Prognostic value of nectin-4 in human cancers: A meta-analysis

Background Many reports have described that abnormal nectin-4 expression may be used as a prognostic marker in many tumors. However, these studies failed to reach a consensus. Here, we performed a meta-analysis to comprehensively evaluate the prognostic value of nectin-4 in cancers. Methods Relevant studies were identified through a comprehensive search of PubMed, EMBASE and Web of science until August 31, 2022. Pooled hazard ratios (HRs) with 95% confidence intervals (CIs) were used to evaluate the relationship between nectin-4 expression and overall survival (OS) and disease-free survival/progression-free survival/relapse-free survival (DFS/PFS/RFS). Odds ratios (ORs) with 95% CIs were applied to assess the relationship between nectin-4 expression and clinicopathologic features. Subgroup analysis was performed to explore the sources of heterogeneity. Sensitivity analysis and funnel plot were used to test the reliability of the results. All data analyses were performed using STATA version 12.0 software. Results Fifteen articles involving 2245 patients were included in the meta-analysis. The pooled analysis showed that high nectin-4 expression was significantly associated with poor OS (HR: 1.75, 95% CI: 1.35–2.28). There was no relationship between high nectin-4 expression and DFS/PFS/RFS (HR: 178, 95% CI: 0.78–4.08).Subgroup analyses revealed that that high nectin-4 expression mainly presented adverse OS in esophageal cancer (EC) (HR: 1.78, 95% CI: 1.30–2.44) and gastric cancer (GC) (HR: 1.92, 95% CI: 1.43–2.58). We also found that high nectin-4 expression was associated with tumor diameter (big vs small) (OR: 1.96, 95% CI: 1.02–3.75), tumor stage (III-IV vs I-II) (OR: 2.04, 95% CI: 1.01–4.12) and invasion depth (T3+T4 vs T2+T1) (OR: 3.95, 95% CI: 2.06–7.57). Conclusions Nectin-4 can be used as an effective prognostic indicator for specific cancers.


Introduction
Cancer has surpassed all other diseases and become the leading cause of death worldwide. The morbidity and mortality of cancer have shown an upward trend and placed huge burdens on the medical system. There were 18.1 million new cancer cases and 9.6 million cancer deaths worldwide in 2018 (1). Infiltration and metastasis are vital biological behaviors that affect tumor progression. Cell adhesion molecules play crucial roles in this process. Understanding the mechanism of tumor invasion and metastasis can help to adopt appropriate strategies for timely prevention and treatment.
In recent years, the nectin family has been found to be involved in cell adhesion (2). Nectins belong to the immunoglobulin superfamily of Ca+ independent cellular adhesion molecules, and can bind to the actin cytoskeleton through a filamentous actin binding protein called afadin (2,3). Nectin family members are involved in cell-cell connections and regulate cell activities including polarization, differentiation, migration, proliferation and survival (4).Nectin family members include nectin-1, -2, -3, and -4. Nectin-1, -2, and -3 are widely expressed in human tissues, while nectin-4 is mainly found in embryos and tumor tissues (5).
The clinical value of nectin-4 as a potential tumor target in human tumors has not been fully described. Therefore, we performed a comprehensive meta-analysis to evaluate the prognostic role of nectin-4 in human tumors.

Search strategy
Three researchers(Rongqiang Liu, Kailiang Zhao and Kunpeng Wang) independently searched Pubmed, Embase and Web of science databases to find relevant articles about the prognostic significance of nectin-4 in human cancers. The search deadline was August 31, 2022. The following keywords were used: "Nectin-4" OR "PVRL4" OR "poliovirus-receptor-like 4" AND "cancer" OR "carcinoma" OR "neoplasm" OR "tumor" OR "tumour". Language restriction was set in English. We manually searched the references included in the studies.

Study selection
Articles were included in this meta-analysis if they met the following criteria: (1) investigated the association between nectin-4 expression and survival outcomes of cancer patients; (2) provided sufficient data to compute hazard ratios (HRs) or odds ratios (ORs) with 95% confidence intervals (CIs); (3) detected nectin-4 expression in human tumor tissue or serum; (4) articles published in English. The exclusion criteria were: (1) insufficient data to calculate HRs and 95% CIs; (2) case reports, conference papers, reviews, letters and articles published in non-English languages; (3) animal or cell experiments; (4) data from public databases.

