1中國醫藥大學臨床醫學研究所; 2Terry Fox癌症研究室; 3國軍台中總醫院泌尿外科;
4國防醫學院臨床醫學研究所; 5中國醫藥大學附設醫院泌尿外科; 6彰化基督教醫院眼科部
Significant assosciation of cyclooxygenase 2 genotypes with upper tract urothelial cancer
Cheng-Hsi Liao1,3*, Chin-Mu Hsu2*, Chung-Yu Huang4, Chia-Wen Tsai2, Hsi-Chin Wu5, Pei-Shin Hu6, Hao-Ai Shui4 and Da-Tian Bau1,2
1 Graduate Institute of Clinical Medical Science, China Medical University, Taichung, Taiwan, R.O.C.;
2 Terry Fox Cancer Research Laboratory and 5 Department of Urology, China Medical University Hospital, Taichung, Taiwan, R.O.C.;
3 Department of Urology, Taichung Armed Forces General Hospital, Taichung, Taiwan, R.O.C.;
4 Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei, Taiwan, R.O.C.;
6 Department of Ophthalmology, Changhua Christian hospital, Changhua, Taiwan, R.O.C.
* These Authors contribute equally to this work
Purpose: Aim: Reliable biomarkers are in urgent need for diagnosis, outcome prediction or treatment effect monitoring for upper tract urothelial carcinomas (UTUC). Although up-regulation of cyclooxygenase 2 (COX-2) are found in stroma and tumor cells in more than half of the UTUC patients investigated, the genomic contribution of COX-2 to UTUC is never studied. The study was to evaluate the association of six polymorphic genotypes of COX-2 with UTUC within a Taiwanese population. Materials and Methods: A total of 218 UTUC patients and 580 healthy controls were genotyped for six COX-2 polymorphisms, A-1195G, G-765C, T+8473C, intron 1, intron 5, and intron 6, of their association with UTUC risk. Results: The distribution of genotypes of COX-2 G-765C and intron 5 were significantly different between UTUC patient and control groups (p=0.0001 and 0.0016, respectively), while others were not (p>0.05). The haplotype analysis showed that compared with GG/TT haplotype of COX-2 G-765C/intron 5, people carrying GG/AT variants have a significant increased risk of UTUC (odds ratio, OR=4.83, 95% confidence interval, 95%CI=1.79-13.06) while people carrying CG/TT variants have a decreased risk (OR=0.26, 95%CI=0.14-0.49). Conclusion: Our results suggested that individual and combined COX-2 G-765C/intron 5 genotypes play a role in controlling the COX-2 expression and UTUC development.
Materials and Methods: Sample collection. A total of 218 patients with UTUC were recruited at China Medical University and Kaohsiung Medical University medical centers, all of whom were diagnosed by pathologic examination of specimens obtained by biopsy or surgical resection. The clinical and histopathologic information were collected from patient charts and pathologic reports. The information was reviewed, and the data were entered into the database. The tumor stage was assigned according to the TNM staging system (21), and the pathologic grade was determined according to the World Health Organization criteria (22). Five hundreds and eighty healthy individuals, who had been matched with the patients with age, admitted to the same hospital for health checkup and who had no previous diagnosis of urologic neoplastic disease or other malignancy were enrolled as controls. All the subjects enrolled were provided an informed consent and Human Research Committees have approved this study.
Genotyping conditions. The total genomic DNA of each subject was extracted from the leucocytes of peripheral blood and stored as previously published (23-25). The polymerase chain reaction (PCR) cycling conditions were: one cycle at 94°C for 5 min; 35 cycles of 94°C for 30 sec, 55°C for 30 sec, and 72°C for 30 sec, and a final extension at 72°C for 10 min. Pairs of PCR primer sequences and restriction enzyme for each DNA product of COX-2 genotyping work are all listed in Table I. The PCR products were cut by proper restriction enzymes and the reaction was incubated for 2 h at 37°C. Then, 10 ml of each PCR product was loaded into a 3% agarose gel for electrophoresis.
Statistical analyses. Two hundreds and eighteen cases and five hundreds and eighty controls were analyzed in the presented Tables. To ensure that the controls used were representative of the general population and to exclude the possibility of genotyping error, the deviation of the genotype frequencies of COX-2 single nucleotide polymorphism in the controls from those expected under the Hardy-Weinberg equilibrium was assessed using the goodness-of-fit test. Pearson’s chi-square test or Fisher’s exact test (when the expected number in any cell was less than five) was used to compare the distribution of the COX-2 genotypes between cases and controls. Cancer risk associated with the genotypes was estimated as odds ratio (ORs) and 95% confidence intervals (CIs) using unconditional logistic regression. All statistical tests were performed using SPSS for Windows (version 14.0; SPSS Inc., Chicago, IL, USA) on two-sided probabilities. The correlation between categorical variables was calculated for statistical significance using Pearson’s chi-square test and the threshold for significance was p<0.05.
Results: The frequency distributions of clinical characteristics for the subjects (218 UTUC patients and 580 healthy controls) are shown in Table II. Epidemiologically, there was no difference in the frequency distribution between the gender (p=0.4256) or age (p=0.8518) since the population was well matched (Table II). From the clinical and pathological viewpoints, tumors distributed in renal pelvic, ureter and multiple sites were 38.5%, 34.9% and 26.6%, respectively. Among the patients, 60.6% were high grade, and 77.1% were of stages lower than pT3 (Table II).
