Review article
A systematic review and meta-analysis of lymphovascular invasion in patients treated with radical cystectomy for bladder cancer

https://doi.org/10.1016/j.urolonc.2018.03.018Get rights and content

Highlights

  • LVI is an important step in BCa cell dissemination.

  • Overall, LVI increases the risk of disease recurrence and CSM after RC.

  • In node-negative patients LVI doubles the risk of disease recurrence and CSM.

  • Even in N+ patients LVI is a significant predictor of disease recurrence and CSM.

Abstract

Purpose

Lymphovascular invasion (LVI) is an important step in bladder cancer cell dissemination. We aimed to perform a systematic review and meta-analysis of the literature to assess the prognostic value of LVI in radical cystectomy (RC) specimens.

Patients and methods

A systematic review and meta-analysis of the last 10 years was performed using the MEDLINE, EMBASE, and the Cochrane libraries in July 2017. The analyses were performed in accordance with the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) statement.

Results

We retrieved 65 studies (including 78,107 patients) evaluating the effect of LVI on oncologic outcomes in patients treated with RC. LVI was reported in 35.4% of patients. LVI was associated with disease recurrence (pooled hazard ratio [HR] = 1.57; 95% CI: 1.45–1.70) and cancer-specific mortality (CSM) (pooled HR = 1.59; 95% CI: 1.48–1.73) in all studies regardless of tumor stage and node status (pT1–4 pN0–2). LVI was associated with recurrence and CSM in patients with node-negative bladder cancer (BC). In patients with node-negative BC, LVI rate increased and was associated with worse oncologic outcome. LVI had a lower but still significant association with disease recurrence and CSM in node-positive BC.

Conclusions

LVI is a strong prognostic factor of worse prognosis in patients treated with RC for bladder cancer. This association is strongest in node-negative BC, but it is also in node-positive BC. LVI should be part of all pathological reporting and could provide additional information for treatment-decision making regarding adjuvant therapy after RC.

Introduction

Radical cystectomy (RC) and pelvic lymph node (LN) dissection is the standard treatment in patients with muscle-invasive bladder cancer (MIBC) and in patients with very high risk non-MIBC disease [1], [2]. Despite adequate surgery, up to 50% patients experience disease recurrence and mortality after RC [3]. Randomized clinical trials failed to demonstrate a benefit to adjuvant chemotherapy (AC). However, most guidelines recommend a discussion regarding AC for patients with pT3–4 or node-positive disease who are able to tolerate cisplatin-based chemotherapy and did not receive neoadjuvant cisplatin-based chemotherapy. Due of challenging adjuvant systemic therapy is to identify the patients who are likely to benefit from it and spare those who are not the side effects from an unnecessary therapy. Current prognostic tools based on TNM staging, however, did not reach sufficient accuracy to change our clinical decision making. Inclusion of easily available pathologic features that are associated with the biological aggressiveness of BC may facilitate clinical decision making regarding adjuvant systemic therapy, patient counseling, and follow up intensity.

Lymphovascular invasion (LVI) is an important step in cancer cell dissemination [4], [5], [6], [7]. In BC, the detection of LVI in the RC specimens has been shown to be associated with adverse outcomes in several large retrospective studies. In 2014, a meta-analysis by Kim et al. [8] examined 21 studies and confirmed the significant prognostic role of LVI in RC specimens despite a certain degree of asymmetry between studies. In the recent years, numerous papers added new evidence to the topic [9], [10], [11], [12], [13], [14]. These studies add information regarding the prognostic role of LVI in specific subcohorts based on stage and perioperative treatment. We performed a systematic review and meta-analysis to assess the prognostic value of LVI in RC specimens of BC patients focusing on subgroups such as organ confined BC.

