© The Author 2007. Published by Oxford University Press.
ARTICLES |
Alcohol Intake and Renal Cell Cancer in a Pooled Analysis of 12 Prospective Studies
Affiliations of authors: Channing Laboratory (JEL, DJH, EC, EG, WCW) and Division of Preventive Medicine (JEB, SMZ), Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA; Departments of Nutrition (DJH, EG, WCW, SASW), Epidemiology (DJH, DS, HOA, JEB, EG, WCW, SMZ, SASW), and Biostatistics (DS), Harvard School of Public Health, Boston, MA; Department of Medical Epidemiology and Biostatistics (HOA) and Division of Nutritional Epidemiology, National Institute of Environmental Medicine (AW), Karolinska Institutet, Stockholm, Sweden; Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Department of Health and Health Services, Bethesda, MD (DA, MFL, AS); Department of Preventive Medicine and USC/Norris Comprehensive Cancer Center, University of Southern California, Los Angeles, CA (LB); Department of Epidemiology, Nutrition and Toxicology Research Institute Maastricht, Maastricht University, Maastricht, The Netherlands (PAVDB, LJS); Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, MN (ARF); Department of Social and Preventive Medicine, University at Buffalo, State University of New York, Buffalo, NY (JLF, SG); Northern California Cancer Center, Fremont, CA (PLHR); Epidemiology and Surveillance Research, American Cancer Society, Atlanta, GA (MLM, CR); Department of Public Health Sciences, University of Toronto, Toronto, ON, Canada (ABM); Department of Urology, Mayo Clinic College of Medicine, Jacksonville, FL (ASP); Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY (TER); Department of Health Promotion and Chronic Disease Prevention, National Public Health Institute, Helsinki, Finland (MV)
Correspondence to: Jung Eun Lee, ScD, Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA 02115 (e-mail: jung.lee{at}channing.harvard.edu).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsSubjects and MethodsResultsDiscussionReferencesNotes |
|---|
Background: The association between alcohol intake and risk of renal cell cancer has been inconsistent in case–control studies. An inverse association between alcohol intake and risk of renal cell cancer has been suggested in a few prospective studies, but each of these studies included a small number of cases.
Methods: We performed a pooled analysis of 12 prospective studies that included 530469 women and 229575 men with maximum follow-up times of 7–20 years. All participants had completed a validated food-frequency questionnaire at baseline. Using the primary data from each study, the study-specific relative risks (RRs) for renal cell cancer were calculated using Cox proportional hazards models and then pooled using a random-effects model. All statistical tests were two-sided.
Results: A total of 1430 (711 women and 719 men) cases of incident renal cell cancer were identified. The study-standardized incidence rates of renal cell cancer were 23 per 100000 person-years among nondrinkers and 15 per 100000 person-years among those who drank 15 g/day or more of alcohol. Compared with nondrinking, alcohol consumption (
15 g/day, equivalent to slightly more than one alcoholic drink per day) was associated with a decreased risk of renal cell cancer (pooled multivariable RR = 0.72, 95% confidence interval = 0.60 to 0.86; Ptrend<.001); statistically significant inverse trends with increasing intake were seen in both women and men. No difference by sex was observed (Pheterogeneity = .89). Associations between alcohol intake and renal cell cancer were not statistically different across alcoholic beverage type (beer versus wine versus liquor) (P = .40).
Conclusion: Moderate alcohol consumption was associated with a lower risk of renal cell cancer among both women and men in this pooled analysis.
Prior knowledge Results from studies of the relationship between alcohol consumption and renal cell cancer risk have been inconsistent. Study design Pooled analysis of 12 prospective studies to estimate relative risks of renal cell cancer by alcohol intake. Contribution Intake of approximately one drink per day was associated with a reduced risk of renal cell cancer, compared with no alcohol intake. The association was observed among both men and women. Implications Moderate intake of alcohol may be associated with reduced risk of renal cell cancer. Study limitations The study was survey based, and there may have been inaccuracies in reporting intake. Although the study was large, the majority of the population was white; thus, it is unknown whether the associations apply to other ethnic groups or other populations.
|
The incidence of renal cell cancer has increased during the past 30 years in the United States (1), Canada (2), and Northern Europe (3). This rapid increase is not fully explained by the increase in tumor detection (1); environmental factors, including diet, may have contributed to the upward trend. Although obesity (4,5), hypertension (5), and smoking (6) are all associated with an increased risk of renal cell cancer, the independent roles of obesity and hypertension in renal cell tumorigenesis remain controversial. Furthermore, the relationships between dietary factors and renal cell cancer risk remain unclear. In 1997, a working group of the World Cancer Research Fund and the American Institute for Cancer Research (7) concluded that associations between fat, cholesterol, and fish intakes and renal cell cancer risk remain unclear due to inconsistent and limited evidence but that a positive link between meat intake and risk of renal cell cancer may exist. However, a more recent review (8) found no clear support for the hypothesis that increased intakes of meat, milk, fat, and protein may increase the risk of renal cell cancer. In addition, in 2003, an international review panel sponsored by the International Agency for Research on Cancer (9) concluded that there was only limited evidence that consuming increased quantities of fruit and vegetables is associated with reduced risk for renal cell cancer.
The association between alcohol intake and renal cell cancer risk has been evaluated in at least 16 studies (10–25). Some case–control studies (10–14) have found that alcohol consumption is associated with a lower risk of renal cell cancer, although others (15–21) have not. Inverse associations between alcohol intake and risk of renal cell cancer have also been observed in recent prospective studies (22–25). However, evaluation of high intake of alcohol, intake of different types of alcoholic beverages, and whether the association varied by other potential risk factors was restricted due to the small number of patients (less than 250) in each of these prospective studies.
