Gastric cancer (GC) is one of the most common malignant tumors in the digestive system. It has become the third leading cause of cancer death worldwide. Over 70% of GC cases occur in developing countries, and roughly half of the world’s total occurs in eastern Asia (chiefly in China) . Most GC patients present with advanced tumor due to the inconspicuous symptoms of early onset GC and the limited diagnostic conditions, making the sufferers lose the optimal opportunity for radical cure . For these patients, systemic treatments, such as chemotherapy, are the main treatment option. Although great progress has been made for advanced GC chemotherapy in recent years, mortality is still unacceptably high. Therefore, searching for ideal diagnostic biomarkers and novel therapeutic targets remains critical for the treatment of GC.
Hippo signaling pathway is an evolutionarily conserved regulator for organ size control and tissue growth. Accumulating literature suggests that dysregulation of Hippo pathway leads to proliferation and anti-apoptosis associated with increased cancer risk   . As a pivotal downstream effector of Hippo signal cascade, Yes-associated protein (YAP) was considered as an oncoprotein, and its overexpression and accumulation in the nucleus were closely related to the poor clinical outcomes of various tumors including gastric cancer  . Moreover, YAP was even reported as a potential target for GC therapy . Nevertheless, the published clinical studies showed the data were still controversial, and the opposite role of YAP in GC was also reported . Therefore, we conducted this meta-analysis to comprehensively assess the relationship between YAP overexpression and gastric cancer.
2.1. Search Strategy
We performed a systematic literature search in PubMed, Embase, Web of Science databases and the Chinese National Knowledge Infrastructure (CNKI) from inception up to December 2018. Relevant studies were identified using a combination of the following terms: “YAP” or “Yes-associated protein” or “Yes protein” or “Hippo” and “gastric cancer” or “gastric carcinoma” or “gastric neoplasm” or “stomach cancer” or “GC”. To availably identify relevant studies, we also manually searched for references cited in the eligible articles. When two studies had partial overlaps, both studies should be considered.
2.2. Inclusion and Exclusion Criteria
The eligible literature in this study fulfill the following inclusion criteria: 1) patients were diagnosed as gastric cancer; 2) YAP expression was quantified by immunohistochemistry or other adequate methods; 3) sufferers were categorized into high YAP (or YAP-positive) and low YAP (or YAP-negative) groups; 4) the association between YAP expression and clinicopathological features was described, or YAP expression in human tumor and adjacent tissues was detected; 5) the search was restricted to human studies published in English or Chinese. Studies were excluded if they met any of the following exclusion criteria: a) reviews, case reports, comments, conference abstracts, letters, or laboratory studies; b) published in a language other than English or Chinese; c) insufficient data available to estimate the association.
2.3. Data Extraction
Required data were extracted by two investigators independently based on the inclusion criteria listed above. Any discrepancies in data extraction were evaluated by discussion to reach a consensus. The extracted data included the first author’s name, year of publication, country of origin, distribution of gender and age in patients, number of patients, staining location, method of detection, YAP expression in gastric cancer and adjacent normal tissue, and clinicopathological features.
2.4. Assessment of Study Quality
Two investigators independently evaluated the quality of each study according to Newcastle-Ottawa Scale (NOS) . The NOS includes three parameters of quality for studies: selection of the study population, comparability of subjects, and exposure assessment, with scores ranging from 0 to 9 (Additional file 1: TableS1). NOS scores > 5 was considered as high-quality studies.
2.5. Statistical Analysis
We implemented the meta-analysis based on the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) Checklist (Additional file 2: TableS2). The odds ratios (ORs) and corresponding 95% confidence intervals (95% CIs) were used to assess the association between YAP expression and gastric cancer or the clinicopathological features of gastric cancer. The heterogeneity among eligible studies was evaluated by the Q-test. P-value < 0.1 indicated that the heterogeneity was significant. I2 values were also calculated to quantify the heterogeneity: I2 < 25%, 25% < I2 < 50%, 50% < I2 < 75%, and I2 > 75%, indicated no heterogeneity, moderate heterogeneity, large heterogeneity, and extreme heterogeneity, respectively. When P-value > 0.1 and I2 < 25%, the heterogeneity was considered not significant, and then the pooled OR and 95% CI were assessed by the fixed-effects model; otherwise, the random-effects model was performed . The sensitivity analysis was carried out by sequential omission of individual studies to test the stability of meta-analysis results. Moreover, Begg’s test and Egger’s test were used to evaluate the publication bias; P-value < 0.05 indicated the presence of publication bias. All statistical analyses were performed using the software STATA version 12.0 (Stata Corporation, College Station, TX, USA).
