The prognostic utility and clinical outcomes of MNX1-AS1 expression in cancers: a systematic review and meta-analysis

Background: Recently, emerging studies have identified that MNX1-AS1 highly expressed among variety of cancers and related with worse prognosis of cancer patients. The purpose of this study was to evaluate the relationship between MNX1-AS1 expression with clinical features and prognosis in different cancers. Methods: In this study, we searched the Web of Science, PubMed, CNKI, and Wanfang databases to find relevant studies of MNX1-AS1. Pooled hazard ratios (HRs) and odds ratios (ORs) with 95% confidence intervals (CIs) were applied to explore the prognostic and clinical significance of MNX1-AS1. Results: A total of 9 literatures were included in this study, including 882 cancer patients. The results showed that patients with increased MNX-AS1 expression were more likely to develop lymph node metastasis (OR = 5.616, 95%CI: 3.093 - 10.199, P = 0.000) and advanced TNM stage (OR = 4.625, 95%CI: 2.366-9.040, P = 0.000). Moreover, we demonstrated that patients with high expression of MNX1-AS1 had poor OS in different cancers (HR = 1.976, 95%CI: 1.653 - 2.361, P=0.000). In addition, patients with high expression of MNX1-AS1 suffer from worse prognosis in gastric cancer (HR = 2.385, 95% CI: 1.838 - 3.094, P = 0.000) and lung cancer (HR= 1.959, 95%CI: 1.353 - 2.835, P = 0.000). Conclusions: High MNX1-AS1 expression is significantly associated with unfavorable clinical outcomes and it has the potential to serve as a prognostic biomarker in cancer patients. homeobox cell lymph node competitive


Background
Long non-coding RNA (LncRNA) is a novel type of non-coding RNA with longer than 200 nucleotides, and has the characteristics of lacking protein-coding capacity and open reading frames [1][2][3]. Growing evidence from recent studies has indicted that in different oncogenic signaling pathways, lncRNAs mediate multiple molecular mechanisms, including alternative splicing of precursor mRNAs, functions of competing for endogenous RNA, promotion of epithelial-mesenchymal transition (EMT), epigenetic modification and cells phenotypic changes, etc., and finally promote the occurrence and progression of malignant tumors [1,2,4,5].
In addition, numerous independent studies have investigated the prognostic performance of MNX1-AS1 expression among various cancers. However, most published findings were limited by sample size and research methods. Therefore, it is imperative to conduct a comprehensive meta-analysis to systematically assess relationship of MNX1-AS1 expression with patient clinical characteristics and prognosis in different malignant cancers by reviewing all related published studies.

Materials And Methods Retrieval Strategy
This study searched the Web of Science, PubMed, CNKI, and Wanfang databases for getting related articles up to January 31, 2020. Search by matching the following keywords: lncRNA, MNX1-AS1, cancer, tumor, malignancy, neoplasm, lymphoma, leukemia. Two researchers (JL and ZYZ) screened strictly the related literatures according to the inclusion and exclusion criteria. Exclusion criteria: (a) the study did provide Kaplan-Meier survival curves. (b) incomplete data made it impossible to obtain hazard ratio (HR) data. (c) sample size of the original study less than 40. (d) nonhuman research, letters, case reports, meta-analysis, review articles and other studies.

Quality Evaluation And Data Extraction
Two investigator (JL and ZYZ) evaluated the quality of the included literature independently in accordance with quality evaluation criteria of Newcastle-Ottawa (NOS). The qualitied articles with NOS ≥ 6 scores were ultimately regard as high-quality or eligible studies. The contents of the data extraction are as follows: first author, publication year, country, number of cases, and patient clinical parameters including: age, gender, TNM stage, differentiation, distant metastasis (DM), tumor size and lymph node metastasis (LNM), overall survival rate, HR (hazard ratio) and 95% CI. Engauge Digitizer v4.1 software extracted data of Kaplan-Meier curve to obtaining HR and 95%CI by the study of Tierney et al. [19]. Two researchers (JL and ZYZ) extracted the relevance data independently. When they encounter disagreements, they would discuss with a third researcher (WJ) to decide the final results.

