The Hippo Signaling Core Components YAP and TAZ as New Prognostic Factors in Lung Cancer

Background The Hippo pathway is an essential signaling cascade that regulates cell and organ growth. However, there is no consensus about (i) the expression levels of the Hippo signaling core components yes-associated protein (YAP) and transcriptional co-activator with PDZ-binding motif (TAZ) in lung cancer, especially in small cell lung cancer (SCLC), or (ii) their association with the prognosis of patients with SCLC. Methods We screened relevant articles and identified eligible studies in the PubMed, EMBASE, COCHRANE, and WanFang databases. A combined analysis was performed to investigate (i) the expression levels of the major effectors, YAP and TAZ, in lung cancer and its subsets and (ii) their prognostic role in lung cancer, especially in SCLC. Results In total, 6 studies related to TAZ and 13 studies concerning YAP were enrolled in this meta-analysis. We found that high TAZ expression was significantly associated with poor overall survival (OS) of patients with non-small cell lung cancer (NSCLC) in the overall population [Ph < 0.001, crude hazard ratio (HR) = 1.629, 95% CI = 1.199–2.214 for TAZ expression; Ph = 0.029, adjusted HR = 2.127, 95% CI = 1.307–3.460 for TAZ], the Caucasian population (Ph = 0.043, crude HR = 1.233, 95% CI = 1.030–1.477 for TAZ expression), and the Asian population (Ph = 0.551, adjusted HR = 2.676, 95% CI = 1.798–3.982 for TAZ). Moreover, there was a significant negative association between YAP expression and an unsatisfactory survival of patients with lung cancer (Ph = 0.327, crude HR = 1.652, 95% CI = 1.211–2.253 for YAP expression) and patients with NSCLC [disease-free survival (DFS): Ph = 0.693, crude HR = 2.562, 95% CI = 1.876–3.499 for YAP expression; Ph = 0.920, crude HR = 2.617, 95% CI = 1.690–4.052 for YAP-mRNA; OS: Ph = 0.878, crude HR = 1.777, 95% CI = 1.233–2.562 for YAP expression], especially in the Asian population (DFS: Ph = 0.414, crude HR = 2.515, 95% CI = 1.755–3.063; OS: Ph = 0.712, crude HR = 1.772, 95% CI = 1.214–2.587). However, no association was observed in the multivariate combined analysis. High YAP expression was significantly associated with short OS of patients with SCLC in our combined multivariate analysis in the Asian population (Ph = 0.289, crude HR = 4.482, 95% CI = 2.182–9.209), but not with crude data (Ph = 0.033, crude HR = 1.654, 95% CI = 0.434–6.300). Conclusion The Hippo pathway is involved in carcinogenesis and progression of NSCLC and SCLC, and high expression levels of YAP and TAZ are independent and novel prognostic factors for lung cancer.


INTRODUCTION
Lung cancer is a primary malignancy with one of the highest rates of morbidity and mortality in China (1). According to its histological classification, lung cancer is stratified into small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC), of which the latter subgroup accounts for ∼80% of lung cancer cases (2). While emerging molecular-targeted and oncoimmunological therapeutics are increasingly used in the clinic (2, 3), the survival rates of patients need to be further improved.
The Hippo pathway is an important signaling cascade related to cell proliferation, apoptosis, and differentiation. Yes-associated protein 1 (YAP1) and transcriptional co-activator with PDZbinding motif (TAZ) are transcription co-activators and the core downstream effectors of the Hippo signaling pathway (4). When the Hippo pathway is activated, the two molecules are phosphorylated and retained in the cytoplasm in normal cells. On the contrary, YAP/TAZ can translocate into the nucleus to activate the transcription of multiple oncogenes in tumor cells (5). Meanwhile, they are frequently upregulated and biologically function as oncogenes via participation in multiple cancerrelated processes, including cancer cell growth, migration, and distal metastasis in various malignancies, such as colorectal, gastric, and lung cancer (6)(7)(8).
Several studies have shown that the Hippo pathway is activated upon carcinogenesis and lung cancer progression (9)(10)(11). However, different YAP and TAZ expression levels have been reported in lung cancer and its subsets. Moreover, no consensus has been reached with respect to their prognostic values in patients with lung cancer. Yang et al. reported that YAP was not associated with the disease stage and survival of patients with SCLC (12). In another study, higher expression of YAP1 was associated with worse progression-free survival in patients with epidermal growth factor receptor (EGFR)-mutant NSCLC treated with first-line EGFR tyrosine kinase inhibitors (13). Although higher TAZ mRNA and protein levels were associated with shorter survival of patients with NSCLC (14), the nuclear localization of TAZ was increased, which was correlated with poor prognosis in lung squamous cell carcinomas, but not lung adenocarcinomas (15).
To better understand the prognostic role of the Hippo pathway in lung cancer, we investigated YAP and TAZ expression in lung cancer and analyzed their associations with the prognosis of patients with lung cancer in this meta-analysis.