3.1. Study Characteristics
The flow chart of the study selection process was presented in Figure 1. A total
Figure 1. Flow chart of study selection.
of 432 relevant publications were retrieved after the initial database searches. According to the inclusion and exclusion criteria, data from 16 studies including 2229 patients were included in this meta-analysis   - . As shown in Table 1, the eligible studies were published between 2009 and 2018, and sample sizes ranged from 53 to 302. 12 studies (75.00%) reported on Chinese, 3 studies (18.75%) on Koreans, and only 1 study (6.25%) on Japanese. YAP expression was detected by immunohistochemistry in all eligible studies, among them, formalin-fixed and paraffin-embedded (FFPE) samples were used for immunohistochemical staining in 8 studies, while the other 8 studies used tissue microarray (TMA). Of the eligible publications, 13 studies reported nuclear and cytoplasmic staining of YAP (overall YAP expression), and 3 studies detected YAP expression only by nuclear staining. Additionally, the quality of each eligible study was assessed according to the NOS, and all articles were of high quality (NOS score > 5).
3.2. The Difference of YAP-Positive Expression between GC and Adjacent Tissues
YAP expression in GC and adjacent non-tumor tissues was detected simultaneously in 8 studies        . As depicted in Table 2, the eight studies included 925 GC samples (case) and 728 paracancerous samples (control). YAP-positive rate in GC was much higher than that in normal tissues (OR = 8.08, 95% CI = 4.41 - 14.80). Moreover, the results of meta-analysis showed that 95% confidence intervals of combined study, as well as each independent
Table 1. Characteristics of the studies included in the meta-analysis.
Abbreviations: YAP, Yes-associated protein; TMA, tissue microarray; FFPE, formalin-fixed and paraffin-embedded; NOS, Newcastle-Ottawa Scale.
Table 2. YAP expression in gastric cancer and adjacent non-tumor tissue.
study, exceeded “1”, indicating that positive YAP expression was correlated with gastric cancer but not adjacent non-tumor tissue (Figure 2(a)).
3.3. Association of YAP Overexpression with GC Clinicopathological Features
The main results of meta-analysis and heterogeneity test for the association study of YAP overexpression with gastric cancer clinicopathological features were summarised in Table 3. The elevated YAP expression was correlated with more advanced TNM stage (OR = 2.68, 95% CI = 1.61 - 4.48) (Figure3(a)), deeper invasion depth (OR = 2.05, 95% CI = 1.32 - 3.19) (Figure3(b)), and lymph node metastasis (OR = 1.95, 95% CI = 1.29 - 2.96) Figure3(c)) in gastric cancer. Moreover, no significant correlation was observed between YAP overexpression and degree of differentiation (OR = 1.17, 95% CI = 0.63 - 2.16) (Figure3(d)), as well as gender of patients (OR = 1.12, 95% CI = 0.91 - 1.37) (FigureS1(a)) or tumor size (OR = 1.11, 95% CI = 0.82-1.49) (FigureS1(b)) of gastric cancer.
Figure 2. The difference of YAP-positive expression between gastric cancer and adjacent tissues. (a) Forest plot for the association between YAP-positive expression and gastric cancer; (b) Sensitivity analysis for the association between YAP-positive expression and gastric cancer.
Table 3. The correlation of YAP overexpression with GC clinicopathological features.
Notes: aMale/female; b>5 cm/<5 cm; cIII + IV/I + II; dT3 + T4/T1 + T2; eYes/No; fLow differentiation/middle and high differentiation.
Figure 3. The association of elevated YAP expression with clinicopathological features of GC. Forest plots for the association between elevated YAP expression and (a) TNM stage, (b) invasion depth, (c) lymph node metastasis, or (d) degree of differentiation in GC.
