Abstract
The role of alcohol consumption in the etiology of endometrial cancer has not been clarified. To examine the association between alcohol consumption and endometrial cancer risk, we conducted a case‐control study with 148 histologically diagnosed incident endometrial cancer cases and 1468 matched non‐cancer controls. Median consumption of alcohol was only 19.3 g/week among cases who drank and 28.2 g/week among controls who drank. These values are lower than in Western countries. Relative risk was analyzed in subjects classified into four groups according to weekly alcohol consumption (non‐drinkers, 1–24 g/week, 25–175 g/week, and >175 g/week). Confounder‐adjusted odds ratios for those consuming alcohol at <25 g/week, 25–175 g/week, and >175 g/week compared to non‐drinkers were 0.79 (95% confidence interval (CI), 0.49–1.28), 0.42 (95% CI, 0.23–0.79), and 0.47 (95% CI, 0.14–1.58), respectively. Further analysis was conducted concerning self‐reported physical reaction to alcohol. Among women without flushing after drinking, a significant inverse association between risk and alcohol intake was seen (trend P = 0.001). In contrast, no protective effect of alcohol was seen among women who experience flushing after drinking. These results suggest the presence of an inverse association between alcohol drinking and endometrial cancer risk among Japanese women, and that this association is evident among those without flushing. Further investigation of these findings is warranted. (Cancer Sci 2008; 99: 1195–1201)
Endometrial cancer is a common gynecologic cancer in Japan, and its incidence is increasing, possibly due to the recent Westernization of the Japanese lifestyle.( 1 ) The development of endometrial cancer has been related to exposure to unopposed estrogens.( 2 , 3 , 4 ) Several studies have shown a positive association between alcohol intake and estrogen level in postmenopausal women.( 5 , 6 ) Although alcohol intake could therefore be expected to increase the risk of endometrial cancer by elevating estrogen levels, epidemiologic studies of this association have been inconsistent. Most previous studies have indicated that alcohol consumption is either weakly or not associated with the risk of endometrial cancer.( 7 , 8 , 9 , 10 , 11 ) However, several others have shown an increased risk in heavy drinkers( 12 , 13 ) while a case‐control study by Swanson et al. suggested an inverse association between moderate alcohol consumption and endometrial cancer risk among young women (<55 years).( 14 ) These inconsistent findings, as well as uncertainties regarding the etiology of endometrial cancer, hamper any coherent understanding of this association.
Here, we conducted a hospital‐based case‐control study to examine the association between alcohol consumption and endometrial cancer risk among Japanese women, considering other predisposing characteristics, such as body mass index and a history of hormone replacement therapy. In addition, given recent findings that a genetic polymorphism in aldehyde dehydrogenase2 (ALDH2), which has a strong impact on alcohol metabolism, was associated with several cancer risks,( 15 , 16 , 17 ) we also analyzed this risk using self‐reported reactions after drinking as a surrogate for ALDH2 genotyping.
Materials and Methods
Subjects. The subjects were 148 patients newly and histologically diagnosed with endometrial carcinoma between January 2001 and June 2005 at Aichi Cancer Center Hospital (ACCH) in Japan. The distribution of histological subtypes among 148 cases was 93 type I tumor (low‐grade endometrioid adenocarcinoma) (62.8%), and 55 type II tumor (high‐grade endometrioid adenocarcinoma and other adenocarcinomas) (37.2%). Mixed epithelial and mesenchymal tumors were excluded due to the paucity of knowledge on their etiology. Controls (n = 1476) were randomly selected and matched by age (± 3 years) and menopausal status (premenopause or postmenopause) to cases with a 1:10 case‐control ratio from 11 814 women who were diagnosed as cancer‐free (four cases were matched with nine controls). All subjects were recruited in the framework of the Hospital‐based Epidemiologic Research Program at Aichi Cancer Center (HERPACC), as described elsewhere.( 18 , 19 ) In brief, information on lifestyle factors was collected using a self‐administered questionnaire for all first‐visit outpatients at Aichi Cancer Center Hospital aged 20–79 who were enrolled in HERPACC between January 2001 and November 2005. Patients were also asked about lifestyle when healthy or before the current symptoms developed. Responses were checked by a trained interviewer. Approximately 90% of eligible subjects completed the questionnaire. Outpatients were also asked to provide blood samples. Our previous study showed that the lifestyle patterns of first‐visit outpatients accorded with those in a randomly selected sample of the general population of Nagoya City.( 20 ) The data were loaded into the HERPACC database and routinely linked with the hospital‐based cancer registry system to update the data on cancer incidence. All participants gave written informed consent and the study was approved by Institutional Ethical Committee of Aichi Cancer Center.
Assessment of alcohol intake and alcohol reaction. All subjects were asked about their average frequency, beverage type, and amount of drinking per day during the 1‐year period before onset of the present disease or before being interviewed. Usual alcohol intake was first reported as frequency of consumption in the five categories of non‐drinker, <1 day/week, 1–2 days/week, 3–4 days/week, and 5 or more days per week. Consumption of each type of beverage (Japanese sake, beer, shochu, whiskey, and wine) was determined by the average number of drinks per day, which was then converted into a Japanese sake (rice wine) equivalent. One Japanese drink equates to one ‘go’ (180 mL) of Japanese sake, which contains 23g of ethanol, equivalent to one large bottle (633 mL) of beer, two shots (57 mL) of whiskey, or 2.5 glasses of wine (200 mL). One drink of shochu (distilled spirit), which contains 25% ethanol, was rated as 108 mL. Total alcohol consumption was estimated as the summed amount of pure alcohol consumption (g/drink) of Japanese sake, beer, shochu, whiskey, and wine among current regular drinkers. Weekly ethanol consumption was calculated by combining the amount of ethanol per day and frequency per week. In this study, we used self‐reported flushing (yes/no) after a small amount of drinking (a glass of beer) as a stratification factor in the examination of alcohol impact.