MATERIALS AND METHODS
The Medical Ethics Committee of Fuzhou Second Hospital affiliated with Xiamen University approved this study. To obtain studies for our meta-analysis, we screened and identified the relevant articles according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (16). First, the following words and phrases were used to identify candidate articles in the PubMed, EMBASE, COCHRANE, and WanFang databases: "lung cancer, " "lung squamous cell carcinomas, " "lung adenocarcinomas, " "non-small cell lung cancer, " "small cell lung cancer, " "YAP1, " "YAP, " "Yes-associated protein, " "Hippo pathway, " "transcriptional coactivator with PDZ-binding motif, " "TAZ, " "WWTR1, " "survival, " "outcome, " "prognosis, " "progression-free survival, " "recurrencefree survival, " "disease-free survival, " and "overall survival." The retrieval deadline was August 30, 2021. We also screened references of the relevant reports to obtain additional studies. After reading the title and abstract of the candidate articles, we identified the relevant studies. Finally, we identified eligible studies by reading the full texts of the relevant articles. The inclusion criteria were as follows: (1) the study reported the relationship between YAP or TAZ and prognosis of lung cancer and (2) the study provided clinical baseline characteristics and reported the results in hazard ratios (HRs) and 95% CIs. Articles that did not provide detailed survival data and comments were excluded, along with letters, reviews, and meta-analyses.
The following clinical baseline characteristics were extracted from each study: the first author's name, study design, country, race, recruitment time, disease, number of included cases, treatment, outcome, HR, and 95% CI. All data were rechecked.
The HR and 95% CI were selected as parameters to assess the strength of the association between YAP1 or TAZ and the prognosis of the cases with lung cancer. The Q test and estimated I 2 were selected to evaluate the heterogeneity of included studies in this meta-analysis, where P h < 0.1 or I 2 > 50% was considered to indicate substantial heterogeneity. The combined analysis of eligible studies was assessed by the Z test based on a fixed (P h > 0.1) or random (P h < 0.1) model. Publication bias within the included studies was evaluated by funnel plots (17). The Stata v.11.0 software (Stata Corporation, College Station, TX, USA) was used for all statistical analyses, and p < 0.05 was considered to indicate statistical significance.
The combined HR and 95% CI of TAZ expression and overall survival (OS) of patients with NSCLC are described in Figure 2 and Table 2. TAZ expression and OS of patients with NSCLC were reported for a total of 1,977 patients over five studies. Two studies reported TAZ mRNA levels, and three studies were conducted in the Asian population. The crude and adjusted HR and 95% CI were extracted from five and three studies, respectively. The combined results showed that high TAZ expression was significantly associated with poor OS in the overall population (P h < 0.001, I 2 = 87.8%, crude HR = 1.629, 95% CI = 1.199-2.214 for TAZ and TAZ; P h = 0.029, I 2 = 66.7%, adjusted HR = 2.217, 95% CI = 1.307-3.460 for TAZ), the Caucasian population (P h = 0.043, I 2 = 68.3%, crude HR = 1.233, 95% CI = 1.030-1.477), and the Asian population (P h = 0.551, I 2 = 0.00%, adjusted HR = 2.676, 95% CI = 1.798-3.982).
In the combined analysis of YAP expression and survival of patients with lung cancer, the OS of the patients with lung cancer with high YAP expression was significantly shorter than in patients with low YAP expression in the overall population (P h = 0.327, I 2 = 13.6%, crude HR = 1.652, 95% CI = 1.211-2.253) (Figure 3; Table 3), and even more so in the Asian population (P h = 0.218, I 2 = 30.6%, crude HR = 1.642, 95% CI = 1.195-2.257). The significant negative association between YAP expression and survival of patients  . However, no association between them was observed in the multivariate combined analysis. High YAP expression was significantly associated with short OS of patients with SCLC in our combined multivariate analysis in the Asian population (P h = 0.289, I 2 = 11.10%, crude HR = 4.482, 95% CI = 2.182-9.209), but not with the crude data (P h = 0.033, I 2 = 78.00%, crude HR = 1.654, 95% CI = 0.434-6.300) (Figure 3; Table 3). In our study, relatively symmetric funnel plots were observed in our prognostic comparisons of patients with high-and low-TAZ NSCLC. Moreover, symmetric funnel plots were also found in the combined survival analysis of the patients with high-and low-YAP lung cancer, NSCLC, and SCLC (Figure 4).