A systematic review and meta-analysis of the English-language literature was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and the Cochrane Handbook for Systematic Reviews of Interventions [15], [16]. We systematically searched MEDLINE, EMBASE, and the Cochrane Library to identify studies published between January 2007 and July 2017 (date last search: 30/Jul/2017) that examined the effect of LVI on oncologic outcomes in patients treated with RC for BC. After a first screening based on study title and abstract, all papers were assessed based on full text and excluded with reasons when appropriate. Two reviewers (A.M. and S.K.) carried out this process independently. Disagreement was solved by a third party (B.F.). The following string terms were used: (((“bladder”) AND (“cancer” OR “urothelial carcinoma” OR “urothelial neoplasm” OR “carcinoma” OR “transitional cell carcinoma”)) AND (“radical cystectomy” OR “cystectomy”)) AND (“LVI” OR “lymphovascular invasion” OR “lymphatic invasion” OR “vascular invasion”). Disease recurrence and cancer-specific mortality (CSM) were the primary endpoints of interest.

As proposed by the PRISMA guidelines, we used the Population, Intervention, Comparator, Outcome, and Study design approach to specify the inclusion criteria. Reports were considered relevant when included patients diagnosed with BC (P), recorded LVI (C), and underwent RC (I) to independently determine the prognostic value of LVI on disease recurrence and CSM (O) using multivariable Cox proportional hazard regression analyses (S). If more than 1 report of the same cohort of patients existed, we selected the most recent regarding a specific survival outcome. Review articles, editorials, comments, and meeting abstracts were excluded. References of included manuscripts were scanned for additional studies of interest. The PRISMA 2009 checklist is reported in Supplementary Table 1.

After full text evaluation, data was independently extracted by 2 authors (A.M. and S.K.) for further assessment of qualitative and quantitative analyses. All extracted variables were crosschecked to ensure their reliability. We recorded the baseline characteristics, of the included participants, the use of chemotherapy and the median/mean follow-up duration. Subsequently, the HRs and 95% CIs of LVI associated with each outcome were retrieved. Furthermore, we searched for methods and important confounders to establish comparability. The quality of the studies was assessed using the Newcastle-Ottawa Scale (NOS) for cohort studies [17]. The NOS assesses the quality of studies using a star system based on 3 domains: selection of cohorts, comparability of cohorts, assessment of outcomes. The NOS ranges from 0 to 9. A threshold ≥7 was identified to distinguish studies with higher quality. All discrepancies regarding data extraction were generally resolved by consensus or finally decided by the senior author (S.F.S.).

Due to the observational nature of included studies, we extracted adjusted HR and 95% confidence interval for cumulative effect size calculation. Studies with Kaplan-Meier log-rank, univariable Cox proportional hazard regression, or general logistic regression analyses were not considered for meta-analysis. Effect summary estimation methods were not used in these cases; as high level of additional selection bias would have been introduced. Statistical pooling of effect measures was based on the level of heterogeneity among studies, which was assessed with the Cochrane Q test and I2 statistic. Significant heterogeneity was indicated by a P<0.05 in Cochrane Q tests and a ratio>50% in I2 statistics, which led to the use of random-effect models according to the DerSimonian and Laird method [18], [19], [20]. When these tests were negative for heterogeneity, fixed-effect models were chosen for calculation of pooled hazard ratios through the inverse-variance method. Publication bias including small-study effect were evaluated by visual inspection of funnel plots for all assessed comparisons. Statistical analyses were performed using STATA/MP 14.2 (Stata Corp., College Station, TX).

The “risk-of-bias” (RoB) evaluation of included trials was assessed according to the Cochrane Handbook for Systematic Reviews of Interventions for including nonrandomized studies [21], [22]. Due to only nonrandomized comparative studies, RoB was determined the pragmatic approach by examining the risk of preassigned confounders. The main confounding factors were identified as the most important prognostic factors of disease recurrence and CSM after RC. We therefore reviewed the articles adjusting for the effects of age, pathologic tumor and node stage, tumor grade, number of positive nodes, and of overall nodes removed, NAC and AC within their used Cox models. The presence of confounders was determined by consensus among A.M. and B.F. The RoB summary and graph were built using the Cochrane Review Manager 5.3.

