The role of long intergenic noncoding RNA 00511 in malignant tumors: a meta-analysis, database validation and review

ABSTRACT Increasing studies suggested that long intergenic noncoding RNA 00511 (LINC00511) could facilitate the progression of various malignancies and correlates with prognosis of patients with malignant tumors. However, its clinical significance is still not completely clarified. Therefore, we performed a meta-analysis and bioinformatics analysis to further evaluate the correlation of LINC00511 expression level with prognosis and metastasis in patients with tumors. The pooled hazard ratio (HR) with 95% confidence interval (CI) was used to evaluate the prognostic significance of LINC00511 expression level. The pooled odds ratio (OR) with 95% CI was applied to assess the association between LINC00511 expression level and tumor metastasis. A total of 12 studies involving 1040 tumor patients were included in this meta-analysis. The pooled analyses suggested that higher LINC00511 expression level correlated with worse overall survival (OS) (HR = 1.93, 95% CI 1.49–2.49, P < 0.001) and higher incidence of lymph node metastasis (OR = 3.07, 95% CI 2.23–4.23, P < 0.001). Additionally, bioinformatics analysis based on TCGA datasets also showed that increased LINC00511 expression level may predict poor OS and disease-free survival (DFS) in patients with malignant tumors. Taken together, our finding suggested that high LINC00511 expression level may be correlated with poor prognosis and high incidence of metastasis. Nevertheless, further large-scale and high-quality studies are needed to validate our findings.

unfavorable [3]. Therefore, it remains necessary and urgent to explore novel therapeutic targets and prognostic biomarkers for cancers [4].
However, the association between LINC00511 expression and prognosis of patients with malignant tumor is still controversial. For example, Qiao et al. [13] reported that LINC00511 was downregulated in osteosarcoma, which was related to improved survival. Additionally, they further demonstrated that upregulation of LINC00511 could impede the progression of osteosarcoma through inducing apoptosis and suppressing the migration and invasion of osteosarcoma cells [13]. Moreover, the conclusions concerning the clinical role of LINC00511 expression in patients with malignancies in individual studies may be limited by small sample size. Therefore, in this study, we further evaluated the clinical significance of LINC00511 expression in malignant tumors through performing a meta-analysis of the current literatures and conducting a bioinformatics analysis using The Cancer Genome Atlas (TCGA) data. Meanwhile, we also made a comprehensive review of the mechanisms underlying the biological functions of LINC00511 in malignant tumors.

Search strategy and study selection
We searched PubMed, EMBASE, Web of science, Wanfang and China National Knowledge Infrastructure (CNKI) databases from inception up to July 2020 for eligible studies. The search terms included 'Long intergenic noncoding RNA 00511' or 'LINC00511'. The studies according with the following criteria simultaneously were included: 1) patients were divided into two groups based on LINC00511 expression level; 2) Hazard ratios (HRs) with 95% confidence intervals (CIs) for the association between LINC00511 expression and overall survival (OS) were available; and 3) studies were published in English or Chinese. The exclusion criteria were as following: 1) studies were letters, conference abstracts, meta-analysis or review articles; 2) studies evaluated the prognostic value of LINC00511 expression level using The Cancer Genome Atlas (TCGA) data; or 3) studies enrolled the overlapped patients.

Data extraction and quality assessment
Two authors independently extracted data from eligible studies. Any disagreement was removed by discussion among all the authors. The extracted information covered first author, publication date, country, tumor type, treatment type, sample size, detection method, reference controls, cutoff values, HRs with 95% CIs for OS, survival analysis type and follow-up duration. When HRs were generated using both univariate and multivariate, the HRs from multivariate analysis were preferentially selected due to less confounding factors. If not directly provided in paper, HRs with 95% CIs would be calculated from Kaplan-Meier survival curves using Engauge Digitizer software. The methodological quality of studies was evaluated using Newcastle-Ottawa scale (NOS) criteria, which is a 'star' rating system from 0 to 9 stars [25].

Validation by analyzing public data
This study conformed to the publication guidelines established by The Cancer Genome Atlas (TCGA). Gene Expression Profiling Interactive Analysis (GEPIA) was applied to assess the correlations of LINC00511 expression level with OS and DFS. The survival analysis was calculated using the K-M method and log-rank test, and the HRs and p value were presented in the figures of K-M curve as previously described [4].

Statistical analyses
All data analyses were conducted using Stata software version 12.0 (Stata Corporation, College Station, TX, USA). HRs with 95% CIs were used to assess the correlation of LINC00511 expression level with OS or DFS. X 2 -based Cochran Q test and Higgins I 2 statistic were applied to evaluate the heterogeneity among the eligible studies. P-value < 0.05 or I 2 -value > 50% indicated the presence of substantial heterogeneity [25]. The random-effect model was chosen when substantial heterogeneity existed. Sensitivity analyses were performed to explore the source heterogeneity and the robustness of the pooled HR for OS. Publication bias was examined by Begg's funnel plots and Egger's linear regression tests [25].

