Common variants in toll-like receptor family genes and risk of gastric cancer: a systematic review and meta-analysis

Background: An increasing number of studies have suggested the relationship between single-nucleotide polymorphisms (SNPs) in toll-like receptor (TLR) genes and gastric cancer (GC) susceptibility; however, the available evidence is contradictory. This meta-analysis aimed to comprehensively evaluate whether the SNPs within the TLR family are related to GC development. Methods: PubMed, Scopus, and China National Knowledge Infrastructure (CNKI) were systematically searched up to May 2023 to obtain the pertinent publications. Pooled odds ratios (ORs) with 95% confidence intervals (CIs) were applied to examine the associations using the random-effects model. Results: A total of 45 studies with 25,831 participants (cases: 11,308; controls: 14,523) examining the relation of 18 different SNPs in the TLR family to GC were analyzed. Variations in TLR-4 rs4986790, TLR-4 rs4986791, TLR-5 rs5744174, and TLR-9 rs187084 were significantly associated with increased risk of GC in different genetic models. No significant association was detected for TLR-2-196 to -174de (Delta22), TLR-2 rs3804100, TLR-4 rs11536889, TLR-4 rs11536878, TLR-4 rs2770150, TLR-4 rs10116253, TLR-4 rs1927911, TLR-4 rs10983755, TLR-4 rs10759932, TLR-4 rs1927914, and TLR-10 rs10004195. Conclusion: These findings indicate that variations in TLR-4, TLR-5, and TLR-9 genes were found to be potential risk factors for GC.


Introduction
Gastric cancer (GC) is the fifth most prevalent kind of malignancy and the third main cause of cancer-related death globally (Castano-Rodriguez et al., 2014), accounting for 8.8% of total cancer-associated mortalities, with 5-year survival rates less than 30% (Castaño-Rodríguez et al., 2014).GC is a complex disease with a multifactorial etiology that includes the combined impacts of lifestyle, bacteria, and host and environmental factors (Gao et al., 2020).However, Helicobacter pylori infection is an established risk factor for GC, but only 0.1%-4% of infected people develop GC, highlighting the etiological involvement of environmental predisposing factors and host genetics in GC development (Companioni et al., 2014).
Changes in the host immune elements, such as toll-like receptors (TLRs), may affect the development of GC via their crucial involvement in triggering adaptive and innate immune responses (Sultan et al., 2022).Ten TLRs have been identified on the surfaces of immune cells as well as gastric epithelial cells in humans (Eed et al., 2020).TLRs deliver the first line of host defense against damaging pathogens by recognizing pathogenassociated molecular patterns (PAMPs) expressed by the majority of microorganisms, such as unmethylated CpG motifs (TLR-9), flagella (TLR-5), lipopolysaccharides (LPSs) and lipoproteins (TLR-1,-2,-4,-5, and -6), peptidoglycan (TLR-2), and microbial nucleic acids   (Castano-Rodriguez et al., 2014;Li et al., 2021).Moreover, TLR-10 expressed by the gastric epithelial cells plays a key role in the identification of multiple patterns of H. pylori LPSs (Li et al., 2021).The contribution of TLRs in gastric tumorigenesis has been suggested by previous studies (de Oliveira and Silva, 2012;Castaño-Rodríguez et al., 2014).It has been detected that TLR-3 and TLR-4 are highly expressed in cancer cells (Li et al., 2021).Evidence has also shown that the overexpression of TLR-5 is associated with increased GC cell proliferation and may act as a biomarker for GC (Song et al., 2011;Kasurinen et al., 2019).Furthermore, TLR-7 expression has been negatively linked to the viability of GC cells (Jiang et al., 2016), and the TLR-7 agonist has been examined as a possible vaccine formulation in GC immunotherapy (Wang X-D. et al., 2015).TLR-9 is also expressed aberrantly in GC, and H. pylori DNA could trigger TLR-9-mediated GC cell invasion (Gao et al., 2020).Singlenucleotide polymorphisms (SNPs) within TLR genes may lead to a change in their expression along with dysregulation of their signaling pathways, resulting in disturbed secretion of inflammatory mediators and enabling H. pylori to cause persistent infection and affecting gastric immunopathology and GC susceptibility (Eed et al., 2020;Sultan et al., 2022).
In recent studies, polymorphisms in TLR genes have been suggested to be associated with the risk of GC (de Oliveira and Silva, 2012;Companioni et al., 2014;Dargiene et al., 2018).However, the results have been inconsistent among various ethnic populations, and the small sample sizes of individual studies limit their statistical power to detect associations.This systematic review and meta-analysis was performed to comprehensively and quantitatively evaluate the relationship between TLR gene polymorphisms and the risk of GC.