We evaluated the association between alcohol intake and risk of renal cell cancer by reanalyzing the primary data from 12 prospective studies (22–32) using standardized criteria. This pooled analysis included the five prospective studies that had previously examined the association between alcohol intake and risk of renal cell cancer (22–25) as well as seven studies that had not previously reported on this association.
|
Subjects and Methods |
|---|
|
TopAbstractContext and CaveatsSubjects and MethodsResultsDiscussionReferencesNotes |
|---|
Study Population
The Pooling Project of Prospective Studies of Diet and Cancer (herein referred to as the Pooling Project) has been described elsewhere (33). For the renal cell cancer analyses, each study that we included (22–32) (Table 1) met the following prespecified criteria: identification of at least 25 incident renal cell cancers, assessment of long-term intake of a variety of foods and beverages, and validation of the dietary assessment method used in the study or a closely related questionnaire. Studies that included both women and men were treated as two separate cohorts (one of women and one of men), and the inclusion criteria were applied to each cohort. Overall, the renal cell cancer analyses included 530469 women and 229575 men from 12 prospective studies. In the Pooling Project, the Nurses’ Health Study was analyzed in two parts corresponding to the 1980–1986 follow-up and the 1986–2000 follow-up periods to take advantage of the increased comprehensiveness of the 1986 food-frequency questionnaire. In the renal cell cancer analyses, we only used data from the Nurses’ Health Study starting in 1986 because fewer than 25 renal cell cancer cases were identified between 1980 and 1986. Each of the studies included here were reviewed and approved by the institutional review board of the institution at which the study was conducted.
|
Ascertainment of Renal Cell Cancer Patients
Renal cell cancer patients were ascertained by follow-up questionnaires and subsequent review of medical records (25,27), linkage to cancer registries (22,23,26,28–30), or both (24,31,32). Some studies also used linkage to mortality registries to identify incident and/or fatal outcomes (22–27,30–32). The follow-up rate of each cohort generally exceeded 90% (33). We defined renal cell cancer patients as those with histologically confirmed renal cell cancer (ICD-O-2 code = C64.9), using histologic codes based on the International Classification of Diseases for Oncology (34) or the morphologic classification provided by the study investigators. The most frequently described (61%) type of renal cell cancer in our data was renal cell carcinoma (not otherwise specified) (morphology code = 8312); the second most frequent (18%) was clear cell carcinoma. The higher proportion of renal cell carcinoma, not otherwise specified, in our database compared with surgical series that have reported clear cell carcinoma as the most common type of renal cell cancer (35) may be partially due to the large number of renal cell cancer patients in our analyses who were ascertained before 1997, when a workshop on the diagnosis and prognosis of renal cell cancer was held by the World Health Organization (35). This workshop prompted more emphasis on obtaining information on histologic types and hence more widespread use of the current classification system. Due to the insufficient number of patients with specific histologic types of renal cell cancer, we have combined all histologically confirmed renal cell cancers for our analyses.
Assessment of Alcohol and Other Dietary Factors
Each study assessed baseline dietary intake of a variety of foods and beverages with a validated food-frequency questionnaire. For each food or beverage, each study assessed the frequency of intake of specified portions (22,25,27), the frequency of intake of unspecified portions (23,26), or both the frequency of intake and the portion consumed (24,28,30–32,36). The consumption of beer, wine, and liquor was assessed separately in all studies except the New York State Cohort, in which only total alcohol consumption was queried; the New York State Cohort was not included in the beverage-specific analyses.
Each study provided alcohol intake data as either the number of servings of alcoholic beverages consumed per day or the grams of alcohol consumed per day. For studies that quantified intake in servings per day, the intake of alcohol from each beverage was converted to grams per day using the reported frequency of intake of each beverage, study-specific serving sizes of each beverage, and study-specific conversion factors for the amount of alcohol in beer, wine, and liquor. For example, information on conversion factors appropriate for the US populations was obtained from the US Department of Agriculture (37) (12.8 g of alcohol for a 12-oz can or bottle of beer, 11.0 g for a 4-oz glass of wine, and 14.0 g for one standard drink of liquor). Total alcohol intake was calculated by summing the alcohol intake from each alcoholic beverage.
Each study investigator provided data on the intakes of other nutrients, which were calculated in their study. For those nutrients, units were standardized across studies and intakes were energy adjusted within each study using the residuals from the regression of nutrient intake on total energy intake (38).
To determine how accurately the questionnaires estimated alcohol intake, alcohol intake from the food-frequency questionnaires used in the studies or a closely related questionnaire was compared with intake estimated by either multiple diet records or 24-hour recalls. The correlation coefficients comparing the two methods generally exceeded 0.8 (39–43) (Wolk A, Horn-Ross PL: personal communication).
Assessment of Nondietary Factors
Information on nondietary factors was collected in each study using self-administered questionnaires at baseline. Information on age, height, and weight was provided in all studies; we used height and weight information to calculate body mass index (BMI [weight in kilograms/height in square meters]). All the cohort studies among women assessed parity, age at first birth, oral contraceptive use, and hormone replacement therapy use. Most studies assessed smoking history and history of hypertension.
Statistical Analysis
After applying the study-specific exclusion criteria, we further excluded participants if they consumed an unreasonable energy intake (±3 standard deviations from the study-specific loge-transformed mean energy intake), had a history of cancer at baseline (except for nonmelanoma skin cancer), or had missing data on alcohol consumption. Each study was analyzed using the Cox proportional hazards model (44). Age at baseline (in days) and the year the baseline questionnaire was returned were used as stratification variables, thereby creating a time metric that simultaneously accounted for age, calendar time, and time since entry into the study. Person-years of follow-up time were calculated from the date of the baseline questionnaire until the date of renal cell cancer diagnosis, death, loss to follow-up (if applicable), or end of follow-up, whichever came first. We tested whether the assumption of the proportional hazards was satisfied by adding interaction terms between age and alcohol intake and found that terms were not statistically significant; thus, the assumption was satisfied. The studies were analyzed using SAS PROC PHREG (45).
