Background: Fruits and vegetables rich in antioxidants have been proposed to reduce the risk of renal cell cancer. However, few prospective studies have examined the intakes of fruits, vegetables, and antioxidant vitamins in relation to the risk of renal cell cancer.

Methods: We prospectively examined the associations between the intakes of fruits, vegetables, vitamins A, C, and E, and carotenoids and risk of renal cell cancer in women and men. We followed 88,759 women in the Nurses' Health Study from 1980 to 2000, and 47,828 men in the Health Professionals Follow-up Study from 1986 to 2000. We assessed dietary intake every 2 to 4 years using a validated semiquantitative food frequency questionnaire. The Cox proportional hazards model was used to estimate study-specific multivariate relative risks (RR), which were pooled using a random effects model.

Results: A total of 248 (132 women and 116 men) incident renal cell cancer cases were ascertained during 2,316,525 person-years of follow-up. The consumption of fruits and vegetables was associated with a decreased risk of renal cell cancer in men (multivariate RR, 0.45; 95% CI, 0.25-0.81, for ≥6 servings of fruit and vegetable intake/d versus <3 servings/d; P test for trend = 0.02), but not in women (multivariate RR, 1.17; 95% CI, 0.66-2.07, for the same contrast; P test for trend = 0.25; P test for between-studies heterogeneity = 0.02). Intakes of vitamins A and C from food and carotenoids were inversely associated with the risk of renal cell cancer in men only, but we cannot exclude the possibility that this was due to other factors in fruit and vegetables. No clear association was observed for vitamin E in women or men.

Conclusions: Fruit and vegetable consumption may reduce the risk of renal cell cancer in men. (Cancer Epidemiol Biomarkers Prev 2006;12(24):2445–52)

The incidence rates of kidney cancer, and specifically, renal cell cancer, which accounts for >80% of all kidney cancers, have been increasing globally (1). Obesity and hypertension have contributed to this upward trend (2, 3), but the association of diet with risk of renal cell cancer is unclear.

Although antioxidants and other bioactive compounds in fruits and vegetables have been proposed to prevent cancers, evidence of the effects of fruit, vegetable, and antioxidant consumption on the risk of renal cell cancer is limited, particularly from prospective studies. An international review panel sponsored by the IARC concluded that the evidence for a cancer-preventive effect of fruit and vegetable consumption on kidney cancer was limited (4). Recent prospective studies generally have found no association between fruit and vegetable intakes and risk of renal cell cancer (5-7), although a cohort of Swedish women found a suggestive inverse association (8). For intakes of vitamins A, C, or E, or carotenoids in relation to renal cell cancer risk (5, 9-14), the results have also been inconsistent.

We prospectively examined whether the intake of fruits, vegetables, vitamins A, C, and E, and carotenoids were associated with a lower risk of renal cell cancer in two large cohorts, the Nurses' Health Study (NHS) and the Health Professionals Follow-up Study (HPFS).

Study Population

The NHS was established in 1976, when 121,700 female registered nurses who were 30 to 55 years of age returned a mailed questionnaire. The HPFS was initiated in 1986 when 51,529 male health professionals, consisting of dentists, pharmacists, optometrists, osteopath physicians, podiatrists, and veterinarians, aged 40 to 75 years, returned a mailed questionnaire. Participants in these two cohorts provided detailed information about medical history, lifestyle, and various risk factors for chronic diseases; follow-up questionnaires were sent at 2-year intervals. The follow-up rates to questionnaires for both studies exceeded 90% of the potential person-years.

Dietary Assessment

Dietary information was collected from the NHS participants using validated semiquantitative food frequency questionnaires (SFFQ) in 1980, 1984, 1986, and every 4 years thereafter, and from the HPFS participants every 4 years since 1986. In the NHS, the 1980 SFFQ included 61 food items, including 6 questions on fruits, 11 questions on vegetables, and 3 on potatoes. The SFFQ used in 1984 in the NHS was expanded to include 15 questions on fruits, 28 questions on vegetables, and 3 on potatoes. The SFFQs used in 1986, 1990, 1994, and 1998 were similar to the 1984 SFFQ. In the HPFS, SFFQs with 116 food items, including 15 questions on fruits, 30 questions on vegetables, and 3 on potatoes were administered in 1986, and subsequent SFFQs have had a similar number of questions on fruit and vegetable consumption.

Participants were asked how frequently, on average, during the past year they consumed one standard serving of a specific food item in nine categories (<1/mo, 1-3/mo, 1/wk, 2-4/wk, 5-6/wk, 1/d, 2-3/d, 4-5/d, or ≥6/d). Responses on the frequencies of a specified serving size for each food item were converted to average daily intake. We then calculated the intakes of total fruits and vegetables as well as specific fruit and vegetable groups by summing the average daily intake for each food item in each group. The specific groups examined included total fruits, citrus fruits, total vegetables, green leafy vegetables, cruciferous vegetables, vitamin C–rich fruits, and vegetables (i.e., fruits and vegetables that contain ≥30 mg of vitamin C per serving), legumes, and potatoes. We grouped fruits and vegetables based on the criteria used by Smith et al. (15), and modified it to correspond with our questionnaires (16). Fruit and vegetable juice in relation to risk of renal cell cancer has been examined elsewhere and no association was observed in these two cohorts (17). We excluded beans and potatoes from the total fruit and vegetable group and total vegetable group because of their high protein and starch contents, respectively (18). We calculated the intakes of vitamins and β-carotene from food (dietary intake) and from both food and vitamin supplements (total intake). The quantities of vitamins and carotenoids from foods were calculated by multiplying the reported frequency of each food by the nutrient content of one serving of that food (19). We took into account the participants' current use and dose of vitamin supplements, and the brand and type of multivitamins were asked every 2 years to calculate the supplemental intake.

The average correlation coefficients between intakes of specific fruit and vegetable items from the SFFQs and those from multiple 1-week diet records were 0.79 (range, 0.74-0.84) for fruits and 0.53 (range, 0.30-0.73) for vegetables in women (20), and 0.75 (range, 0.38-0.95) for fruits and 0.46 (range, 0.19-0.80) for vegetables in men (21), following correction for attenuation due to random error in diet records (22). For vitamins, correlation coefficients between intakes from the SFFQs and the average of two to four 1-week diet records were 0.49 for total vitamin A and 0.75 for total vitamin C in women (23), and 0.61, 0.92, and 0.92 for total vitamins A, C, and E, respectively, in men (24). The correlations of the average of the two SFFQs with plasma concentration of α-carotene, β-carotene, β-cryptoxanthin, lutein, and lycopene were 0.21 to 0.48 in male and female nonsmokers (25).

Assessment of Renal Cell Cancer

We obtained self-reported information on the occurrence of kidney cancer on each questionnaire, and asked participants (or next-of-kin for those who died) who reported a diagnosis of kidney cancer for permission to access medical records related to the diagnosis. The deaths in the cohort were ascertained by reports of family members in response to the follow-up questionnaires. In addition, the National Death Index (26) was used to identify fatalities. Physicians blinded to the participants' risk factor status reviewed medical records. We included only those with renal cell cancer as cases because transitional cell cancer, which arises from the renal pelvis, may have a different etiology. After a review of medical records, we included clear cell carcinoma (n = 148), papillary carcinoma (n = 25), chromophobe carcinoma (n = 3), and renal cell carcinoma not otherwise classified (n = 72), based on the classification developed by the workshop held by the WHO (27). A total of 132 cases in the NHS and 116 cases in the HPFS were included after applying exclusion criteria.

