10-30, and >30% (P<0.001). The proportion of involvement of the presacral and common iliac nodes increased according to the risk of LNI and was 0.1 vs. 5.2 vs. 10.4 and 0.3 vs. 5.0 vs. 18.9 for individuals with a risk of LNI <10, 10-30, and >30% (all P<0.001). The risk of LNI was an independent predictor of the number of positive nodes and of involvement of the presacral and common iliac nodes (all P<0.001).

Conclusions

The presacral and common iliac lymph nodes are involved by nodal metastases in more than 10% of patients with a calculated risk of LNI >30%. An ePLND that includes the dissection of the presacral and common iliac nodes should then be considered in men with a risk of LNI >30%.

Funding

none