We categorized alcohol intake using uniform cut points (none, 0.1–4.9 g/day, 5.0–14.9 g/day, and 15 g/day) across studies based on multiples of one drink per day (
15 g/day) and based on the distribution of alcohol intake in each study. If no participants diagnosed with renal cell cancer were in the highest intake category in a study, the participants in the highest category in that study were included in the second highest intake category. To test for trend across alcohol intake, participants were assigned the median value of their intake category. This variable was entered as a continuous term in the model, the coefficient for which was evaluated by the Wald test. In multivariable analyses, we further adjusted for history of hypertension (yes, no), pack-years of smoking (continuous), energy intake (kcal/day, continuous), BMI (kg/m2, continuous), and, among women, parity and age at first birth (age at first birth < 25 years and parity of 1 or 2, age at first birth 25 years and parity of 1 or 2 or nulliparous, age at first birth < 25 years and parity of 3, and age at first birth 25 years and parity of 3. Cut points for parity and age at first birth were based on their overall distribution across studies and on the magnitude of the associations with renal cell cancer that we observed in each cohort. An indicator variable was used for missing responses for each measured covariate within a study, if needed.
After calculating study- and sex-specific relative risks for each category, we combined the loge relative risks using a random-effects model (46–48). The individual study estimates were weighted by the inverse of their variance. We tested for heterogeneity across studies using the Q statistic, which follows an approximate chi-square distribution (df = number of studies in that analysis – 1) (48,49). Two-sided 95% confidence intervals (CIs) were calculated.
In additional analyses, we adjusted for smoking history using categories of never smoking, past smoking duration (<30, 30 years), and current smoking dose (<15, 15 to <25, 25 cigarettes/day) instead of pack-years of smoking to evaluate the effect of different parameterizations of smoking variables on the risk estimates observed for alcohol intake. Because patients who were diagnosed near the time when they complete their food-frequency questionnaire may have altered their diet due to prediagnostic disease symptoms, we also conducted analyses in which patients who were diagnosed during the first 2 years of follow-up were excluded. To address whether detection bias could have affected our results, we also conducted analyses in which we included only patients who died less than 2 years after diagnosis. We examined whether the risk estimates for alcohol from beer, wine, and liquor varied using a contrast test (50); the null hypothesis was that there was no difference in the pooled estimates across the three alcoholic beverages. This test statistic has an approximate chi-square distribution with 2 df.
To assess the linearity of the association between alcohol intake and risk of renal cell, we examined nonparametric regression curves using restricted cubic splines (51,52). To test for nonlinearity, the likelihood ratio test was used to compare the model fit including the linear and cubic spline terms selected by a stepwise regression procedure with the model fit including only the linear term; visual inspection of the restricted cubic spline graphs was also used. For these analyses, all studies were combined into a single dataset and then stratified by age, the year that the questionnaire was returned, and study and then adjusted for other covariates in the model. Participants with alcohol intake of 60 g/day or more (1% of overall study population) were excluded in the cubic spline analysis to avoid excessive influence of extreme intakes.
The study-specific relative risks for alcohol intake were corrected for measurement error by regressing intake from the reference dietary assessment methods (multiple 24-hour recalls or dietary records) (39–43) (Wolk A, Horn-Ross PL: personal communication) on intake from the food-frequency questionnaires (53,54). We then computed the corrected estimates of the loge relative risks by dividing the uncorrected estimates by the obtained regression coefficients. The corrected estimates were then pooled using a random-effects model (46–48). Under moderate measurement error as found in the validation studies for the studies included in this article, normality of residuals of the measurement error is not required (55). Although heteroscedasticity of the measurement error model can be a problem, in the case of mismeasurement of alcohol intake in cancer incidence models, the linear regression calibration method has been found to provide an accurate estimate of effect (56) (Spiegelman D, Logan R, Grove D: unpublished technical report; the manuscript is available at http://www.hsph.harvard.edu/faculty/spiegelman/manuscripts/beta_RCH.pdf).
We also assessed whether the associations for alcohol intake varied by sex, BMI (<25, 25 kg/m2), history of hypertension (no, yes), smoking status (never, ever), age at diagnosis (<68, 68; the median age at diagnosis), multivitamin use (user, nonuser), total folate intake (tertile), hormone replacement therapy use among postmenopausal women (ever, never), and, among women, parity (<3 children, 3 children) and oral contraceptive use (yes, no) using a mixed effects meta-regression model (57). A two-sided Wald test statistic was used to test the null hypothesis that there was no modification of the alcohol–renal cell cancer association by levels of the potential effect modifiers. In these analyses, we divided alcohol intake into three categories (none, 0.1–4.9 g/day, and 5 g/day), instead of four categories, because only a low proportion of participants drank 15 g/day or more of alcohol.
All statistical tests were two-sided. P values less than .05 were considered to be statistically significant.
|
Results |
|---|
|
TopAbstractContext and CaveatsSubjects and MethodsResultsDiscussionReferencesNotes |
|---|
During maximum follow-up periods of 7–20 years across studies, 1430 patients (711 women and 719 men) were diagnosed with renal cell cancer among 530469 women and 229575 men (Table 1). Alcohol was consumed more commonly and consumed in greater quantities among men than women. The proportion of women in the combined dataset of all studies (referred to as the aggregated dataset) who reported drinking a specific alcoholic beverage was 26% for beer, 49% for wine, and 30% for liquor; the corresponding proportions among men were 55%, 43%, and 52%, respectively. Among drinkers of each specific alcoholic beverage, median daily alcohol intakes were 1.8 g from beer, 3.3 g from wine, and 4.5 g from liquor among women and 3.3 g from beer, 1.9 g from wine, and 6.0 g from liquor among men. Median intakes of each beverage among drinkers of each specific alcoholic beverage varied at least 2.4-fold across the studies for women and 1.6-fold across the studies for men.