Statistical Analyses

We excluded participants who did not return the baseline SFFQ, had been previously diagnosed with cancer (except non–melanoma skin cancer prior to baseline), left extensive items blank on the baseline SFFQs for each analysis (>10 in 1980, >11 in 1984, and ≥70 in 1986), or reported implausible energy intake at baseline (<500 or >3,500 kcal/d for women and <800 or >4,200 kcal/d for men). As a result, 88,759 women in the NHS and 47,828 men in the HPFS were included for the analyses, for which follow-up started in 1980 and 1986, respectively. Person-years of follow-up were estimated from the date that the baseline questionnaire was returned to the date of renal cell cancer diagnosis, date of death, or end of follow-up (May 31, 2000, for women, and January 31, 2000, for men), whichever came first.

Participants were categorized using study-specific absolute cut-points or quartiles on the basis of the distribution of intakes in the study populations. As the main analytic strategy, cumulative average intakes were calculated from all available questionnaires up to the end of each 2-year follow-up interval. For example, in the NHS, fruit intake in 1980 was used for the analyses of renal cell cancer diagnosed from 1980 to 1984, and the average of fruit intake in 1980 and 1984 was used for the analyses of renal cell cancer diagnosed from 1984 to 1986. Additional analyses were done using a simple update of dietary intake and using only baseline intake. Simple updated intakes were calculated using the most recent SFFQs prior to the diagnosis. For example, fruit intake in 1980 was used for the analyses of renal cell cancer diagnosed from 1980 to 1984, and fruit intake in 1984 was used for the analyses of renal cell cancer diagnosed from 1984 to 1986. In the baseline analyses, we used only the baseline SFFQs. Vitamins and carotenoids were energy-adjusted using the residuals from the regression of nutrient intake on total energy intake (28). The relative risks (RR) were calculated using the Cox proportional hazards model (29) with SAS PROC PHREG (30). By stratifying according to age in months at the start of each questionnaire cycle and calendar year of the current questionnaire cycle, we controlled as finely as possible for confounding by age, calendar time, and any possible two-way interactions between these two time scales. In the multivariate models, we also adjusted for possible risk factors, including body mass index (BMI; <25, 25-26.9, 27-29.9, ≥30 kg/m2), smoking (never, 1-19, 20-39, ≥40 pack-years of smoking), history of hypertension (yes, no), multivitamin use (never, past, current <5 years, and current ≥5 years of use), alcohol intake (continuous), and total energy intake (continuous) in both men and women. We adjusted for parity (continuous) and history of diabetes (yes, no) in women. Because history of diabetes in men was not associated with risk of renal cell cancer, we did not include it in the study-specific multivariate models. Pack-years of smoking were calculated by multiplying the duration and dose of smoking; 1 pack-year is equivalent to having smoked one pack per day for 1 year. Information on BMI, smoking, history of hypertension, history of diabetes and multivitamin use were updated every 2 years. Parity among women was updated until 1984. Duration of multivitamin use was calculated based on the reported duration at baseline and the subsequent responses to current multivitamin use. Two-sided 95% confidence intervals (95% CIs) were obtained for the RRs. To calculate the P values for the test for trend, participants were assigned the median value of their intake category, and this variable was treated as a continuous term in the model.

After obtaining RRs from each cohort, we combined loge RRs using a random effects model (31, 32). Heterogeneity between the two studies was assessed by the Q statistic (33, 34). We also did several subgroup analyses and examined whether the association varied by BMI (<25, ≥25 kg/m2), history of hypertension (yes, no), smoking habits (ever, never), multivitamin use (ever users, never users), or age (<65, ≥65 years). Tests for heterogeneity were conducted using a Wald test based on the pooled estimates of the cross-product term of a continuous exposure variable with a binary modifier variable.

During 1,708,260 person-years of follow-up in the NHS, and 608,265 person-years of follow-up in the HPFS, we documented 132 incident renal cell cancer cases in the NHS, and 116 cases in the HPFS. Women and men who consumed more fruit and vegetables were generally older, smoked less, consumed less animal fat, were more physically active, and were more likely to use multivitamins compared with those who consumed less fruit and vegetables (Table 1). Median intakes of total fruits in 1986 were 2.3 servings/d for women and 2.1 servings/d for men, and median intakes of total vegetables in 1986 were 2.9 servings/d for women and 3.0 servings/d for men.

Table 1.

Characteristics of participants by fruit and vegetable consumption in 1986

Total fruits (servings/d)
Total vegetables (servings/d)
<1/d1-1.9/d2-2.9/d≥3/d<2/d2-2.9/d3-3.9/d≥4/d
NHS women         
    Age (y) 51 52 53 54 52 52 53 54 
    BMI (kg/m225 25 25 25 25 25 25 25 
    Physical activity (METS/wk)* 10 12 14 18 11 13 17 
    Alcohol intake (g/d) 14 12 11 10 11 12 12 
    Animal fat (g/d) 37 34 32 30 35 34 33 31 
    Current smoker (%) 37 24 17 15 31 26 21 18 
    History of hypertension (%) 24 25 26 27 26 25 26 26 
    History of diabetes (%) 2.9 3.8 3.8 4.5 3.8 3.7 3.7 4.1 
    Multivitamin user (%) 36 41 44 47 39 40 42 45 
HPFS men         
    Age (y) 52 54 55 56 54 54 55 55 
    BMI (kg/m226 26 25 25 26 26 26 26 
    Physical activity (METS/wk)* 15 19 22 27 17 19 21 25 
    Alcohol intake (g/d) 14 12 11 10 11 12 12 
    Animal fat (g/d) 47 43 40 35 44 42 41 38 
    Current smoker (%) 18 11 12 10 
    History of hypertension (%) 22 22 22 22 21 22 22 23 
    History of diabetes (%) 2.4 2.9 3.2 3.8 2.8 2.9 3.3 3.6 
    Multivitamin user (%) 36 40 44 46 40 40 41 45 
Total fruits (servings/d)
Total vegetables (servings/d)
<1/d1-1.9/d2-2.9/d≥3/d<2/d2-2.9/d3-3.9/d≥4/d
NHS women         
    Age (y) 51 52 53 54 52 52 53 54 
    BMI (kg/m225 25 25 25 25 25 25 25 
    Physical activity (METS/wk)* 10 12 14 18 11 13 17 
    Alcohol intake (g/d) 14 12 11 10 11 12 12 
    Animal fat (g/d) 37 34 32 30 35 34 33 31 
    Current smoker (%) 37 24 17 15 31 26 21 18 
    History of hypertension (%) 24 25 26 27 26 25 26 26 
    History of diabetes (%) 2.9 3.8 3.8 4.5 3.8 3.7 3.7 4.1 
    Multivitamin user (%) 36 41 44 47 39 40 42 45 
HPFS men         
    Age (y) 52 54 55 56 54 54 55 55 
    BMI (kg/m226 26 25 25 26 26 26 26 
    Physical activity (METS/wk)* 15 19 22 27 17 19 21 25 
    Alcohol intake (g/d) 14 12 11 10 11 12 12 
    Animal fat (g/d) 47 43 40 35 44 42 41 38 
    Current smoker (%) 18 11 12 10 
    History of hypertension (%) 22 22 22 22 21 22 22 23 
    History of diabetes (%) 2.4 2.9 3.2 3.8 2.8 2.9 3.3 3.6 
    Multivitamin user (%) 36 40 44 46 40 40 41 45 

NOTE: All values except age were standardized according to the age distribution of cohorts. All values are means except where otherwise noted. Animal fat and dietary fiber values represent the mean energy-adjusted intake.

