Evaluation of lymphovascular invasion as a prognostic predictor of overall survival after radical prostatectomy

OBJECTIVE: To assess the prognostic ability of lymphovascular invasion (LVI) as a predictor of overall survival (OS).

MATERIALS AND METHODS: We included 126,682 prostate cancer (CaP) cM0 patients who underwent radical prostatectomy with lymph node dissection between 2010 and 2015, within the National Cancer Database. Patients who received androgen deprivation therapy were included. Patients were divided into four sub-cohorts based on LVI and lymph node invasion (LNI) status: pL0N0, pL1N0, pL0N1, and pL1N1. Kaplan-Meier curves estimated OS and Cox-regression analysis tested the relationship between LVI and OS.

RESULTS: Median (IQR) age and PSA at diagnosis were 62 (57-66) years and 5.7 (4.5-8.9) ng/ml, respectively. Most patients had pT2 stage (68.5%), and pathological Gleason 3+4 (46.7%). 10.0% and 4.0% patients had LVI and LNI, respectively. Median follow-up was 42 months (27-58). At 5-years, OS was 96.5% in pL0N0 patients vs 93.1% pL1N0 patients vs 93.3% in pL0N1 patients vs 86.6% pL1N1 patients. LVI was an independent predictor of OS (hazard ratio [HR]:1.28). LVI showed interaction with LNI, as LVI was associated with a higher overall-mortality in patients with LNI (HR:1.66), than in patients without LNI (HR:1.22). (all P<0.0001) CONCLUSIONS: Our report highlights the detrimental impact of LVI on OS. Patients with LVI alone fared similarly to patients with LNI alone. Patients with both LVI and LNI had worse OS than those with only LVI or LNI, implying a synergetic detrimental interaction. Our findings demonstrate an important utility that LVI can provide in deciding patients' prognoses.