Total alcohol intake was inversely associated with risk of renal cell cancer. In the age-adjusted model, a modest inverse association was observed compared with nondrinkers (for alcohol intake of 0.1–4.9 g/day, pooled age-adjusted RR = 0.94, 95% CI = 0.81 to 1.08; for intake of 5.0–14.9 g/day, RR = 0.78, 95% CI = 0.67 to 0.92; for intake of 15 g/day, RR = 0.75, 95% CI = 0.62 to 0.89; Ptrend<.001). The associations became slightly stronger when we adjusted for pack-years of smoking (consumers of 15 g/day of alcohol versus nondrinkers, pooled age- and smoking-adjusted RR = 0.69, 95% CI = 0.58 to 0.82; Ptrend<.001). When we alternatively adjusted for smoking history using categories of never smoking, past smoking duration (<30, 30 years), and current smoking dose (<15, 15 to <25, 25 cigarettes/day) instead of pack-years of smoking, the results differed little (data not shown). The results from the multivariable model (comparing consumers of 15 g/day versus nondrinkers, RR = 0.72, 95% CI = 0.60 to 0.86; Ptrend<.001; Table 2) were similar to those from the age- and smoking-adjusted model. The study-standardized incidence rates of renal cell cancer were 23 per 100000 person-years among nondrinkers and 15 per 100000 person-years among those who drank 15 g/day or more of alcohol. A statistically nonsignificant lower risk of renal cell cancer was observed in nearly all studies for this comparison (Fig. 1). Furthermore, there was no evidence of heterogeneity across studies (Pheterogeneity = .99), indicating that the differences in results across studies were compatible with random variation. Results were similar for women and men (Pheterogeneity = .89). When we limited the analyses to alcohol drinkers, an inverse association was also observed (Ptrend<.001).
|
|
When patients who were diagnosed during the first 2 years of follow-up were excluded (n = 161), the results were similar (
15 g/day versus nondrinkers, the pooled multivariable RR = 0.71, 95% CI = 0.58 to 0.87; Ptrend<.001). The multivariable relative risks obtained in the aggregated dataset were not different from the pooled multivariable relative risks (data not shown). Because detection bias is less likely to be present among patients who died soon after diagnosis compared with those who had a longer survival, we conducted a sensitivity analysis in which we limited the analyses to patients who died less than 2 years after diagnosis (the 378 patients included represented 26% of the total). Because the smaller number of patients available for this analysis prevented calculation of study-specific results, we examined the association in the aggregated dataset only. In addition, due to the small number of patients with higher alcohol intake, we set the highest alcohol intake category at 5 g/day or more of alcohol. The relative risks in comparisons of individuals in the highest intake category with nondrinkers from analyses only including patients who died less than 2 years after diagnosis (multivariable RR = 0.77, 95% CI = 0.67 to 0.88; Ptrend<.001) were similar to those observed when all patients were included (pooled multivariable RR = 0.75; 95% CI = 0.65 to 0.87; Ptrend<.001).
When we assessed associations separately for specific alcoholic beverages, alcohol intakes from beer, wine, and liquor were each associated with a lower risk of renal cell cancer among women and men combined (Table 2). When these associations were examined separately by sex, the statistically significant inverse association was limited to alcohol from wine among women and to alcohol from beer and from liquor among men, although the differences in the risk estimates between women and men for each beverage were not statistically significant. In addition, the risk estimates for the three types of alcoholic beverages were not significantly different from each other for the category of 5 g/day or more of alcohol (P = .40).
The nonparametric regression curve and a formal test showed statistically significant nonlinearity in the association between alcohol consumption and renal cell cancer risk in the age- and calorie-adjusted (Pcurvature = .02) and multivariable (Pcurvature = .03) analyses. A linear inverse association was observed primarily in those with alcohol intakes of less than 30 g/day, and the relationship appeared flat above approximately 30 g/day (Fig. 2).
|
We corrected the study-specific relative risks for bias due to measurement error in alcohol intake in the nine cohort studies that assessed alcohol intake in their respective food-frequency questionnaires’ validation studies (39–43) (Wolk A, Horn-Ross PL: personal communication). For these analyses, we excluded participants who drank more than 30 g/day of alcohol (6% of the participants), so we could model alcohol consumption as a continuous variable over the intake range for which the association appeared linear on the log scale. The pooled corrected age- and calorie-adjusted relative risk for a 10 g/day increment (RR = 0.79, 95% CI = 0.70 to 0.89) was similar to the pooled uncorrected age- and calorie-adjusted relative risk (RR = 0.81, 95% CI = 0.74 to 0.90).
Because alcohol consumption was higher among men than women, we performed additional analyses in which we divided the 15 g/day or more alcohol intake category into three categories. As observed in the nonparametric regression curve analyses for the entire study population, the association was similar for these additional categories (all compared with nondrinkers) (for 0.1–4.9 g/day, pooled multivariable RR = 1.08, 95% CI = 0.87 to 1.35; n = 203; for 5.0–14.9 g/day, RR = 0.82, 95% CI = 0.61 to 1.10; n = 162; for 15.0–29.9 g/day, RR = 0.68, 95% CI = 0.51 to 0.90; n = 93; for 30.0–44.9 g/day, RR = 0.75, 95% CI = 0.54 to 1.06; n = 49; and for 45 g/day, RR = 0.76, 95% CI = 0.53 to 1.08; n = 44). When we additionally analyzed alcohol from beer and liquor using 15 g/day or more of alcohol intake as the highest category among men, the associations comparing 15 g/day or more of alcohol versus nondrinkers were inverse, but not statistically significant (for alcohol from beer, n = 36, pooled multivariable RR = 0.76, 95% CI = 0.53 to 1.09; Ptrend = .05; and for alcohol from liquor, n = 93, RR = 0.79, 95% CI = 0.56 to 1.13; Ptrend = .25).
We examined whether the association between alcohol intake and risk of renal cell cancer was modified by several risk factors for renal cell cancer (Table 3). The associations did not vary appreciably by BMI, history of hypertension, smoking status, or age at diagnosis. In addition, multivitamin use, total folate intake, hormone replacement therapy use among postmenopausal women, and parity and oral contraceptive use among women did not modify the association (data not shown).
|
|
Discussion |
|---|
|
TopAbstractContext and CaveatsSubjects and MethodsResultsDiscussionReferencesNotes |
|---|
In this pooled analysis of 1430 renal cell cancer patients from 12 prospective studies, we found that moderate alcohol intake was associated with a statistically significantly lower risk of renal cell cancer. Inverse, but non–statistically significant, associations were observed in nearly all the individual studies, including five (27–29,31,32) of the seven prospective studies that had not previously published their findings on alcohol intake and renal cell cancer risk. The lack of statistical significance of the study-specific associations could be due to the small number of cases in each study as only one of which included more than 200 patients. The association between alcohol intake and risk of renal cell cancer in the pooled analysis was not modified by several renal cell cancer risk factors, including age, BMI, history of hypertension, and smoking status. We found a stronger inverse association for alcohol from wine among women and for alcohol from beer and liquor among men, although the difference between the multivariable relative risks for the three beverages was not statistically significant. Non–statistically significant associations observed for alcohol from beer among women, alcohol from liquor among women, and alcohol from wine among men may be due to a lack of power because of low consumption of these alcoholic beverages. In these analyses of specific alcoholic beverages, we were restricted to using 5.0 g/day or more (2–3 drinks per week) as the highest category, especially for women, because of the limited number of patients with higher alcohol intakes for each beverage.
