Various angiogenic factors secreted by tumour cells stimulate the microvessel formation around the tumour cells1.Vascular endothelial growth factor (VEGF) is one of the most important angiogenic factors which stimulates tumour progression by the induction of endothelial cell migration and division, inhibition of the apoptosis of endothelial cells, induction of serine protease activity and enhancement of vascular permeability2.VEGF has been reported to be implicated in lymphatic metastasis and dysplastic lesions in oesophageal carcinoma3 and over-expression of VEGF has been reported in about 30–60 per cent of oesophageal cancer (EC) patients.

VEGF or VEGFA is localized on 6p21.3, having a 14kb coding region and comprises eight exons and seven introns. VEGFA is highly polymorphic with several potentially functional single nucleotide polymorphisms (SNPs) in the 5′ untranslated region (UTR), 3′ UTR and promoter. UTRs inhabit genomic regions that may serve as one of the key mediators of cancer pathogenesis by causing post-transcriptional deregulation4. Functional genetic variations in the VEGF have been reported to be involved in the progression and development of EC, due to which it can be used as a potential prognostic and predictive marker5. It has been documented that VEGF +405G/C (VEGF -634G/C) 5′ UTR polymorphism affects the translation efficiency of VEGFA and is correlated with microvessel density and prognosis of many cancers6. Association of GG genotype of VEGF +405G/C polymorphism with increased risk of EC has been reported in Kashmiri patients from north India7.

The association of VEGF -7C/T polymorphism located in 5′ UTR has been explored in gastrointestinal tract (GIT) cancers such as gastric cancer in northern Chinese8, colorectal in Japanese9, hepatocellular in Korean10 and Italian population11. So far, no study has been published in the literature on VEGF -7C/T polymorphism in EC.

The 3′UTR polymorphism VEGF +936C/T affects VEGF plasma levels by reducing the activity of activator protein (AP-4) in tumour tissue12. Correlation of VEGF +936C/T polymorphism with EC risk has been studied in different populations. Association of VEGF +936T allele with better survival was reported in Caucasian EC patients5. Increased risk of EC was observed in Caucasians13 and north Indian7 patients carrying the combined CT+TT genotype.

The prognostic and predictive significance of some VEGF polymorphisms have been documented in oesophageal cancer. CC and CG genotype of VEGF +405G/C polymorphism was associated with improved overall survival time in Japanese oesophageal squamous cell carcinoma patients treated with 5-fluorouracil/cisplatin14. GG genotype of VEGF +405GG genotype showed improved overall survival in Taiwanese EC patients treated with cisplatin14. In the German population, VEGF +936CT and TT genotypes were reported to influence the response to cisplatin or oxaliplatin with worse event-free survival in oesophagogastric junction adenocarcinoma patients16.

As per author’s knowledge, there is only one study on VEGF +936C/T and VEGF +405G/C polymorphisms in oesophageal cancer from north India7. VEGF is an important target in anticancer therapy, and identification of the role of VEGF variants in oesophageal cancer will help physicians to tailor therapy and for monitoring the therapy response. Therefore, the present study aimed to investigate the correlation of VEGF +405G/C (rs2010963), -7C/T (rs25648) and +936C/T (rs3025039) polymorphisms with EC risk in patients from Punjab, northwest India. As far as we know, this is the first study reporting the combined role of these three VEGF variants in EC.

Material & Methods

This case-control study was conducted in the Department of Human Genetics, Guru Nanak Dev University, Amritsar, Punjab, India from October 2020- March 2022. The study was carried out according to ICMR’s Ethical guidelines. The study protocol was approved by the institutional ethics committee.

Study subjects

A total of 464 individuals, including 231 sporadic EC affected individuals and 233 controls, were investigated in this case-control study. The study participants were clinically diagnosed at Sri Guru Ram Das Institute of Medical Sciences and Research, Amritsar, Punjab. The control group comprised randomly selected healthy, unrelated gender and age-matched individuals without any malignant and chronic disease from the same geographical area as EC affected individuals. Characteristics including gender, age, habits, habitat, family and disease history of all participants were recorded on proforma. From each study participant, a 5 ml blood sample was collected in an EDTA-coated vial after obtaining their signed informed consent. The blood sample was stored at -20°C till further use.

