UGT2B17 Deletion Polymorphism is a Risk Factor for Upper Aero digestive-Tract Cancer in Japanese: A Case-Control Study

Background: The UDP-glucuronosyltransferase 2 family polypeptide B17 (UGT2B17) detoxifies carcinogens found in tobacco smoke and ethanol in alcoholic drinks. Tobacco carcinogens and ethanol synergistically raise the risk for upper aero digestive tract (UADT) cancer (head and neck squamous carcinoma, and esophageal cancer). Deletion polymorphism of the UGT2B17 gene (UGT2B17-deletion) is a much more common copy number variant among Japanese than other populations. Thus, we conducted a nested and age/gender-matched case-control (1:1) study to determine if UGT2B17-deletion associates with cancer risk, including UADT-cancer in Japanese. Methods: Polymerase chain reaction was used to determine UGT2B17-deletion using DNA samples derived from peripheral blood or tumor tissue. Cases were cancer patients and controls were non-cancer patients. Nonconditional and conditional logistic regression analyses were performed. To overcome the issue of multiple-testing, Bonferroni correction was applied to set p < 0.003 as statistically significant. Results: A total of 3,092 patients were enrolled. UGT2B17-deletion was detected in 74% of the 1,887 non-cancer patients and 77% of the 1,205 cancer patients. UGT2B17-deletion was a significant risk factor for UADT-cancer development: odds ratio, 2.07; 95% confidence interval, 1.34 to 3.20, p=0.001. In contrast, UGT2B17-deletion was not a risk for any other type of cancers represented in our study population. Conclusions: These results suggest that UGT2B17-deletion may associate to increase the risk of UADT-cancer in Japanese.


Introduction
UDP-glucuronosyltransferase (UGT) belongs to family of enzymes glucuronidate carcinogens, C19 steroids, and xenobiotics [1][2][3]. UGT2B17 is a member of the UGT family, and deletion polymorphism of the UGT2B17 gene (UGT2B17-deletion) is a much more common copy number variant (CNV) among Japanese than other populations [4][5][6][7]. UGT2B17 metabolize androgen hormones. The UGT2B17-deletion may therefore decrease androgen elimination and thus may associate with the development of prostate cancer, since the growth of these cancerous cells is facilitated by androgens [8]. However, it remains controversial whether UGT2B17-deletion is a genuine risk factor for prostate cancer [9] or not [10]. UGT2B17deletion is associated with decreased glucuronidation of tobaccospecific carcinogens [11] and may be associated with increased risk of lung adenocarcinoma in women [12], although the latter was not confirmed in a recent study [13]. In addition, UGT2B17 was shown to be responsible for 32% of all ethanol glucuronidation [14]. Tobacco and alcohol are major risk factors for upper aerodigestive tract (UADT) cancer, which includes head and neck cancer, and esophageal cancer [15], because these mucosal membranes are exposed directly to Epidemiology: Open Access Urashima, et al., Epidemiology (sunnyvale)  tobacco smoke and alcoholic drink. Multiple alcohol dehydrogenase gene polymorphisms were reported to associate with UADT cancer in the Japanese population [16]. However, the effects of UGT2B17deletion have not been studied, despite UGT2B17-deletion being highly prevalent in Japanese. Therefore, we aimed to examine the association between UGT2B17-deletion and UADT cancer as well as other types of cancers by conducting age-and gender-matched and nested case-control studies.

Study design
As a posthoc analyses, we retrospectively used DNA samples of nine prospective cohort studies conducted separately and designed for other objectives: 1) cardiosurgery (n=287), recruited in 2010 at the intensive care unit of Jichi Medical University Saitama Medical Center to study associations between circulating 25-hydroxyvitamin D (25OHD) levels, single-nucleotide polymorphisms (SNPs) of vitamin D receptor and predicted operative mortality of patients with cardiovascular disease [17]; 2) dialysis (n=1,042), ongoing project started from 2011; 3) diabetes mellitus (n=422), recruited between 2011 and 2012 at outpatient clinics of Jikei University school of medicine and Shin-Kashiwa clinic to study associations between 25OHDs, SNPs of vitamin D receptor and renal function of patients with diabetes [18] Both case and control were nested in the study cohorts by matching age and gender. Specifically, the case cohorts comprised patients with any kind of cancer, or specific for each kind of cancer. For each case, one control of the same age and gender was randomly selected from patients without cancer from the mixture of study cohorts.