Our results confirm the suggestive inverse associations observed in the five published prospective studies (22–25) (all of which were included in our analysis) and help to resolve the inconsistency of the results from case–control studies (10–21), which may be subject to recall and selection biases. Inconsistent results regarding the association between alcoholism or heavy drinking and kidney cancer have been reported in some cohort studies (58–61), but these studies were limited by the absence of information on confounding factors, the inclusion of only alcoholics or heavy drinkers, and a small number of renal cell cancer patients. Due to the small proportion (1%) of participants in the studies we analyzed who reported drinking more than 60 g of alcohol per day, we were unable to evaluate associations with heavy drinking.
A potential mechanism by which alcohol may reduce the risk of renal cell cancer is by improving insulin sensitivity. Light to moderate alcohol consumption is associated with enhanced insulin sensitivity (62–65) and inversely associated with the risk of diabetes (66), which is usually characterized by impaired insulin sensitivity (67). At least two lines of evidence suggest that improving insulin sensitivity may lower the risk of renal cell cancer. First, persons who are obese, a known risk factor for renal cell cancer, have higher insulin levels than people who are not obese (68). Second, individuals with diabetes are more likely to develop renal cell cancer than individuals without diabetes (69,70).
The diuretic effect of alcohol intake could also be, in theory, hypothesized to lower the risk of renal cell cancer by decreasing the time that carcinogenic solutes are in contact with renal epithelial cells. This possible mechanism could be investigated by examining whether the diluting effect caused by high fluid intake is associated with a reduced risk of renal cell cancer. However, a pooled analysis of the Nurses’ Health Study and the Health Professionals Follow-up Study, both of which were included in our analyses, reported that total fluid intake was not associated with a lower risk of renal cell cancer (25).
Alcoholic beverages contain antioxidant phenolic compounds, which also may help to decrease the risk of renal cell cancer by removing oxidized carcinogenic agents, reducing lipid peroxidation, reducing cell proliferation, or promoting apoptosis (71,72). Although we also cannot exclude the possibility that these compounds are responsible for the apparent favorable association between alcoholic beverage intake and renal cell cancer risk, the finding that all three types of alcoholic beverages were associated with lower risk suggests that alcohol per se is most likely the responsible factor.
This pooled analysis had several limitations. We had only a baseline measure of alcohol intake and could not investigate changes in intake, intakes at earlier ages, or lifetime consumption. We also could not examine patterns or timing of alcohol consumption or the effects of high alcohol intake, particularly for specific alcoholic beverages. The amount of alcohol consumed may have been underreported, particularly by heavy drinkers, even though, overall, the validation studies for the food-frequency questionnaires used in these studies or closely related questionnaires showed that alcohol intake was measured accurately compared with the referent methods (39–43) (Wolk A, Horn-Ross PL: personal communication). However, we cannot rule out the possibility that any errors in the food-frequency questionnaires and in the referent methods were correlated, which would result in incomplete removal of all the bias due to measurement error (54,73–75). We did not have information on some risk factors for renal cell cancer, including family history of renal cell cancer, environmental exposures such as asbestos, medications such as phenacetin, or advanced kidney disease and thus were unable to control for these factors in our analyses. However, the inverse associations for alcohol intake that we observed are not likely to be fully explained by confounding because these exposures would need to be both common and strongly associated with alcohol consumption. In our analyses, smoking was the strongest confounder for the association between alcohol intake and risk of renal cell cancer among the potential known risk factors that we controlled for in our study, and adjustment for smoking strengthened the association. With more thorough control for smoking, the true inverse relationship may have been stronger. We were not able to examine whether the associations differed by ethnicity because more than 90% of the participants in our study were white. Finally, we did not have sufficient information or power to examine associations separately by histologic type of renal cell cancer or by tumor characteristics.
This analysis also has important strengths. Because of the prospective design and high follow-up rates of the studies, recall bias or selection bias are very unlikely to account for our findings. In addition, because our study had more patients than the single prospective studies that were included, we achieved better precision than the individual prospective studies. Because we analyzed the primary data from each study, we were able to use uniform categories of alcohol intake and covariates across studies in the analyses to remove potential sources of heterogeneity across studies.
In conclusion, we found that modest intake of alcohol was associated with a lower risk of renal cell cancer. Also, our findings did not suggest that intake of any particular alcoholic beverage was more strongly associated with a lower risk of renal cell cancer. Future investigations are needed to provide information on potential mechanisms supporting this association. However, because alcohol drinking is associated with increased risks of cancers of the oral cavity, larynx, pharynx, esophagus, liver (76), and breast (77), and probably the colon and rectum (7,78), maintaining a healthy weight and avoiding smoking are the principal known means to reduce the risk of renal cell cancer that should be encouraged and doing so may also reduce the risk of many other cancers as well as cardiovascular disease.
|
NOTES |
|---|
|
TopAbstractContext and CaveatsSubjects and MethodsResultsDiscussionReferencesNotes |
|---|
The Pooling Project is supported by National Institutes of Health grant CA55075 (to W. C. Willett). The sponsor had no role in the design, data collection, data analysis, interpretation of the results, the preparation of the manuscript, or the decision to submit the manuscript for publication.
The authors thank Ruifeng Li, Christine C. Rivera, and Shiaw-Shyuan Yaun for assistance in data management and Carol Piñeiro for assistance in preparing the manuscript.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsSubjects and MethodsResultsDiscussionReferencesNotes |
|---|
(1) Chow WH, Devesa SS, Warren JL, Fraumeni JF Jr. Rising incidence of renal cell cancer in the United States. JAMA (1999) 281:1628–31.