Materials and methods
This meta-analysis was performed according to the PRISMA protocol for systematic review and meta-analysis (Moher et al., 2015).

Search strategy
Without language restriction, we systematically searched PubMed, Scopus, and China National Knowledge Infrastructure (CNKI) up to 30 May 2023 to find studies examining the relation of polymorphisms in the TLR family to the risk of GC using the following terms: (((((Toll-like  ).The lists of references within the reviews and pertinent studies were manually searched for possible additional publications.Initially, all studies obtained through PubMed and Scopus were entered into a citation manager software (EndNote) to screen the identified studies according to the inclusion/exclusion criteria.Then, a complementary systematic search was conducted in CNKI for additional studies.Some authors were experts in multiple languages, including Chinese.Nevertheless, except for one non-English study (Chinese) (Zeng et al., 2011a), all other articles published in non-English languages were excluded from the meta-analysis based on their English titles/abstracts because they had unrelated exposures or outcomes.

Eligibility criteria
Studies with the following criteria were eligible to be included in the present analysis: 1) investigated the association between gene polymorphisms in TLRs (TLR-1, TLR-2, TLR-3, TLR-4, TLR-5, TLR-6, TLR-7, TLR-8, TLR-9, or TLR-10) as the exposure and GC as the outcome; 2) studies with cohort or case-control design; and 3) provided genotype frequency in both cases and controls.We excluded studies that did not have a control group, studies with irrelevant exposure/outcomes, gene expression papers, letters, conference papers, case reports, book chapters, publications with unextractable data, and studies on precancerous gastric lesions.Moreover, studies were excluded from the meta-analysis if the genotypic frequency of the control group deviated from the Hardy-Weinberg equilibrium (HWE).However, for the studies that investigated several polymorphisms, polymorphisms that were not in HWE were excluded from the analysis, and only polymorphisms that were in HWE were included in the metaanalysis.Two reviewers screened titles/abstracts and full texts of the potentially related studies for eligibility assessment based on the inclusion/exclusion criteria.The disagreements regarding the eligibility of studies were resolved by a group discussion among all authors.

Data extraction and quality assessment
Two investigators independently extracted the data using a standardized data extraction sheet, and disagreements were resolved by discussion between the two investigators, and if necessary, a third reviewer was consulted.The following data were obtained for each publication: the first author's name, country, sample size of cases and controls, ethnicity, minor allele frequency (MAF) for the control group, HWE, year of publication, and genotype frequencies in controls and cases.The quality of the included publications was assessed using the Q-Genie tool.This tool includes 11 questions to be marked on a 7-point Likert scale for assessing several aspects of the genetic association studies.Overall scores ≤35 indicate poor-quality studies, those >35 and ≤45 indicate moderate-quality studies, and those >45 indicate good-quality studies (Sohani et al., 2015).

Statistical analysis
The HWE in control groups was tested using the chi-squared test, and p < 0.05 indicated significant disequilibrium (Alizadeh et al., 2016).The strength of the relation between the TLR polymorphisms and the odds of GC risk was examined using the pooled odds ratio (OR) with a 95% confidence interval (CI) using the random-effects model (DerSimonian-Laird approach) (DerSimonian and Laird, 1986) under five genetic models, namely, the allelic, recessive, dominant, homozygote contrast, and heterozygote contrast models.The heterogeneity among studies was evaluated by the Q-test and I 2 statistics; p-value <0.1 was considered a remarkable evidence of heterogeneity (Abyadeh et al., 2019;Nazari et al., 2022).Subgroup analyses based on ethnicity, age, and source of controls (community-based or hospital-based) were conducted to assess the possible sources of heterogeneity.Publication bias was assessed by funnel plots and Egger's test, and p-values less than 0.05 were considered statistically significant (Alizadeh et al., 2017).All analyses were conducted using Stata software (version 13.0; Stata Corporation, College Station, TX).