Genomic DNA isolation and genotyping of VEGF+936C/T, +405G/C and -7C/T polymorphisms

From the EDTA anticoagulated blood sample, the genomic DNA was extracted using the standard phenol-chloroform method. The quality and quantity of DNA samples were checked on 1% agarose gel. Genotyping of VEGF +936C/T and +405G/C polymorphisms was done using the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. The PCR reaction for VEGF +405G/C and +936C/T polymorphisms was carried out in atotal volume of 15 μl having 50 ng of DNA sample, 1X Taq buffer with 1.5 mM MgCl2, 0.3 μl of dNTPs mix, 5 pmol of forward and reverse primer (Sigma-Aldrich, Bangalore) and 1U of TaqDNA polymerase (Sigma-Aldrich, Bangalore).The details of screening conditions and genotyping methodology have been described in Supplementary Table. The PCR products of VEGF +405G/C (304 bp) and +936C/T (207 bp) polymorphisms were digested overnight with BsmFI and NlaIII restriction enzymes, respectively, as per manufacturer’s instructions (New England Biolabs, Beverly, MA). The genotying detail and gel photographs showing pattern of genotyping has been mentioned in Supplementary Table and Supplementary Figure 1 and 2, respectively.

Genotyping of VEGF-7C/T polymorphism was done by using amplification refractory mutation system-PCR (ARMS-PCR). Two different PCR master mixes were prepared in separate tubes with C and T allele-specific reverse primers. The PCR reaction was carried out in a 10 μl volume containing 50 ng genomic DNA, 1X Taq buffer with 1.5 mM MgCl2, 0.15 μl dNTPs mixture, 4 pmol of each primer (Sigma-Aldrich, Bangalore) and 1U Taq DNA polymerase (Sigma-Aldrich, Bangalore). A negative control without a DNA sample was included in each batch of reaction. The amplified PCR products were separated on 2.4% agarose gel. A sharp band of size 183bp represents C and T alleles while a band size of 425 bp represents control (Supplementary Table and Supplementary Fig. 3). To ensure quality control, genotyping was carried out without knowing the status of the case/control. Data were revalidated by reanalysing 10per cent of samples and both analyses’ results were 100 per cent concordant.

Statistical analysis

The data were presented as percentages or as mean±standard deviation. The allele and genotype frequencies of VEGF +936C/T, +405G/C and -7C/T polymorphisms in EC affected individuals and controls were tested for deviation in Hardy-Weinberg equilibrium (HWE) by using the Chi-square test. The strength of the association of alleles and genotypes with EC risk was measured by overall odds ratio (OR) with a 95% confidence interval (95% CI). Lastly, we investigated the associations of haplotypes of VEGF +936C/T, +405G/C and -7C/T polymorphisms with EC risk using online SNP stats software ( https://www.snpstats.net/start.htm). A probability value of P<0.05 was considered statistically significant. The age, gender, diet, smoking status and alcohol consumption were selected as adjustment variables.

Results Demographic characteristics

In total, 464 study participants comprising 231 sporadic EC affected individuals (94 males and 137 females) and 233 controls (94 males and 139 females) were included in the present study. The number of female cancer patients (n=137) was higher in comparison to male (n=94). The mean age (±SD) of EC affected individuals at diagnosis was 56.83±12.55 yr and that of the controls was 54.15 ± 13.51 yr. About 64 per cent of affected individuals developed cancer of oesophagus after the age of 50 yr. The demographic characteristics of EC affected individuals and controls are given in Table I.

