Prognostic value of perinodal lymphovascular invasion following radical cystectomy for lymph node-positive urothelial carcinoma

Eur Urol. 2013 Apr;63(4):739-44. doi: 10.1016/j.eururo.2012.09.053. Epub 2012 Sep 28.

Abstract

Background: Metastasis of urothelial carcinoma of the bladder (UCB) into regional lymph nodes (LNs) is a key prognosticator for cancer-specific survival (CSS) after radical cystectomy (RC). Perinodal lymphovascular invasion (pnLVI) has not yet been defined.

Objective: To assess the prognostic value of histopathologic prognostic factors, especially pnLVI, on survival.

Design, setting, and participants: A total of 598 patients were included in a prospective multicentre study after RC for UCB without distant metastasis and neoadjuvant and/or adjuvant chemotherapy. En bloc resection and histopathologic evaluation of regional LNs were performed based on a prospective protocol. The final study group comprised 158 patients with positive LNs (26.4%).

Intervention: Histopathologic analysis was performed based on prospectively defined morphologic criteria of LN metastases.

Outcome measurements and statistical analysis: Multivariable Cox proportional hazard regression models determined prognostic impact of clinical and histopathologic variables (age, gender, tumour stage, surgical margin status, pN, diameter of LN metastasis, LN density [LND], extranodal extension [ENE], pnLVI) on CSS. The median follow-up was 20 mo (interquartile range: 11-38).

Results and limitations: Thirty-one percent of patients were staged pN1, and 69% were staged pN2/3. ENE and pnLVI was present in 52% and 39%, respectively. CSS rates after 1 yr, 3 yr, and 5 yr were 77%, 44%, and 27%, respectively. Five-year CSS rates in patients with and without pnLVI were 16% and 34% (p<0.001), respectively. PN stage, maximum diameter of LN metastasis, LND, and ENE had no independent influence on CSS. In the multivariable Cox model, the only parameters that were significant for CSS were pnLVI (hazard ratio: 2.47; p=0.003) and pT stage. However, pnLVI demonstrated only a minimal gain in predictive accuracy (0.1%; p=0.856), and the incremental accuracy of prediction is of uncertain clinical value.

Conclusions: We present the first explorative study on the prognostic impact of pnLVI. In contrast to other parameters that show the extent of LN metastasis, pnLVI is an independent prognosticator for CSS.

MeSH terms

  • Aged
  • Cystectomy / methods*
  • Disease-Free Survival
  • Female
  • Follow-Up Studies
  • Humans
  • Lymphatic Metastasis
  • Lymphatic Vessels / pathology*
  • Male
  • Middle Aged
  • Neoplasm Invasiveness
  • Neoplasm Recurrence, Local / pathology*
  • Neoplasm Recurrence, Local / surgery
  • Prognosis
  • Prospective Studies
  • Survival Rate
  • Treatment Outcome
  • Urinary Bladder Neoplasms / pathology*
  • Urinary Bladder Neoplasms / surgery
  • Urothelium / pathology*