Systematic review of rare polymorphisms in TLR genes and GC
The characteristics of studies on less common SNPs of TLRs and GC that were not included in the meta-analysis but were included in  Sum of homozygous and heterozygote variants.

Subgroup analyses
In the stratified analysis, the association of TLR-4 rs4986790 with GC was supported by the subgroups of Caucasian, Latino, people of age ≤60 years, and studies with the community-based control group (Table 2).Moreover, the association of TLR-4 rs4986791 with GC was supported by the subgroups of Caucasian, people of age ≤60 years, and studies with the community-based control group (Table 2).In different genetic models, there was a positive relationship between TLR-5 rs5744174 and GC susceptibility (recessive model: OR = 1.71, 95% CI: 1. allelic model: OR = 1.25,; homozygote model (CC vs. TT): OR = 1.74, 95% CI: 1.28-2.37)(Figure 5), which was also observed in the Asian population (Table 2).
In the stratified analysis, the association of TLR-9 rs187084 with GC was supported by the subgroups of Caucasian, Latino, people of age ≤60 years, and studies with the community-based control group (Table 2).

Heterogeneity and publication bias
For the SNPs of TLR-2 rs3804099, TLR-2 rs3804100, TLR-4 rs11536889, TLR-4 rs4986791, TLR-4 rs11536878, TLR-4 rs2770150, TLR-5 rs5744174, and TLR-9 rs352140, no significant heterogeneity was detected across the studies, but the test of heterogeneity was significant for other investigated SNPs (Table 2).Egger's test of publication bias revealed no evidence of publication bias for most SNPs (Table 2).Funnel plots for publication bias for SNPs with ≥10 studies under the dominant model are given in Supplementary File S2.

