A total of 296 subjects from 90 gastric cancer case families, and 319 subjects from 90 control families were included. The individuals in this study were selected from four villages of two counties in the Henan Province of China. Samples were obtained through cluster sampling in this highly cancer-prevalent area following a previous cancer screening. The gastric cancer patient samples were collected from these four villages between August 2003 and March 2007, including samples obtained from patients who had suffered from non-cardiac adenocarcinoma of the stomach during the five years prior to the screening. The patients were histopathologically diagnosed, and were considered as the probands. Their first- and second-degree relatives were further investigated. The control families were selected to correspond to the number of case family members and the age of the probands (±5 years), gender and address. Controls were healthy community-based individuals without any similar digestive diseases and had no marital ties with the case family. Subjects in this study provided informed consent prior to participating in the study, and the private data were kept as coded data.

Blood samples were collected from the subjects and DNA was extracted using the salting-out procedure and stored in a −80°C freezer for subsequent studies.

The COX-2-1195G>A and -765G>C polymorphisms were genotyped using polymerase chain reaction-restriction fragment length polymorphism. Primers were designed and synthesized by Beijing SBS Genetech Co., Ltd. The sequences were as follows: COX-2-1195G>A: 5′-CCCTGAGCACTACCCATGAT-3′ (forward) and 5′-GCC CTTCATAGGAGATACTGG-3′ (reverse). COX-2-765G>C: 5′-TATTATGAGGAGAATTTACCTTTCGC-3′ (forward) and 5′-GCTAAGTTGCTTTCAACAGAAGAAAT-3′ (reverse). Thermo cycling conditions of PCR were: COX-2-1195G>A: 95°C for 5 min, 35 cycles of 95°C for 30 sec, 54°C for 40 sec, 72 °C for 45 sec and finally 72°C for 10 min; COX-2-765G>C: 95°C for 5 min, 35 cycles of 95°C s for 30 sec, 55.5°C for 40 sec, 72°C for 45 sec and finally 72°C for 10 min. The PCR products were digested with restriction enzymes PvuII (Takara Biotechnology Co., Ltd.) at 37°C overnight for analysis of COX-2-1195G>A and with HhaI (Takara Biotechnology Co., Ltd.) for -765G>C. The digested products were separated using 3% agarose gel electrophoresis, stained with ethidium bromide and visualized under ultraviolet light. Subsequently, an altered pattern of bands was observed: a single band of 273 bp indicated that COX-2-1195A/A was not digested; three bands of 273, 220 and 53 bp indicated that G/A was only partially digested; two bands of 220 and 53 bp indicated that G/G was completely digested (Fig. 1). Analogically, COX-2-765C/C appeared as a single band of 100 bp; G/C, three bands of 100, 75 and 25 bp, and G/G, two bands of 75 and 25 bp (Fig. 2). The 25-bp fragment was not distinguishable from the primer-dimer band in the agarose gel. Each type of PCR product was sequenced to ascertain the genotype. Finally, 10% of the specimens were genotyped in duplicate using the same assay and the results were 100% concordant.

The H. pylori immunoglobulin G antibody was determined on the serum specimens of all case and control family members using ELISA. The relevant reagents were provided by the National Institute for Communicable Disease Control and Prevention of the Chinese Center for Disease Control and Prevention.

The Hardy-Weinberg equilibrium equation was used to determine whether the proportion of each genotype obtained was in agreement with the expected values as calculated from allele frequencies. The Chi-square test was used to examine the differences in demographic variables. Mean ages of case and control family members were compared using the t-test. Unconditional logistic regression analysis was used to estimate the odds ratio (OR) and its 95% confidence interval (CI) as a measure of the association between the genotype carrier and risk of gastric cancer. The interaction between H. pylori infection and polymorphisms of COX-2 was analyzed using stratification analysis. P-values were two-sided, and P<0.05 was considered as statistically significant. Data analysis was performed by Statistical Product and Service Solutions software (Version 15.0, SPSS, Inc., Chicago, IL, USA) unless otherwise specified.