Data extraction and quality assessment
Three researchers (Rongqiang Liu, Kailiang Zhao and Kunpeng Wang) independently extracted the data from the included articles. Disagreement was settled by discussion. The following information was extracted: name of the first author, publication year, country, study type, tumor type, detected sample, detection method, survival analysis, analysis type and source of HRs. Multivariate analysis was adopted due to its higher accuracy. The quality of each study was assessed using the Newcastle-Ottawa Quality Assessment Scale (NOS) (24). NOS scores ranged from 0 to 9. NOS score from 0-4 was considered as low quality, 5-6 as medium quality, and 7-9 as high quality.

Statistical analysis
HRs or ORs with corresponding 95% CIs were applied to analyze survival outcomes and clinicopathological features. We preferred to use the results of multivariate analysis. For literatures that could not directly extract survival data, we extracted survival results from survival plots according to Tierney et al. method (25).The forest plot was used to explore the prognostic role of nectin-4 in cancers. The p value <0.05 or I 2 >50% was defined as significant heterogeneity, and the random-effects model was used to pooled HRs. Otherwise, the fixed-effect model was used. Sources of heterogeneity were explored through subgroup analysis. Sensitive analysis was applied to test the stability of the results. Begg's test and Egger's test were used to assess publication bias. P < 0.05 denoted statistical significance. STATA version 12.0 software (Stata Corporation, College Station, TX, USA) was used to perform all the statistical analysis.

Search results
Articles about the prognosis significance of nectin-4 in tumors were searched by retrieving the specified databases. We primarily identified 814 articles. After removal of 437 duplicate publications, 377 articles were remained. After browsing the title and abstract, 351 articles were excluded. 11 articles were further excluded after viewing the full text. Finally, we identified fifteen articles published from 2011 to 2020. The search flow diagram was displayed in Figure 1.

Study characteristics
Fifteen articles containing 2245 patients were included in the metaanalysis. Ten articles were from Asia, and five were from the West. Fourteen articles included overall survival (OS) data. Four articles reported disease-free survival/progression-free survival/relapse-free survival (DFS/PFS/RFS)data. All articles detected nectin-4 expression in tumor tissues and used immunohistochemistry to identify nectin-4 expression. Nine studies directly provided the HRs and 95% CIs. The HRs and 95% CIs of the other articles were extracted from the survival curves. A total of seven different types of tumors were included, namely BC (9,13,15,19,23),esophageal cancer(EC) (10,11,18),hepatocellular carcinoma(HCC) (12),pancreatic cancer(PC) (14),lung cancer(LC) (16), upper tract urothelial carcinoma(UBC) (17),gastric cancer(GC) (20,21) and gallbladder cancer(GBC) (22). The NOS scores of the included studies ranged from 6 to 7 (mean: 6.26). Basic information of the included studies was summarized in Table 1.

Nectin-4 and OS
Fourteen articles involving 2048 patients displayed OS data. Due to obvious heterogeneity (I 2 = 71.9%), a random model was used to calculate the pooled HRs with 95% CIs. The pooled results revealed that high nectin-4 expression was significantly associated with poorer OS (HR: 1.75, 95% CI: 1.35-2.28) (Figure 2).

Subgroup analysis for OS
In order to better detect the predictive effect of nectin-4, subgroup analysis was implemented based on the cancer type, analysis type, race, country, source of HRs and sample size ( . We also observed the absence of heterogeneity in some subgroups (I 2 = 0), including cancer type (EC and GC), analysis type (univariate analysis), race (Asian), country (China and Japan) and sample size (< 100). These factors may be sources of heterogeneity. We further performed a subgroup analysis of BC. We found that nectin-4 expression was not associated with prognosis of patients with triple-negative BC. Due to the small numbers of included articles, more studies were needed to explore the relationship between nectin-4 expression and different subtypes of BC. In addition, we evaluated protein that bound to the nectin-4 in BC using the PINA database (26).We found that 12 proteins (PICK1,FBXO10,CDH1,PIN4,EIF2B4,EVL,ZNF582,MRPL18, ENAH,NECTN1,PDSS1 and RRP1) were associated with nectin-4 in BC (Figure 3).

Sensitivity analysis
Sensitivity analysis was performed by removing one study at a time. The results were consistent with the comprehensive analysis, confirming that the outcomes were stable (Figures 5, 6).