Table I. Baseline data of oesophageal cancer affected individuals and healthy controls

Variable

Affected individuals

n=231, n(%)

Controls

n=233, n(%)

Gender Males 94 (40.7) 94 (40.34) Females 137 (59.3) 139 (59.66) Age ≤50 yr 84 (36.36) 88 (37.77) >50 yr 147 (63.64) 145 (62.64) Total (Mean±SD) 56.83±12.55 54.15±13.51 Males (Mean±SD) 58.46±12.94 56.07±12.63 Females 55.77±12.24 53.921±13.05 Pathological type Squamous cell carcinoma 214 (92.64) - Adenocarcinoma 17 (7.36) - VEGF +936C/T, +405G/C and -7C/T Polymorphisms and risk of oesophageal cancer

The genotype distributions of VEGF +405G/C and -7C/T polymorphisms were in HWE in the control group (P > 0.05). The distribution of genotypes and alleles of VEGF polymorphisms is given in Table II. A statistically significant association of CT genotype of VEGF-7C/T polymorphism with reduced risk of EC was observed in total study participants (OR: 0.54, 95% CI: 0.34–0.85; P=0.02) and also in the female group (Table III). T allele of VEGF-7C/T polymorphism was marginally associated with reduced risk of EC in total study participants (P =0.05) and the female group (P=0.06). VEGF-7C/T polymorphism was significantly associated with reduced risk of oesophageal cancer under dominant, co-dominant, over-dominant and log-additive genetic models in total study participants and also in the female group (Table III). No significant difference was found in genotype, allele frequencies (Table II) and genetic models (Table IV) of VEGF +405G/C and +936C/T polymorphisms in total EC affected individuals and also in male and female groups (P>0.05).

Table II. Association of VEGF +936C/T, +405G/C and -7C/T polymorphisms with oesophageal cancer risk

Affected individuals (n=231)

n(%)

Controls (n=233)

n(%)

OR (95% CI) P value# VEGF +936C/T (rs3025039) Genotype CC 183 (79.22) 192 (82.4) Reference CT 47 (20.35) 41 (17.6) 1.11 (0.68–1.82) 0.65 TT 1 (0.43) 0 - - Allele C 413 (89.39) 425 (91.2) Reference T 49 (10.61) 41 (8.8) 1.23 (0.79–1.9) 0.35 VEGF +405C/G (rs2010963) [HWE: Affected individuals (P=0.63); Controls (P=0.29); Both (P=0.23)] Genotype CC 17 (7.36) 20 (8.58) Reference CG 98 (42.42) 110 (47.21) 1.05 (0.52–2.11) 0.23 GG 116 (50.22) 103 (44.21) 1.32 (0.66–2.66) 0.43 Allele C 132 (28.57) 150 (32.19) Reference G 330 (71.43) 316 (67.81) 1.19 (0.89–1.57) 0.23 VEGF-7C/T (rs25648) [HWE: Affected individuals (P=0.51); Controls (P=0.22); Both (P=0.69)] Genotype CC 182 (78.79) 162 (69.53) Reference CT 45 (19.48) 68 (29.18) 0.54 (0.34–0.85) 0.02* TT 4 (1.73) 3 (1.29) 1.19 (0.26–5.4) 0.82 Allele C 409 (88.53) 392 (84.12) Reference T 53 (11.47) 74 (15.88) 0.69 (0.47–1) 0.05

Table III. Gender-wise stratification analysis on the association between VEGF polymorphisms and oesophageal cancer risk.

Males

Affected individuals (94); Controls (94)

Females

Affected individuals (137); Controls (139)

Genotype

Affected individuals

n(%)

Control

n(%)

OR 95% CI) P value#

Affected individuals

n(%)

Control

n(%)