The demographic distribution of case family members and healthy controls is shown in Table I. The mean age of the gastric cancer case families was 44.0±16.6 years [mean ± standard deviation (SD)] and the controls was 44.3±15.9 years (mean ± SD, P=0.81), indicating that there was no difference between the two groups. Males comprised 42.9% of case family members compared with 43.6% among controls (P=0.87). Regarding H. pylori infection, there was a significant difference between the case family groups and the controls (72.3% versus 62.7%; P=0.01). With regard to familial history of gastric cancer, case groups exhibited a higher freqeuncy than the controls (19.6% versus 5.3%; P<0.01).

The genotype distribution and allele frequencies of COX-2 polymorphisms in gastric cancer case and control families are shown in Table II. The genotype distribution of COX-2-1195G>A and -765G>C in the case and control family members did not significantly deviate from the expected Hardy-Weinberg equilibrium (P>0.05). For COX-2-1195G>A, the frequencies of GG, GA and AA genotypes were 17.9, 49 and 33.1% in the case family group and 25.1, 52.0 and 22.9% in the control group, respectively. Compared with the frequency of the GG genotype, an association was found between the COX-2-1195AA genotype and the risk of gastric cancer (AA genotype: OR = 2.03; 95% CI, 1.27–3.22). A statistical difference was observed between the A carrier and A allele frequencies among the case and control family members (A carriers: OR = 1.55; 95% CI, 1.02–2.28; A allele: OR = 1.42; 95% CI, 1.09–1.81). With regard to COX-2-765G>C, the frequencies of GG, GC and CC genotypes in the case and control family group were 81.4, 17.9 and 0.7%, and 86.2, 13.5 and 0.3%, respectively. Due to the low occurrence (<1%) of the CC genotype, heterozygous and homozygous variants were combined for the analysis (C carriers: G/C + C/C). Compared with the GG homozygous variant or G allele, C carriers or C allele were not shown to be statistically different between the two groups (C carriers: OR=1.44; 95% CI, 0.91–2.23; C allele: OR = 1.45; 95% CI, 0. 92–2.25).

As shown in Table III, the genotype distribution of COX-2-1195G>A and -765G>C in different relative grades between the case and control family members was further examined. Since the spouse had no genetic affinity with the host of the family, the spouse was disregarded while analyzing the gastric cancer genetic susceptibility. The OR value of COX-2-1195 AA genotype in the first-degree relatives (OR=3.35; 95% CI, 1.69–6.12) was higher than that in the second-degree relatives (OR=2.66; 95% CI, 1.06–6.82). Similarly, the OR values of A carriers (GA + AA) and A allele exhibited a diminished trend from closer degrees of relationship (A carriers: first-degree relatives: OR=2.11; 95% CI, 1.19–3.67; second-degree relatives: OR=1.98; 95% CI, 0.89–4.63; A allele: first-degree relatives: OR=1.77; 95% CI, 1.29–2.41; second-degree relatives: OR=1.69; 95% CI, 0.99–2.68). Compared with the GG genotype, the mutant genotype distribution of -765G>C was not correlated with gastric cancer in the first- or second-degree relatives, due to the fact that the OR values included 1.00 (C carriers: first-degree relatives: OR=1.62; 95% CI, 0.89–2.87; second-degree relatives: OR=1.35; 95% CI, 0.54–3.22).

As COX-2-1195G>A and H. pylori were involved in gastric cancer, the interaction between this functional SNP and H. pylori infection status was examined using stratified analysis (Table IV). Taking H. pylori-seronegative AA genotype and A carriers together as a reference, H. pylori-seropositive COX-2-1195 AA genotype and A carriers exhibited significantly increased susceptibility to gastric cancer (AA genotype: OR=2.96; 95% CI, 1.57–5.58; A carriers: OR=2.04; 95% CI, 1.18–3.52).