Investigation of IGF1, IGF2BP2, and IGFBP3 variants with lymph node status and esophagogastric junction adenocarcinoma risk

Abstract Esophagogastric junction adenocarcinoma (EGJA) may be associated with obesity and overweight. Thus, any variant in energy metabolism–related gene may influence the development of EGJA. In this study, we recruited 720 EGJA cases and 1541 noncancer controls. We selected IGF2BP2 rs4402960 G > T, rs1470579 A > C, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T and rs6953668 G > A loci and assessed the relationship of these polymorphisms with lymph node status and susceptibility of EGJA. We found that IGF2BP2 rs1470579 A > C and IGFBP3 rs6953668 G > A polymorphisms were associated with the decreased risk of EGJA ( IGF2BP2 rs1470579: CC vs AA: adjusted odds ratio [OR] = 0.65, 95% confidence interval [CI] = 0.43‐0.98, P = 0.041 and CC vs AA/AC: adjusted OR = 0.62, 95% CI = 0.41‐0.93, P = 0.021 and IGFBP3 rs6953668: GA vs GG: adjusted OR = 0.66, 95% CI = 0.47‐0.93, P = 0.019 and GA/AA vs GG: adjusted OR = 0.68, 95% CI = 0.48‐0.95, P = 0.026). However, we also found that IGF1 rs5742612 A > G polymorphism increased the risk of LNM among patients with EGJA (GG vs AA: adjusted OR = 1.88, 95% CI = 1.02‐3.46, P = 0.042 and GG vs AA/AG: adjusted OR = 1.92, 95% CI = 1.06‐3.47, P = 0.032). This study suggests that IGF2BP2 rs1470579 A > C and IGFBP3 rs6953668 G > A polymorphisms may decrease genetic susceptibility to EGJA in eastern Chinese Han population. In addition, our findings also indicate that IGF1 rs5742612 A > G polymorphism may increase the susceptibility of LNM among patients with EGJA.

A > G polymorphism may increase the susceptibility of LNM among patients with EGJA.

| INTRODUCTION
In the past few decades, the incidence of esophagogastric junction adenocarcinoma (EGJA) has been increasing worldwide. 1,2 According to its anatomical region relative to the esophagogastric junction (EGJ), EGJA has been divided into three subtypes by the Siewert classification. Siewert type I and type III of EGJA are usually considered as esophageal and gastric cancer, respectively. Siewert type II malignancies are treated as "true" EGJA. However, the etiology and potential risk factor remain unclear. Recently, obesity and overweight have been known cancer risk factors. In addition, EGJA has been considered as an obesity and overweight-related cancer. [3][4][5] Thus, any variant and abnormal expression in energy metabolism gene may influence the development of EGJA.
Insulin-like growth factor-1 (IGF1), a growth hormone similar in molecular structure and function to insulin, may be implicated in growth during childhood and continue to have metabolism-related influences in adults. IGF1 is generally produced by the liver. Most of the IGF1 bind to insulin-like growth factor binding proteins (IGFBPs). IGFBP3 is the most abundant protein and binds to IGF1. It is found that the IGF signaling pathway plays an important role in some cancers. 6 Gallagher et al 7 have reported that patients with Laron syndrome have a decreased susceptibility of developing cancer. Dietary interventions and modifications may downregulate IGF1 activity and reduce the susceptibility of cancer by promoting increased glucagon activity. 8 Recently, some case-control studies have focused on the relationship of IGFBP3 and IGF1 single nucleotide polymorphisms (SNPs) with the risk of cancer. [9][10][11] A previous case-control study indicated that IGFBP3 rs2270628 C > T was associated with an increased risk of ovarian cancer. 12 Also, significant association with the survival of breast cancer in Chinese premenopausal women was identified for IGFBP3 rs3110697 G > A. 13 Liu et al 14 reported that IGFBP3 rs2270628 C > T and rs3110697 G > A SNPs were associated with a significantly decreased risk of esophageal squamouscell carcinoma (ESCC). In addition, some case-control studies focused on the relationship of IGF1 SNPs and gastric cancer. 15,16 IGF1 rs5742612 A > G polymorphism was found to be associated with tumor response to chemotherapy in patients with advanced gastric cancer. 17 Insulin-like growth factor 2 mRNA-binding protein 2 (IGF2BP2) is encoded by the IGF2BP2 gene and acts as an RNA-binding protein of IGF2 mRNA. 18 Functions of IGF2BP2 are associated with insulin resistance, lipid metabolism, and tumorigenesis. 19,20 Dai et al 21 reported that IGF2BP2 is a tumor promoter, which drives tumor proliferation through HMGA1 and mRNAs IGF2. Results of the previous case-control study demonstrated that IGF2BP2 rs4402960 G > T was involved in the risk of cancer. 22,23 In addition, Liu et al 24 found that IGF2BP2 variants might be an independent predictor of chemotherapeutic response in patients with metastatic gastric cancer.
However, the associations of IGFBP3, IGF2BP2 and IGF1 SNPs with EGJA risk were unknown. In this study, with an aim to explore the relationship of IGF1, IGFBP3, and IGF2BP2 SNPs with the development of EGJA, IGF2BP2 rs4402960 G > T, rs1470579 A > C, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T and rs6953668 G > A loci were selected and genotyped in 720 EGJA cases and 1541 controls.