Discussion
This meta-analysis examined the association of 18 common SNPs in the TLR family to GC risk.The results indicated that variations in TLR-4 rs4986790, TLR-4 rs4986791, TLR-5 rs5744174, and TLR-9 rs187084 were significantly associated with increased risk of GC in different genetic models.
Given the close relationship between inflammation and carcinogenesis, several studies have attempted to disclose the role of TLRs in GC progression.However, the available data were contradictory.The protein function of TLRs might be damaged by SNPs in genes coding TLRs, resulting in altered vulnerability to malignancies (Wang et al., 2023).Previous meta-analyses on the relation of TLR SNPs to GC have only focused on four common SNPs in the TLR family, namely, TLR-2-196 to -174de (Delta22), TLR-4 rs11536889, TLR-4 rs4986790, and TLR-4 rs498679, with inconclusive results.While the study by Castano-Rodriguez indicated a significant association of TLR-2-196 to -174de (Delta22) with GC (Castaño-Rodríguez et al., 2013), the metaanalysis by Chen et al. (2013) failed to find an association.In line with our findings, a previous meta-analysis by Zhao et al. identified a significant direct association between TLR-4 rs4986790 and GC, while, in contrast to the present meta-analysis, no association was found for TLR-4 rs4986791 (Zhao et al., 2013).Confirming our results, the significant relation of TLR-4 rs4986790 and TLR-4 rs4986791 variations to GC has been proposed by another metaanalysis (Zhou et al., 2014).Moreover, similar to our study, two previous meta-analyses by Wang X. et al. (2015) and Chen et al. (2013) did not find a significant relationship between TLR-4 rs11536889 and GC.The present study is the first meta-analysis showing a significant relationship between TLR-5 rs5744174 and TLR-9 rs187084 and genetic susceptibility to GC.The differences in the results of the previous meta-analyses may be due to the small number of the analyzed studies and, thus lack of sufficient statistical power to detect the true associations.
Mechanistically, genetic variations in TLRs may result in impaired TLR signaling and alter their affinity to the ligands, leading to variations in immune responses of the gastric mucosa against carcinogens and dysregulated inflammation, leading to modulation of H. pylori infection and, thus, GC risk (Zhou et al., 2014;Tongtawee et al., 2019).TLR signaling plays a crucial role in shaping the tumor microenvironment (Wang et al., 2023).Polymorphisms in TLR genes could influence the recruitment and activation of immune cells, tumor-associated inflammation, and immune evasion mechanisms employed by cancer cells, which all affect cancer susceptibility (Paulos et al., 2007).TLR-2 activation results in the activation of nuclear factor-κB (NF-κB) and functions as an innate immune response (de Matos Lourenço et al., 2020).It has been suggested that NF-κB is a main mediator of inflammation-induced tumor progression (Wang et al., 2023).The rs3804099 variants may reduce TLR-2-related chronic inflammation and induce apoptosis (Kim and Karin, 2011;Semlali et al., 2017;Mohamed et al., 2020), resulting in decreased risk of GC.Furthermore, the TLR-5 rs5744174 may affect the recognition and binding affinity of TLR-5 toward bacterial flagellin, leading to compromised antibacterial responses and prolonged exposure to pathogens, thus promoting gastric oncogenesis (Castaño-Rodríguez et al., 2014).TLR-9 impacts tumor development and progression through various molecular mechanisms, such as activation of immune response, induction of apoptosis, modulation of angiogenesis, and regulation of tumor cell proliferation and survival.TLR-9 activation in GC cells leads to the production of pro-inflammatory cytokines and chemokines, such as TNFalpha, IL-6, and IL-8 (Karapetyan et al., 2020).These molecules attract immune cells, such as macrophages and dendritic cells, to the tumor microenvironment.This immune response helps in tumor recognition, elimination, and control (Karapetyan et al., 2020).TLR-9 activation can induce apoptosis in GC cells.This is mediated by the activation of caspases, which are enzymes involved in the execution of apoptosis (Sindhava et al., 2017).TLR-9 can influence the angiogenesis that is required for tumor growth and metastasis.Studies have shown that TLR-9 activation can inhibit angiogenesis by suppressing the production of pro-angiogenic factors, such as vascular endothelial growth factor (VEGF) (He et al., 2018b).Activation of TLR-9 impacts the proliferation and survival of GC cells (Tang et al., 2022).It can activate signaling pathways, such as the nuclear factor kappa-light-chain-enhancer of activated B-cell (NF-κB) pathway, which is involved in promoting cell proliferation and survival (Tsujimura et al., 2004).Regarding TLR-5, induction of TLR-5 leads to the production of pro-inflammatory cytokines and chemokines, such as interleukin 1 beta (IL-1β), IL-6, and IL-8.These molecules attract immune cells, including macrophages and natural killer cells, to the tumor microenvironment.This immune response aids in tumor recognition and elimination (Tallant et al., 2004).TLR-5 plays a role in the integrity and barrier function of the epithelial layer in the stomach (Feng et al., 2023).This is achieved through the activation of various signaling pathways, such as the NF-κB pathway (Tallant et al., 2004), which regulates the expression of genes involved in tight junction formation and epithelial cell adhesion (Al-Sadi et al., 2016).A well-functioning epithelial barrier helps prevent the infiltration of potentially harmful agents and pathogens into the underlying tissues (Mitamura et al., 2021).TLR-5 activation has been shown to affect the invasive potential and metastatic spread of GC cells (Castaño-Rodríguez et al., 2014).This effect is mediated by the suppression of various molecular pathways involved in invasive behavior, such as the matrix metalloproteinase (MMP) family enzymes that degrade the extracellular matrix (Jiang et al., 2017), a critical step in tumor invasion and metastasis (Zhuyan et al., 2020).Studies have also demonstrated that activation of TLR-5 promotes apoptosis in cancer cells through the activation of caspases and the upregulation of pro-apoptotic factors (Zeng et al., 2006).This helps in reducing tumor burden and inhibiting tumor progression.Upon binding to molecular patterns, TLR-4 triggers the activation of inflammatory pathways, leading to the release of pro-inflammatory cytokines, chemokines, and various immune cells (Escoubet-Lozach et al., 2011).Chronic inflammation can promote tumor growth, angiogenesis, and metastasis in GC (Landskron et al., 2014).TLR-4 signaling can activate the NF-κB pathway (Verstrepen et al., 2008).Activation of NF-κB promotes the expression of genes involved in cell survival, proliferation, and angiogenesis, which can contribute to the development and progression of GC (Rastogi et al., 2023).TLR-4 activation has been shown to induce epithelial-mesenchymal transition (EMT), a process in which epithelial cells acquire a mesenchymal phenotype (Shi et al., 2016).EMT is associated with increased invasive potential and metastasis in cancer cells (Huang et al., 2015).TLR-4-induced EMT in GC is mediated through signaling pathways such as Snail, Twist, and ZEB1, which are known regulators of EMT markers (Jing et al., 2012).TLR-4 activation can contribute to the activation of cancer-associated fibroblasts (CAFs) and the development of fibrosis in the GC microenvironment (Janovec et al., 2021).CAFs can promote GC progression through the secretion of growth factors, extracellular matrix remodeling, and immune modulation (Kobayashi et al., 2019).TLR-4 signaling can directly impact tumor cell behavior.Activation of TLR-4 promotes cell survival and proliferation through signaling pathways such as phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT), mitogen-activated protein kinase/ extracellular signal-regulated kinase (MAPK/ERK), and the Janus kinase (JAK)/signal transducer and activator of transcription (STAT) (JAK/STAT) (Vivarelli et al., 2004;Korneev et al., 2017).These pathways regulate various aspects of cell growth, survival, and apoptosis resistance, ultimately influencing GC cell behavior (Hu et al., 2014;Sun et al., 2015).It is important to note that the molecular mechanisms of TLRs in GC are still an area of active research, and further studies are needed to fully understand the complexities of the TLR-mediated effects on GC development and progression.Accordingly, the SNPs of TLRs might change the aforementioned pathways, resulting in cancer proneness.These findings can contribute to the development of targeted preventive strategies and personalized therapeutic interventions for individuals at high risk of GC.
As a strength, no evidence of publication bias was detected, and studies that deviated from the HWE were excluded from the analyses.Nevertheless, some limitations of the present metaanalysis should be considered.First, there was significant heterogeneity across the analyzed publications.To improve the reliability of the results, the random-effects model was used for analyses.Moreover, stratified analysis revealed that ethnicity, the source of controls, and the age of participants were the sources of the observed heterogeneity.The association of the TLR polymorphisms with GC was modified by ethnicity, the source of controls, and the age of participants, indicating that these factors have contributed to the differences in the results of the previous studies.Therefore, future studies should consider these factors when investigating the association between TLR polymorphisms and gastric cancer.Second, the interactions between gene-gene and gene-environmental factors, including diet, alcohol intake, smoking, and physical activity, potentially affect GC carcinogenesis; due to the unavailability of data, we could not consider interaction analyses in this study.Third, for some SNPs, a small number of studies were included, and the findings of some subgroups may be at risk of bias because of relatively low statistical power to detect associations.Therefore, findings from the stratified analyses should be interpreted cautiously and need additional validation.We also could not perform subgroup analysis by the sex of subjects because of the lack of sex-specific data in original studies.Moreover, data for the effects of the haplotype analyses and gene-environmental factor interactions on GC were not reported sufficiently in the included original studies; thus, we could not report haplotype analyses and gene-environment interactions.Lastly, no sufficient information was available for African populations, and our result, therefore, may not be expandable to these populations.Thus, further research is required to assess the impact of these SNPs on GC in various ethnicities, particularly those of African ethnicities.

