The prognostic signicance of NLR in non-metastatic renal cell carcinoma undergoing nephrectomy, A meta-analysis

INTRODUCTION Neutrophil to Lymphocyte ratio (NLR) has been reported to correlate with poor survivals in many tumors. However, the association between preoperative NLR elevation and survival outcome in non-metastatic renal cell carcinoma (RCC) underdoing nephrectomy remains controversial. The aim of this meta-analysis was to investigate the prognostic signicance of elevated NLR in non-metastatic RCC. EVIDENCE ACUISITION We systematically searched PubMed, EmBase, and the Cochrane Library databases in may 2018. Cancer specic survival (CSS), disease-free survival (DFS) and overall survival (OS) were pooled by hazard ratio (HR) with corresponding 95% condence interval. EVIDENCE SYNTHESIS A total of 3,175 patients from 8 studies were analyzed. The results demonstrated that elevated pretreatment NLR was signicantly related to poor CSS (HR 1.91, 95% CI=1.53-2.40), DFS (HR 1.38, 95% CI=1.09-1.74), and OS (HR 1.84, 95% CI=1.58-2.14) in patients with non-metastatic RCC. CONCLUSION Elevated NLR indicates a poor long-term survival (CSS, DFS and OS) in non-metastatic RCC. Patients with elevated NLR are more likely to have poor prognosis than those with lower NLR. reflects a strong association between high NLR value and poor survival outcome. Between-study heterogeneity was evaluated using Cochran’s Q test,P value and


Introduction
Renal cell carcinoma (RCC) was the most common malignant tumor, accounting for 2~3% of adult malignancies [ 1] . Based on ultrasound and computed tomography, Some incidental and early stage RCC can be detected. About 20% to 30% patients occur distant metastases after nephrectomy Because of unresponsive to chemotherapy and radiotherapy, some of them even died [2,3] . Therefore, to identify a significant biomarkers for clinical dignaostic and therapy of RCC is important.
It has been demonstrated that inflammatory response plays an essential roles in tumor development and progression [4] . The systemic inflammation have intimated connection with poor survivals of many tumors in human [5] . Increased of pro-inflammatory cytokines and signaling molecules in cancer patients might reflect both disease activity and the innate response of the host to the tumor [6] . There is accumulating evidence for the correlation between pretreatment systemic inflammation and worse survival in patients with RCC [7] .
Peviouse research have shown that Neutrophil to Lymphocyte ratio (NLR) being used as a biomarker for the prognosis of RCC [8] [9] . NLR was an indicator of general immune response to various stress, it has been proved to be a prognosis related marker in various Page 3/17 malignancies, including RCC [10] [11] [12] . Earlier meta-analysis on NLR in patients with metastasis renal cancer described that elevated baseline NLR remained an independent predictor of OS and PFS [13] . However, the association between NLR and CSS, DFS, OS in non-metastasis RCC are in consistent. Some studies demonstrated that NLR represented an independent risk factor for prognosis [14] [15] , while others did not [16] [17] . Thus, it is necessary to perform a comprehensive systematic review and meta-analysis of published studies to determine predictive value of NLR for RCC.

Study Identification
A systematic literature search were performed in May 2018. Three electronic databases (PubMed, Embase, and the Cochrane Library) were searched to select relevant articles. The following key words were used: ("neutrophil lymphocyte ratio" or "neutrophil-lymphocyte ratio", or "neutrophil to lymphocyte ratio", or "neutrophil-to-lymphocyte ratio", or "NLR") and ("kidney neoplasm" or "kidney cancer" or "renal neoplasm" or "renal cancer" or "renal carcinoma" or "RCC"). At the same time, a MESH/EMTREE search for "Kidney neoplasms" were also performed. Two reviewers independently scanned the potential references for titles and abstracts to exclude irrelevant articles. When disagreements or questions appeared, it would be resolved by consulting another author. In addition, the reference list was assessed for additional relevant articles. We evaluated the remaining articles to identify research that covered the topic of interest. Then the full texts were estimated comprehensively. The selection process of the articles is shown in  The studies included in this meta-analysis must be met the following criteria: (a) the patients must have been diagnosed with non-metastatic RCC by a pathological and imaging examination. (b) Provided pretreatment NLR as a variable in outcome analysis and cut off value. (c) HR and 95% CI were reported for pretreatment NLR in CSS, Page 4/17 DFS and OS or could be calculated from the article data. A search must meet all three inclusion criteria for inclusion.
The exclusion criteria were: (a) review articles, case reports, letters, editorial comments,conference abstract and studies irrelevant to our topic. (b) Overlapping or duplicate reports. (c) Lacking required outcomes data that could not be calculated from other information presented. (d) Those nonhuman research. A search meeting any of the four exclusion criteria was excluded.