Statistical analysis. To assess the strength of associations between alcohol consumption and risk of endometrial cancer, odd ratios (OR) with 95% confidence intervals (CI) were estimated using unconditional logistic models adjusted for potential confounders. For subgroup analysis, subjects were classified by alcohol intake into the four groups of non‐drinkers, and weekly ethanol intake of 1–24, 25–175, and >175 g. Among controls, median weekly intake in current drinkers was 25 g. Potential confounders considered in the multivariate analyses were age, smoking habit (never smokers or ever smokers), body mass index (BMI; <25 or ≥25 kg/m2 based upon our previous study),( 21 ) regular exercise (yes or no), menstrual status (premenopausal or postmenopausal), age at menarche (≤ 12, 13–14, or ≥ 15), duration of menstruation (years, quartiles), parity (0, 1–2, ≥ 3), diabetes history (yes or no), hypertension history (yes or no), contraceptive usage history (yes or no), hormone replacement therapy history (yes or no), flushing after drinking (yes or no), and histological subtype (type I or type II). Missing values for any covariate were treated as a dummy variable in the logistic model. Differences in categorized demographic variables between the cases and controls were tested by the χ2‐test. Age, age at menarche, duration of menstruation, BMI, and parity between cases and controls were compared by the Mann–Whitney test. Stratification analysis was used to estimate risk for subgroups by drinking habit. P‐values less than 0.05 were considered statistically significant. All analyses were conducted using STATA version 9 (Stata, College Station, TX, USA).
Results
Baseline characteristics of the 148 endometrial cancer patients and 1476 controls are shown in Table 1. Median age was 56 years for both patients and controls. Smoking status did not differ between the two groups. Prevalence of ever smokers was 16.2% and 16.5% in case and controls, respectively. BMI was higher among cases than controls (P < 0.001). Regarding reproductive factors, only parity showed a significant difference between two groups. Low experience of delivery was more prevalent among cases than controls (P < 0.001). A history of diabetes was more common in cases, although with only marginal statistical significance. Although contraceptive usage did not differ, hormone replacement therapy was more prevalent in cases.
Table 1.
Characteristics of subjects
| Characteristic | Cases | Controls | P‐values | ||
|---|---|---|---|---|---|
| Number | 148 | 1476 | |||
| Age (median, [min–max]) | 56.0 (26–79) | 56.0 (23–80) | 0.846 | ||
| ≤39 (%) | 22 | (14.9) | 223 | (15.1) | 0.986 |
| 40–49 (%) | 13 | (8.8) | 136 | (9.2) | |
| 50–59 (%) | 64 | (43.2) | 610 | (41.3) | |
| 60–69 (%) | 36 | (24.3) | 385 | (26.1) | |
| ≥ 70 (%) | 13 | (8.8) | 122 | (8.3) | |
| Smoking status | |||||
| Ever (%) | 24 | (16.2) | 244 | (16.5) | 0.942 |
| Never (%) | 123 | (83.1) | 1225 | (83.0) | |
| Unknown (%) | 1 | (0.7) | 7 | (0.5) | |
| Body mass index (median, [min–max]) | 23.2 (13.4–40.9) | 21.9 (13.2–42.7) | <0.001 | ||
| <25 kg/m2 (%) | 104 | (70.3) | 1211 | (82.1) | <0.001 |
| ≥25 kg/m2 (%) | 40 | (27.0) | 257 | (17.4) | |
| Unknown (%) | 4 | (2.7) | 8 | (0.5) | |
| Regular exercise | |||||
| No (%) | 46 | (31.1) | 388 | (26.3) | 0.252 |
| Yes (%) | 101 | (68.2) | 1057 | (71.6) | |
| Unknown (%) | 1 | (0.7) | 31 | (2.1) | |
| Menstrual status | |||||
| Premenopausal (%) | 51 | (34.5) | 506 | (34.3) | 0.965 |
| Postmenopausal (%) | 97 | (65.5) | 970 | (65.7) | |
| Age at menarche (median, [min–max]) | 14.0 (10–20) | 14.0 (10–21) | 0.963 | ||
| ≤12 (%) | 38 | (25.7) | 379 | (25.7) | 0.729 |
| 13–14 (%) | 75 | (50.7) | 701 | (47.5) | |
| ≥15 (%) | 31 | (21.0) | 365 | (24.7) | |
| Unknown (%) | 4 | (2.7) | 31 | (2.1) | |
| Duration of menstration (median, [min–max]) | 37.0 (0–49) | 36.0 (11–43) | 0.390 | ||
| ≤32 (%) | 38 | (25.7) | 395 | (26.8) | 0.822 |
| 33–36 (%) | 33 | (22.3) | 367 | (24.9) | |
| 37–39 (%) | 38 | (25.7) | 388 | (26.3) | |
| ≥40 (%) | 34 | (23.0) | 284 | (19.2) | |
| Unknown (%) | 5 | (3.4) | 42 | (2.9) | |
| Parity (median, [min–max]) | 2 (0–4) | 2 (0–6) | <0.001 | ||
| 0 (%) | 41 | (27.7) | 207 | (14.0) | <0.001 |