OR (95% CI) P value## VEGF +936C/T Genotype CC 72 (76.6) 75 (79.79) Reference 111 (81.02) 117 (84.17) Reference CT 22 (23.4) 19 (20.21) 1.04 (0.46–2.35) 0.93 25 (18.25) 22(15.83) 1.17 (0.62–2.21) 0.52 TT 0 0 - - 1(0.73) - - - Allele C 166 (88.3) 169 (89.89) Reference 247 (90.15) 256 (92.09) Reference T 22 (11.7) 19 (10.11) 1.18 (0.61–2.25) 0.62 27 (9.85) 22 (7.91) 1.27 (0.7–2.3) 0.42 VEGF+405C/G Genotype CC 3 (3.19) 7 (7.45) Reference 14 (10.22) 13(9.35) Reference CG 45 (47.87) 45 (47.87) 2.33 (0.57–9.59) 0.24 53 (38.69) 65(46.76) 0.76 (0.33–1.75) 0.51 GG 46 (48.94) 42 (44.68) 2.56 (0.62–10.52) 0.19 70 (51.09) 61(43.88) 1.06 (0.46–2.44) 0.88 Allele C 51 (27.12) 59 (31.38) Reference 81 (29.56) 91 (32.73) Reference G 137 (72.87) 129 (68.69) 1.23 (0.79–1.92) 0.36 193 (70.44) 187 (67.27) 1.16 (0.8–1.66) 0.42 VEGF-7C/T Genotype CC 70 (74.47) 64 (68.09) Reference 112 (81.75) 98(70.5) Reference CT 21 (22.34) 27 (28.72) 0.66 (0.3–1.42) 0.56 24 (17.52) 41(29.5) 0.49 (0.28–0.88) 0.02* TT 3 (3.19) 3 (3.19) - - 1 (0.73) - - - Allele C 161 (85.64) 155 (82.45) Reference 248 (90.51) 237 (85.25) Reference T 27 (14.36) 33 (17.55) 0.79 (0.45–1.37) 0.39 26 (9.49) 41 (14.75) 0.6 (0.36–1) 0.06

Table IV. Genetic models of VEGF polymorphisms and oesophageal cancer risk

Polymorphism Total P value# Males Females Model OR (95% CI) OR (95% CI) P value## OR (95% CI) P value### VEGF +936C/T Co-dominant 1.11 (0.68–1.82) 0.65 - - 1.27 (0.62–2.21) 0.52 Dominant 1.13 (0.69–1.84) 0.63 - - 1.21 (0.64–2.27) 0.55 Over dominant 1.11 (0.68-1.81) 0.68 - - 1.16 (0.62–2.19) 0.64 Log-additive 1.14 (0.71–1.85) 0.59 - - 1.24 (0.67–2.30) 0.48 VEGF +405C/G Dominant 0.80 (0.54–1.19) 0.55 1.24 (0.76–2.02) 0.38 0.73 (0.44–1.21) 0.44 Dominant 0.81 (0.55–1.19) 0.28 2.44 (0.61–9.74) 0.19 0.77 (0.48–1.25) 0.29 Recessive 0.96 (0.48–1.94) 0.91 1.14 (0.64–2.02) 0.66 1.36 (0.5–2.58) 0.72 Over dominant 0.82 (0.56–1.20) 0.31 1 (0.56–1.77) 1.0 0.73 (0.45–1.18) 0.2 Log-additive 0.87 (0.64–1.18) 0.37 0.78 (0.48–1.27) 0.32 0.89 (0.62–1.28) 0.52 VEGF-7C/T Co-dominant 0.54 (0.34–0.85) 0.02* 0.66 (0.3–1.42) 0.56 0.49 (0.28–0.88) 0.02* Dominant 0.56 (0.36–0.88) 0.01* 0.67 (0.32–1.4) 0.29 0.52 (0.29–0.92) 0.02* Recessive 1.54 (0.29–8.22) 0.61 0.87 (0.12–6.32) 0.89 - - Over dominant 0.53 (0.34–0.84) 0.01* 0.66 (0.31–1.43) 0.29 0.49 (0.27–0.87) 0.01* Log-additive 0.63 (0.42–0.95) 0.03* 0.73 (0.39–1.39) 0.34 0.56 (0.32–0.98) 0.04* Association of VEGF haplotypes with oesophageal cancer risk

Analysis of haplotypes of VEGF +936C/T, +405G/C and -7C/T polymorphisms showed that C+936G+405T-7 haplotype was significantly associated with decreased risk (P=0.01)of EC in total study participants (Table V). Haplotype C+936C+405C-7 was marginally associated (P=0.07) with decreased EC risk in total study participants. When we performed a gender-stratified analysis, a significant association of C+936G+405T-7 haplotype (P=0.02) with decreased risk of EC was observed in the male group (Table V). There was no association between other haplotypes and EC risk.