Samples
With each patient's consent, peripheral blood samples or tumor tissue were collected during a visit to an outpatient clinic and during an operational procedure, respectively. QIAamp DNA Micro Kits 50 (Qiagen, Tokyo, Japan) were used to purify extracted DNA, and NanoVue Plus (General Electric Healthcare Japan, Tokyo, Japan) was used to measure DNA concentration in each sample. The samples were then frozen at -80°C until use.

Polymerase chain reaction to differentiate CNVs of the UGT2B17 gene
There is a high level of sequence identity between the UGT2B17 and UGT2B15 genes. Therefore, gene-specific polymerase chain reaction (PCR) primers were used to distinguish UGT2B17 from UGT2B15, and to distinguish among zero, one and two copies of the UGT2B17 gene. Marker D (Forward primer 5'-TCACAAGTCAATCTCCCATCC-3', Reverse primer 5'-CTGCAGAATATGTCAATAATTGGC-3') is positive for one copy and two copies (100 bp), and Marker J (Forward primer 5'-TGCACAGAGTTAAGAAATGGAGAGATGTG-3', Reverse primer 5'-GATCATCCTATATCCTGACAGAATT-3') is positive for only one copy (900 bp) [20,21]. PCR reactions were carried out in 25-μl mixtures containing 1 μg of genomic DNA, 2.5 μL of 10xLA PCR buffer II, 2 μL of dNTP (400 μM), 0.25 μL of LA Taq (Takara Bio Inc., Shiga, Japan), 18.25 μL of nuclease-free water, and 0.5 μL of each of the two primers (100 pmol/μL). Each reaction mixture was incubated at 94°C for 3 min and then subjected to 30 cycles of 94°C for 20s, 60°C for 30s, and 72°C for 90s. Each reaction was then incubated at 16°C until analysis.

Statistical analysis
The chi-square test was used to evaluate significant differences in the frequency of UGT2B17-deletion polymorphism. Study populations were composed of cases and controls and we used conditional logistic regression to calculate odds ratios (OR) and 95% confidence intervals (95%CI) as estimates of the cancer risk. Each p<0.05 was considered statistically significant. Moreover, the Bonferroni correction was used to correct for multiple testing. Logistic regression was performed 12 times for all kinds of cancer, namely, UADT-cancer (head and neck cancer, and esophageal cancer) and 13 other kinds of cancers. The difference was considered significant when the p-value was less than 0.004 that was obtained as p 0.05 divided by 12 time-analyses. All statistical analyses were performed using STATA 13.1 (STATA Corp., College Station, TX).

Characteristics of cohorts
The characteristics of the patients in the cohorts are shown in Table  1. Mean age was 64 years and males comprised 58% of the study population. The frequency of the UGT2B17-deletion was 75% in total, and ranged from 70% to 80% in different cohorts, which was not significantly different among the cohorts (p=0.088).

Cancers in these cohorts
In the mixture of nine cohorts, 1,887 patients did not have cancer and 1,205 did have cancer ( Table 2). The frequency of UGT2B17deletion was 74% in non-cancer patients and 77% in cancer patients, which was not significantly different (p=0.15). Next, patients with various cancers were categorized by type of cancer ( Table 3). The frequency of UGT2B17-deletion ranged from 67% to 87% depending on cancer type.

All cancers
A nested case-control study was performed for all cancer patients and all non-cancer patients. For each of 996 cases, age and gender were completely matched with controls randomly selected from the non-cancer patients (Table 4). Analysis by conditional logistic regression showed a similar trend: OR, 1.24; 95%CI, 1.02 to 1.51, p=0.035. However, these p-values were considered not significant on the basis of multiple testing theory.

