The evaluation of monocyte lymphocyte ratio as a preoperative predictor in urothelial malignancies: a pooled analysis based on comparative studies

In recent years, several studies have reported monocyte lymphocyte ratio (MLR) to predict prognosis in various tumors. Our study was performed to evaluate the association between preoperative MLR between prognostic variables in urothelial carcinoma patients. Systematic literature search was conducted in PubMed, Embase, Web of science. The correlation between preoperative MLR and overall survival (OS), cancer specific survival (CSS), disease free survival (DFS)/relapse free survival (RFS), progression free survival(PFS) was evaluated in urothelial carcinoma patients. Meanwhile, the association between MLR and clinicopathological characteristics was assessed. Finally, 12 comparative studies comprising a total of 6209 patients were included for pooled analysis. The hazard ratios (HRs), odds ratios (ORs)and 95% confidence intervals (CIs) were further analyzed as effect measures. The pooled results demonstrated that elevated preoperative MLR indicated unfavorable OS (HR = 1.29, 95%CI = 1.18-1.39, I2 = 33.6%), DFS/RFS (HR = 1.42, 95%CI = 1.30–1.55, I2 = 0.0%) and CSS (HR = 1.41, 95%CI = 1.29–1.52, I2 = 0.0%). Moreover, the pooled results also suggested that elevated preoperative MLR was correlated with high tumor stage (OR = 1.22, 95%CI = 1.07–1.37, I2 = 0.0%) in urothelial carcinoma patients. No significant association was found between preoperative MLR and PFS in upper urinary tract urothelial carcinoma (UUTUC) patients. Collectively, elevated preoperative MLR predicted poor prognosis in urothelial carcinoma and have the potential to be a feasible and cost-effective prognostic predictor for management of urothelial carcinoma.

Prognostic role of MLR in CSS. 5 studies including 5301 patients presented the data about MLR and CSS, and 4 of 5 studies focused the relationship between MLR and CSS in UCB patients. The pooled results showed that elevated preoperative MLR was significantly associated with poor CSS in urothelial patients (HR = 1.41, 95%CI = 1.29-1.52), without significant heterogeneity among the 5 studies (I 2 = 0.0%, p = 0.775). Only 1 of 5 studies reported the prognostic role of MLR in CSS in UUTUC patients (HR = 1.85, 95%CI = 0.91-3.73) (Fig. 4).    www.nature.com/scientificreports www.nature.com/scientificreports/ was found between preoperative MLR and Diabetes, tumor necrosis, multifocality, tumor grade and lymphovascular invasion (LVI). The detailed results were shown in Table 2.

Discussion
Urothelial carcinoma was the most common malignant tumors in urinary system, which derived from the lining surface epithelium termed as "urothelium" with the same embryologic origin 1 . UCB and UUTUC were the two main types in urothelial carcinoma with respect to anatomical location, however, the two types behave identically, and much of clinical decision-making could be similar between UCB and UUTUC 29,30 . In order to improve the management of urothelial carcinoma, it is pivotal to identify novel, economic and feasible predictors for individual therapy. Emerging evidences have shown immune cells were involved with cancer initiation, progression and invasion in multiple tumors 31,32 . The hematological inflammatory cells and factors have been widely investigated and identified as important prognostic indicators in urothelial carcinoma 10,33 . More recently, preoperative MLR, as an inflammation related marker, was reported as a potential predictor for urothelial maligancies in several studies. However, there was no related meta-analysis, which can increase the statistical power by methodologically combining the reported results from varying studies. To the best of our knowledge, this is the first pooled analysis to assess the preoperative MLR as a kind of predictor in urothelial maligancies.
According to the pooled analysis of comparative studies from meta-analysis, our results indicated elevated preoperative MLR was significantly associated with reduced OS, DFS/RFS and CSS in urothelial carcinoma patients. However, no significant association was found between preoperative MRL and PFS in urothelial carcinoma patients. Furthermore, subgroup analysis was performed according to the different sites of urothelial carcinoma(UCB, UUTUC), and the results also suggested elevated preoperative MLR predicted worse OS, DFS/  www.nature.com/scientificreports www.nature.com/scientificreports/ RFS and CSS in UCB or UUTUC, respectively. The relationship between elevated preoperative MLR and selected clincopathological features was also assessed in this pooled analysis, and the results demonstrated elevated preoperative MLR was an independent risk for high tumor stage (≥T2 vs. <T2) in urothelial maligancies (OR = 1.201, 95%CI = 1.047-1.356, I 2 = 0). Collectively, the pooled data from this meta-analysis showed that MLR may serve as a prognostic indicator in urothelial carcinoma patients.
The underlying mechanisms involved with the prognostic role of hematological MLR in urothelial carcinoma remain to be addressed. Mounting studies have indicated tumor infiltrating lymphocytes (TILs) were associated with prognosis of urothelial carcinoma 34,35 , and the prognostic role may vary according to the subpopulations of TILs 35 . Low lymphocyte counts can lead to the deficiency of host immune response, which may cause the progression and metastasis of cancer cells and result in poor survival in malignancies 36,37 . Circulating monocytes can be recruited into tumor microenviroment and further polarize into tumor associated macrophages (TAMs). Elevated tumor infiltrating TAMs were always involved with worse survival in various tumors 38,39 . Taken together, we postulated that elevated preoperative MLR could reflect the disorders of immune cells in tumor miroenvironment, which generated poor survival in urothelial carcinoma patients. This pooled analysis also had some disadvantages. First, considerable heterogeneity was found in our pooled analysis. In order to detect and decrease the potential heterogeneity, we have performed sensitivity analysis and subgroup analyses. Second, the number of included studies was limited, and the pooled results should be cautiously interpreted. Moreover, 8 included studies had reported HRs and 95%CIs derived from multivariate analysis, while HRs and 95%CIs in 4 studies were extracted from univariate analysia, which may overestimate the prognostic role of MLR. In addition, all the included studies were retrospective design. In future, large-scale prospective studies are still needed to validate our results in this pooled analysis.
Collectively, our priamry results derived from pooled analysis suggested elevated preoperative MLR was significantly correlated with poor survival, which may contribute to the risk stratifications before surgery and clinical decision-making for individual therapeutic strategies in urothelial carcinoma patients.