Statistical analysis
The odds ratios (ORs) and hazard ratios (HRs) and their 95% confidence intervals (CIs) were combined to quantitatively evaluate the correlation of MNX1-AS1 expression with clinical parameters and prognostic implication of cancer. The heterogeneity across the studies was assessed by using Cochrane Q-test and I 2 statistics. For P > .05 as well as I 2 < 50%, the fixed-effect model was adopted to pool the related results. Otherwise, the random-effect model was chosen. Sensitivity analysis assessed the stability of pooled HRs/ORs value. Begg and Egger tests, as well as funnel plots were used to appraise publication bias. All P values less than 0.05 was considered statistically significant.
Statistical analysis was performed using Stata 11 software.

Results
Search results of eligible studies After preliminary retrieval, we retrieved 44 literatures in total from online databases. We did not obtain the related studies among Chinese databases. Ultimately, 9 articles were contained in the current study. The selection process of 9 eligible articles presented in Fig. 1. A total of 882 patients from the all eligible articles were included to evaluate prognostic value of MNX1-AS1 (presented in Table 1). All articles covered gastric cancer(GC) (n = 3) [8][9][10], lung cancer (LC) (n = 3) [6,20,21], other cancer(n = 3) including: esophageal squamous cell carcinoma (ESCC), cervical cancer (CC), epithelial ovarian cancer (EOC) [14,18,22]. The summary of SNHG12 expression and clinical features presented Table 2. As shown in Table 3, pooled analyses showed that MNX1-AS1 expression did not correlate significantly with age or gender, which implied that the original study gender and age were comparable. The heterogeneity among studies was not statistically significant (Age: I 2 = 0.0%, P = 0.888; Gender: I 2 = 0.0%, P = 0.696).

Publish Bias And Sensitivity Analysis
In this study, funnel plot, Egger regression and Begg rank correlation tests were used to evaluate potential publication bias. As shown in Table S1, there was no publication bias among studies on clinical parameters. Begg and Egger tests also demonstrated that no significant publication bias was found for OS (Egger's test: P = 0.058; Begg's test: P = 0.150), and the corresponding funnel plot is shown in Fig. 3. The results of sensitivity analysis provided that after excluding any of the studies, the value of the combined HRs or ORs did not change significantly, indicating that the pooling results were stability and robustness.
MNX1-AS1 also induced proliferation and repressed apoptosis of cervical cancer cells through MAPK signaling pathway [18]. Overall, MNX1-AS1 might facilitate carcinogenesis of different cancer by exerting its carcinogenic molecular function to regulate the expression of miRNA and other related genes or protein and activate the corresponding signal pathway [23,25]. Therefore, future research can deeply explore the potential of MNX1-AS1 as a biomarker, which is valuable for evaluating prognosis, early diagnosis, and targeted therapy of different cancers.
In recent years, several published studies reported that the association of MNX1-AS1 expression with prognosis, tumor size, lymph node metastasis, distant metastasis, differentiation, as well as TNM stage. Consequently, we performed a meta-analysis with the goal of comprehensively assessing the prognostic value of MNX1-AS1. The study is the first study on the expression of MNX1-AS1 and the prognosis of cancer. The combined results presented that patients with MNX1-AS1 over-expression were susceptible to suffer from unfavorable prognosis in gastric cancer and lung cancer. The pooling results of the study are consistent with most of the original literatures, which recommend that MNX1-AS1 high expression might be associated with worse prognosis of malignant cancer patients [6,8,9].
Moreover, the sensitivity analysis proved the stability and reliability of the pooled results. The heterogeneity of gastric cancer (I-squared = 0.0%, P = 0.943), lung cancer (I-squared = 0.0%, P = 0.508), and sample sizes ≥ 100 (I-squared = 0.0%, 0.559) subgroups was lower than the overall heterogeneity in the study. Therefore, subgroup analysis of this study demonstrated that the heterogeneity among the included studies stems from cancer type and sample size. Additionally, MNX1-AS1 expression was significantly related to lymph node metastasis and TNM stage. However, further studies on the possible relationship of MNX1-AS1 expression with tumor size, distant metastasis or differentiation are needed. Therefore, based on the above evidence, MNX1-AS1 may be a novel candidate biomarker of cancer prognosis.
There are several limitations in the study. First, the cut-off values of MNX1-AS1 expression were different across studies. Second, cancer type and sample size might explain source of heterogeneity.
Finally, HRs and 95%CIs of several studies were extracted from the Kaplan-Meier curves. Therefore, larger-scale well-designed studies are needed to confirm the result of this study.

Conclusions
In conclusion, we indicated that high MNX1-AS1 expression is associated with unfavorable OS and poor clinical outcomes in patients with different types of cancer.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.
Supplementary file Table S1 The summary of publish bias.docx