Data extraction
The following data included in this meta-analysis were extracted by two independent reviewers: first author, year of publication, country,sample size, age, cut-off value of NLR. Disagreements or questions in data extraction were resolved by joint discussion. The basic features of the 8 studies were summarized in table 1.
Quality assessment for the included studies was conducted using the Newcastle-Ottawa Scale (NOS) [18] . This scale mainly concerned with three aspect (selection of patients, comparability of group, and assessment of outcomes).
Studies with NOS scores ≥6 were considered to be high quality, and studies with NOS scores <6 were considered to be low quality. Studies from conference abstracts were defined as low quality. All P value were two tailed, and P <0.05 was considered statistically significant. Page

Search result
The

NLR and OS
Six cohorts covering 2,465 patients described the association between NLR and OS in RCC. As illustrated in Figure 4.
The pooled HR for high pretreatment NLR value group was found to be 1.84 (95% CI=1.58-2.14) with low heterogeneity (I 2 =33%, P=0.19), indicating that there was significant difference between high and low pretreatment NLR values in RCC patients.  Table 2.
A sensitivity analysis was performed in which each study was successively deleted to assess the influence of individual studies on the pooled HRs. This analysis shows that no obvious change for OS , DFS and CSS when we removed each article in turn . The detailed resuts were shown in in Table 3.
Funnel plots of the studies used in the meta-analysis were conducted for assessing the publication bias, and we could roughly assess the publication bias by seeing wether their shapes were of any obxious asymmetry. As showed in Figure   5, there were no evident publication bias by funnel plots. However, because the number of included studies was just six, the funnel plots mayot be significant.

Discussion
The association between NLR and prognosis has been identified in various tumors.
Semeniuk-Wojtas A have reported a meta-analysis about the prognostic significance of NLR in metastasis RCC [13] . However, prognostic significance of preoperative NLR in nonmetastasis RCC still uncertain. So we conduct this meta-analysis to investigate the clincal value of pretreatment NLR in non-metastasis RCC.  [15] from the CSS and OS estimate, the I 2 decreased from 44% to 0%and 33% to 0% respectively. The results proved that Chen Z et al [15] might be the source of heterogeneity of CSS and OS in this study.
The potential reasons for the correlation between NLR elevation and low survival outcome in RCC patients are complex and have not been well clarified. It has been widely accepted that inflammatory mediators can destabilization the cancer cell genome by directly inducing DNA damage or affecting DNA repair systems and altering cell cycle checkpoint, leading to accumulation of cancer initiation and progression [23] . The association between chronic inflammation and cancer are duplex, inflammation can induce tumorigenesis and metastasis by affect cell proliferation, death, and angiogenesis through elevate the levels of cytokines, nuclear factor κB (NF-κB), prostaglandins and micro RNAs.et al, tumors also can induce inflammatory reactions through express cytokines that recruit neutrophils and macrophages [24] . Recent study have indicated that NLR is an indicator of systemic inflammation and general immune response of the cancer [25,26] . Although there are some other systemic inflammation reaction biomarkers, such as CRP , cystatin-C and PLR , were reported to be used for prognostic indexes in RCC [9][27] . Our meta-analysis indicated that an increased pretreatment NLR predict worse long-term survival in patients with RCC.
Heterogeneity is the quality or state of consisting of dissimilar or diverse elements,it was very impotrtant in meta-analysis process. In this study, there are no significant heterogeneity when pooled HR for CSS and OS, but when pooled HR for DFS, the heterogeneity proved high diversity. Page 11/17 In this study, we carefully calculated the possible association between elevated NLR and prognosis of non-metastasis RCC patients from the most recent studies. The aim of the study was to drive reliable conclusions for clinical applications of NLR. But certain limitation should be taken into consideration for the results. The main limitation was that only 8 studies (among them, 3, 4 and 6 studies investigated the relationship of NLR and CSS, DFS and OS) being analysis, so the conclusion of the study are preliminary and need to examine in the future studies. Besides, we can not avoid the selection bias, all the studies we selected were retrospective, which may influence the accuracy of the results.
Furthermore, the heterogeneity and variability could be influence by NLR cut-off value.
Although we proved that elevated NLR was associated with a poor prognosis in nonmetastasis RCC patients. Further well-designed studies with a large sample size are needed to correct our limitations. Whether the NLR can be use as a potential biomarkers in clinical applications are needing further investigation.

Conflicts of interest
The authors declare no conflicts of interest.

Figure 1
Flow chart of study selection from the PubMed, the Embase, and the Cochrane Library online databases for the NLR as prognostic in non-metastatic renal cell carcinoma patients, published in English from January 1, 2015 to May 31,2018.

Figure 2
Forests plot of the association between neutrophil-lymphocyte ratio (NLR ) and cancer speci c survival (CSS) of non-metastatic renal cell carcinoma patients.

Figure 3
Forest plot for the association between neutrophil-lymphocyte ratio (NLR ) and disease-free survival (DFS) of non-metastatic renal cell carcinoma patients.

Figure 4
Forest plot for the association between neutrophil-lymphocyte ratio (NLR ) and overall surviva (OS) of non-metastatic renal cell carcinoma patients.