3.4. Sensitivity Analysis
We conducted sensitivity analysis by sequential omission of individual studies to probe the change in the odds ratio and 95% confidence interval of meta-analysis. As shown in Figure2(b), Figures 4(a)-(d) and FigureS1(c) & FigureS1(d), no significant difference was observed when any of the studies was excluded in all correlation assessments, indicating the reliability and stability of the meta-analysis.
3.5. Publication Bias
Begg–Mazumdar adjusted rank correlation test and Egger’s regression test were performed to assess the publication bias. The results showed that the shape of Begg’s funnel plots appeared to be symmetrical (data not shown). Meanwhile, the P-values were all greater than 0.05 in both Begg’s test and Egger’s test (Table 3). These results suggested the absence of significant publication bias in the overall meta-analysis.
Figure 4. Sensitivity analysis for the association of elevated YAP expression with clinicopathological features of GC. Sensitivity analysis for the association between elevated YAP expression and (a) TNM stage, (b) invasion depth, (c) lymph node metastasis, or (d) degree of differentiation in GC.
Gastric cancer is characteristic of poor prognosis and high death rate, and there is still a great need to identify diagnostic markers as well as develop novel therapeutic strategies for GC therapy. Hippo pathway regulates tissue growth and organ size via YAP-TEAD complex. Inactivation of Hippo cascade leads to the elevated expression and nucleus accumulation of YAP, which is significantly associated with poor clinical outcomes of most cancers. Therefore, Hippo cascade acts as a tumor suppressor pathway, and YAP is considered to be an oncoprotein in multiple cancers including GC. There are many published clinical data supporting this conclusion. Zhang reported that YAP was strongly expressed in GC, and knockdown of YAP could inhibit the proliferation and metastasis of GC cells . The similar results could also be observed in the study of Zhou and his colleagues . Moreover, Yan showed that YAP acted as a tumor promoter in gastric cancer, and involved in the survival and migration of GC cells through the activation of the SIRT1/Mfn2/mitochondrial autophagy axis . However, there are still inconsistent results published. Suh and colleagues supported that YAP functioned as a tumor suppressor in GC . Zhang also indicated that there was no significant correlation between YAP expression and clinicopathological characteristics in GC, and YAP was not a potential marker for diagnosis or prognosis of GC . In this account, a meta-analysis including 16 studies was performed to comprehensively evaluate the relevance of elevated YAP expression with GC and its pathological parameters. The results of pooled data showed a significant correlation between positive YAP expression and GC, but not adjacent non-tumor tissue. Furthermore, the elevated YAP expression in GC was closely related to more advanced TNM stage, deeper invasion depth and lymph node metastasis.
The heterogeneity and several limitations of this meta-analysis should be acknowledged. First, all of the eligible studies in this meta-analysis were carried out in Asian population, therefore, it might be insufficient to provide support for other ethnic groups. Second, the evaluation criteria of positive YAP expression differed among included studies, which might influence the results of pooled data and contribute to the heterogeneity. Third, the data we extracted contain both YAP nuclear staining and overall YAP expression, the difference in staining location is also a potential heterogeneous source due to the expression characteristics of YAP in different cell states.
In conclusion, this meta-analysis demonstrated that YAP-positive expression in gastric cancer was significantly higher than that in adjacent non-tumor tissues. Additionally, the overexpression of YAP was closely correlated with more advanced TNM stage, deeper invasion depth and lymph node metastasis. Therefore, YAP may be a promising diagnostic marker and even a therapeutic target for GC. However, the results of this study should be interpreted cautiously due to the existence of heterogeneity and limitations. Hence, well-designed prospective studies based on larger sample sizes, as well as the corresponding basic research are still warranted to validate the present findings.
This work was supported by the Fundamental Research Funds of Science & Technology Department of Sichuan Province (2017YSKY0001), the Scientific Research Subject of Health and Family Planning Commission of Sichuan Province (17PJ582), and the Applied Basic Research Fund of Science & Technology Department of Sichuan Province (2018JY0652).
Table S1. Newscastle-ottawa quality assessment scale.
Table S2. PRISMA 2009 Checklist.
Figure S1. The association of elevated YAP expression with clinicopathological features of GC. Forest plots for the association between elevated YAP expression and (a) gender, or (b) tumor size in GC. Sensitivity analysis for the association between elevated YAP expression and (c) gender, or (d) tumor size in GC.