Conclusion
In conclusion, this meta-analysis revealed that variations in TLR-4, TLR-5, and TLR-9 genes were potential risk factors of GC.Additional well-designed, large-scale studies, with more detailed information concerning gene-gene and gene-environmental interactions, should be conducted on different races, especially for less investigated SNPs, to yield a better understanding of the relationship between TLR SNPs and GC risk.

FIGURE 1
FIGURE 1Flow diagram of the study.

TABLE 1
Characteristics of eligible studies considered for the association between toll-like receptor family gene polymorphism and GC risk in the meta-analysis.

TABLE 1 (
Continued) Characteristics of eligible studies considered for the association between toll-like receptor family gene polymorphism and GC risk in the meta-analysis.

TABLE 1 (
Continued) Characteristics of eligible studies considered for the association between toll-like receptor family gene polymorphism and GC risk in the meta-analysis.

TABLE 1 (
Continued) Characteristics of eligible studies considered for the association between toll-like receptor family gene polymorphism and GC risk in the meta-analysis.
(Continued on following page)

TABLE 1 (
Continued) Characteristics of eligible studies considered for the association between toll-like receptor family gene polymorphism and GC risk in the meta-analysis.

TABLE 2
Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.

TABLE 2 (
Continued) Main results and subgroup analysis by ethnicity for the meta-analysis of various types of toll-like receptors and gastric cancer susceptibility.