Clinical and Prognostic Role of Vasohibin-1 Expression in Solid Tumors

The role of Vasohibin-1 (VASH1) expression in solid tumors remains controversial. We thus performed this meta-analysis to elucidate the associations between VASH1 expression and the prognosis of solid tumors. We searched relevant literature in PubMed, Web of Science and EMBASE. The hazard ratio (HR) or odds ratio (OR) and 95% condence intervals (CIs) were measured by xed-effects or random-effects models. Publication bias was assessed using funnel plots and Egger's regression test. The results showed that VASH1 expression exhibited a signicantly decreased overall survival (OS) time (HR = 1.85; 95% Cl = 1.27-2.69) and disease-free survival (DFS) (HR = 1.80; 95% Cl = 1.41-2.29) time. Meanwhile, VASH1 expression was found signicantly associated with TNM stage (OR = 1.96; 95% CI = 1.57-2.46), tumor stage (OR = 2.35; 95% CI = 1.88-2.93), lymph node metastasis (OR = 2.02; 95% CI = 1.37-2.98), venous invasion (OR = 1.63; 95% CI = 1.00-2.65), tumor grade (OR = 1.74; 95% CI = 1.13-2.67) and microvessel density (MVD) (OR = 4.30; 95% CI = 2.31-8.03). However, no signicant association was found between the VASH1 expression and distant metastasis (OR = 1.81; 95% Cl = 0.74-1.41). This study demonstrated that VASH1 and might inferior in solid tumor patients. 33 cancer types including ten rare cancers. VASH1 mRNA expression levels across normal tissues were retrieved from Genotype-Tissue Expression (GTEx) Portal. it is downregulated in adrenocortical carcinoma (TCGA-ACC), bladder urothelial carcinoma (TCGA-BLCA), cervical squamous cell carcinoma and endocervical adenocarcinoma (TCGA-CESC), colon adenocarcinoma (TCGA-COAD), lung adenocarcinoma (TCGA-LUAD), lung squamous cell carcinoma (TCGA-LUSC), ovarian serous cystadenocarcinoma (TCGA-OV), prostate adenocarcinoma (TCGA-PRAD), rectum adenocarcinoma (TCGA-READ), testicular germ cell tumors (TCGA-TGCT), uterine corpus endometrial carcinoma (TCGA-UCEC) and uterine carcinosarcoma (TCGA-UCS).

All records were independently evaluated by three investigators (Ye Tian, Chenkui Miao, Xiaohan Ren) through title, abstract and full-text screening. Any discrepancies about record eligibility were resolved by a discussion between two members as well as a fourth author (Qi Gu). All studies meeting inclusion criteria were retained in the analysis.
For each included record, two investigators independently extracted relevant data to rule out any discrepancies. Disagreements were resolved by a discussion between two investigators. The following data were recorded from all eligible studies: (1) the rst author's name and year of publication, (2) the study nationality, (3) cancer types, (4) sample and pathology type, (5) the cut-off value and assay method, (6) following-up months, (7) the case number of VASH1 relatively-high expression and prognostic outcomes. Those indirectly reported HRs and 95% CIs were extracted from graphical survival plots using Engauge Digitizer version 4.1 (markmitch, Boston, USA) [13].

Quality Assessment
We used the Newcastle-Ottawa scale (NOS) score to independently evaluated the study quality, which allows for the evaluation of methods of patient selection, comparability of study groups, and reporting of essential outcomes [14]. With the score ranging from 0 to 9, studies with scores ≥6 were graded as high quality. Details are available in Table 1.

Statistical analysis
Stata version 16.0 (Stata Corporation, College Station, TX, USA) was used to calculate all statistical analyses. All reported P-values were two-sided, with p<0.05 de ned as statistically signi cant. Testing for the heterogeneity of pooled effects (HR; OR), Cochrane Q-test [15] and the inconsistency index value (I2) [16] were performed in the meta-analysis. The xed-effects model or random-effects model was selected following the heterogeneity of included studies. The funnel plot and Egger's test were used to judge the publication bias.

Acquisition of data from TCGA and GTEx database
The data of gene expression pro le and survival information were obtained from the TCGA database (https://portal.gdc.cancer.gov/), which characterizes 33 cancer types including ten rare cancers. VASH1 mRNA expression levels across normal tissues were retrieved from Genotype-Tissue Expression (GTEx) Portal.

Characteristics of included studies
The owchart shows the literature retrieval and selection process ( Figure 1). Using the search strategy in PubMed, Web of Science, EMBASE and the Cochrane Register of Controlled Trials databases, we initially collected a total of 132 studies, but 63 of them were excluded after screening the titles and abstracts. By further reviewing assessing the remaining studies, 30 studies were excluded for review articles, letters, research with non-human or not relevant to the current analysis. Eventually, 18 studies were identi ed as meeting our inclusion criteria [17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34], whose detailed information was shown in Table 2. The VASH1 relatively-low expression group (1225 individuals; 52.8%) was regarded as the control group. The main assay method of all studies was immunohistochemistry (IHC).