Publication bias
We used funnel plots to qualitatively evaluate publication bias. Begg's test and Egger's test were used to quantitatively assess the publication bias. P values of Begg's and Egger's tests were 0.09 and 0.289 for OS (Figure 7

Discussion
Cancer is considered as the important obstacle to increase life expectancy worldwide (27). Promoting the early diagnosis and treatment of cancer is significant for the prognosis of patients with cancers. Therefore, it is important to find effective and appropriate prognostic markers. Nectin-4 as an important cellular adhesion molecule mediates intercellular and cell-cell matrix adhesion and cross-linking. Nectin-4 is considered as a tumor promoter (28).Cumulative studies have shown that nectin-4 is closely related to tumor prognosis, but no consensus has been reached. Therefore, we conducted the first meta-analysis to determine whether nectin-4 was a suitable prognostic marker in cancers.
Our meta-analysis included fifteen studies involving 2245 patients. The results showed that high nectin-4 expression was significantly associated with poorer OS. Subgroup analysis for OS showed that high nectin-4 expression mainly predicts poor prognosis in EC (HR: 1.78; 95% CI: 1.30-2.44) and GC (HR: 1.82; 95% CI: 1.50-2.22), suggesting that nectin-4 may have better predictive power for the two kinds of tumors. Our meta-analysis FIGURE 2 Forest plot of the relationship between high Nectin-4 expression and OS. also revealed no association between high nectin-4 expression and DFS/PFS/RFS. We found that high nectin-4 expression significantly positively correlated with tumor diameter (big vs small), tumor stage (III-IV vs I-II) and infiltration depth (T3+T4 vs T2+T1). Moreover, we revealed that nectin-4 was closely associated with 12 proteins in BC. Taken together, these results indicated that nectin-4 may be a suitable and effective prognostic indicator for cancers. Nectin-4 plays a crucial role in carcinogenesis and participates in tumor progression in different ways. Pavlova et al. showed that low expression of nectin-4 in BC could attenuate the adhesion between cancer cells and promote tumor cell apoptosis and distant metastasis through the b4/SHP-2/c-Src signaling pathway (29).Moreover, overexpression nectin-4 induced BC angiogenesis via endothelial integrin-beta4 (30). Nectin-4 could upregulate vascular endothelial growth factor (VEGF) expression to promote angiogenesis in PC (14). In LC, takano et al. revealed that interfering with nectin-4 expression could significantly inhibit tumor cell proliferation (16). They also found that exogenous nectin-4 expression increased the formation of lamellar lipoprotein and the invasive ability of mammalian cells by activating Rac1 signaling (16). In gallbladder and gastric cancers, nectin-4 promoted tumor cell proliferation, migration and invasion by activating PI3K/AKT signaling (20,22). In colon cancer, overexpression nectin-4 significantly enhanced tumor cell proliferation, migration, colony formation and resistance to 5fluorouracil (30).Mechanistically, nectin-4 may function in tumorigenesis by affecting afadin protein expression and the normal operation of the AKT-NF-kB signaling pathway (31).The epithelialmesenchymal transition (EMT) is an important biological process in carcinogenesis (32). EMT is involved in the occurrence and development of numerous tumor types (33). Nectin-4 may promote tumor EMT by affecting a variety of adhesion proteins, including Ecadherin, N-cadherin and Vimentin (34,35). The understanding of the specific mechanism of nectin-4 regulating tumors is insufficient. Singlecell technology and spatial transcriptomics can help us to further fully explore the role of nectin-4 in tumors (36,37).
Although our study revealed that high nectin-4 expression could be a valuable prognostic biomarker in various cancers, some limitations need be noted. Firstly, all included studies had small sample sizes and were retrospective studies. Secondly, FIGURE 3 Protein that binds to the nectin-4 in BC.  Despite these limitations, our meta-analysis had some advantages. Firstly, this was the first meta-analysis to assess the association between nectin-4 expression and survival outcomes in different cancers. Secondly, we had identified potential causes of heterogeneity. Thirdly, sensitivity analysis revealed that the results were stable. Additionally, there was no obvious publication bias in this study.
In conclusion, we suggested that high nectin-4 expression predicted unfavorable OS. We also confirmed that nectin-4 can be applied as an effective prognostic indicator in cancers, especially for EC and GC.Nectin-4 may be a potentially important tumor target. More randomized controlled studies are required to validate our findings and further explore the association between nectin-4 and cancers.

Data availability statement
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

Author contributions
WW and ZQ contributed to the study inception and design. RL, KZ, and KW equally contributed to the literature search, analysis and writing of the manuscript. Other authors contributed to the study design and study supervision. All authors contributed to the article and approved the submitted version.

Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Sensitivity analysis for DFS/PFS/RFS. The publication bias for OS. Sensitivity analysis for OS. The publication bias for DFS/PFS/RFS.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.