(2) Liu S, Semenciw R, Morrison H, Schanzer D, Mao Y. Kidney cancer in Canada: the rapidly increasing incidence of adenocarcinoma in adults and seniors. Can J Public Health (1997) 88:99–104.[Web of Science][Medline]
(3) Bray F, Sankila R, Ferlay J, Parkin DM. Estimates of cancer incidence and mortality in Europe in 1995. Eur J Cancer (2002) 38:99–166.[CrossRef][Web of Science][Medline]
(4) Bergstrom A, Hsieh CC, Lindblad P, Lu CM, Cook NR, Wolk A. Obesity and renal cell cancer—a quantitative review. Br J Cancer (2001) 85:984–90.[CrossRef][Web of Science][Medline]
(5) Chow WH, Gridley G, Fraumeni JF Jr, Jarvholm B. Obesity, hypertension, and the risk of kidney cancer in men. N Engl J Med (2000) 343:1305–11.
(6) Hunt JD, van der Hel OL, McMillan GP, Boffetta P, Brennan P. Renal cell carcinoma in relation to cigarette smoking: meta-analysis of 24 studies. Int J Cancer (2005) 114:101–8.[CrossRef][Web of Science][Medline]
(7) World Cancer Research Fund, American Institute for Cancer Research. Food, nutrition and the prevention of cancer: a global perspective. (1997) Washington (DC): American Institute for Cancer Research.
(8) McLaughlin JK, Lipworth L, Tarone RE, Blot WJ. Renal cancer. In: Cancer epidemiology and prevention—Schottenfeld D, Fraumeni JF Jr, eds. (2006) 3rd ed. New York: Oxford University Press. 1087–100.
(9) IARC. Fruit and Vegetables. In: IARC handbooks of cancer prevention—Vainio H, Bianchini F, eds. (2003) Lyon (France): IARC Press. 98–237.
(10) Wolk A, Gridley G, Niwa S, Lindblad P, McCredie M, Mellemgaard A, et al. International renal cell cancer study. VII. Role of diet. Int J Cancer (1996) 65:67–73.[CrossRef][Web of Science][Medline]
(11) Parker AS, Cerhan JR, Lynch CF, Ershow AG, Cantor KP. Gender, alcohol consumption, and renal cell carcinoma. Am J Epidemiol (2002) 155:455–62.
(12) Asal NR, Risser DR, Kadamani S, Geyer JR, Lee ET, Cherng N. Risk factors in renal cell carcinoma: I. Methodology, demographics, tobacco, beverage use, and obesity. Cancer Detect Prev (1988) 11:359–77.[Web of Science][Medline]
(13) Goodman MT, Morgenstern H, Wynder EL. A case-control study of factors affecting the development of renal cell cancer. Am J Epidemiol (1986) 124:926–41.
(14) Hu J, Mao Y, White K. Diet and vitamin or mineral supplements and risk of renal cell carcinoma in Canada. Cancer Causes Control (2003) 14:705–14.[CrossRef][Web of Science][Medline]
(15) Brownson RC. A case-control study of renal cell carcinoma in relation to occupation, smoking, and alcohol consumption. Arch Environ Health (1988) 43:238–41.[Web of Science][Medline]
(16) Yu MC, Mack TM, Hanisch R, Cicioni C, Henderson BE. Cigarette smoking, obesity, diuretic use, and coffee consumption as risk factors for renal cell carcinoma. J Natl Cancer Inst (1986) 77:351–6.[Web of Science][Medline]
(17) Talamini R, Baron AE, Barra S, Bidoli E, La Vecchia C, Negri E, et al. A case-control study of risk factor for renal cell cancer in northern Italy. Cancer Causes Control (1990) 1:125–31.[CrossRef][Web of Science][Medline]
(18) Maclure M, Willett W. A case-control study of diet and risk of renal adenocarcinoma. Epidemiology (1990) 1:430–40.[Medline]
(19) Benhamou S, Lenfant MH, Ory-Paoletti C, Flamant R. Risk factors for renal-cell carcinoma in a French case-control study. Int J Cancer (1993) 55:32–6.[Web of Science][Medline]
(20) Lindblad P, Wolk A, Bergstrom R, Adami HO. Diet and risk of renal cell cancer: a population-based case-control study. Cancer Epidemiol Biomarkers Prev (1997) 6:215–23.[Abstract]
(21) Pelucchi C, La Vecchia C, Negri E, Talamini R, Franceschi S. Alcohol drinking and renal cell carcinoma in women and men. Eur J Cancer Prev (2002) 11:543–5.[CrossRef][Web of Science][Medline]
(22) Nicodemus KK, Sweeney C, Folsom AR. Evaluation of dietary, medical and lifestyle risk factors for incident kidney cancer in postmenopausal women. Int J Cancer (2004) 108:115–21.[CrossRef][Web of Science][Medline]
(23) Rashidkhani B, Åkesson A, Lindblad P, Wolk A. Alcohol consumption and risk of renal cell carcinoma—a prospective study of Swedish women. Int J Cancer (2005) 117:848–53.[CrossRef][Web of Science][Medline]
(24) Mahabir S, Leitzmann MF, Virtanen MJ, Virtamo J, Pietinen P, Albanes D, et al. Prospective study of alcohol drinking and renal cell cancer risk in a cohort of Finnish male smokers. Cancer Epidemiol Biomarkers Prev (2005) 14:170–5.
(25) Lee JE, Giovannucci E, Smith-Warner SA, Spiegelman D, Willett WC, Curhan GC. Total fluid intake and use of individual beverages and risk of renal cell cancer in two large cohorts. Cancer Epidemiol Biomarkers Prev (2006) 15:1204–11.