*

METS, metabolic equivalents. One MET is the energy expended by sitting quietly. Each activity was assigned a MET value; e.g., swimming = 7 METS, and running = 12 METS. Total METS per week were calculated by summing all METS per week of each activity.

The consumption of total fruits and vegetables was inversely associated with the risk of renal cell cancer in men, but not in women (Table 2). For men, compared with <3 servings/d, the multivariate RR was 0.45 (95% CI, 0.25-0.81; P test for trend = 0.02) for ≥6 servings/d of fruit and vegetable intake. A similar pattern was observed when fruit and vegetable intake was modeled as a continuous variable; the multivariate RR for an increment of one serving of total fruits and vegetables per day was 0.91 (95% CI, 0.84-0.99) in men. Baseline intake of total fruits and vegetables was also associated with a lower risk of renal cell cancer only in men (multivariate RR for ≥6 servings/d versus <3 servings/d = 0.43; 95% CI, 0.24-0.76; P test for trend = 0.004). In the simple updated analyses, which used intakes closest to but prior to diagnosis, a significant inverse association was observed in men (data not shown). Neither baseline intake nor recent intake (simple update analysis) of total fruits and vegetables was associated with renal cell cancer risk in women (data not shown). Additional adjustment for physical activity (tertiles) and meat intake (<1, 1-1.9, ≥2 servings/d) did not markedly change the results in either men or women (data not shown).

Table 2.

Multivariate RRs for consumption of fruits and vegetables

Food groupCategories (servings)P test for trendP test for between-studies heterogeneity*
Total fruits and vegetables, no. of cases (women; men), <3/d (24; 20) 3-4.9/d (41; 37) 5-5.9/d (18; 17) ≥6/d (49; 42)   
RR (95% CI)       
    NHS women 1.00 0.73 (0.44-1.23) 0.69 (0.36-1.32) 1.17 (0.66-2.07) 0.25  
    HPFS men 1.00 0.61 (0.35-1.07) 0.51 (0.26-1.00) 0.45 (0.25-0.81) 0.02  
    Pooled 1.00 0.67 (0.46-0.98) 0.60 (0.38-0.95) 0.73 (0.28-1.87) 0.76 0.02 
       
Total fruits, no. of cases (women; men) <1/d (23; 19) 1-1.9/d (32; 37) 2-2.9/d (40; 36) ≥3/d (37; 24)   
RR (95% CI)       
    NHS women 1.00 0.54 (0.31-0.94) 0.68 (0.39-1.17) 0.78 (0.43-1.40) 0.94  
    HPFS men 1.00 0.80 (0.45-1.41) 0.76 (0.42-1.35) 0.47 (0.24-0.91) 0.02  
    Pooled 1.00 0.65 (0.44-0.97) 0.71 (0.48-1.06) 0.62 (0.38-1.02) 0.35 0.26 
       
Total vegetables, no. of cases (women; men) <2/d (33; 31) 2-2.9/d (44; 32) 3-3.9/d (37; 27) ≥4/d (18; 26)   
RR (95% CI)       
    NHS women 1.00 1.26 (0.78-2.01) 1.61 (0.94-2.67) 1.17 (0.62-2.20) 0.36  
    HPFS men 1.00 0.73 (0.44-1.22) 0.69 (0.40-1.18) 0.44 (0.25-0.77) 0.005  
    Pooled 1.00 0.97 (0.57-1.65) 1.06 (0.46-2.44) 0.71 (0.27-1.86) 0.63 0.02 
       
Total citrus fruits, no. of cases (women; men) <4/wk (44; 47) 4-6.9/wk (29; 18) 1-1.9/d (51; 44) ≥2/d (8; 7)   
RR (95% CI)       
    NHS women 1.00 0.84 (0.52-1.34) 1.06 (0.70-1.61) 1.00 (0.46-2.18) 0.83  
    HPFS men 1.00 0.59 (0.34-1.03) 0.86 (0.56-1.33) 0.59 (0.26-1.34) 0.26  
    Pooled 1.00 0.72 (0.50-1.04) 0.96 (0.71-1.30) 0.78 (0.44-1.37) 0.51 0.36 
       
Cruciferous vegetables, no. of cases (women; men) <2/wk (39; 54) 2-2.9/wk (43; 25) 3-4.9/wk (33; 18) ≥5/wk (17; 19)   
RR (95% CI)       
    NHS women 1.00 1.55 (1.00-2.40) 1.16 (0.72-1.86) 1.04 (0.58-1.86) 0.92  
    HPFS men 1.00 0.79 (0.49-1.28) 0.48 (0.27-0.83) 0.67 (0.39-1.16) 0.06  
    Pooled 1.00 1.12 (0.58-2.15) 0.75 (0.31-1.79) 0.82 (0.54-1.26) 0.25 0.28 
       
Green leafy vegetables, no. of cases (women; men) <3/wk (42; 36) 3-4.9/wk (34; 35) 5-6.9/wk (29; 25) ≥1/d (27; 20)   
RR (95% CI)       
    NHS women 1.00 1.26 (0.78-2.03) 1.45 (0.88-2.39) 1.49 (0.90-2.48) 0.09  
    HPFS men 1.00 1.24 (0.77-2.00) 1.09 (0.65-1.84) 0.72 (0.41-1.27) 0.23  
    Pooled 1.00 1.25 (0.89-1.75) 1.26 (0.88-1.81) 1.05 (0.51-2.13) 0.87 0.06 
       
Vitamin C–rich fruits and vegetables, no. of cases (women; men) <4/wk (22; 25) 4-6.9/wk (17; 16) 1-1.9/d (69; 52) ≥2/d (24; 23)   
RR (95% CI)       
    NHS women 1.00 0.52 (0.27-0.98) 0.83 (0.50-1.37) 0.78 (0.41-1.45) 0.97  
    HPFS men 1.00 0.48 (0.25-0.90) 0.53 (0.32-0.87) 0.43 (0.24-0.79) 0.04  
    Pooled 1.00 0.49 (0.32-0.78) 0.66 (0.43-1.02) 0.58 (0.32-1.03) 0.32 0.18 
       
Legumes, no. of cases (women; men) <1/wk (47; 47) 1-2.9/wk (72; 53) ≥3/wk (13; 16)    
RR (95% CI)       
    NHS women 1.00 1.27 (0.86-1.86) 1.26 (0.66-2.40)  0.41  
    HPFS men 1.00 0.97 (0.64-1.47) 0.61 (0.34-1.12)  0.11  
    Pooled 1.00 1.12 (0.84-1.48) 0.87 (0.43-1.76)  0.76 0.11 
       
Potatoes, no. of cases (women; men) <2/wk (39; 34) 2-3.9/wk (59; 45) 4-4.9/wk (15; 16) ≥5/wk (19; 21)   
RR (95% CI)       
    NHS women 1.00 1.09 (0.71-1.66) 0.84 (0.45-1.57) 0.73 (0.40-1.34) 0.26  
    HPFS men 1.00 1.00 (0.63-1.58) 0.93 (0.49-1.75) 0.69 (0.38-1.26) 0.22  
    Pooled 1.00 1.05 (0.77-1.43) 0.88 (0.56-1.38) 0.71 (0.46-1.09) 0.09 0.89 
Food groupCategories (servings)P test for trendP test for between-studies heterogeneity*
Total fruits and vegetables, no. of cases (women; men), <3/d (24; 20) 3-4.9/d (41; 37) 5-5.9/d (18; 17) ≥6/d (49; 42)   
RR (95% CI)       
    NHS women 1.00 0.73 (0.44-1.23) 0.69 (0.36-1.32) 1.17 (0.66-2.07) 0.25  
    HPFS men 1.00 0.61 (0.35-1.07) 0.51 (0.26-1.00) 0.45 (0.25-0.81) 0.02  
    Pooled 1.00 0.67 (0.46-0.98) 0.60 (0.38-0.95) 0.73 (0.28-1.87) 0.76 0.02 
       