| 1–2 (%) | 82 | (55.4) | 911 | (61.7) | |
| ≥ 3 (%) | 24 | (16.2) | 348 | (23.6) | |
| Unknown (%) | 1 | (0.7) | 10 | (0.7) | |
| Diabetes history | |||||
| No (%) | 137 | (92.6) | 1416 | (95.9) | 0.056 |
| Yes (%) | 11 | (7.4) | 60 | (4.1) | |
| Hypertension history | |||||
| No (%) | 121 | (81.8) | 1273 | (86.3) | 0.135 |
| Yes (%) | 27 | (18.2) | 203 | (13.8) | |
| Contraceptive usage history | |||||
| No (%) | 138 | (93.2) | 1377 | (93.3) | 0.934 |
| Yes (%) | 8 | (5.4) | 74 | (5.0) | |
| Unknown (%) | 2 | (1.4) | 25 | (1.7) | |
| Hormone replacement therapy history | |||||
| No (%) | 132 | (89.2) | 1355 | (91.8) | 0.247 |
| Yes (%) | 15 | (10.1) | 100 | (6.8) | |
| Unknown (%) | 1 | (0.7) | 21 | (1.4) | |
Median consumption of alcohol among cases and controls who drank was only 19.3 and 28.2 g/week, respectively. Table 2 shows the impact of drinking habit on endometrial cancer risk. Frequent drinkers showed a reduced risk: compared with non‐drinkers, the age‐adjusted OR of those who drank 5 or more days per week was 0.39 (95% CI, 0.18–0.85). Although without significance, all groups except non‐drinkers showed OR below unity and their point estimates decreased as frequency increased (P‐trend = 0.011). This trend was consistently observed in the multivariate model. Similarly, with regard to the amount of alcohol consumed, those who consumed less than 25 g per week, those who consumed 25–175 g per week, and those who consumed 175 g or more per week showed a lower risk of endometrial cancer than non‐drinkers, with OR of 0.79 (95% CI, 0.49–1.27), 0.44 (95% CI, 0.24–0.81), and 0.54 (95% CI, 0.16–1.76), respectively. The multivariate model again showed consistent results.
Table 2.
Odds ratios (OR) and 95% confidence intervals (CI) for endometrial cancer according to frequency and quantitiy of alcohol intake
| Category | Cases (n = 148) | Controls (n = 1476) | Age‐adjusted OR (95% CI) | Multivariate OR (95% CI)† |
|---|---|---|---|---|
| Frequency of alcohol intake | ||||
| None | 108 | 929 | 1.00 (Reference) | 1.00 (Reference) |
| <1/week | 14 | 166 | 0.72 (0.40–1.29) | 0.71 (0.39–1.29) |
| 1–2/week | 11 | 119 | 0.79 (0.41–1.52) | 0.77 (0.40–1.50) |
| 3–4/week | 8 | 99 | 0.69 (0.33–1.46) | 0.67 (0.31–1.43) |
| 5‐/week | 7 | 154 | 0.39 (0.18–0.85) | 0.37 (0.17–0.82) |
| unknown | 0 | 9 | ||
| P‐trends | 0.011 | 0.009 | ||
| Amount of alcohol consumption | ||||
| None | 109 | 933 | 1.00 (Reference) | 1.00 (Reference) |
| <25 g/week | 23 | 246 | 0.79 (0.49–1.27) | 0.79 (0.49–1.28) |
| (median, range) (eta g/week) | (8.6, 2.9–24.2) | (8.6, 1.7–24.2) | ||
| 25–175 g/week | 12 | 232 | 0.44 (0.24–0.81) | 0.42 (0.23–0.79) |
| (median, range) (eta g/week) | (54.3, 25.9–96.6) | (69, 25.3–172.5) | ||
| >175 g/week | 3 | 47 | 0.54 (0.16–1.76) | 0.47 (0.14–1.58) |
| (median, range) (eta g/week) | (201.3, 179.4–552) | (276, 177.1–805) | ||
| unknown | 1 | 18 | ||
| P‐trends | 0.006 | 0.005 | ||
Multivariate models adjusted for age, smoking, body mass index, regular exercise, menstrual status, age at menarche, duration of menstruation, parity, diabetes history, hypertension history, contraceptive usage history, hormone replacement therapy, and flushing after drinking.
Table 3 shows a stratified analysis according to potential confounders designed to examine the consistency of association and to explore the possible interaction with weekly alcohol consumption. The inverse association between endometrial cancer risk and alcohol intake persisted after stratification by BMI, regular exercise, menstrual status, age at menarche, duration of menstruation, parity, diabetes history, hypertension history, and type I tumor. In contrast, no associations were seen for ever smokers, oral contraceptive users, hormone replacement therapy users, and type II tumor. Regarding BMI, obese women (BMI ≥ 25) showed a stronger protective effect by alcohol than leaner women (BMI < 25). Among postmenopausal women, the OR for weekly drinking of less than 25, 25–175, and 175 g or more for EC were 0.83 (95% CI, 0.46–1.52), 0.46 (95% CI, 0.21–1.02), and 0.72 (95% CI, 0.17–3.15), respectively, but the P‐trend was marginally significant (P = 0.069). Generally, endometrial cancer risk was lowest among women with weekly consumption of 25–175 g.
Table 3.