(26) Bandera EV, Freudenheim JL, Marshall JR, Zielezny M, Priore RL, Brasure J, et al. Diet and alcohol consumption and lung cancer risk in the New York State Cohort (United States). Cancer Causes Control (1997) 8:828–40.[CrossRef][Web of Science][Medline]
(27) Higginbotham S, Zhang ZF, Lee IM, Cook NR, Giovannucci E, Buring JE, et al. Dietary glycemic load and risk of colorectal cancer in the Women's Health Study. J Natl Cancer Inst (2004) 96:229–33.
(28) van Dijk BA, Schouten LJ, Kiemeney LA, Goldbohm RA, van den Brandt PA. Vegetable and fruit consumption and risk of renal cell carcinoma: results from the Netherlands cohort study. Int J Cancer (2005) 117:648–54.[CrossRef][Web of Science][Medline]
(29) Rohan TE, Jain M, Howe GR, Miller AB. Alcohol consumption and risk of breast cancer: a cohort study. Cancer Causes Control (2000) 11:239–47.[CrossRef][Web of Science][Medline]
(30) Horn-Ross PL, Canchola AJ, West DW, Stewart SL, Bernstein L, Deapen D, et al. Patterns of alcohol consumption and breast cancer risk in the California Teachers Study cohort. Cancer Epidemiol Biomarkers Prev (2004) 13:405–11.
(31) Flood A, Caprario L, Chaterjee N, Lacey JV Jr, Schairer C, Schatzkin A. Folate, methionine, alcohol and colorectal cancer in a prospective study of women in the United States. Cancer Causes Control (2002) 13:551–61.[CrossRef][Web of Science][Medline]
(32) Calle EE, Rodriguez C, Jacobs EJ, Almon ML, Chao A, McCullough ML, et al. The American Cancer Society Cancer Prevention Study II Nutrition Cohort: rationale, study design, and baseline characteristics. Cancer (2002) 94:500–11.[CrossRef][Web of Science][Medline]
(33) Smith-Warner SA, Spiegelman D, Ritz J, Albanes D, Beeson WL, Bernstein L, et al. Methods for pooling results of epidemiologic studies. Am J Epidemiol (2006) 163:1053–64.
(34) Percy C, Van Holten V, Muir C. International Classification of Diseases for Oncology (1990) Geneva (Switzerland): World Health Organization.
(35) Storkel S, Eble JN, Adlakha K, Amin M, Blute ML, Bostwick DG, et al. Classification of renal cell carcinoma: Workgroup No. 1. Cancer (1997) 80:987–9.[CrossRef][Web of Science][Medline]
(36) Rohan TE, Jain M, Howe GR, Miller AB. A cohort study of dietary carotenoids and lung cancer risk in women (Canada). Cancer Causes Control (2002) 13:231–7.[CrossRef][Web of Science][Medline]
(37) U.S. Department of Agriculture, Agricultural Research Service. USDA Nutrient Data Laboratory. USDA Nutrient Database for Standard Reference, Release 13. (1999) Washington (DC): U.S. Department of Agriculture.
(38) Willett W, Stampfer M. Implications of total energy intake for epidemiologic analysis. In: Nutritional epidemiology—Willett W, ed. (1998) 2nd ed. New York: Oxford University Press. 273–301.
(39) Giovannucci E, Colditz G, Stampfer MJ, Rimm EB, Litin L, Sampson L, et al. The assessment of alcohol consumption by a simple self-administered questionnaire. Am J Epidemiol (1991) 133:810–7.
(40) Goldbohm RA, van den Brandt PA, Brants HAM, van't Veer P, Al M, Sturmans F, et al. Validation of a dietary questionnaire used in a large-scale prospective cohort study on diet and cancer. Eur J Clin Nutr (1994) 48:253–65.[Web of Science][Medline]
(41) Munger RG, Folsom AR, Kushi LH, Kaye SA, Sellers TA. Dietary assessment of older Iowa women with a food frequency questionnaire: nutrient intake, reproducibility, and comparison with 24-hour dietary recall interviews. Am J Epidemiol (1992) 136:192–200.
(42) Pietinen P, Hartman AM, Haapa E, Räsänen L, Haapakoski J, Palmgren J, et al. Reproducibility and validity of dietary assessment instruments. I. A self-administered food use questionnaire with a portion size picture booklet. Am J Epidemiol (1988) 128:655–66.
(43) Flagg EW, Coates RJ, Calle EE, Potischman N, Thun MJ. Validation of the American Cancer Society Cancer Prevention Study II Nutrition Survey Cohort food frequency questionnaire. Epidemiology (2000) 11:462–8.[CrossRef][Web of Science][Medline]
(44) Cox DR. Regression models and life-tables. J R Stat Soc (1972) 34:187–220.
(45) Allison PD. Survival analysis using the SAS system: a practical guide (1995) Cary (NC): SAS Institute Inc.
(46) Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics (1982) 38:963–74.[CrossRef][Web of Science][Medline]
(47) Harville DA. Maximum likelihood approaches to variance component estimation and to related problems. J Am Stat Assoc (1977) 72:320–40.[CrossRef][Web of Science]
(48) DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials (1986) 7:177–88.[CrossRef][Web of Science][Medline]
(49) Cochran WG. The combination of estimates from different experiments. Biometrics (1954) 10:101–129.[Medline]
(50) Anderson TW. Introduction to multivariate statistics (1984) New York: John Wiley Sons.
(51) Smith PL. Splines as a useful and convenient statistical tool. Am Stat (1979) 33:57–62.[CrossRef]
(52) Durrleman S, Simon R. Flexible regression models with cubic splines. Stat Med (1989) 8:551–61.[Web of Science][Medline]
(53) Rosner B, Spiegelman D, Willett WC. Correction of logistic regression relative risk estimates and confidence intervals for measurement error: the case of multiple covariates measured with error. Am J Epidemiol (1990) 132:734–45.
(54) Spiegelman D, Schneeweiss S, McDermott A. Measurement error correction for logistic regression models with an "alloyed gold standard". Am J Epidemiol (1997) 145:184–96.
