High BANCR expression is associated with worse prognosis in human malignant carcinomas: an updated systematic review and meta-analysis

BRAF-activated noncoding RNA (BANCR) is aberrantly expressed in various tumor tissues and has been confirmed to function as a tumor suppressor or oncogene in many types of cancers. Considering the conflicting results and insufficient sampling, a meta-analysis was performed to explore the prognostic value of BANCR in various carcinomas. A comprehensive literature search of PubMed, Web of Science, EMBASE, Cochrane Library and the China National Knowledge Infrastructure (CNKI) was conducted to collect relevant articles. The pooled results showed a strong relationship between high BANCR expression and poor overall survival (OS) (HR (hazard ratio) =1.60, 95% confidence interval (CI): 1.19–2.15, P = 0.002) and recurrence-free survival (RFS) (HR = 1.53, 95% CI: 1.27–1.85, P < 0.00001). In addition, high BANCR expression predicted advanced tumor stage (OR (odds ratio) =2.39, 95% CI: 1.26–4.53, P = 0.008), presence of lymph node metastasis (OR = 2.03, 95% CI: 1.08–3.83, P = 0.03), positive distant metastasis (OR = 3.08, 95% CI: 1.92–4.96, P < 0.00001) and larger tumor sizes (OR = 1.63, 95% CI: 1.09–2.46, P = 0.02). However, no associations were found for smoking status (OR = 1.01, 95% CI: 0.65–1.56, P = 0.98), age (OR = 0.88, 95% CI: 0.71–1.09, P = 0.236) and sex (OR = 0.91, 95% CI: 0.72–1.16, P = 0.469). The sensitivity analysis of OS showed that the results of each publication were almost consistent with the combined results, and the merged results have high robustness and reliability. The results showed that elevated BANCR expression was associated with unfavorable prognosis for most cancer patients, and BANCR could serve as a promising therapeutic target and independent prognostic predictor in most of cancer types.


Background
Currently, cancer remains one of the major public health concerns worldwide [1]. Approximately 1,762,450 new cancer cases and 606,880 cancer deaths were predicted to occur in the United States in 2019 [2]. Notably, due to the rapid advancement of cancer research, treatment and diagnostic methods, cancer mortality has continuously decreased by a total of 27% in the last two decades [3]. In spite of this, the 5-year relative survival rate of patients is still unsatisfactory [4]. When cancer is diagnosed, many patients are already in the middle and late stages of the disease, and there is still no ideal effective treatment. Therefore, it is critical to explore specific and sensitive therapeutic targets and promising prognostic biomarkers for the effective treatment of cancer.
Increasing studies have suggested that long noncoding RNAs (lncRNAs), which are transcripts longer than 200 nucleotides that do not have the ability to code proteins, play vital roles in multifarious biological processes, including cell differentiation, growth, apoptosis, cell cycle and metabolism [5]. Moreover, abnormal lncRNA expression has been observed in various tumor tissues and is involved in the proliferation, invasion and metastasis of tumor cells [6][7][8]. A growing number of publications have revealed the great application value of long noncoding RNAs, including MALAT1 [9], CRNDE [10], ZEB1-AS1 [11], etc., in targeted treatment and cancer prognosis.
By using RNA-sequencing, Flockhart et al. originally found that BRAF-activated noncoding RNA (BANCR), a 693-bp lncRNA located on chromosome 9, was overexpressed in melanoma cells. Additionally, accumulating studies have suggested that BANCR is correlated with the metastasis and invasion of multiple tumor cells and could function as a prognostic biomarker for cancers such as gastric cancer [12,13], hepatocellular carcinoma [14][15][16][17], renal cell carcinoma and non-small cell lung cancer [18,19]. However, due to the small sample size and discrepant conclusions among those studies, the association of BANCR expression with the prognosis of patients is still undefined. Thus, a meta-analysis was performed to investigate the prognostic value of BANCR in various cancers.

Literature search strategies
A literature search was conducted in the electronic databases of PubMed, Cochrane Library, EMBASE, Web of Science and the Chinese National Knowledge Infrastructure (CNKI) by using the following terms: ("BANCR" OR "Lnc RNA BANCR" OR "lncBANCR" OR "BRAF-activated non-coding RNA") AND ("neoplasm" OR "carcinoma" OR "tumor" OR "cancer"). The latest literature search was performed up to July 25, 2019.

Inclusion and exclusion criteria
The selection of studies was completed independently by two researchers. The inclusion criteria were as follows: (a) studies investigated the correlation of BANCR expression with the survival outcomes and clinical prognosis of cancer patients; (b) patients were classified into a high expression group and a low expression group in accordance with the primary literature; (c) the expression level of BANCR was detected by validated techniques; (d) publications provided sufficient and usable data to calculate the OR and HR; and (e) studies published in English or Chinese. The exclusion criteria were as follows: (a) publications exploring the molecular biological mechanisms of BANCR but not investigating the relationship between the expression level of BANCR and the prognosis of cancer patients; (b) reviews and meta-analyses, letters, animal studies, and conference literature; (c) studies without enough data to perform prognostic analysis; and (d) duplicate publications.