Odds ratios (OR) and 95% confidence intervals (CI) for endometrial cancer stratified according to weekly alcohol consumption and lifestyle factors
| Category | Alcohol consumption | ||||
|---|---|---|---|---|---|
| None | <25 g/week | 25–175 g/week | >175 g/week | P‐trends | |
| Total (case/control) † | 109/933 | 23/246 | 12/232 | 3/47 | |
| OR (95% CI) | 1.00 (Reference) | 0.79 (0.49–1.27) | 0.44 (0.24–0.81) | 0.54 (0.16–1.76) | 0.006 |
| Smoking | |||||
| Never (case/control) | 98/829 | 18/213 | 5/157 | 1/16 | |
| OR (95% CI) | 1.00 (Reference) | 0.70 (0.41–1.18) | 0.26 (0.11–0.66) | 0.51 (0.07–3.87) | 0.002 |
| Ever (case/cotrol) | 11/98 | 4/33 | 7/75 | 2/31 | |
| OR (95% CI) | 1.00 (Reference) | 1.25 (0.36–4.40) | 0.89 (0.33–2.46) | 0.63 (0.13–3.04) | 0.586 |
| Unknown (case/cotrol) | 0/6 | 1/0 | 0/0 | 0/0 | |
| Body mass index | |||||
| <25 kg/m2 (case/control) | 73/757 | 17/197 | 11/202 | 2/40 | |
| OR (95% CI) | 1.00 (Reference) | 0.92 (0.53–1.61) | 0.58 (0.30–1.12) | 0.54 (0.13–2.31) | 0.090 |
| ≥25 kg/m2 (case/control) | 32/168 | 6/49 | 1/30 | 1/7 | |
| OR (95% CI) | 1.00 (Reference) | 0.55 (0.21–1.43) | 0.15 (0.02–1.13) | 0.48 (0.05–4.34) | 0.035 |
| Unknown (case/control) | 4/8 | 0/0 | 0/0 | 0/0 | |
| Regular exercise | |||||
| No (case/control) | 36/257 | 7/40 | 2/63 | 1/22 | |
| OR (95% CI) | 1.00 (Reference) | 1.27 (0.53–3.05) | 0.23 (0.05–0.97) | 0.34 (0.04–2.57) | 0.047 |
| Yes (case/control) | 72/654 | 16/201 | 10/167 | 2/25 | |
| OR (95% CI) | 1.00 (Reference) | 0.70 (0.40–1.24) | 0.53 (0.27–1.05) | 0.69 (0.16–3.00) | 0.053 |
| Unknown (case/control) | 1/22 | 0/5 | 0/2 | 0/0 | |
| Menstrual status | |||||
| Premenopausal (case/control) | 35/280 | 9/99 | 5/98 | 1/23 | |
| OR (95% CI) | 1.00 (Reference) | 0.72 (0.34–1.57) | 0.41 (0.15–1.07) | 0.35 (0.05–2.65) | 0.038 |
| Postmenopausal (case/control) | 74/653 | 14/147 | 7/134 | 2/24 | |
| OR (95% CI) | 1.00 (Reference) | 0.83 (0.46–1.52) | 0.46 (0.21–1.02) | 0.72 (0.17–3.15) | 0.069 |
| Age at menarche | |||||
| ≤12 (case/control) | 28/236 | 8/61 | 1/64 | 1/13 | |
| OR (95% CI) | 1.00 (Reference) | 1.04 (0.45–2.40) | 0.12 (0.02–0.92) | 0.56 (0.07–4.49) | 0.053 |
| 13–14 (case/control) | 53/428 | 11/127 | 9/114 | 1/22 | |
| OR (95% CI) | 1.00 (Reference) | 0.72 (0.36–1.42) | 0.65 (0.31–1.37) | 0.38 (0.05–2.90) | 0.120 |
| ≥15 (case/control) | 26/249 | 2/54 | 2/48 | 1/11 | |
| OR (95% CI) | 1.00 (Reference) | 0.39 (0.09–1.73) | 0.44 (0.10–1.91) | 1.07 (0.13–8.88) | 0.260 |
| Unknown (case/control) | 2/20 | 2/4 | 0/6 | 1/0 | |
| Duration of menstruation | |||||
| ≤32 years (case/control) | 27/219 | 7/77 | 4/71 | 0/22 | |
| OR (95% CI) | 1.00 (Reference) | 0.69 (0.28–1.67) | 0.43 (0.15–1.29) | NE | 0.029 |
| 33–36 years (case/control) | 27/246 | 5/51 | 1/54 | 0/9 | |
| OR (95% CI) | 1.00 (Reference) | 0.93 (0.34–2.55) | 0.18 (0.02–1.35) | NE | 0.063 |
| 37–39 years (case/control) | 29/249 | 3/71 | 4/57 | 1/8 | |
| OR (95% CI) | 1.00 (Reference) | 0.36 (0.11–1.23) | 0.60 (0.20–1.78) | 1.07 (0.13–8.88) | 0.249 |
| ≥40 years (case/control) | 23/189 | 6/43 | 3/43 | 2/7 | |
| OR (95% CI) | 1.00 (Reference) | 1.13 (0.43–2.95) | 0.56 (0.16–1.95) | 2.23 (0.43–11.49) | 0.932 |
| Unknown (case/control) | 3/30 | 2/4 | 0/7 | 0/1 | |
| Parity | |||||
| 0 (case/control) | 30/115 | 6/36 | 4/42 | 1/10 | |
| OR (95% CI) | 1.00 (Reference) | 0.63 (0.24–1.65) | 0.36 (0.12–1.09) | 0.38 (0.05–3.10) | 0.046 |
| 1–2 (case/control) | 58/599 | 15/147 | 6/129 | 2/25 | |
| OR (95% CI) | 1.00 (Reference) | 1.12 (0.61–2.05) | 0.50 (0.21–1.20) | 0.90 (0.21–3.93) | 0.271 |
| ≥3 (case/control) | 21/213 | 2/61 | 1/59 | 0/12 | |
| OR (95% CI) | 1.00 (Reference) | 0.37 (0.08–1.64) | 0.19 (0.02–1.43) | NE | 0.035 |
| Unknown (case/control) | 0/6 | 0/2 | 1/2 | 0/0 | |
| Diabetes history | |||||
| No (case/control) | 99/894 | 22/237 | 12/224 | 3/45 | |