Total fruits, no. of cases (women; men) <1/d (23; 19) 1-1.9/d (32; 37) 2-2.9/d (40; 36) ≥3/d (37; 24)   
RR (95% CI)       
    NHS women 1.00 0.54 (0.31-0.94) 0.68 (0.39-1.17) 0.78 (0.43-1.40) 0.94  
    HPFS men 1.00 0.80 (0.45-1.41) 0.76 (0.42-1.35) 0.47 (0.24-0.91) 0.02  
    Pooled 1.00 0.65 (0.44-0.97) 0.71 (0.48-1.06) 0.62 (0.38-1.02) 0.35 0.26 
       
Total vegetables, no. of cases (women; men) <2/d (33; 31) 2-2.9/d (44; 32) 3-3.9/d (37; 27) ≥4/d (18; 26)   
RR (95% CI)       
    NHS women 1.00 1.26 (0.78-2.01) 1.61 (0.94-2.67) 1.17 (0.62-2.20) 0.36  
    HPFS men 1.00 0.73 (0.44-1.22) 0.69 (0.40-1.18) 0.44 (0.25-0.77) 0.005  
    Pooled 1.00 0.97 (0.57-1.65) 1.06 (0.46-2.44) 0.71 (0.27-1.86) 0.63 0.02 
       
Total citrus fruits, no. of cases (women; men) <4/wk (44; 47) 4-6.9/wk (29; 18) 1-1.9/d (51; 44) ≥2/d (8; 7)   
RR (95% CI)       
    NHS women 1.00 0.84 (0.52-1.34) 1.06 (0.70-1.61) 1.00 (0.46-2.18) 0.83  
    HPFS men 1.00 0.59 (0.34-1.03) 0.86 (0.56-1.33) 0.59 (0.26-1.34) 0.26  
    Pooled 1.00 0.72 (0.50-1.04) 0.96 (0.71-1.30) 0.78 (0.44-1.37) 0.51 0.36 
       
Cruciferous vegetables, no. of cases (women; men) <2/wk (39; 54) 2-2.9/wk (43; 25) 3-4.9/wk (33; 18) ≥5/wk (17; 19)   
RR (95% CI)       
    NHS women 1.00 1.55 (1.00-2.40) 1.16 (0.72-1.86) 1.04 (0.58-1.86) 0.92  
    HPFS men 1.00 0.79 (0.49-1.28) 0.48 (0.27-0.83) 0.67 (0.39-1.16) 0.06  
    Pooled 1.00 1.12 (0.58-2.15) 0.75 (0.31-1.79) 0.82 (0.54-1.26) 0.25 0.28 
       
Green leafy vegetables, no. of cases (women; men) <3/wk (42; 36) 3-4.9/wk (34; 35) 5-6.9/wk (29; 25) ≥1/d (27; 20)   
RR (95% CI)       
    NHS women 1.00 1.26 (0.78-2.03) 1.45 (0.88-2.39) 1.49 (0.90-2.48) 0.09  
    HPFS men 1.00 1.24 (0.77-2.00) 1.09 (0.65-1.84) 0.72 (0.41-1.27) 0.23  
    Pooled 1.00 1.25 (0.89-1.75) 1.26 (0.88-1.81) 1.05 (0.51-2.13) 0.87 0.06 
       
Vitamin C–rich fruits and vegetables, no. of cases (women; men) <4/wk (22; 25) 4-6.9/wk (17; 16) 1-1.9/d (69; 52) ≥2/d (24; 23)   
RR (95% CI)       
    NHS women 1.00 0.52 (0.27-0.98) 0.83 (0.50-1.37) 0.78 (0.41-1.45) 0.97  
    HPFS men 1.00 0.48 (0.25-0.90) 0.53 (0.32-0.87) 0.43 (0.24-0.79) 0.04  
    Pooled 1.00 0.49 (0.32-0.78) 0.66 (0.43-1.02) 0.58 (0.32-1.03) 0.32 0.18 
       
Legumes, no. of cases (women; men) <1/wk (47; 47) 1-2.9/wk (72; 53) ≥3/wk (13; 16)    
RR (95% CI)       
    NHS women 1.00 1.27 (0.86-1.86) 1.26 (0.66-2.40)  0.41  
    HPFS men 1.00 0.97 (0.64-1.47) 0.61 (0.34-1.12)  0.11  
    Pooled 1.00 1.12 (0.84-1.48) 0.87 (0.43-1.76)  0.76 0.11 
       
Potatoes, no. of cases (women; men) <2/wk (39; 34) 2-3.9/wk (59; 45) 4-4.9/wk (15; 16) ≥5/wk (19; 21)   
RR (95% CI)       
    NHS women 1.00 1.09 (0.71-1.66) 0.84 (0.45-1.57) 0.73 (0.40-1.34) 0.26  
    HPFS men 1.00 1.00 (0.63-1.58) 0.93 (0.49-1.75) 0.69 (0.38-1.26) 0.22  
    Pooled 1.00 1.05 (0.77-1.43) 0.88 (0.56-1.38) 0.71 (0.46-1.09) 0.09 0.89 

NOTE: In the NHS, multivariate RRs were adjusted for BMI (<25, 25-26.9, 27-29.9, or ≥30 kg/m2), history of hypertension (yes, no), parity (continuous), history of diabetes (yes, no), smoking status (never, 1-19, 20-39, ≥40 pack-years), multivitamin use (never, past, current <5 yr, current ≥5 yr of use), alcohol intake (continuous), and total energy intake (continuous). In the HPFS, multivariate RRs were adjusted for BMI (<25, 25-26.9, 27-29.9, or ≥30 kg/m2), history of hypertension (yes, no), smoking status (never, 1-19, 20-39, ≥40 pack-years), multivitamin use (never, past, current <5 yr, current ≥5 yr of use), and total energy intake (continuous).

*

For the highest category.

Number of cases (women; men): a number of cases for women in the NHS and for men in the HPFS.

Potatoes and legumes were excluded from total fruit and vegetable group and total vegetable group.

To address the possibility that the inverse association observed for total fruit and vegetable consumption in men was due to confounding by other healthy lifestyles, we excluded men (58 renal cell cancer cases were excluded in the analysis) who had vigorous physical activity [>36 metabolic equivalents (METS) of physical activity/wk], and never smoked. The association for fruit and vegetable consumption was still inverse, although the confidence intervals were wider based on the smaller sample size. Compared with <3 servings/d, the multivariate RRs were 0.71 (95% CI, 0.32-1.55) for 3 to 4.9 servings/d, 0.39 (95% CI, 0.14-1.11) for 5 to 5.9 servings/d, and 0.57 (95% CI, 0.24-1.31) for ≥6 servings/d (P test for trend = 0.23).