(55) Kuha J. Corrections for exposure measurement error in logistic regression models with an application to nutritional data. Stat Med (1994) 13:1135–48.[Web of Science][Medline]
(56) Schmid CH, Rosner B. A Bayesian approach to logistic regression models having measurement error following a mixture distribution. Stat Med (1993) 12:1141–53.[Web of Science][Medline]
(57) Stram DO. Meta-analysis of published data using a linear mixed-effects model. Biometrics (1996) 52:536–44.[CrossRef][Web of Science][Medline]
(58) Monson RR, Lyon JL. Proportional mortality among alcoholics. Cancer (1975) 36:1077–9.[CrossRef][Web of Science][Medline]
(59) Pell S, D'Alonzo CA. A five-year mortality study of alcoholics. J Occup Med (1973) 15:120–5.[Medline]
(60) Jensen OM. Cancer morbidity and causes of death among Danish brewery workers. Int J Cancer (1979) 23:454–63.[Web of Science][Medline]
(61) Adami HO, McLaughlin JK, Hsing AW, Wolk A, Ekbom A, Holmberg L, et al. Alcoholism and cancer risk: a population-based cohort study. Cancer Causes Control (1992) 3:419–25.[CrossRef][Web of Science][Medline]
(62) Facchini F, Chen YD, Reaven GM. Light-to-moderate alcohol intake is associated with enhanced insulin sensitivity. Diabetes Care (1994) 17:115–9.[Abstract]
(63) Davies MJ, Baer DJ, Judd JT, Brown ED, Campbell WS, Taylor PR. Effects of moderate alcohol intake on fasting insulin and glucose concentrations and insulin sensitivity in postmenopausal women: a randomized controlled trial. JAMA (2002) 287:2559–62.
(64) Lazarus R, Sparrow D, Weiss ST. Alcohol intake and insulin levels. The Normative Aging Study. Am J Epidemiol (1997) 145:909–16.
(65) Kiechl S, Willeit J, Poewe W, Egger G, Oberhollenzer F, Muggeo M, et al. Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional population study (Bruneck study). BMJ (1996) 313:1040–4.
(66) Howard AA, Arnsten JH, Gourevitch MN. Effect of alcohol consumption on diabetes mellitus: a systematic review. Ann Intern Med (2004) 140:211–9.
(67) Fujimoto WY. The importance of insulin resistance in the pathogenesis of type 2 diabetes mellitus. Am J Med (2000) 108(Suppl 6a):9S–14S.[Medline]
(68) Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer (2004) 4:579–91.[CrossRef][Web of Science][Medline]
(69) Lindblad P, Chow WH, Chan J, Bergstrom A, Wolk A, Gridley G, et al. The role of diabetes mellitus in the aetiology of renal cell cancer. Diabetologia (1999) 42:107–12.[CrossRef][Web of Science][Medline]
(70) Wideroff L, Gridley G, Mellemkjaer L, Chow WH, Linet M, Keehn S, et al. Cancer incidence in a population-based cohort of patients hospitalized with diabetes mellitus in Denmark. J Natl Cancer Inst (1997) 89:1360–5.
(71) Kehrer JP, Smith CV. Free radicals in biology: sources, reactivities, and roles in the etiology of human diseases. In: Natural antioxidants in human health and disease—Frei B, ed. (1994) San Diego (CA): Academic Press. 25–62.
(72) Halliwell B. Free radicals, antioxidants, and human disease: curiosity, cause, or consequence? Lancet (1994) 344:721–4.[CrossRef][Web of Science][Medline]
(73) Plummer M, Clayton D. Measurement error in dietary assessment: an investigation using covariance structure models. Part I. Stat Med (1993) 12:925–35.
(74) Plummer M, Clayton D. Measurement error in dietary assessment: an investigation using covariance structure models. Part II. Stat Med (1993) 12:937–48.
(75) Spiegelman D, Zhao B, Kim J. Correlated errors in biased surrogates: study designs and methods for measurement error correction. Stat Med (2005) 24:1657–82.[CrossRef][Web of Science][Medline]
(76) IARC. IARC monographs on the evaluation of carcinogen risk to humans. In: Alcohol drinking (1988) Vol 44. Lyon (France): International Agency for Research on Cancer.
(77) World Health Organization (WHO). Diet, nutrition and the prevention of chronic diseases: report of a joint WHO/FAO expert consultation. (2003) Geneva (Switzerland): WHO. Report No.: WHO Technical Report Series No. 916.
(78) Cho E, Smith-Warner SA, Ritz J, van den Brandt PA, Colditz GA, Folsom AR, et al. Alcohol intake and colorectal cancer: a pooled analysis of 8 cohort studies. Ann Intern Med (2004) 140:603–13.
Manuscript received November 2, 2006; revised March 6, 2007; accepted April 2, 2007.
Related Article in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 741.
This article has been cited by other articles:
|
|
 |
|
  C. Galeone, C. Pelucchi, L. D. Maso, E. Negri, R. Talamini, M. Montella, V. Ramazzotti, R. Bellocco, S. Franceschi, and C. La Vecchia Glycemic index, glycemic load and renal cell carcinoma risk Ann. Onc., November 1, 2009; 20(11): 1881 - 1885. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. E. Allen, V. Beral, D. Casabonne, S. W. Kan, G. K. Reeves, A. Brown, J. Green, and on behalf of the Million Women Study Collaborators Moderate Alcohol Intake and Cancer Incidence in Women J Natl Cancer Inst, March 4, 2009; 101(5): 296 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Schouten, B. A.C. van Dijk, E. Oosterwijk, M. van Engeland, C. A. Hulsbergen-van de Kaa, L. A.L.M. Kiemeney, R. A. Goldbohm, A. Kester, S. de Vogel, J. A. Schalken, et al. Alcohol Consumption and Mutations or Promoter Hypermethylation of the von Hippel-Lindau Gene in Renal Cell Carcinoma Cancer Epidemiol. Biomarkers Prev., December 1, 2008; 17(12): 3543 - 3550. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Pelucchi, C. Galeone, M. Montella, J. Polesel, A. Crispo, R. Talamini, E. Negri, V. Ramazzotti, M. Grimaldi, S. Franceschi, et al. Alcohol consumption and renal cell cancer risk in two Italian case-control studies Ann. Onc., May 1, 2008; 19(5): 1003 - 1008. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||