Clinicopathological parameters and VASH1 expression
The pooled OR from 18 studies, including 1092 high VASH1 and 1225 low VASH1 individuals, shown in

Prognostic value of VASH1 expression for DFS and OS
The correlation of VASH1 expression and survival outcomes is shown in Figure 4 and Table 3 Sensitivity analysis aimed at assessing whether the exclusion of any individual study in uenced the overall results. Figure S1 revealed that the results were reliable, indicating that no individual study affected the pooled HR signi cantly.
The expression of VASH1 has a signi cant difference between tumor tissue and normal tissue of multiple cancers

Discussion
As an endogenous inhibitor of angiogenesis, VASH1 may play a pivotal role in maintaining vascular homeostasis, preventing pathological angiogenesis [35]. VASH1 may possess the duality functions -an experiment based on the mice model stated that the decrease of VASH1 resulted in a long lifespan, which contrary to the prime expectation of the experimenters [36]. Meanwhile, the feasible suppression and facilitation of VASH1 on tumors have of late years investigated by scholars and the results remain controversial [37,38].
Meta-analysis can provide more reliable results compared with a single study and serves as a powerful tool to explain controversial conclusions. For this reason, we performed a meta-analysis to clarify whether VASH1 expression has a signi cant impact on the clinical and prognostic outcomes of cancer. To our knowledge, this is the rst complete meta-analysis concerning the role of VASH1 in solid tumors. In our analysis, we found that VASH expression was relevant to more aggressive clinicopathological parameters and might predict inferior DFS and OS in cancer patients.
The result of our analysis showed that high expression of VASH1 might be associated with more inferior TNM stage, tumor stage, lymph node metastasis, distant metastasis and tumor grade, same as the conclusion with many studies, for example, Cao et al. and Zhang et al. [39,40]. Despite the concrete mechanism not being clearly understood, an accountable explanation is that -in the process of tumor angiogenesis, the out-off-balance between VASH1 and VEGF may lead to the formation of numerous new blood vessels with fragmentary structure or lack of basement membrane, although the inhibition role of VASH1 in tumor vascularization. And the noteworthy increase of vascular permeability caused by this imbalance could further accelerate the proliferation and metastasis of cancer cells [41]. Put another way, VASH1 suppresses tumor angiogenesis but not the tumor cells.
MVD, as an independent predictor of poor prognosis, is generally regarded connected with clinical outcome, for instance, tumor grade, pathologic stage et al. [42,43]. Additionally, from the present analysis, an advanced relevance between venous invasion, MVD, and VASH1 expression was found, which was opposite to our expectations. Interestingly, considering that the indispensable effect of VASH1 on maintaining vascular health and inhibiting angiogenesis, this unexpected result is worth a thorough discussion. Firstly, as we know, VASH1 is induced by angiogenic factors, to illustrate, VEGF and FGF-2 [4]. There is a positive relevance between the intensity of VASH1 and VEGFA in tumor cells [31,44]. Nevertheless, with VASH1 degraded and inactivated after its secretion in the tumor microenvironment, the IHC staining of VASH1 in vascular ECs may not demonstrate its anti-angiogenic activity, but only re ect the response of ECs to angiogenic stimulation [45]. Secondly, as a feedback regulator of angiogenesis, VASH1 is up-regulated with VEGFA expression. Yet this endogenous upregulation is not enough to inhibit neovascularization. Researchers found that in tumor tissue, not only MVD expression increased but also the more ripe microvasculature where the VASH1 mainly expressed [46]. These mature vasculature may supply more nutrition for tumors, and therefore facilitate tumor growth and distant metastasis. Consequently, the increase of VASH1 expression may be associated with the enhancement of tumor invasiveness [25,27].
Thirdly, VASH1 might be an activator or inhibitor of angiogenic factor mRNA translation, and this dual functional role may be the reason for VASH1 to inhibit angiogenesis and to pro t ECs survival [47,48].
Finally, our results showed that the higher expression of VASH1 associated with shorter DFS and OS time, which was in line with conclusions from most studies [33,34]. From our perspective, the high expression of VASH1 is related to worse clinicopathological features and thus shorten survival time. However, the results of the OS time showed high heterogeneity, which might be caused by different types of solid tumors.
Furthermore, despite the overall robust statistical evidence generated through this analysis, some limitations have been identi ed. Firstly, owing to the limited quantity of the studies included, further subgroup analysis is unavailable, for example, cancer types. Secondly, IHC based VASH1 detection has limitations for its subjectivity in determining a clear de nition of "positive (or high)" tumor VASH1 staining. Meanwhile, different studies had different de nitions of positive expression of VASH1 ( nal scores, VASH1 density, IHC expression level, et al. ), and thus more uniform standards need to be established. Thirdly, only the Asian population was applicable in this meta-analysis, which might minimize the analyzing value to some level. Moreover, the difference in samples from different institutions might reduce the credibility of our conclusion, which serves as an unavoidable factor in the study.

Conclusion
In conclusion, our meta-analysis suggested that high expression of VASH1 could act as a common maker to predict a high risk of clinical outcomes and poor OS or DFS in patients with a solid tumor. However, based on the limitations of this meta-analysis, large size, and better design studies are needed to validate the clinical role of VASH1.

Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and its additional les.

Competing interests
The author declares that they have no competing interests.

Funding
The study was supported by the National Natural Science YT and MC collected the data and performed the meta-analysis. YT wrote the manuscript. All the authors participated in the data analysis and approved the nal version of the manuscript.      Supplementary Files