Fruit intake and vegetable intake were individually associated with a lower risk of renal cell cancer in men, but not in women, in analyses using cumulative average intakes (Table 2), baseline intake (data not shown), and most recent intake (data not shown). For men, when we simultaneously adjusted for total fruit and total vegetable intakes, similar patterns were still observed for total fruit intake (multivariate RR for ≥3 serving/d versus <1 serving/d = 0.58; 95% CI, 0.29-1.14; P test for trend = 0.11) and for total vegetable intake (multivariate RR ≥4 serving/d versus <2 servings/d = 0.50; 95% CI, 0.28-0.89; P test for trend = 0.03). We also examined associations with specific fruit and vegetable groups. For men, a significant inverse association was seen only for vitamin C–rich fruits and vegetables; the other groups examined had nonsignificant inverse associations (P test for trend >0.1) and cruciferous vegetables had a borderline significant inverse association (P test for trend = 0.06). No clear associations for specific fruit and vegetable groups were observed in women.

To explore a potential explanation for the different results between women and men, we examined whether reproductive factors such as postmenopausal hormone therapy use (past and current use, never use and premenopausal), number of children (0-2, ≥3 children), and oral contraceptive use (ever user, nonuser) modified the risk of renal cell cancer in women. We did not find any clear association of total fruit and vegetable consumption with renal cell cancer risk in each subgroup, although we had limited statistical power (data not shown). Also, when we limited our analyses to only postmenopausal women, and when we used 1986 as the starting point of follow-up time in the NHS to make the age distribution and SFFQs used in the NHS (86 renal cell cancer cases were included) compatible with those of the HPFS, no clear associations were observed in women (data not shown).

As observed for fruit and vegetable consumption, intakes of dietary and total vitamin A were inversely associated with renal cell cancer risk only in men (Table 3). When we examined total retinol (preformed vitamin A) intake, of which the primary contributors in our cohorts were multivitamins, liver, vitamin A–fortified milk, and cereals, in relation to renal cell cancer risk, no association was found (data not shown). Given our findings and a high correlation between dietary vitamin A and vegetable intake (r ≈ 0.7 across the follow-up period), we cannot rule out the possibility that the inverse association observed for vitamin A in men could be due to other nutrients or bioactive compounds in vegetables that we did not examine.

Table 3.

Multivariate RRs for intakes of vitamins A, C, and E, and carotenoids

Quartile 1Quartile 2Quartile 3Quartile 4P test for trendP test for between-studies heterogeneity*
Total vitamin A       
Median, IU/d (women; men) (6,644; 7,306) (10,210; 11,284) (13,852; 15,872) (20,320; 24,856)   
RR (95% CI)       
    NHS women 1.00 1.03 (0.58-1.81) 1.48 (0.86-2.52) 1.40 (0.80-2.47) 0.17  
    HPFS men 1.00 0.78 (0.47-1.29) 0.82 (0.49-1.35) 0.41 (0.22-0.78) 0.009  
    Pooled 1.00 0.88 (0.60-1.28) 1.09 (0.61-1.95) 0.77 (0.23-2.54) 0.78 0.005 
       
Dietary vitamin A       
Median, IU/d (women; men) (5,852; 6,451) (8,582; 9,368) (11,358; 12,628) (16,134; 18,724)   
RR (95% CI)       
    NHS women 1.00 1.04 (0.61-1.76) 1.12 (0.67-1.87) 1.37 (0.83-2.27) 0.17  
    HPFS men 1.00 1.05 (0.65-1.71) 0.71 (0.42-1.19) 0.47 (0.26-0.84) 0.003  
    Pooled 1.00 1.05 (0.73-1.49) 0.89 (0.57-1.39) 0.81 (0.28-2.33) 0.72 0.006 
       
Total vitamin C       
Median, mg/d (women; men) (96; 114) (154; 190) (242; 328) (633; 907)   
RR (95% CI)       
    NHS women 1.00 1.05 (0.63-1.76) 0.91 (0.53-1.57) 0.89 (0.51-1.56) 0.60  
    HPFS men 1.00 0.75 (0.45-1.26) 0.77 (0.44-1.35) 0.97 (0.55-1.71) 0.64  
    Pooled 1.00 0.89 (0.62-1.29) 0.84 (0.57-1.24) 0.93 (0.62-1.38) 0.95 0.84 
       
Dietary vitamin C       
Median, mg/d (women; men) (79; 91) (116; 137) (148; 177) (196; 243)   
RR (95% CI)       
    NHS women 1.00 0.81 (0.48-1.38) 0.94 (0.57-1.55) 1.20 (0.74-1.95) 0.29  
    HPFS men 1.00 0.58 (0.34-0.98) 0.70 (0.43-1.15) 0.51 (0.30-0.88) 0.03  
    Pooled 1.00 0.69 (0.47-0.99) 0.81 (0.57-1.15) 0.79 (0.35-1.82) 0.76 0.02 
       
Total vitamin E       
Median, mg/d (women; men) (6; 8) (9; 11) (14; 18) (103; 162)   
RR (95% CI)       
    NHS women 1.00 0.94 (0.54-1.65) 1.16 (0.65-2.08) 1.20 (0.66-2.16) 0.55  
    HPFS men 1.00 0.98 (0.59-1.63) 0.65 (0.36-1.19) 0.67 (0.36-1.24) 0.38  
    Pooled 1.00 0.96 (0.66-1.40) 0.88 (0.49-1.55) 0.90 (0.51-1.60) 0.74 0.18 
       
Dietary vitamin E       
Median, mg/d (women; men) (6; 8) (7; 9) (8; 11) (10; 13)   
RR (95% CI)       
    NHS women 1.00 1.46 (0.85-2.49) 1.66 (0.99-2.81) 1.31 (0.75-2.28) 0.41  
    HPFS men 1.00 1.32 (0.78-2.23) 0.88 (0.50-1.54) 0.97 (0.56-1.70) 0.56  
    Pooled 1.00 1.38 (0.95-2.02) 1.22 (0.65-2.28) 1.13 (0.76-1.67) 0.97 0.46 
       
α-Carotene       
Median, μg/d (women; men) (254; 351) (444; 577) (707; 898) (1,327; 1,668)   
RR (95% CI)       
    NHS women 1.00 0.82 (0.49-1.38) 1.23 (0.77-1.96) 0.90 (0.54-1.49) 0.89  
    HPFS men 1.00 1.36 (0.83-2.22) 0.99 (0.59-1.67) 0.53 (0.29-0.98) 0.007  
    Pooled 1.00 1.07 (0.65-1.74) 1.12 (0.79-1.58) 0.71 (0.42-1.19) 0.26 0.19 
       
β-Carotene       
Median, μg/d (women; men) (1,963; 2,371) (3,244; 2,718) (4,678; 5,293) (7,242; 8,518)   
RR (95% CI)       
    NHS women 1.00 1.08 (0.63-1.84) 1.38 (0.83-2.29) 1.24 (0.74-2.09) 0.36  
    HPFS men 1.00 0.91 (0.55-1.50) 0.81 (0.49-1.36) 0.57 (0.32-0.99) 0.04  
    Pooled 1.00 0.98 (0.68-1.42) 1.06 (0.63-1.78) 0.84 (0.39-1.82) 0.70 0.04 
       
β-Cryptoxanthin       
Median, μg/d (women; men) (24; 21) (53; 52) (86; 94) (152; 179)   
RR (95% CI)       
    NHS women 1.00 0.87 (0.52-1.47) 1.11 (0.68-1.82) 1.02 (0.61-1.69) 0.76  
    HPFS men 1.00 0.84 (0.52-1.38) 0.72 (0.44-1.19) 0.48 (0.27-0.84) 0.008  
    Pooled 1.00 0.86 (0.60-1.23) 0.90 (0.59-1.37) 0.70 (0.34-1.47) 0.43 0.05 
       