| OR (95% CI) | 1.00 (Reference) | 0.81 (0.50–132) | 0.47 (0.25–0.87) | 0.57 (0.17–1.89) | 0.015 |
| Yes (case/control) | 10/39 | 1/9 | 0/8 | 0/2 | |
| OR (95% CI) | 1.00 (Reference) | 0.48 (0.05–4.33) | NE | NE | 0.212 |
| Hypertension history | |||||
| No (case/control) | 87/797 | 21/225 | 10/200 | 2/38 | |
| OR (95% CI) | 1.00 (Reference) | 0.85 (0.51–1.40) | 0.45 (0.23–0.89) | 0.47 (0.11–2.00) | 0.016 |
| Yes (case/control) | 22/136 | 2/21 | 2/32 | 1/9 | |
| OR (95% CI) | 1.00 (Reference) | 0.54 (0.12–2.47) | 0.36 (0.08–1.62) | 0.64 (0.08–5.32) | 0.178 |
| Contraceptive usage history | |||||
| No (case/control) | 101/871 | 23/231 | 12/216 | 1/43 | |
| OR (95% CI) | 1.00 (Reference) | 0.85 (0.53–1.38) | 0.47 (0.26–0.88) | 0.20 (0.03–1.45) | 0.005 |
| Yes (case/control) | 6/44 | 0/11 | 0/15 | 2/4 | |
| OR (95% CI) | 1.00 (Reference) | NE | NE | 3.63 (0.53–24.92) | 0.892 |
| Unknown (case/control) | 2/18 | 0/4 | 0/1 | 0/0 | |
| Hormone replacement therapy history | |||||
| No (case/control) | 101/860 | 18/227 | 10/212 | 2/40 | |
| OR (95% CI) | 1.00 (Reference) | 0.66 (0.39–1.12) | 0.39 (0.20–0.77) | 0.41 (0.10–1.72) | 0.002 |
| Yes (case/control) | 7/59 | 5/15 | 2/19 | 1/7 | |
| OR (95% CI) | 1.00 (Reference) | 2.79 (0.78–10.05) | 0.89 (0.17–4.64) | 1.21 (0.13–11.31) | 0.826 |
| Unknown (case/control) | 1/14 | 0/4 | 0/1 | 0/0 | |
| Histological subtype | |||||
| Type I (case/control) | 68/933 | 17/246 | 6/232 | 1/47 | |
| OR (95% CI) | 1.00 (Reference) | 0.71 (0.51–1.57) | 0.34 (0.14–0.79) | 0.27 (0.04–1.97) | 0.007 |
| Type II (case/control) | 41/933 | 6/246 | 6/246 | 2/47 | |
| OR (95% CI) | 1.00 (Reference) | 0.60 (0.25–1.43) | 0.63 (0.26–1.50) | 1.09 (0.25–4.69) | 0.323 |
One case and 18 controls were excluded from analyses due to lack of information on alcohol drinking.
NE, not estimated because of no case in this category.
Table 4 shows a stratified analysis according to self‐reported reaction to alcohol. Flushing after drinking depends mainly on the activity of aldehyde dehydrogenase, particularly ALDH2, and might therefore reflect lower ALDH2 activity. Among women who did not experience flushing after drinking, an inverse association was seen between endometrial cancer risk and alcohol intake. The age‐adjusted OR for weekly drinking of less than 25, 25–175, and 175 g or more for endometrial cancer were 0.51 (95% CI, 0.26–0.98), 0.24 (95% CI, 0.11–0.56), and 0.49 (95% CI, 0.14–1.69), respectively, and the P‐trend was statistically significant (P = 0.001). By contrast, the protective effect of alcohol was not observed among women who had flushing after drinking (age‐adjusted P‐trend = 0.560). The multivariate model again showed consistent results.
Table 4.
Impact of alcohol consumption according to self‐reported reaction to alcohol
| Category | Alcohol consumption | ||||
|---|---|---|---|---|---|
| None | <25 g/week | 25–175 g/week | >175 g/week | P‐trends | |
| Total (case/control) † | 109/933 | 23/246 | 12/232 | 3/47 | |
| Age‐adjusted OR (95% CI) | 1.00 (Reference) | 0.79 (0.49–1.27) | 0.44 (0.24–0.82) | 0.54 (0.16–1.76) | 0.006 |
| Multivariate OR (95% CI) | 1.00 (Reference) | 0.79 (0.49–1.28) | 0.42 (0.23–0.79) | 0.47 (0.14–1.58) | 0.005 |
| Flushing after drinking | |||||
| No (case/control) | 44/292 | 13/157 | 7/175 | 3/36 | |
| Age‐adjusted OR (95% CI) | 1.00 (Reference) | 0.51 (0.26–0.98) | 0.24 (0.11–0.56) | 0.49 (0.14–1.69) | 0.001 |
| Multivariate OR (95% CI) | 1.00 (Reference) | 0.53 (0.27–1.05) | 0.25 (0.11–0.59) | 0.48 (0.14–1.67) | 0.002 |
| Yes (case/control) | 61/574 | 9/86 | 5/55 | 0/10 | |
| Age‐adjusted OR (95% CI) | 1.00 (Reference) | 1.03 (0.49–2.15) | 0.89 (0.34–2.30) | NE | 0.560 |
| Multivariate OR (95% CI) ‡ | 1.00 (Reference) | 1.07 (0.51–2.27) | 0.97 (0.37–2.57) | NE | 0.677 |
| Unknown (case/control) | 4/67 | 1/3 | 0/2 | 0/1 | |
One case and 18 controls were excluded from analyzes due to lack of information on alcohol drinking.