Firstly, all studies examining the prognostic role of LVI with multivariable Cox regression analyses that adjusted for the effects of tumor and nodal status (pT1–4 and pN0–2), and perioperative chemotherapy used were considered for the meta-analysis. Therefore, 2 additional meta-analyses on the node-negative (pT1–4 and pN0) and node-positive (pT1–4 and pN1–2) stage population examining the effect of LVI on multivariable Cox regression analyses after adjusting for the effect of AC and with the previous exclusion of patients treated with neoAC (NAC).

Section snippets

Study selection

Overall, we identified 760 articles from the search query and 7 from the reference lists. Of these, 145 duplicates were removed. Further 423 articles were excluded after title and abstract screening and 154 after full-text evaluation. The remaining 68 articles were included in the systematic review. The flow-chart of study selection process is showed in Fig. 1.

Study population

Studies characteristics are reported in Table 1. Overall, the 63 studies enrolled 76,488 patients treated with RC. The population

Discussion

MIBC is a debilitating disease with a high rate of disease recurrence and eventual death despite seemingly optimal surgical therapy. The use of adjuvant therapy after RC and LN dissection based on final pathology is a sensitive approach to treat patients who are at an increased risk of failure. In BC, AC is indicated in patients with NOC disease or LN involvement who did not undergo cisplatin-based NAC [85]. However, only 10.7% to 11.6% of RC patients receive AC because of various reasons

Conclusions

This meta-analysis demonstrates that LVI in the RC specimen is a significant prognostic factor for disease recurrence and CSM in patients with BC. LVI has a strong effect on outcomes in patients with node-negative disease. In these patients, an inverse relation between the tumor stage and the association between LVI and worse oncologic outcome was seen. LVI has a lower, but still significant, association with disease recurrence and CSM in patients with node-positive disease. This readily

Referencs (92)

  • K. Turkolmez et al.

    Muscle-invasive bladder cancer: predictive factors and prognostic difference between primary and progressive tumors

    Urology

    (2007)
  • S.F. Shariat et al.

    Combination of multiple molecular markers can improve prognostication in patients with locally advanced and lymph node positive bladder cancer

    J Urol

    (2010)
  • A.J. Stephenson et al.

    Aggregate lymph node metastasis diameter and survival after radical cystectomy for invasive bladder cancer

    Urology

    (2010)
  • M. Zhang et al.

    Lymphovascular invasion and the presence of more than three tumors are associated with poor outcomes of muscle-invasive bladder cancer after bladder-conserving therapies

    Urology

    (2010)
  • G. Sonpavde et al.

    Prognostic risk stratification of pathological stage T3N0 bladder cancer after radical cystectomy

    J Urol

    (2011)
  • E. Park et al.

    Prediction of prognosis after radical cystectomy for pathologic node-negative bladder cancer

    Int Urol Nephrol

    (2011)
  • A. Buchner et al.

    Prediction of outcome in patients with urothelial carcinoma of the bladder following radical cystectomy using artificial neural networks

    Eur J Surg Oncol

    (2013)
  • R.D. da Silva et al.

    Impact of statin use on oncologic outcomes in patients with urothelial carcinoma of the bladder treated with radical cystectomy

    J Urol

    (2013)
  • H. Fajkovic et al.

    Extranodal extension is a powerful prognostic factor in bladder cancer patients with lymph node metastasis

    Eur Urol

    (2013)
  • Y. Lotan et al.

    Prospective evaluation of a molecular marker panel for prediction of recurrence and cancer-specific survival after radical cystectomy

    Eur Urol

    (2013)
  • T. Klatte et al.

    Effect of ABO blood type on mortality in patients with urothelial carcinoma of the bladder treated with radical cystectomy

    Urol Oncol Semin Orig Investig

    (2014)
  • M. May et al.

    EORTC progression score identifies patients at high risk of cancer-specific mortality after radical cystectomy for secondary muscle-invasive bladder cancer

    Clin Genitourin Cancer

    (2014)
  • G. Ploussard et al.

    Conditional survival after radical cystectomy for bladder cancer: evidence for a patient changing risk profile over time

    Eur Urol

    (2014)
  • L.C. Wang et al.