Lutein/zeaxanthin       
Median, μg/d (women; men) (1,552; 1,523) (2,623; 2,629) (4,111; 3,794) (7,248; 6,044)   
RR (95% CI)       
    NHS women 1.00 1.22 (0.71-2.08) 1.54 (0.93-2.57) 1.36 (0.81-2.29) 0.30  
    HPFS men 1.00 0.77 (0.46-1.27) 0.92 (0.57-1.48) 0.46 (0.26-0.82) 0.02  
    Pooled 1.00 0.96 (0.61-1.51) 1.18 (0.70-1.97) 0.80 (0.28-2.30) 0.62 0.006 
       
Lycopene§       
Median, μg/d (women; men) (3,668; 4,192) (5,936; 7,082) (8,308; 10,220) (12,296; 16,180)   
RR (95% CI)       
    NHS women 1.00 0.90 (0.51-1.57) 0.88 (0.51-1.54) 0.90 (0.51-1.57) 0.76  
    HPFS men 1.00 0.90 (0.55-1.49) 0.77 (0.46-1.29) 0.71 (0.42-1.20) 0.18  
    Pooled 1.00 0.90 (0.62-1.31) 0.82 (0.56-1.20) 0.79 (0.54-1.16) 0.20 0.56 
Quartile 1Quartile 2Quartile 3Quartile 4P test for trendP test for between-studies heterogeneity*
Total vitamin A       
Median, IU/d (women; men) (6,644; 7,306) (10,210; 11,284) (13,852; 15,872) (20,320; 24,856)   
RR (95% CI)       
    NHS women 1.00 1.03 (0.58-1.81) 1.48 (0.86-2.52) 1.40 (0.80-2.47) 0.17  
    HPFS men 1.00 0.78 (0.47-1.29) 0.82 (0.49-1.35) 0.41 (0.22-0.78) 0.009  
    Pooled 1.00 0.88 (0.60-1.28) 1.09 (0.61-1.95) 0.77 (0.23-2.54) 0.78 0.005 
       
Dietary vitamin A       
Median, IU/d (women; men) (5,852; 6,451) (8,582; 9,368) (11,358; 12,628) (16,134; 18,724)   
RR (95% CI)       
    NHS women 1.00 1.04 (0.61-1.76) 1.12 (0.67-1.87) 1.37 (0.83-2.27) 0.17  
    HPFS men 1.00 1.05 (0.65-1.71) 0.71 (0.42-1.19) 0.47 (0.26-0.84) 0.003  
    Pooled 1.00 1.05 (0.73-1.49) 0.89 (0.57-1.39) 0.81 (0.28-2.33) 0.72 0.006 
       
Total vitamin C       
Median, mg/d (women; men) (96; 114) (154; 190) (242; 328) (633; 907)   
RR (95% CI)       
    NHS women 1.00 1.05 (0.63-1.76) 0.91 (0.53-1.57) 0.89 (0.51-1.56) 0.60  
    HPFS men 1.00 0.75 (0.45-1.26) 0.77 (0.44-1.35) 0.97 (0.55-1.71) 0.64  
    Pooled 1.00 0.89 (0.62-1.29) 0.84 (0.57-1.24) 0.93 (0.62-1.38) 0.95 0.84 
       
Dietary vitamin C       
Median, mg/d (women; men) (79; 91) (116; 137) (148; 177) (196; 243)   
RR (95% CI)       
    NHS women 1.00 0.81 (0.48-1.38) 0.94 (0.57-1.55) 1.20 (0.74-1.95) 0.29  
    HPFS men 1.00 0.58 (0.34-0.98) 0.70 (0.43-1.15) 0.51 (0.30-0.88) 0.03  
    Pooled 1.00 0.69 (0.47-0.99) 0.81 (0.57-1.15) 0.79 (0.35-1.82) 0.76 0.02 
       
Total vitamin E       
Median, mg/d (women; men) (6; 8) (9; 11) (14; 18) (103; 162)   
RR (95% CI)       
    NHS women 1.00 0.94 (0.54-1.65) 1.16 (0.65-2.08) 1.20 (0.66-2.16) 0.55  
    HPFS men 1.00 0.98 (0.59-1.63) 0.65 (0.36-1.19) 0.67 (0.36-1.24) 0.38  
    Pooled 1.00 0.96 (0.66-1.40) 0.88 (0.49-1.55) 0.90 (0.51-1.60) 0.74 0.18 
       
Dietary vitamin E       
Median, mg/d (women; men) (6; 8) (7; 9) (8; 11) (10; 13)   
RR (95% CI)       
    NHS women 1.00 1.46 (0.85-2.49) 1.66 (0.99-2.81) 1.31 (0.75-2.28) 0.41  
    HPFS men 1.00 1.32 (0.78-2.23) 0.88 (0.50-1.54) 0.97 (0.56-1.70) 0.56  
    Pooled 1.00 1.38 (0.95-2.02) 1.22 (0.65-2.28) 1.13 (0.76-1.67) 0.97 0.46 
       
α-Carotene       
Median, μg/d (women; men) (254; 351) (444; 577) (707; 898) (1,327; 1,668)   
RR (95% CI)       
    NHS women 1.00 0.82 (0.49-1.38) 1.23 (0.77-1.96) 0.90 (0.54-1.49) 0.89  
    HPFS men 1.00 1.36 (0.83-2.22) 0.99 (0.59-1.67) 0.53 (0.29-0.98) 0.007  
    Pooled 1.00 1.07 (0.65-1.74) 1.12 (0.79-1.58) 0.71 (0.42-1.19) 0.26 0.19 
       
β-Carotene       
Median, μg/d (women; men) (1,963; 2,371) (3,244; 2,718) (4,678; 5,293) (7,242; 8,518)   
RR (95% CI)       
    NHS women 1.00 1.08 (0.63-1.84) 1.38 (0.83-2.29) 1.24 (0.74-2.09) 0.36  
    HPFS men 1.00 0.91 (0.55-1.50) 0.81 (0.49-1.36) 0.57 (0.32-0.99) 0.04  
    Pooled 1.00 0.98 (0.68-1.42) 1.06 (0.63-1.78) 0.84 (0.39-1.82) 0.70 0.04 
       
β-Cryptoxanthin       
Median, μg/d (women; men) (24; 21) (53; 52) (86; 94) (152; 179)   
RR (95% CI)       
    NHS women 1.00 0.87 (0.52-1.47) 1.11 (0.68-1.82) 1.02 (0.61-1.69) 0.76  
    HPFS men 1.00 0.84 (0.52-1.38) 0.72 (0.44-1.19) 0.48 (0.27-0.84) 0.008  
    Pooled 1.00 0.86 (0.60-1.23) 0.90 (0.59-1.37) 0.70 (0.34-1.47) 0.43 0.05 
       
Lutein/zeaxanthin       
Median, μg/d (women; men) (1,552; 1,523) (2,623; 2,629) (4,111; 3,794) (7,248; 6,044)   
RR (95% CI)       
    NHS women 1.00 1.22 (0.71-2.08) 1.54 (0.93-2.57) 1.36 (0.81-2.29) 0.30  
    HPFS men 1.00 0.77 (0.46-1.27) 0.92 (0.57-1.48) 0.46 (0.26-0.82) 0.02  
    Pooled 1.00 0.96 (0.61-1.51) 1.18 (0.70-1.97) 0.80 (0.28-2.30) 0.62 0.006 
       
Lycopene§       
Median, μg/d (women; men) (3,668; 4,192) (5,936; 7,082) (8,308; 10,220) (12,296; 16,180)   
RR (95% CI)       
    NHS women 1.00 0.90 (0.51-1.57) 0.88 (0.51-1.54) 0.90 (0.51-1.57) 0.76  
    HPFS men 1.00 0.90 (0.55-1.49) 0.77 (0.46-1.29) 0.71 (0.42-1.20) 0.18  
    Pooled 1.00 0.90 (0.62-1.31) 0.82 (0.56-1.20) 0.79 (0.54-1.16) 0.20 0.56 

NOTE: Multivariate RRs were adjusted for the same covariates listed in Table 2.