The frequencies of the genotypes for COX-2 A-1195G, G-765C, T+8473C, intron 1, intron 5, and intron 6 among the UTUC patients and healthy controls are summarized and analyzed in Table III. Among the six polymorphic genotypes investigated, two of them, G-765C (OR=0.32, 95%CI=0.19-0.55, p=0.0001) and intron 5 (OR=3.91, 95%CI=1.71-8.95, p=0.0016), were found to be differentially distributed between UTUC case and control groups (Table III). First, the frequencies of GG and CG genotypes of COX-2 G-765C were 92.7 and 7.3% among UTUC cases, and 80.2 and 19.8% among healthy controls, respectively. Second, the frequencies of TT and AT genotypes of COX-2 intron 5 were 93.6 and 6.4% among UTUC cases, and 98.1 and 1.9% among healthy controls, respectively (Table III). As for other polymorphic sites of COX-2, there was no difference in the distribution of genotypes among UTUC cases and controls (Table III).
In the next step, we have performed the allelic frequency analysis, and the frequencies of the alleles for COX-2 A-1195G, G-765C, T+8473C, intron 1, intron 5, and intron 6 among the UTUC cases and healthy controls are summarized in Table IV. Consistent to the findings of Table III, G-765C and intron 5 of COX-2 were found to be associated with UTUC risk in Table IV. In detail, higher frequencies of G allele in G-765C and A allele in intron 5 in the UTUC case group than the control group, associating to higher risk of UTUC (Table IV; p=0.0001 and 0.0017, respectively). Regarding to the other four COX-2 polymorphic sites, no distribution of their allelic frequencies was found to be significantly different between the control and UTUC case groups (Table IV).
Considering the possible interactions between the two determinant COX-2 genotypes for UTUC susceptibility, the haplotypic distributions of COX-2 G-765C and intron 5 were further analyzed (Table V). We have set the most abundant genotypes for both G-765C and intron 5 genotypes as wild-type for haplotypic combination. Under this criteria, the “GG” genotype for COX-2 G-765C and “TT” for COX-2 intron 5 were selected, resulting in setting GG/TT combinative genotype in G-765C/intron 5 as the reference haplotype. Compared with the reference haplotype of COX-2 G-765C/intron 5, the GG/AT group has a significant higher risk of UTUC (OR=4.83, 95%CI=1.79-13.06, p=0.0014) while CG/TT has a lower risk (OR=0.26, 95%CI=0.14-0.49, p=0.0001) (Table V). After adjusting for age and gender, the significances became more obvious for the GG/AT and CG/TT groups with their individual ORs altered to 4.86 and 0.32, remained statistically significant (Table V). The combination of CG/AT did not confer significantly altered cancer risk compared to the wild-type haplotype before or after adjusting for age and gender (Table V).
Conclusions: Urothelial carcinoma is the second most common cancer which usually arises from the urothelium with transitional cell differentiation, including that of the renal pelvis, ureter and bladder. In literature, there were at least 600 papers investigating the contribution of individual genomic variations to bladder cancer (26-30), while few to UTUC (7, 31, 32). Following the rule of molecular central dogma, the single nucleotide variations on COX-2 among people may determine their differential expression of COX-2 and personal susceptibility to cancers. The supporting data comes from that COX-2 is often undetectable in normal tissues, whereas overexpression of COX-2 has been observed in neoplastic cells of canine (33) and human RCC (34-36). In our previous report, COX-2 was found to be up-regulated in both stromal and tumor cells of more than half of the UTUC patients and the positive expression of COX-2 in stromal cells may be a biomarker for UTUC-specific death and recurrence (10). However, the contribution of COX-2 genotypes to UTUC has never been studied. In this study, the genotypes of COX-2 of 218 UTUC patients together with 580 controls (Table III) were examined. Statistically, the distributions of COX-2 genotypes for G-765C and intron 5 were differentially distributed among the UTUC and healthy control groups (Table III). In addition, the allelic frequencies of the two polymorphisms were also differentially distributed between the two groups (Table IV). The results showed that the G allele of G-765C and A allele of intron 5 were associated with higher risk for UTUC and the haplotype analysis suggested that individuals with GG/AT and CG/TT haplotypes at G-765C/intron 5 were at altered risk of UTUC before and after adjusting of gender and age (Table V). UTUC is not a common cancer in Taiwan and the world, which may be one of the limitation for genomic study of UTUC. Compared with previous finding, we have enlarged the sample size from 56 cases and 436 controls in 2011 (37) to 218 cases and 580 controls (Table II). The strengthened sample size and the same trend of significant genotype distribution after age and gender adjustments highlighted the value, accuracy and reliability of the overall findings (Table V).
This is the first UTUC genomic study showing COX-2 G-765C and intron 5 genotypes were associated with UTUC risk. The G allele of Cox-2 G-765C and A allele of COX-2 intron 5 were genomic risk factors and may serve as early screening and prediction biomarkers for UTUC in Taiwan and all over the world.