Data extraction and quality assessment
The data were independently extracted by two investigators (FSX and LZ), including first author's name, publication date, cancer type, sample size, overall survival (OS), recurrence-free survival (RFS), disease-free survival (DFS), TNM stage, tumor size, distant metastasis (DM), histological grade, lymph node metastasis (LNM), depth of invasion, smoking status, follow-up time of patients, detection methods of BANCR and HR, age and sex. The Newcastle-Ottawa Scale (NOS) was used to assess the quality of the included articles, and high-quality studies had NOS scores greater than 6 [20].

Statistical analysis
The meta-analysis was conducted to calculate the pooled ORs and HRs with corresponding 95% CIs by using Review Manager 5.3 software (Cochrane Collaboration, London, UK) and STATA 12.0 software (Stata Corp., College Station, TX). A random-effects model was adopted when I 2 >50%, which indicated significant heterogeneity among the enrolled studies, otherwise, a fixed-effects model was applied. Publication bias was assessed by using funnel plots and Begg's test. When significant heterogeneity existed, subgroup analysis was conducted to explore the source of heterogeneity. Sensitivity analysis was carried out to test the reliability and stability of the results by excluding each of the included studies one by one and then combining the effect sizes to determine whether the result of a single study significantly affected the overall result. Especially, when survival data could not be directly extracted and only Kaplan-Meier curves were provided in the primary articles, the Engauge Digitizer tool (Version 4.1) was used to extract the time-dependent survival rate from the Kaplan-Meier curves, and the HRs and 95% CIs were calculated according to the method in [21]. Statistical significance was considered when P<0.05.

Study characteristics
A total of 386 studies were identified from the databases; among them, 174 duplicate studies were excluded, and 158 studies were omitted after reading the abstracts and full texts. Furthermore, 16 publications did not investigate the association between BANCR expression and the prognosis of patients, 6 publications did not divide patients into high and low BANCR expression groups, and 12 publications lacked usable data. Finally, 20 eligible studies were included for qualitative and quantitative synthesis (Fig. 1).

The association of BANCR with OS
A total of 10 studies comprising 1151 patients were included in the analysis of the relationship between BANCR and OS. The random-effects model was applied due to marked heterogeneity (I 2 = 60%, P = 0.008). The pooled results supported the conclusion that patients with high BANCR expression tended to have shorter overall survival (HR = 1.60, 95% CI: 1.19-2.15, P = 0.002, Fig. 2a). Moreover, subgroup analysis was conducted to explore the sources of heterogeneity based on cancer type, the level of BANCR expression (high BANCR expression vs. low BANCR expression), the method of HR extraction (direct / indirect extraction), sample size (less / more than 100 patients) and NOS score (score of 9 / less than 9). A strong correlation was revealed between high BANCR expression and poor OS for cancers in the digestive system (HR =     Table 3. The poor prognosis related to BANCR was also identified by the positive association between high BANCR expression and short DFS (HR = 1.21, 95% CI: 0.33-4.41, P = 0.77) and RFS (HR = 1.53, 95% CI: 1.27-1.85, P < 0.00001) (Fig. 2b).

The association of BANCR with TNM stage
Fourteen studies including 1378 patients were enrolled to investigate the association of BANCR expression level with TNM stage. The random-effects model was adopted, and subgroup analysis was carried out due to significant heterogeneity (I 2 = 83.9%, P < 0.00001). The pooled OR showed a strong association between high   (Fig. 3).