UGT2B17 copy number variants Total
Number of copies 0 copy 1 or 2 copies

UADT-cancers
The UGT2B17 gene is associated with gene-protection due to its glucuronidation of carcinogens in tobacco smoke and of ethanol. Therefore, we focused on UADT cancers: head and squamous cell carcinoma, and esophageal cancer. For each case in the UADT-cancer cohort, one age-and gender-matched patient was selected from the non-cancer cohort as a control ( Table 5). The analysis by conditional logistic regression showed a significant result: OR, 2.07; 95%CI, 1.34 to 3.20, p=0.001. This p-value was significantly smaller than the predefined cutoff point of p=0.004 and thus was considered as significant.

Other cancers
An age/gender-matched case-control study was applied to each kind of cancer in a manner similar to UADT-cancer: renal cell cancer, colorectal cancer, breast cancer, gastric cancer, thyroid cancer, ovarian cancer, lung cancer, hematologic malignancy, prostate cancer, and hepatic cell cancer. In colorectal cancer, UGT2B17-deletion was more frequent in cases than in controls: OR, 1.85; 95% CI, 1.07 to 3.22, p=0.020. Analysis by conditional logistic regression showed a similar trend: OR, 1.90; 95% CI, 1.11 to 3.27, p=0.020. These p-values were larger than the pre-defined cutoff point and thus the associations were considered as not significant. Furthermore, there were no significant differences between the other types of cancers and UGT2B17.

Discussion
This study found that UGT2B17-deletion was a significant risk only for the development of UADT-cancer, which is the first report to our knowledge. Head and neck squamous cell carcinomas (HNSCC) arise from mucosa lining the paranasal sinuses, nasal cavities, oral cavity, nasopharynx, oropharynx, hypopharynx and larynx. These anatomical sites can be directly exposed to extremely high levels of carcinogens via tobacco and alcohol use. Indeed, at least 75% of HNSCC from the oral cavity, pharynx and larynx are attributed to smoking and/or alcohol use [22]. In addition, tobacco and alcohol use synergistically raise the risk of HNSCC [23]. Moreover, a synergistic effect between smoking tobacco and consuming alcohol has been reported for esophageal cancer [24]. UGT2B17 at least partially glucuronides carcinogens in tobacco smoke and ethanol in alcoholic drinks; consequently, UGT2B17-deletion can result in decreased elimination of carcinogens and ethanol, which explains the main result of this study that UGT2B17-deletion may be a risk for UADT-cancer.
On the other hand, UGT2B17-deletion was not a significant factor for cancer itself in this study. Recent meta-analysis also showed that UGT2B17-deletion was not associated with cancer susceptibility [25], consistent with our results. However, this meta-analysis included only four kinds of cancer: prostate cancer, lung cancer, colorectal cancer, and breast cancer, whereas our study included many types of cancers.
With the exception of UADT-cancer, the other type of cancer had no significant association with UGT2B17-deletion in this study. A previous study showed significant association in women with lung adenocarcinoma [12], but another recent study could not detect significant association in lung cancer [13]. On the other hand, UGT2B17-deletion was reported to associate with a decreased colorectal cancer risk in a Caucasian population [26], which is opposite direction to our and previous reports, and has not been supported by other reports. Currently, therefore, there is no clear evidence showing association between UGT2B17-deletion and cancer risk.
There are several limitations to this study. First, controls were not healthy adults but patients with diabetes, cardiac disease, and neurological diseases. However, UGT2B17-deletion was observed in 74% of healthy adults [5], which is equal to the frequency of 74% observed in patients without cancer in this study. Second, ovarian, head and neck, colorectal, gastric, lung, and thyroid cancers were dominant in our sample population, but the distribution did not reflect the real frequency of each cancer in the Japanese population. Third, the number of patients with each cancer type was too small to allow the detection of statistically significant differences, although a total number of 1,402 cancer patients were included in this study. Fourth, the results obtained in our study may only be applicable to the Japanese population, in which UGT2B17-deletion is much more prevalent than in other populations.

Conclusion
Deletion polymorphism of the UGT2B17 gene may associate with UADT-cancer risk in the Japanese population.