Multivariate models adjusted for age, smoking, body mass index, regular exercise, menstrual status, age at menarche, duration of menstruation, parity, diabetes history, hypertension history, contraceptive usage history, and hormone replacement therapy.
CI, confidence interval; NE, not estimated because of no case in this category; OR, odds ratio.
Discussion
In this study, we found that a small amount of alcohol consumption was protective against endometrial cancer among Japanese women. This association was consistently observed regardless of potential confounders. OR were lowest among those who consumed 25–175 g per week. In addition, the protective effect of alcohol drinking decreased among women who reported flushing after drinking.
Results to date regarding the relationship between alcohol intake and endometrial cancer risk are inconsistent. Although most previous studies have indicated a null association,( 7 , 8 , 9 , 11 , 22 , 23 , 24 , 25 ) three have shown a protective effect of alcohol,( 10 , 14 , 26 ) while three others have reported that alcohol intake was a risk factor of endometrial cancer.( 12 , 13 , 27 ) Newcomb et al. suggested a significant inverse association in premenopausal women consuming one drink per day or more (RR = 0.20; 95% CI, 0.06–0.71)( 10 ) while Swanson et al. showed an inverse association between moderate consumption and endometrial cancer risk among young women (<55 years), with relative risks for three levels of drinking (<1, 1–4, >4 drinks per week) from lowest to highest of 0.78, 0.64, and 0.41 compared to non‐drinkers.( 14 ) Webster et al. showed that non‐drinkers aged 20–54 years had a higher relative risk (RR = 1.83; 95% CI, 1.11–3.01) than women who consumed an average of 150 g or more of alcohol per week.( 26 ) These results may indicate that light alcohol consumption decreases endometrial cancer risk in younger women. In contrast, Setiawan et al. suggested that alcohol consumption equivalent to two or more drinks per day increased the risk of endometrial cancer in postmenopausal women.( 12 ) The other two case‐control studies showed similar positive associations between increased alcohol consumption and risk.( 13 , 27 )
Here, our study has added to the evidence for a protective effect of alcohol on endometrial cancer. The degree of consumption may be an important consideration in determining the impact of alcohol. Average consumption in our study was very low compared with previous studies. Relatively high consumption (≥175 g/week) was seen in only three cases and 99 controls, who showed a protective effect compared with non‐drinkers (multivariate OR = 0.47; 95% CI, 0.14–1.58). The provision of stable estimates for this subgroup is hampered by their small sample size.
One possible explanation for these results is that a small amount of drinking might be protective against cancer, as suggested in several prospective cohort studies.( 28 , 29 , 30 , 31 , 32 ) The biological mechanism of this protective effect for cancer among light‐moderate drinkers is not clear. Tsugane et al. considered the background characteristics of moderate drinkers to be healthier than those of either non‐drinkers or heavy drinkers.( 32 ) It has been reported that alcohol intake increases endogenous serum levels of estrogen in postmenopausal women,( 5 , 6 ) but it is unclear whether this is due to either a decrease in metabolic clearance or an increase in production.( 33 ) It has thus been hypothesized that alcohol drinking might lead to an increased risk of endometrial cancer risk due via the increased mitotic proliferation of endometrial cells, resulting in increased DNA replication errors and somatic mutations.( 34 ) Our findings here contradict this hypothesized mechanism; nevertheless, we assume that the amount of drinking may differentiate the impact of alcohol on endometrial cancer risk, as stated above.
Of interest was the combined effect of the amount of consumption and physical reaction to alcohol.( 19 ) Subjects who reported flushing did not show the protective effect observed in the non‐flushing group. It has been suspected that the oxidative metabolite of ethanol, acetaldehyde, is carcinogenic for humans due to its binding to cellular proteins and DNA, thus leading to carcinogenesis.( 35 , 36 ) Further, in individuals with ALDH2 encoded by ALDH2 Glu/Lys, the blood acetaldehyde level after drinking is approximately six‐fold that in individuals with active ALDH2.( 37 ) Taking results from our previous study demonstrating sensitivity and specificity of self‐reported flushing for ALDH2 genotype as 83.5% and 87.8%,( 38 ) our findings may have resulted from a decrease in the protective effect of alcohol owing to exposure to high levels of acetaldehyde.
Several potential limitations of our study warrant consideration. First, because it was a hospital‐based case‐control study, the threat of inadequate comparability between cases and controls rested on whether the control population was the source population from which cases arose. In the ACCH, it is assumed that those who are diagnosed as not having cancer at a particular period of time will visit the ACCH in the event that they do develop malignant disease. Our source of controls is therefore assumed to be appropriate for the drawing of causal inferences. Second, as with other case‐control studies, this study may have suffered from recall bias. Although the questionnaires, including that on alcohol intake, were completed before diagnosis in our hospital, some case patients referred to the hospital might have known their diagnosis. The fact that alcohol intake is not a well‐accepted risk factor for endometrial cancer among the public might preclude this possibility of information bias regarding alcohol Third, our study had a modest sample size, and replication in other studies is required.