| Subjects
In this case-control study, we examined 720 patients (188 female, 532 male, mean age 64.21 ± 8.82 years) with EGJA diagnosed according to gastroscope and pathology. Consenting patients with EGJA treated between January 2014 and May 2016 in the Fujian Medical University Cancer Hospital and Union Hospital were enrolled in this study. In addition, 440 patients with EGJA were included in this study from Affiliated People's Hospital of Jiangsu University from November 2010 to November 2016. The patients with autoimmune disease history, prior chemoradiotherapy, and a history of another malignancy were excluded. All patients with EGJA were Asians from the east region of China. The noncancer controls were selected randomly from the population of the same region of China and consisted of healthy Asian 1541 subjects (404 female, 1137 male, mean age 64.30 ± 10.19 years). Each subject enrolled in this study answered a routine prestructured questionnaire, and height and weight were measured. Body mass index (BMI) ≥ 24 was accepted as the criterion for overweight and obesity. 25,26 The status of lymph node metastasis (LNM) was also collected. The study was approved by the ethics committee at Jiangsu University, Zhenjiang City, China, and a written informed consent was obtained from each participant.

| DNA extraction and genotyping
The genomic DNA was carefully extracted from 2 mL of whole blood samples using a Promega Blood DNA Purification Kit (Promega, Madison, WI). IGF2BP2 rs1470579 A > C, rs4402960 G > T, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T and rs6953668 G > A polymporphisms were genotyped using SNPscan genotyping assays from Genesky Biotechologies Inc (Shanghai City, China). 27,28 Ninety DNA samples were selected randomly for quality control. The genotypes of IGF2BP2 rs1470579 A > C, rs4402960 G > T, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T and rs6953668 G > A SNPs were checked by another laboratory technicians. And the results were not changed.

| Statistical analysis
Continuous variables were expressed as the mean ± standard deviation. The Student t test was applied to compare the differences between patients with EGJA and noncancer controls. Chi-square (χ 2 ) or Fisher's exact tests were used to compare categorical variables (eg, age, sex, weight, height, BMI, and genotype and allele frequencies) between EGJA groups and controls. SAS software (Version 9.4; Cary, NC) was used for data analysis. A P value less than 0.05 was considered statistically significant. Internet-based software (http://ihg.gsf.de/cgi-bin/ hw/hwa1.pl) was harnessed to determine whether the distribution of genotype frequencies was according to Hardy-Weinberg equilibrium (HWE).

| Baseline characteristics
We list the clinical characteristics, selected risk factors, and demographics in Table 1. In our study, 720 patients with EGJA and 1541 noncancer controls were included. T A B L E 3 Logistic regression analyses of association between IGF2BP2 rs4402960 G > T, rs1470579 A > C, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T and rs6953668 G > A polymorphisms and risk of EGJA   Table 3]). However, we found that IGF2BP2 rs4402960 G > T, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T variants might be not associated with the development of EGJA (Table 3).

| Association of IGF2BP2 rs4402960
G > T, rs1470579 A > C, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T and rs6953668 G > A polymorphisms with Lymph node status in EGJA patients Table 4, we found that IGF1 rs5742612 A > G polymorphism had a tendency of increased risk to LNM among EGJA patients (GG vs AA: crude OR = 1.77, 95% CI = 0.97-3.23, P = 0.063 and GG vs AA/AG: crude OR = 1.80, 95% CI = 1.00-3.22, P = 0.050). After adjustment for BMI, gender, sex, alcohol use, and smoking status, this association was more significant (GG vs AA: adjusted OR = 1.88, 95% CI = 1.02-3.46, P = 0.042 and GG vs AA/AG: adjusted OR = 1.92, 95% CI = 1.06-3.47, P = 0.032). some publications reported that obesity and overweight were associated with the development of EGJA. [3][4][5] Thus, the variants in energy metabolism-related gene may influence the susceptibility of EGJA. In this study, we explored the relationship of IGF2BP2 rs4402960 G > T, rs1470579 A > C, IGF1 rs5742612 A > G and IGFBP3 rs3110697 G > A, rs2270628 C > T and rs6953668 G > A SNPs with the development of EGJA in 2261 subjects. We found that IGF2BP2 rs1470579 A > C and IGFBP3 rs6953668 G > A polymorphisms might be protective factors for EGJA. However, we identified that IGF1 rs5742612 A > G polymorphism had an increased risk to LNM among EGJA patients. IGF2BP2 rs1470579 A > C polymorphism is located on intron 2. Recently, a meta-analysis study reported that CC carriers of rs1470579 conferred risk to type 2 diabetes mellitus (T2DM) than IGF2BP2 rs1470579 CA/AA carriers. 29 Several case-control studies assessed the potential association of IGF2BP2 rs1470579 A > C variants with T2DM susceptibility and therapeutic efficacy in the Chinese population. 30,31 In these studies, IGF2BP2 rs1470579 A > C polymorphism were found to be associated with T2DM risk, and this polymorphism may influence the therapeutic efficacy of some oral antidiabetic agents in patients with T2DM. 30,31 It is found that some variants in energy metabolism-related gene may influence the development of cancer. 22,23,32 In the current study, we first explored the association of IGF2BP2 rs1470579 A > C polymorphism with the risk of EGJA. It was found that the rs1470579 CC genotype of IGF2BP2 gene might be a protective factor for the development of EGJA.