Study selection and characteristics
A total of 304 potentially relevant articles were retrieved from electronic databases. To yield eligible articles, we firstly excluded 128 duplicate articles. After screening the titles and abstracts, we further excluded 98 articles for conference abstracts, review articles and irrelevant-topic studies. Then, 78 articles were screened by full-text. During this step, 66 studies were removed due to insufficient data and TCGA data. Finally, a total of 12 eligible studies were included in this meta-analysis [7][8][9][10][11][12][13][14][15]18,19]. The flow diagram summarizing the study selection procedure is presented in Figure 1. The main characteristics of eligible studies are summarized in Table 1. All studies enrolled Chinese patients and were published between 2016 and 2020. The sample sizes of eligible studies ranged from 39 to 140. Ten types of cancer were analyzed in eligible including hepatocellular carcinoma (n = 1), osteosarcoma (n = 1), breast cancer (n = 2), lung cancer (n = 1), glioma (n = 1), ovarian cancer (n = 1), cervical cancer (n = 2), renal cell carcinoma (n = 1), pancreatic cancer (n = 1) and gastric cancer (n = 1). LINC00511 expression levels were detected by RT-qPCR in all the eligible studies. All the included studies were given with no less than six scores based on NOS criteria, suggesting that they were moderate to high quality and eligible for this meta-analysis.

Meta-analysis and bioinformatics analysis of association between LINC00511 expression and survival outcomes
A total of 12 studies, which enrolled 1040 patients with malignant tumors, were included in our meta-analysis to assess the correlation between LINC00511 expression level and OS. Considering the obvious heterogeneity among these studies (I 2 = 81.9%, p < 0.01), we applied the random-effect model to calculate the pooled HR and 95% CI for OS. The result showed that higher LINC00511 expression correlated with worse OS (HR = 1.93, 95% CI 1.49-2.49, P < 0.001) (Figure 2(a)). To date, only two studies reported the relationship between LINC00511 expression and progression-free survival and none of the studies referred to DFS, so we could not conduct a meta-analysis of the prognostic significance of LINC00511 expression for these survival outcomes. To confirm the result of our meta-analysis and evaluate the prognostic significance of LINC00511 expression for the other survival outcomes, we adopted survival plots in GEPIA through merging LINC00511 expression data and survival data of malignancies from all the TCGA dataset, in which 3491 patients were divided into high or low expression group according to LINC00511 expression. The results suggested that higher LINC00511 expression predicted worse OS and DFS (Figure 2(b-c)), further validating the result of our meta-analysis.

Meta-analysis of association between LINC00511 expression and lymph node metastasis
A total of 9 studies with 833 tumor patients were included in this meta-analysis to assess the association between LINC00511 expression and lymph node metastasis. Considering the absence of significant heterogeneity among these studies (I 2 = 34.7%, p = 0.141), we applied the fixedeffect model to calculate the pooled OR and 95% CI for lymph node metastasis. The result showed that higher LINC00511 expression correlated with higher incidence of lymph node metastasis (OR = 3.07, 95% CI 2.23-4.23, P < 0.001) (Figure 3). To date, only two studies reported the relationship between LINC00511 expression and distant metastasis, so we did not conduct a meta-analysis of the association between LINC00511 expression and distant metastasis due to small sample size.

Sensitivity analysis and publication bias
Sensitivity analysis was performed by sequentially omitting single study. As shown in Figure 4(a-b), removal of any individual study had no substantial effects on the pooled HR for OS and the pooled OR for lymph node metastasis. Begg's funnel plot and Egger's test were conducted to evaluate the publication bias. The results showed that Begg's funnel plots for OS and lymph node metastasis were symmetric (Figure 4(c-d)). Moreover, the p values of Egger's tests for OS and lymph node metastasis were 0.509 and 0.052, respectively. These results suggested that there was no significant publication bias in our meta-analysis. Overall, our sensitivity analysis and publication bias assessment indicated that our pooled estimations were robust and reliable.

Discussion
Numerous studies have suggested that LINC00511 was dysregulated in a multiple of tumors and this  correlated with prognosis and metastasis in human malignancies. However, the conclusions regarding the clinical significance of LINC00511 expression level in malignant tumors remain conflicting, so we performed a meta-analysis and bioinformatics analysis based on The Cancer Genome Atlas (TCGA) data to systematically assess the clinical significance of LINC00511 expression level. Additionally, we also made a comprehensive review of the biological functions of LINC00511 in malignant tumors and the relevant mechanisms.
Several limitations should be considered in this meta-analysis. First, all the included studies were retrospective. Second, most of studies did not directly provide HRs for survival outcomes, so we had to calculate HRs based on the Kaplan-Meier survival curves. This manual process may unavoidably cause some calculation errors. Third, there was significant heterogeneity and it may be caused by the complexity and inconsistency in many aspects of the included studies, such as follow-up time, age, gender, and therapy. Fourth, only 12 eligible studies were included into this meta-analysis, suggesting that our conclusions may be still limited by sample size. Additionally, only 1 or 2 eligible studies focused on the specific tumors, so we could not perform the pooled analysis for the specific tumors, which may  [11,12,36] (Continued ) largely discount the clinical significance of our metaanalysis. Last but not least, all the studies included in our meta-analysis were from China. Therefore, our conclusions could not be directly generalized to the other populations.

Conclusion
Increased LINC00511 expression level significantly correlated with worse prognosis and higher incidence of lymph node metastasis in patients with malignant tumors, indicating that LINC00511 may be a prognostic biomarker and therapeutic target. However, our study may overestimate the clinical significance of LINC00511 expression in human malignancies due to several limitations. Therefore, further high-quality studies should be performed to further validate our findings.

Highlight
(1) Role of LINC00511 in tumor

Author contributions
Zhaocong Chen designed this study. Jianlong Ding, Junyan Cao, and Zhiming He were responsible for collecting the data and analyzing and interpreting the data; Jianlong Ding and Zhaocong Chen took charge of drafting the manuscript. All authors read and approved the final manuscript.

Availability of data and material
All data generated or analyzed during this study are included in this published article.

Consent for publication
The manuscript does not contain data from any individual person.