Materials and Methods
This pooled analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA) 40 , and the PRISMA checklist was detailed in Supplementary Table S1. www.nature.com/scientificreports www.nature.com/scientificreports/ search strategies. There databases (Pubmed, Embase and Web of science) were systematically searched for eligible studies involved with preoperative MLR in urotheilal malignancies, up to November 30 th , 2018. The selected keywords were confined in the followings: (urothelial cancer OR bladder cancer OR bladder malignancies OR transitional cancer OR ureteral cancer OR renal cancer) AND (MLR OR monocyte to lymphocyte ratio OR monocyte lymphocyte ratio). References listed in identified studies were also went through to expand the scope of search. study selection. Inclusion criteria. The studies were included according to the criteria listed in the following: (1) patients were pathologically diagnosed as UCB or UUTUC. (2) MLR was collected before surgery. (3) the studies provided HRs or Kaplan-meier curve for evaluating the role of MLR in prognostic variables. (4) studies reported cut-off values for stratified analysis.
Exclusion criteria. The studies were excluded if they met the following criteria: (1) animal studies, editorials, case reports, reviews, comments; (2) duplicates identified by Endnote X7 version; (3) the studies did not focused on the relationship between preoperative MLR and urothelial maligancies or without sufficient data to extract HRs and 95% CIs.
Data management and quality assessment. Data from the included studies were summarized via a kind of predefined form: Acronym of first author (year of publication), design, geographical region, cases number, diagnosis, therapy, cut-off value, outcome and analysis methods. The related HRs and 95%CIs were extracted to further analyze the association between preoperative MLR and urothelial malignancies. OS, CSS, DFS/RFS and PFS were taken as prognostic outcomes for pooled analysis. The clinicopathological characteristics included gender, diabetes, hypertension, concomitant carcinoma in situ, multifocality, tumor grade, tumor necrosis, lymphovascular invasion and tumor stage.
The Newcastle-Ottawa Scale (NOS) system was used to evaluate the quality of included studies 41 . The maximum score is 9, which was involved with the evaluation of subject selection, comparability of groups, and clinical outcome. If the NOS scored more than 7 (included 7) for a study, it was usually taken as a high-quality study. All the processes about data management and quality assessment were reviewed independently by two authors, and any disagreement was settled through discussion with a senior author. statistical analysis. The pooled HRs or ORs were analyzed with Stata version 15.0 (StatCorp, College Station, TX, USA). ORs and 95%CIs regarding the association between preoperative MLR and chinicopathological characteristics were calculated from raw data reported in the included studies. Heterogeneity among the studies was reported via the statistic effects, such as Chi-squared tests, p value and I-square. A fixed-effect model was preferentially used for this pooled analysis, unless there was significant heterogeneity among the studies(I 2 > 50%). Otherwise, a random-effect model was adopted as an alternative method. An observed HR >1, combined with 1 not included in its 95%CIs, indicated elevated preoperative MLR was associated with poor prognosis in urothelial carcinoma patients. We did not perform Begg's and Egger's tests to evaluate the publication bias for that the number of included studies was less than 10 in this pooled analysis [42][43][44] . However, we conducted sensitivity analysis to validate the pooled results by excluding each study.