*

For the highest category.

From food and supplement.

Median (women; men): median for quartile from women in the NHS and median for quartile from men in the HPFS.

§

In the NHS, the analysis for lycopene was started in 1984, and 106 cases were included because the main food sources were not assessed in 1980.

Dietary vitamin C intake from food only was associated with a lower risk of renal cell cancer only in men, but total vitamin C intake (from food and supplements) was not significantly associated with the risk of renal cell cancer in either men or women. Current vitamin C supplement use was not associated with the risk of renal cell cancer; compared with nonusers, the multivariate RRs for vitamin C supplemental users were 1.09 (95% CI, 0.68-1.73) in men and 0.82 (95% CI, 0.54-1.24) in women.

We did not find an association between dietary or total vitamin E intake and the risk of renal cell cancer in either women or men. When examining vitamin E supplements alone, we found no clear associations in either men (multivariate RR for current user versus nonuser, 0.80; 95% CI, 0.49-1.30) or women (multivariate RR for current user versus nonuser, 1.21; 95% CI, 0.80-1.86).

We found that individual carotenoids except lycopene were significantly associated with a lower risk of renal cell cancer in men only (Table 3). The proportion of β-carotene supplemental users was low (average of 3.1% in women and 7.0% in men across the follow-up periods) and the main sources of carotenoids were fruits and vegetables (r ≈ 0.7 with fruits for β-cryptoxanthin; r = 0.6-0.8 with vegetables for α-carotene, β-carotene, lycopene, and lutein/zeaxanthin).

When we examined the most recent intake (closest to but prior to diagnosis for renal cell cancer cases) or baseline intake, we found similar associations as observed for the cumulative average intakes (data not shown). We simultaneously modeled total carotenes (including α-carotene, β-carotene, and β-cryptoxanthin) and dietary vitamin C as continuous variables, and found that the intake of total carotenes was significantly associated with a lower risk of renal cell cancer (for an increment of 4,000 IU/d, multivariate RR, 0.83; 95% CI, 0.71-0.97), but dietary vitamin C was not (for an increment of 100 mg/d, multivariate RR, 0.92; 95% CI, 0.68-1.24).

To examine whether carotenoids and vitamin C accounted for the inverse association of fruit and vegetable consumption in men, we adjusted for total carotenes (continuous), lutein/zeaxanthin, and vitamin C (continuous), although they were correlated with fruit and vegetables. In these analyses, the inverse association for total fruit and vegetable consumption was still observed (multivariate RR comparing ≥6 servings/d versus <3 servings/d = 0.66; 95% CI, 0.32-1.36; P test for trend = 0.31).

We also examined whether the associations between fruit and vegetable intakes and renal cell cancer risk varied by multivitamin use (ever users, never users), smoking status (never, ever), BMI (<25, ≥25 kg/m2), history of hypertension (yes, no), and age (<65, ≥65 years; Table 4). For men, the inverse association was more pronounced in men who were older, who had never smoked, and who had ever used multivitamins.

Table 4.

Multivariate RRs for one increment of serving of fruits and vegetables per day by multivitamin use, smoking status, BMI, history of hypertension, and age

Multivitamin use
Smoking status
BMI (kg/m2)
History of hypertension
Age (y)
EverNeverNeverEver<25≥25NoYes<65≥65
Person-years           
    Women 1,035,515 672,658 784,466 923,707 927,832 780,340 1,206,382 501,790 1,437,421 270,752 
    Men 411,082 197,072 309,200 298,980 264,044 344,136 428,390 179,790 262,317 59,155 
No. of cases           
    Women 93 39 53 79 59 73 63 69 88 44 
    Men 70 46 45 71 43 73 58 58 78 38 
Total fruits and vegetables           
    Women 1.06 (0.96-1.18) 1.14 (0.97-1.33) 1.08 (0.94-1.23) 1.09 (0.97-1.22) 1.16 (1.03-1.30) 1.01 (0.89-1.14) 1.14 (1.02-1.27) 1.01 (0.88-1.15) 1.11 (1.00-1.23) 1.02 (0.88-1.19) 
    Men 0.84* (0.74-0.94) 1.01* (0.91-1.13) 0.82* (0.71-0.95) 0.96* (0.87-1.06) 0.96 (0.84-1.09) 0.89 (0.80-0.99) 0.95 (0.85-1.06) 0.87 (0.77-0.99) 0.96* (0.88-1.05) 0.80* (0.68-0.94) 
    Pooled 0.94 (0.75-1.17) 1.07 (0.94-1.23) 0.94 (0.72-1.22) 1.02 (0.90-1.14) 1.05 (0.87-1.27) 0.94 (0.83-1.06) 1.04 (0.87-1.24) 0.94 (0.82-1.07) 1.03 (0.89-1.18) 0.90 (0.71-1.15) 
Multivitamin use
Smoking status
BMI (kg/m2)
History of hypertension
Age (y)
EverNeverNeverEver<25≥25NoYes<65≥65
Person-years           
    Women 1,035,515 672,658 784,466 923,707 927,832 780,340 1,206,382 501,790 1,437,421 270,752 
    Men 411,082 197,072 309,200 298,980 264,044 344,136 428,390 179,790 262,317 59,155 
No. of cases           
    Women 93 39 53 79 59 73 63 69 88 44 
    Men 70 46 45 71 43 73 58 58 78 38 
Total fruits and vegetables           
    Women 1.06 (0.96-1.18) 1.14 (0.97-1.33) 1.08 (0.94-1.23) 1.09 (0.97-1.22) 1.16 (1.03-1.30) 1.01 (0.89-1.14) 1.14 (1.02-1.27) 1.01 (0.88-1.15) 1.11 (1.00-1.23) 1.02 (0.88-1.19) 
    Men 0.84* (0.74-0.94) 1.01* (0.91-1.13) 0.82* (0.71-0.95) 0.96* (0.87-1.06) 0.96 (0.84-1.09) 0.89 (0.80-0.99) 0.95 (0.85-1.06) 0.87 (0.77-0.99) 0.96* (0.88-1.05) 0.80* (0.68-0.94) 
    Pooled 0.94 (0.75-1.17) 1.07 (0.94-1.23) 0.94 (0.72-1.22) 1.02 (0.90-1.14) 1.05 (0.87-1.27) 0.94 (0.83-1.06) 1.04 (0.87-1.24) 0.94 (0.82-1.07) 1.03 (0.89-1.18) 0.90 (0.71-1.15) 

NOTE: Multivariate RRs were adjusted for the same covariates listed in Table 2.

*

P < 0.05 for interaction by stratification.