Table V. Haplotype frequencies of VEGF polymorphisms in oesophageal cancer affected individuals and controls

Haplotype

Affected individuals

frequency

Controls frequency OR (95% CI) P value# Total C+936G+405C-7 0.547 0.476 Reference C+936C+405C-7 0.26 0.301 0.71 (0.5–1.02) 0.07 C+936G+405T-7 0.087 0.134 0.52 (0.31–0.85) 0.01* T+936G+405C-7 0.06 0.047 0.89 (0.4–2) 0.79 T+936G+405T-7 0.02 0.02 0.60 (0.15–2.33) 0.46 T+936C+405C-7 0.018 0.016 1.32 (0.28–6.26) 0.73 Males C+936G+405C-7 0.531 0.47 Reference C+936C+405C-7 0.249 0.261 0.64 (0.32–1.31) 0.23 C+936G+405T-7 0.103 0.168 0.39 (0.18–0.87) 0.02* T+936G+405C-7 0.055 0.041 0.55 (0.11–2.73) 0.46 T+936C+405C-7 0.022 0.053 0.37 (0.05–2.63) 0.32 Females C+936G+405C-7 0.556 0.497 Reference C+936C+405C-7 0.268 0.317 0.73 (0.48–1.11) 0.15 C+936G+405T-7 0.078 0.107 0.6 (0.31–1.17) 0.13 T+936G+405T-7 0.005 0.035 0.06 (0–21.01) 0.34 T+936G+405C-7 0.066 0.033 1.68 (0.63–4.45) 0.3 Discussion

In the present case-control study, we evaluated the association between VEGF +936C/T, +405G/C and -7C/T functional polymorphisms and the risk of EC. An association of VEGF-7C/T polymorphism with reduced risk of EC was found. In this study, no significant correlation of VEGF +405G/C and +936C/T polymorphisms with EC risk was found. As per authors knowledge, there is no reported study on VEGF-7C/T polymorphism in EC and limited data on VEGF +936C/T and +405G/C polymorphism. Our results agreed with the case-control study conducted by Gu et al17 in a Chinese population in which no association between VEGF +936C/T polymorphism and EC risk was found. Combined CT+TT genotype of VEGF +936C/T polymorphism was associated with increased risk of EC in Caucasian patients13. Our results are contrary to the previous two reports from north India. A study from north India on Kashmiri patients reported an association of the CT genotype and T allele of VEGF +936C/T polymorphism with an increased risk of EC7. In another study, the VEGF +936CT genotype was reported to be associated with an increased risk of oral cancer in north Indian patients18. The currently available results on VEGF +936C/T polymorphism in GIT cancers are inconsistent. A meta-analysis19 of 29 studies including 13,293 cases and 12,308 control individuals reported a significant association of the T allele of VEGF +936C/T polymorphism with increased risk of oral cancer. A significant association of VEGF +936TT genotype with advanced stage of disease and distant metastasis has been reported in Korean colorectal patients20. Correlation of T allele of VEGF +936C/T polymorphism with improved overall survival has been reported in Caucasian EC patients5. Association of VEGF+936TT genotype with worse overall survival was reported in Korean gastric cancer21, with lower overall survival in Greek colorectal adenocarcinoma22 and with increased overall survival in Spanish colorectal cancer patients23. In bevacizumab-treated metastatic colorectal cancer patients, the +936TT genotype was significantly associated with better time to treatment failure as compared to CC and CT genotype24.

The present study has observed no association of VEGF+405C/G (-634 C/G) polymorphism with EC risk. Similarly, no association of VEGF+405C/G polymorphism with EC risk has been reported in Caucasian patients12. A significant association of GG genotype and G allele of VEGF +405C/G polymorphism with increased risk of EC has been reported in north Indian Kashmiri patients7. The VEGF +405C/G polymorphism has been studied in various GIT cancers with mixed results. VEGF +405CC genotype was associated with larger tumour size and advanced stage of disease in Greek gastric cancer patients25. A meta-analysis26 of six studies on VEGF-634C/G polymorphism involving 1504 cases and 1707 controls reported that the-634GG genotype was significantly associated with reduced risk of gastric cancer in Europeans, whereas the -634GC genotype was associated with increased gastric cancer risk in Asians. No association of VEGF +405C/G polymorphism with gastric cancer risk was reported in south Indian patients27.