    Combining smoking information and molecular markers improves prognostication in patients with urothelial carcinoma of the bladder

    Urol Oncol Semin Orig Investig

    (2014)
  • A. Soave et al.

    The impact of tumor diameter and tumor necrosis on oncologic outcomes in patients with urothelial carcinoma of the bladder treated with radical cystectomy

    Urology

    (2015)
  • M.S. Choo et al.

    Effect of sex on prognosis of urothelial carcinoma: propensity score matching analysis

    Clin Genitourin Cancer

    (2015)
  • M.J. Leveridge et al.

    Radical cystectomy and adjuvant chemotherapy for bladder cancer in the elderly: a population-based study

    Urology

    (2015)
  • T.-F. Wu et al.

    Galectin-1 dysregulation independently predicts disease specific survival in bladder urothelial carcinoma

    J Urol

    (2015)
  • D.P. Nguyen et al.

    Factors impacting the occurrence of local, distant and atypical recurrences after robot-assisted radical cystectomy: a detailed analysis of 310 patients

    J Urol

    (2016)
  • A. Salama et al.

    Outcome of muscle-invasive urothelial bladder cancer after radical cystectomy

    Clin Genitourin Cancer

    (2016)
  • Y. Liu et al.

    Galectin-9 as a prognostic and predictive biomarker in bladder urothelial carcinoma

    Urol Oncol

    (2017)
  • A. Mansour et al.

    MP58-15 impact of tumor histopathologic types on pattern of tumor recurrence after radical cystectomy for muscle invasive bladder cancer (MIBC)

    J Urol

    (2017)
  • H. Zargar et al.

    Dense dose MVAC versus GC in patients with CT3-4a bladder cancer treated with radical cystectomy: a real world experience

    J Urol

    (2017)
  • K.A. David et al.

    Low incidence of perioperative chemotherapy for stage III bladder cancer 1998 to 2003: a report from the National Cancer Data Base

    J Urol

    (2007)
  • P.J. Gray et al.

    Use of potentially curative therapies for muscle-invasive bladder cancer in the United States: results from the National Cancer Data Base

    Eur Urol

    (2013)
  • R. Mathieu et al.

    The prognostic role of lymphovascular invasion in urothelial carcinoma of the bladder

    Nat Rev Urol

    (2016)
  • E. Kikuchi et al.

    Lymphovascular invasion predicts clinical outcomes in patients with node-negative upper tract urothelial carcinoma

    J Clin Oncol

    (2009)
  • H. Kim et al.

    Prognostic significance of lymphovascular invasion in radical cystectomy on patients with bladder cancer: a systematic review and meta-analysis

    PLoS One

    (2014)
  • V. Novotny et al.

    Risk stratification for locoregional recurrence after radical cystectomy for urothelial carcinoma of the bladder

    World J Urol

    (2015)
  • C.M. Booth et al.

    Pathological factors associated with survival benefit from adjuvant chemotherapy (ACT): a population-based study of bladder cancer

    BJU Int

    (2015)
  • L.H. Dohn et al.

    Urokinase-type plasminogen activator receptor (uPAR) expression is associated with T-stage and survival in urothelial carcinoma of the bladder

    Urol Oncol Semin Orig Investig

    (2015)
  • G. Gandaglia et al.

    Oncologic outcomes of robot-assisted radical cystectomy: results of a high-volume robotic center

    J Endourol

    (2016)
  • M. Moschini et al.

    Clinical lymphadenopathy in urothelial cancer: a transatlantic collaboration on performance of cross-sectional imaging and oncologic outcomes in patients treated with radical cystectomy without neoadjuvant chemotherapy

    Eur Urol Focus

    (2016)
  • M. Moschini et al.

    The presence of carcinoma in situ at radical cystectomy increases the risk of urothelial recurrence: implications for follow-up schemes

    Urol Oncol

    (2017)
  • A. Liberati et al.