In conclusion, our case‐control study suggested that alcohol drinking decreases the risk of endometrial cancer among Japanese women who consume small amounts. Further, a similar association was observed after stratification by potential confounders. However, this protective effect of alcohol was modified in those who experienced a flushed reaction to it after drinking. Further investigation of these findings is warranted.
Acknowledgments
The authors are grateful to the assistant staff at Division of Epidemiology and Prevention at Aichi Cancer Center Research Institute for their support for this study. This study was supported by a Grant‐in‐Aid for Scientific Research from the Ministry of Education, Science, Sports, Culture, and Technology of Japan and by a Grant‐in‐Aid for the Third Term Comprehensive 10‐Year Strategy for Cancer Control from the Ministry of Health, Labour, and Welfare of Japan.
References
- 1. Inoue M, Okayama A, Fujita M, Enomoto T, Tanizawa O, Ueshima H. A case‐control study on risk factors for uterine endometrial cancer in Japan. Jpn J Cancer Res 1994; 85: 346–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Lukanova A, Lundin E, Micheli A et al . Circulating levels of sex steroid hormones and risk of endometrial cancer in postmenopausal women. Int J Cancer 2004; 108: 425–32. [DOI] [PubMed] [Google Scholar]
- 3. Potischman N, Hoover RN, Brinton LA et al . Case‐control study of endogenous steroid hormones and endometrial cancer. J Natl Cancer Inst 1996; 88: 1127–35. [DOI] [PubMed] [Google Scholar]
- 4. Herrinton LJ, Weiss NS. Postmenopausal unopposed estrogens. Characteristics of use in relation to the risk of endometrial carcinoma. Ann Epidemiol 1993; 3: 308–18. [DOI] [PubMed] [Google Scholar]
- 5. Onland‐Moret NC, Peeters PH, Van Der Schouw YT, Grobbee DE, Van Gils CH. Alcohol and endogenous sex steroid levels in postmenopausal women: a cross‐sectional study. J Clin Endocrinol Metab 2005; 90: 1414–19. [DOI] [PubMed] [Google Scholar]
- 6. Rinaldi S, Peeters PH, Bezemer ID et al . Relationship of alcohol intake and sex steroid concentrations in blood in pre‐ and post‐menopausal women: the European Prospective Investigation into Cancer and Nutrition. Cancer Causes Control 2006; 17: 1033–43. [DOI] [PubMed] [Google Scholar]
- 7. Loerbroks A, Schouten LJ, Goldbohm RA, Van Den Brandt PA. Alcohol consumption, cigarette smoking, and endometrial cancer risk: results from the Netherlands Cohort Study. Cancer Causes Control 2007; 18: 551–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Weiderpass E, Baron JA. Cigarette smoking, alcohol consumption, and endometrial cancer risk: a population‐based study in Sweden. Cancer Causes Control 2001; 12: 239–47. [DOI] [PubMed] [Google Scholar]
- 9. Terry P, Baron JA, Weiderpass E, Yuen J, Lichtenstein P, Nyren O. Lifestyle and endometrial cancer risk: a cohort study from the Swedish Twin Registry. Int J Cancer 1999; 82: 38–42. [DOI] [PubMed] [Google Scholar]
- 10. Newcomb PA, Trentham‐Dietz A, Storer BE. Alcohol consumption in relation to endometrial cancer risk. Cancer Epidemiol Biomarkers Prev 1997; 6: 775–8. [PubMed] [Google Scholar]
- 11. Kalandidi A, Tzonou A, Lipworth L, Gamatsi I, Filippa D, Trichopoulos D. A case‐control study of endometrial cancer in relation to reproductive, somatometric, and life‐style variables. Oncology 1996; 53: 354–9. [DOI] [PubMed] [Google Scholar]
- 12. Setiawan VW, Monroe KR, Goodman MT, Kolonel LN, Pike MC, Henderson BE. Alcohol consumption and endometrial cancer risk: The multiethnic cohort. Int J Cancer 2008; 122: 634–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Parazzini F, La Vecchia C, D’Avanzo B, Moroni S, Chatenoud L, Ricci E. Alcohol and endometrial cancer risk: findings from an Italian case‐control study. Nutr Cancer 1995; 23: 55–62. [DOI] [PubMed] [Google Scholar]
- 14. Swanson CA, Wilbanks GD, Twiggs LB et al . Moderate alcohol consumption and the risk of endometrial cancer. Epidemiology 1993; 4: 530–6. [DOI] [PubMed] [Google Scholar]
- 15. Hiraki A, Matsuo K, Wakai K, Suzuki T, Hasegawa Y, Tajima K. Gene‐gene and gene–environment interactions between alcohol drinking habit and polymorphisms in alcohol‐metabolizing enzyme genes and the risk of head and neck cancer in Japan. Cancer Sci 2007; 98: 1087–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Matsuo K, Hamajima N, Shinoda M et al . Gene–environment interaction between an aldehyde dehydrogenase‐2 (ALDH2) polymorphism and alcohol consumption for the risk of esophageal cancer. Carcinogenesis 2001; 22: 913–16. [DOI] [PubMed] [Google Scholar]
