Introduction

Available recommendations for extended lymph node dissection [eLND] at radical prostatectomy [RP] do not consider patient age, but rely on cancer characteristics only. However, for patients with limited life-expectancy, eLND might be an overtreatment. We hypothesized that limited life-expectancy of older RP candidates might dilute any beneficial effect of eLND in terms of cancer staging and outcomes. Therefore, we aimed at assessing the differential effect of age on the risk of lymph node invasion [LNI] and mortality due to cause other than prostate cancer [OCM] in order to define an age limit above which eLND might be avoided.

Methods

We included 3,906 patients diagnosed with prostate cancer and treated with RP and an anatomically defined eLND at a single Institution. Logistic and Cox regression analyses were used to compute the risk of LNI at eLND and the risk of OCM 10 years after RP. Predictors of LNI were chosen in compliance with guidelines-recommended models and were PSA, primary and secondary biopsy Gleason score and clinical stage. Predictors of OCM were age at surgery, Charlson comorbidity index [CCI] and year of surgery. Locally weighted scatterplot smoothing method was used to graphically examine the differential effect of age on the risk of LNI and OCM.

Results

Median age was 65 years. LNI rate was 12%. 10-year OCM rate was 5%. PSA (odd ratio [OR] 1.06; p<0.001), primary (OR 5.32; p<0.001) and secondary (OR 2.27; p<0.001) biopsy Gleason score ≥4 as well as clinical stage cT2 (OR 2.4; p<0.001) and cT3 (OR 3.24; p<0.001) were associated with higher LNI risk. Age (hazard ratio [HR] 1.11; p<0.001) and CCI (HR 1.28; p=0.03) were associated with higher OCM risk. Year of surgery (HR 0.92; p<0.001) was associated with lower OCM risk. For patients aged 75 or younger, the risk of LNI (8-11%) was higher than the risk of OCM (<1-10%; Figure 1). Conversely, for patients aged 76 or older the risk of LNI (11-17%) was equal or lower than the risk of OCM (11-26%).

Conclusions

For RP candidates older than 75 years, the risk of OCM equals or exceeds the risk of LNI. Such relatively high risk of OCM compared to the relatively low risk of LNI casts relevant doubts on any potential benefit related to eLND at RP. These findings argue against the routine use of eLND for older patients in clinical practice.

Funding

none