DISCUSSION
The association of YAP and TAZ expression and lung cancer survival, particularly in patients with SCLC, remains unclear. This meta-analysis revealed that TAZ is significantly associated with poor OS of patients with NSCLC in the overall population and the Asian population. Moreover, a significant association was observed between high YAP expression and unsatisfactory survival of patients with lung cancer and patients with NSCLC, especially in the Asian population. However, no association between them was observed in the multivariate combined analysis. YAP expression was significantly associated with short OS in our combined analysis in the Asian SCLC population based on multivariate data, but not crude data.
Lung cancer is a complex and challenging disease that interacts with environmental and genetic factors (29). Many signaling pathways, including the Hippo pathway, are involved in lung cancer tumorigenesis and progression (30)(31)(32)(33)(34). YAP and TAZ are transcriptional co-factors. They are considered oncogenes in both NSCLC and SCLC (35,36). YAP was found to mainly regulate lung cancer cells' growth and proliferation. In contrast, TAZ mainly promotes the migration of cancer cells (11). A recent meta-analysis performed by Feng et al. revealed that overexpression of TAZ is a predictive factor of poor prognosis and is associated with advanced TNM stage, poor tumor differentiation, and lymph node metastasis in various cancers (37). However, no association between TAZ expression and survival of patients with NSCLC was observed in a combined analysis of two studies (37). In our meta-analysis, TAZ was highly expressed in patients with NSCLC in most studies. The combined results showed that high TAZ expression is associated with a worse prognosis of patients with NSCLC in the overall and Asian populations. These findings suggest that TAZ is involved in NSCLC carcinogenesis and progression and that it is an independent prognostic factor of NSCLC. Three meta-analyses reported the predictive role of YAP in clinical outcomes in various cancers (38)(39)(40). They found that both overall and nuclear YAP overexpression are intimately associated with worse OS and DFS in patients with malignancies (38,40). Wu et al. reported no association between YAP expression and survival of lung cancer cases in their combined meta-analysis (40). However, a meta-analysis including six eligible studies showed that high nuclear expression of YAP1 was associated with shorter survival outcomes in patients with NSCLC (39). In the present study, we found an association between YAP expression and unsatisfactory survival of patients with lung cancer in the overall Asian population in our combined analysis with crude HR and 95% CI. Moreover, YAP expression at the protein and mRNA levels was significantly associated with the survival of patients with NSCLC using unadjusted HR and 95% CI. However, due to the small sample size and different confounding factors in each included study, we did not observe associations between them in combination with multivariate data. These findings demonstrate that YAP is an essential factor promoting the progression of lung cancer and NSCLC and that it is a novel prognostic factor for the disease. Due to the low number of studies in the Asian population in our combined analysis, we found no relationship between YAP expression and OS of patients with SCLC in our univariate analysis, whereas YAP expression was significantly associated with short OS of patients with SCLC in our combined analysis with multivariate data in the Asian population, indicating that YAP can also be considered a predictor of poor survival of patients with SCLC.
To the best of our knowledge, this meta-analysis used the largest sample size to date to analyze the prognostic roles of TAZ and YAP in lung cancer and patients with NSCLC. In addition, this is the first study to report the role of YAP in the prognosis of patients with SCLC. However, there are several limitations to this study. First, no studies on the association between TAZ expression and survival of patients with SCLC were included, so the prognostic role of TAZ in SCLC remains to be elucidated. Second, the sample size of included studies related to YAP expression in patients with SCLC was small. Our findings should be validated by multi-center clinical trials with larger sample sizes. It is also important to emphasize that most studies were conducted in the Asian population. Third, patients with different TNM stages and various treatment options were enrolled in each study. These factors might influence the combined results and hence, the results should be validated by prospective studies with patients with NSCLC or SCLC in specific TNM stages. Fourth, different expression models of YAP or TAZ, different detection methods, and different genetic backgrounds of the patients enrolled in the included studies might lead to inconsistent results.
In summary, YAP and TAZ function as oncogenes in both NSCLC and SCLC and aberrant protein expression levels of YAP and TAZ are independent and novel prognostic factors for these diseases. Further studies are warranted to validate the findings in the Asian population and explore effective biomarkers to predict the prognosis of patients.

AUTHOR CONTRIBUTIONS
YJ is responsible for the concept or design of the work. W-JX and R-WC are responsible for data collection. W-LY and HC are responsible for drafting the article. W-XC and J-PX are responsible for making important revisions to the article. All authors contributed to the article and approved the submitted version.

FUNDING
The study was supported by the Fujian Provincial Natural Science Foundation (2018J01360).