Publication bias and sensitivity analysis
Sensitivity analysis was performed to assess the OS outcome stability among the included studies. We found that removing each study successively did not influence the overall results significantly (The overall HR value of the sensitivity analysis is: HR = 0.47, 95% CI: 0.18-0.77. The detail HR value with removing each study successively could be seen in Fig. 7, and no HR value exceeds the confidence interval of the combining result (95% CI: 0.18-0.77)), indicating that the results of each publication were almost consistent with the combined results, in other words, the merged results have high robustness Fig. 3 Forest plot of the relationship between BANCR expression and TNM stage. Note: BANCR: BRAF-activated noncoding RNA; CI: confidence interval; Random: random-effects model. The random-effects model was adopted. The square size of individual studies represented the weight of the study. Vertical lines represent 95% CI of the pooled estimate. The diamond represents the overall summary estimate, with the 95% CI given by its width and reliability (Fig. 7). Potential publication bias was estimated by Begg's test. As shown in Fig. 8  Vertical lines represent 95% CI of the pooled estimate. The diamond represents the overall summary estimate, with the 95% CI given by its width the proliferation, migration and invasion of retinoblastoma cells [38]. Wang et al. found that high BANCR expression could be observed in HCC tissues and that high BANCR may induce the proliferation and invasion of liver cancer cells by inhibiting E-cadherin expression and promoting Vimentin expression. Zhang et al. suggested that downregulated BANCR expression drives aggressiveness in papillary thyroid cancer through the MAPK and PI3K pathways [26]. Lou et al. confirmed that the knockdown of BANCR expression could inhibit the proliferation and induce the apoptosis of breast cancer cells by promoting the epithelial-mesenchymal transition (EMT) process [33]. Additionally, it has been reported that the expression of BANCR is increased in colorectal cancer (CRC) and that BANCR could strengthen the migration and proliferation abilities of CRC by inducing epithelial-mesenchymal transition (EMT) via the activation of the MEK/ERK signaling pathway [34,42]. Conversely, Liao et al. discovered that in papillary thyroid cancer (PTC) patients, the expression of BANCR was downregulated, which partially suppressed the proliferation, migration and invasion of PTC cells via the ERK/MAPK signaling pathway [24]. Likewise, Sun et al. observed a decreased expression of BANCR in NSCL C cells, and low BANCR expression may drive NSCLC cell invasion and metastasis by affecting EMT. In summary, the expression level and role of BANCR varies from cancer to cancer, possibly due to the differences between tumors. A comprehensive analysis is therefore needed to accurately assess the prognostic value of BANCR in cancer.
Considering the varied conclusions mentioned above, 20 studies with 1997 patients and 12 types of cancers were finally enrolled in this meta-analysis to explore the relationship between BANCR expression level and the Fig. 6 Forest plot of the relationship between BANCR expression and tumor size. Note: BANCR: BRAF-activated noncoding RNA; CI: confidence interval; Random: random-effects model. The random-effects model was adopted. The square size of individual studies represented the weight of the study. Vertical lines represent 95% CI of the pooled estimate. The diamond represents the overall summary estimate, with the 95% CI given by its width  Random Random-effect model; TNM TNM stage;
On the other hand, through subgroup analysis, we can observe that the heterogeneities of some subgroups reduced significantly heterogeneity (Table 3), such as digestive system (I 2 = 17%), other systems (I 2 = 15%), multivariate analysis (I 2 = 11%), direct HR extraction (I 2 = 0%), and less than 100 subjects (I 2 = 36%). Low heterogeneity suggests reliability, stability and persuasive of results. The unfavorable survival prognosis related to BANCR in cancers was also confirmed for RFS (HR = 1.88, 95% CI: 1.09-3.25). However, no associations were found between BANCR expression and OS for non-digestive system cancers (HR = 1.  Figure S4). In summary, despite serving as both an oncogene and a tumor suppressor gene in different cancers, the pooled results still support the conclusions of most primary studies that have shown that high BANCR expression indicates worse cancer prognosis. The results of the sensitivity analysis for OS showed that the overall results were not significantly affected by the arbitrary deletion of a certain study, which supported the stability of the results. In addition, slight publication bias was observed in the included studies. Therefore, the expression level of BANCR could be used to evaluate the prognosis of tumor patients in most cancers.
Although the relationship between BANCR expression and clinical prognosis has been assessed by Hu et al. and Fan et al. [43,44], there are several differences between these previous investigations and our research. First, the pooled results revealed the significant association between high BANCR expression and worse OS and RFS, advanced TNM stage and a high risk of lymph node metastasis, which failed to be concluded by a previous meta-analysis. Second, larger sample sizes and more cancer types were included in this meta-analysis. Third, comprehensive subgroup analysis was performed, and the correlations between BANCR and tumor size, histological grade, invasion depth, smoking status, number of local tumors, age and sex were first explored in this study, which were not investigated in the previous metaanalysis. Finally, the detailed molecular biological mechanisms of BANCR in various cancers were discussed and summarized. Nevertheless, there are some limitations in this meta-analysis: (a) most of the patients included in this study came from China, which may limit the generalizability of the results; (b) the sample size included was not large enough, which may affect the reliability of the results; (c) only 11 types of cancers were included to investigate the association between BANCR and cancer prognosis; thus, the conclusions of this study could not represent all cancers; (d) some HR values were extracted from survival curves, which may partly lead to extraction bias.

Conclusion
In general, the high expression of BANCR is significantly associated with shorter OS and poor clinical prognosis, and BANCR may be treated as a biomarker and therapeutic target for cancer. High quality, larger sample size and multicenter studies are needed to further confirm the reliability of this conclusion.
Additional file 1: Figure S1. Forest plot of the relationship between BANCR expression and the number of local tumors (multiple/single). Note: BRAF-activated noncoding RNA; OR: odds ratio; CI: confidence interval; Random: random-effects model. The random-effects model was adopted. The square size of individual studies represented the weight of the study. Vertical lines represent 95% CI of the pooled estimate. The diamond represents the overall summary estimate, with the 95% CI given by its width Additional file 2: Figure S2. Forest plot of the relationship between BANCR expression and smoking status (smoker vs. nonsmoker). Note: BANCR: BRAF-activated noncoding RNA; OR: odds ratio; CI: confidence interval; Fixed: fixed-effects model. The fixed-effects model was adopted.