| DISCUSSION
IGFBP-3, a common IGF binding protein, has highly conserved structures and binds IGF-1 and IGF-2 with high affinity. Based on the functional studies, it is believed that IGFBP-3 may be acting as a low-penetrance tumor suppressor. 33 Recently, some case-control studies focused on the relationship between IGFBP3 variants and cancer risk. Liu et al 14 reported that IGFBP3 rs2270628 C > T and rs3110697 G > A variants significantly decreased the risk of ESCC in Chinese Han population. However, in this study, we found that IGFBP3 rs2270628 C > T and rs3110697 G > A SNPs were not associated with the risk of EGJA in the Chinese population. IGFBP3 rs6953668 G > A polymorphism is located on intron. Verheus et al 34 studied the relationship between IGFBP3 rs6953668 G > A polymorphism and mammographic density. And they found a null association. However, we identified that IGFBP3 rs6953668 G > A polymorphism may decrease the risk of EGJA. The current study did not assess the role of this SNP in regulating the expression of the IGFBP3 protein in tissue of patients with EGJA. In the future, a functional study is necessary to be performed.
Several case-control studies focused on the relationship of IGF1 rs5742612 A > G polymorphism with gastrointestinal cancer. 35,36 The results of these studies indicated that IGF1 rs5742612 A > G polymorphism might be not associated with the risk of gastrointestinal cancer. In the current study, we found that IGF1 rs5742612 A > G variants might be not associated with the development of EGJA. Our findings were similar to those studies mentioned above.
A previous study indicated that IGF-1 and IGF-1R are upregulated in tissue of non-small-cell lung cancer (NSCLC), and expression of those factors was associated with the progression and prognosis of NSCLC. 37 In addition, it was found that IGF-1 may induce lymphangiogenesis and facilitates lymphatic metastasis, 38 and be associated with larger tumor size, local LNM, and worse prognosis in cancers. 39,40 Oh et al 17 reported that IGF1 rs5742612 A > G polymorphism was significantly associated with tumor response to patients with gastric cancer treated with 5-fluorouracil, leucovorin, and oxaliplatin. In this study, we found that IGF1 rs5742612 A > G polymorphism might increase the risk of LNM among patients with EGJA. To our knowledge, this is the first study to confirm the relationship between IGF1 rs5742612 A > G polymorphism and the risk of LNM. Wang et al 41 reported that the G allele of rs5742612 was found to be associated with decreased insulin sensitivity and increased insulin secretion. In addition, insulin levels were found to be correlated with LNM risk in both premenopausal and postmenopausal women with endometrial cancer. 42 In view of these findings, it is suggested that IGF1 rs5742612 A > G polymorphism may increase insulin secretion and induce lymphangiogenesis and facilitates lymphatic metastasis. Thus, this SNP may be implicated in the development of EGJA. In this study, some potential limitations should be addressed. First, the included patients with EGJA were limited, which may restrict to draw a strong conclusion. Secondly, only five SNPs were selected and genotyped; the coverage might be insufficient. In the future, for practical reasons, a fine-mapping study is needed to extensively assess the correlation of these genes variants with the development of EGJA. Thirdly, in the current study, the information on other risk factors was lacking. A further analysis on the relationship between these loci and environmental characteristic was not performed. Finally, a functional study was not carried out to further explain the potential role of these SNPs.
In summary, this study suggests that IGF2BP2 rs1470579 A > C and IGFBP3 rs6953668 G > A polymorphisms may be associated with genetic susceptibility to EGJA in eastern Chinese Han population. In addition, our findings also demonstrate that IGF1 rs5742612 A > G polymorphism may increase the risk of LNM among patients with EGJA.