The association of VEGF -634CC genotype with decreased survival rate has been reported in Greek gastric cancer25 and colorectal adenocarcinoma patients22. A case-control study on gastric cancer in the Omani population reported that patients with VEGF +405GG genotype had a lower survival rate as compared to patients with combined VEGF+405CC+CG genotype28.

In the present case-control study, we found that the CT genotype of VEGF-7C/T polymorphism was significantly associated with reduced risk of EC in the total patients and also female group. As per authors knowledge, there is no published study on VEGF-7C/T polymorphism in EC so far. There was no association of VEGF -7C/T polymorphism with colorectal cancer in Japanese9, hepatocellular cancer in Korean10 and gastric cancer in the Chinese8 population.

Furthermore, we evaluated the combined effects of VEGF +936C/T, +405C/G and -7C/T polymorphisms and observed that C+936G+405T-7 haplotype was significantly associated with reduced risk of EC in total study participants and also in the male group. Association of CGT haplotype of VEGF -460T/C, +405C/G and +936C/T polymorphisms with increased risk of oesophageal adenocarcinoma has been reported in Caucasians13. Association of CGC haplotype of VEGF -460T/C, +405C/G and +936C/T polymorphisms with reduced overall survival has been documented in Caucasian oesophageal cancer patients5. TCT haplotype of VEGF -460T/C, +405C/G and +936C/T polymorphisms were associated with decreased risk of gastric cancer in Korean patients29. In surgically resected Korean gastric adenocarcinoma patients, the CACC haplotype of VEGF -460T/C, -116G/A, +405C/G and +936C/T polymorphisms were significantly associated with worse overall survival and disease-free survival21. Haplotype -634C/+936C and -634G/+936T were associated with decreased risk to colorectal cancer in Korean patients20. Romanian surgically treated colorectal adenocarcinoma patients with -2578A/-634G/+936T haplotype had worse overall survival ratesin comparison to cases with -2578C/-634G/+936C haplotype30. Haplotypes -2578C/ -460T/ +405C/ +936C and -2578C/-460T/+405C/+936T have been reported to be associated with inferior response rate in metastatic colorectal cancer31. In Chinese stage I and II gastric cancer patients, -460T/+405G/-7C and -460C/+405G/-7C haplotypes were associated with poor survival8. Haplotype CCGAGCCC of VEGF-2578/-1203/-1190/-1179/-1154/-634/-7/+936 polymorphisms was associated with smaller tumour size in Korean hepatocellular cancer patients10. There was no significant association of genotype, allele and haplotype of VEGF -460T/C, +405G/C and +936C/T polymorphisms with gastric cancer risk in Omani patients28.

In the present study, we observed a gender-specific association of VEGF polymorphisms with EC risk. CT genotype of VEGF-7C/T polymorphism was significantly associated with decreased risk of EC in total study participants and also in female affected individuals, whereas the C+936G+405T-7 haplotype was significantly associated with reduced risk of EC in total affected individuals and the male group. Gender-specific differences in the regulation of vascular remodelling have been demonstrated in the perigonadal adipose tissue of mice subjected toa high-fat diet32. The investigators reported higher levels of VEGFA and VEGFR2 and high vascular density in the visceral adipose tissue of female mice as compared to male mice in response to a high-fat diet. Out of 137 female patients, 107 (78%) were postmenopausal in the present study. It has been documented that oestradiol, the main sex hormone in females, modulates angiogenesis via its effects on endothelial cells and regulates VEGFA expression in adipose tissue33. In non-small cell lung cancer, it has been reported that elevated circulating oestrogen and progesterone in females may promote VEGF secretion and tumour angiogenesis34. Significantly higher adverse effects of chemotherapy have been reported in female oesophagogastric cancer patients in comparison to male patients35. Recently, Hall et al36 suggested that gender-based analysis must be included in cancer research for appropriate interpretation and implication of results in real-world practice.