    The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration

    PLoS Med

    (2009)
  • J.P. Higgins

    Cochrane Handbook for Systematic Reviews of interventions version 5.1.0 (updated March 2011)

    Cochrane Collab

    (2011)
  • Cited by (38)

    • Multi-task deep learning based on T2-Weighted Images for predicting Muscular-Invasive Bladder Cancer

      2022, Computers in Biology and Medicine
      Citation Excerpt :

      The location of the surgical reslice may affect the accuracy of bladder cancer staging so that most of the high-risk NMIBC patients need Repeated Transurethral Reslice of Bladder Tumors (Re-TURBT) to reduce the diagnostic error [9,10]. It not only increases the risk of surgery and the patient's time and surgery cost but may also delay the radical treatment which leads to worse tumor results [11,12]. If clinicians can preoperatively identify patients at high risk of MIBC, then optimal treatment such as Neoadjuvant Chemotherapy and Radical Cystectomy can be implemented in time, which may improve patient prognosis [13,14].

    • The Cancer of the Bladder Risk Assessment (COBRA) score accurately predicts cancer-specific survival after radical cystectomy: external validation and lymphovascular invasion assessment value to improve its performance

      2022, Clinical Genitourinary Cancer
      Citation Excerpt :

      Age, tumor stage and lymph nodes cancer involvement have demonstrated strong correlation with patients’ prognosis after surgery 5. Moreover, the presence of Lympho-Vascular Invasion (LVI) at final pathology has been associated with poorer survival outcomes in several studies 6,7,8,9 and current European Association of Urology (EAU) guidelines strongly recommend reporting LVI 1. In this context, predicting oncological outcomes after RC is of great importance to identify patients who most deserve strict follow-up, adjuvant chemotherapy, radiotherapy, or clinical trials enrollment 10.

    • Does pathological sub stratification of T2 bladder cancer predict outcome in a contemporary pure urothelial carcinoma cohort?

      2022, Urologic Oncology: Seminars and Original Investigations
      Citation Excerpt :

      Lastly, prior studies have also studied lympho vascular invasion and CIS as independent prognostic factors for survival. For LVI, Mari et al. and Kim et al. studied the significant of LVI in radical cystectomy specimens and the systematic review of the literature and meta-analysis concluded that LVI is a strong predictor of worse prognosis regardless of staging [12,13]. For CIS, Kimura et al. showed that patients with organ confined disease after radical cystectomy with CIS, the RFS (HR-1.57, 1.12-2.21) and cancer specific mortality (HR 1.51, 1.001-2.280) was worse in those patient's with CIS compared to those without CIS [14].

    • A 25-year perspective on advances in an understanding of the biology, evaluation, treatment and future directions/challenges of urothelial cancer

      2021, Urologic Oncology: Seminars and Original Investigations
      Citation Excerpt :

      In T1 disease in TUR specimens, LVI has been demonstrated to be an important predictor of disease upstaging to T2 or greater [35]. The presence of LVI in RC specimens has also been shown to be an important independent predictor of disease recurrence and cancer-specific mortality, with the strongest effect seen in node negative patients [36]. The history of variant histology in bladder cancer evolved from first recognizing different subsets of tumors based on histologic morphology, then of matching different clinical behavior and treatment paradigms to these variants, and most recently of understanding the biology that underpins these variants [37,38].

    • Clinical utility of histopathology data: cancers of the testis and urinary bladder

      2021, Diagnostic Histopathology
      Citation Excerpt :

      Similarly, the uterus should be carefully examined for involvement by bladder cancer and any other macroscopic abnormality, but routine sampling is of limited clinical utility. LVI – LVI has been demonstrated to be significantly associated with cancer-specific survival after radical cystectomy in both lymph node negative and lymph node positive patients.26 Practice points

    View all citing articles on Scopus

    Conflicts of interest: The author certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript are the following: Shahrokh F. Shariat is advisory board member of Astellas, Cepheid, Ipsen, Jansen, Lilly, Olympus, Pfizer, Pierre Fabre, Sanofi, and Wolff. He is speaker for Astellas, Ipsen, Jansen, Lilly, Olympus, Pfizer, Pierre Fabre, Sanochemia, Sanofi, and Wolff.

    1

    Andrea Mari and Shoji Kimura contributed equally to this project and fully share the first authorship.

    View full text