- 17. Matsuo K, Hamajima N, Hirai T et al . Aldehyde dehydrogenase 2 (ALDH2) genotype affects rectal cancer susceptibility due to alcohol consumption. J Epidemiol 2002; 12: 70–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Tajima K, Hirose K, Inoue M, Takezaki T, Hamajima N, Kuroishi T. A model of practical cancer prevention for out‐patients visiting a hospital: the Hospital‐based Epidemiologic Research Program at Aichi Cancer Center (HERPACC). Asian Pac J Cancer Prev 2000; 1: 35–47. [PubMed] [Google Scholar]
- 19. Hamajima N, Matsuo K, Saito T et al . Gene–environment interactions and polymorphism studies of cancer risk in the Hospital‐based Epidemiologic Research Program at Aichi Cancer Center II (HERPACC‐II). Asian Pac J Cancer Prev 2001; 2: 99–107. [PubMed] [Google Scholar]
- 20. Inoue M, Tajima K, Hirose K et al . Epidemiological features of first‐visit outpatients in Japan: comparison with general population and variation by sex, age, and season. J Clin Epidemiol 1997; 50: 69–77. [DOI] [PubMed] [Google Scholar]
- 21. Hirose K, Tajima K, Hamajima N et al . Comparative case‐referent study of risk factors among hormone‐related female cancers in Japan. Jpn J Cancer Res 1999; 90: 255–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Hill HA, Austin H. Nutrition and endometrial cancer. Cancer Causes Control 1996; 7: 19–32. [DOI] [PubMed] [Google Scholar]
- 23. Austin H, Drews C, Partridge EE. A case‐control study of endometrial cancer in relation to cigarette smoking, serum estrogen levels, and alcohol use. Am J Obstet Gynecol 1993; 169: 1086–91. [DOI] [PubMed] [Google Scholar]
- 24. Gapstur SM, Potter JD, Sellers TA, Kushi LH, Folsom AR. Alcohol consumption and postmenopausal endometrial cancer: results from the Iowa Women's Health Study. Cancer Causes Control 1993; 4: 323–9. [DOI] [PubMed] [Google Scholar]
- 25. Shu XO, Brinton LA, Zheng W, Gao YT, Fan J, Fraumeni JF Jr. A population‐based case‐control study of endometrial cancer in Shanghai. China Int J Cancer 1991; 49: 38–43. [DOI] [PubMed] [Google Scholar]
- 26. Webster LA, Weiss NS. Alcoholic beverage consumption and the risk of endometrial cancer. Cancer and Steroid Hormone Study Group. Int J Epidemiol 1989; 18: 786–91. [DOI] [PubMed] [Google Scholar]
- 27. La Vecchia C, Decarli A, Fasoli M, Gentile A. Nutrition and diet in the etiology of endometrial cancer. Cancer 1986; 57: 1248–53. [DOI] [PubMed] [Google Scholar]
- 28. Gaziano JM, Gaziano TA, Glynn RJ et al . Light‐to‐moderate alcohol consumption and mortality in the Physicians’ Health Study enrollment cohort. J Am Coll Cardiol 2000; 35: 96–105. [DOI] [PubMed] [Google Scholar]
- 29. Inoue M, Tsugane S. Impact of alcohol drinking on total cancer risk: data from a large‐scale population‐based cohort study in Japan. Br J Cancer 2005; 92: 182–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lin Y, Kikuchi S, Tamakoshi A et al . Alcohol consumption and mortality among middle‐aged and elderly Japanese men and women. Ann Epidemiol 2005; 15: 590–7. [DOI] [PubMed] [Google Scholar]
- 31. Marugame T, Yamamoto S, Yoshimi I, Sobue T, Inoue M, Tsugane S. Patterns of alcohol drinking and all‐cause mortality: results from a large‐scale population‐based cohort study in Japan. Am J Epidemiol 2007; 165: 1039–46. [DOI] [PubMed] [Google Scholar]
- 32. Tsugane S, Fahey MT, Sasaki S, Baba S. Alcohol consumption and all‐cause and cancer mortality among middle‐aged Japanese men: seven‐year follow‐up of the JPHC study Cohort I. Japan Public Health Center. Am J Epidemiol 1999; 150: 1201–7. [DOI] [PubMed] [Google Scholar]
- 33. Ginsburg ES, Walsh BW, Gao X, Gleason RE, Feltmate C, Barbieri RL. The effect of acute ethanol ingestion on estrogen levels in postmenopausal women using transdermal estradiol. J Soc Gynecol Invest 1995; 2: 26–9. [PubMed] [Google Scholar]
- 34. Akhmedkhanov A, Zeleniuch‐Jacquotte A, Toniolo P. Role of exogenous and endogenous hormones in endometrial cancer. review of the evidence and research perspectives. Ann N Y Acad Sci 2001; 943: 296–315. [DOI] [PubMed] [Google Scholar]
- 35. IARC Working Group, Lyon . Alcohol drinking. 13–20 October 1987. IARC Monogr Eval Carcinog Risks Hum 1988; 44: 1–378. [PMC free article] [PubMed] [Google Scholar]
- 36. Seitz HK, Oneta CM. Gastrointestinal alcohol dehydrogenase. Nutr Rev 1998; 56: 52–60. [DOI] [PubMed] [Google Scholar]
- 37. Muto M, Hitomi Y, Ohtsu A, Ebihara S, Yoshida S, Esumi H. Association of aldehyde dehydrogenase 2 gene polymorphism with multiple oesophageal dysplasia in head and neck cancer patients. Gut 2000; 47: 256–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Matsuo K, Wakai K, Hirose K, Ito H, Saito T, Tajima K. Alcohol dehydrogenase 2 His47Arg polymorphism influences drinking habit independently of aldehyde dehydrogenase 2 Glu487Lys polymorphism: analysis of 2299 Japanese subjects. Cancer Epidemiol Biomarkers Prev 2006; 15: 1009–13. [DOI] [PubMed] [Google Scholar]