We found that the consumption of fruits and vegetables was associated with a lower risk of renal cell cancer in men. In contrast, a clear association was not observed in women. Similarly, the inverse associations for vitamins A and C from food and individual carotenoids were limited to men, which may be due to the effects of fruits and vegetables. However, no clear association was observed for vitamin E intake.

A statistically significant inverse association (12, 35-37) or nonsignificant inverse association (10) between total vegetable intake and renal cell cancer risk has been observed in some, but not all (38, 39), case-control studies. Although the results from case-control studies regarding fruit intake have been inconclusive (10, 12, 35-39), case-control studies which included >1,000 RCC cases showed a statistically significant or nonsignificant inverse association for total fruits (10), total vegetables (10), and subgroups of vegetables (11, 12), corresponding with our results.

Five prospective cohort studies (6-8, 14, 40) have examined the associations between total fruit and vegetable consumption. One U.S. prospective study (40) that included 14 cases of RCC found a RR of 0.21 (95% CI, 0.05-1.45) for three or more times a week compared with less than three times a week of fruit intake, and a RR of 0.34 (95% CI, 0.11-1.10) for the same contrast of intake of green salads. The Iowa Women's Health Study (14) reported an age-adjusted RR of 1.56 (95% CI, 0.83-2.92) for >42 servings/wk of fruits and vegetables compared with <28 servings/wk. In the same study with a longer follow-up, there was no association for total fruit and vegetable consumption (but RRs were not reported; ref. 5). The European Prospective Investigation into Cancer and Nutrition study found no association (RR, 1.02; 95% CI, 0.93-1.11) for an increment of 80 g/d of fruit and vegetable consumption (7), and nonsignificant inverse associations for total fruits and vegetables were observed in the Netherlands Cohort Study (RR, 0.78; 95% CI, 0.50-1.21; ref. 6) and in the Swedish Mammography Cohort (RR, 0.59; 95% CI, 0.26-1.34; ref. 8). The inconsistency of our findings with the other studies regarding fruit and vegetable consumption could be due to chance, follow-up time, different types of vegetables, the range of intakes of vegetables, different food frequency questionnaires, and different lifestyle factors such as multivitamin use, smoking habits, or postmenopausal hormone therapy.

However, Swedish women with ≥12 servings of vegetables per month (3 servings/wk) had a lower risk of renal cell cancer compared with those with ≤11 servings per month (8). In the European Prospective Investigation into Cancer and Nutrition study, the cubic spline regression curve showed an increased risk of renal cell cancer at very low vegetable consumption levels (90-130 g/d; ∼1 serving/d).

There are individual fruits and vegetables that have been associated with renal cell cancer risk in these studies. Intake of root vegetables (e.g., carrots, beets) was associated with a lower risk of renal cell cancer in the European Prospective Investigation into Cancer and Nutrition study (7) and the Swedish Mammography Cohort (8), and banana consumption decreased renal cell cancer risk in the Swedish Mammography Cohort (8) and the Netherlands Cohort Study (6). In agreement with other studies, a significant inverse association for carrot intake and a borderline significant inverse association for banana intake were observed in men in our data (data not shown).

We found that the association for vegetables was more pronounced among men who used multivitamins or did not smoke. In three case-control studies (10-12) examining interactions between vegetable consumption and smoking habits, two of these studies found that the inverse association was stronger among nonsmokers (10, 11).

Because the inverse associations for total fruits and vegetables in men were consistently observed in our data after excluding men who had healthy lifestyles and after controlling for physical activity and meat consumption, it is unlikely that residual confounding by healthy lifestyles explains our findings. In addition, whereas men with greater total vitamin C intake were also more likely to have healthy lifestyles, we did not find any clear protective effect of total vitamin C intake.

We did not find clear associations for fruit and vegetable consumption in women. There are several possible explanations for the lack of a clear association. First, it could be a chance finding. Secondly, it is possible that the effects of fruit intake and vegetable intake are modified by hormones such as estrogen and progesterone. More than twice as many men develop renal cell cancer than do women, which suggests the possibility of a gender difference. The difference in magnitude or direction of association by gender varied across studies. The majority of studies adjusted for gender or combined both genders when examining fruit and vegetable intake in relation to renal cell cancer. One case-control study found an inverse association for the intake of vegetables only in women (12), but another study found an inverse association for the intake of fruits and vegetables only in men (35). There is little prior research to support an interaction between diet and hormones. In our data, the association did not vary by postmenopausal hormone therapy use, parity, or oral contraceptive use, although we had limited statistical power in these subgroup analyses. The joint effect of diet and hormones merits further study. Third, fruit and vegetable intakes may have been too high in our female study participants to observe an effect if only low vegetable intake is associated with an increased risk of renal cell cancer in women. We were not able to examine the effect of <1 serving/d of vegetables due to the limited number of participants with low intake.

Furthermore, random misclassification, which is inherent when collecting dietary information, may have obscured an association if the true effect in women is not as strong as that in men. However, because an inverse association between fruit intake and ischemic stroke was found in our female cohort (multivariate RR, 0.69; 95% CI, 0.49-0.98 for the highest quintile versus the lowest; ref. 16) and because the use of repeated assessments of diet reduces measurement error (41), the lack of association observed in women is not likely to be entirely due to random misclassification.

Although our findings in the subgroup analyses are not conclusive due to limited power, fruit intake or vegetable intake may be more beneficial to those who are older and overweight, or to those with renal cell cancer risk factors such as high blood pressure and insulin resistance (2).

Our data suggests that preformed vitamin A, or vitamins C and E are not responsible for the reduction in renal cell cancer risk in men. Because we were not able to separate the effect of each carotenoid from the effects of fruits and vegetables due to high correlations between carotenoids and fruits and vegetables, we cannot exclude the possibility of independent effects of α-carotene, β-carotene, β-cryptoxanthin, and lutein/zeaxanthin on renal cell cancer risk. Information on antioxidants in relation to renal cell cancer risk from prospective studies is sparse. The Iowa Women's Health Study (5) found that vitamin C (from both food and supplements) was positively associated with renal cell cancer risk, but vitamin E (from both food and supplements) was inversely associated with renal cell cancer risk. In the same study with a shorter follow-up period, vitamin A and carotene were not associated with risk (14). The majority of case-control studies have found nonsignificant inverse associations for vitamin C (10-12), and β-carotene (10, 12, 13). The associations from only a few case-control studies for vitamin E (12, 42), β-carotene (11), and vitamin C (among nonsmokers; ref. 10) reached statistical significance.

The strengths of this study include a prospective design with a long and high rate of follow-up. Therefore, we minimized the potential for information and selection bias. Repeated measurements of nondietary and dietary factors allowed for good control for confounding and for assessments of long-term and recent intake. In addition, we were able to reduce the measurement error. Because we had collected comprehensive information on supplement use, we were able to examine whether each of these vitamins had an effect on renal cell cancer risk independent of fruits and vegetables. We had a limited number of cases to examine whether these associations were modified by reproductive factors. Because of the limited number of participants who had low intakes of fruits and vegetables in our study, we could not examine the effect of very low intake.

In conclusion, we found that the consumption of fruits and vegetables was inversely associated with the risk of renal cell cancer in men, but not in women. Carotenoids may partly contribute to the inverse associations observed for fruit and vegetable consumption in relation to renal cell cancer risk, but components other than carotenoids in fruit and vegetables should be considered.

Grant support: Nurses' Health Study grant (CA87969) and Health Professionals Follow-up Study grant (CA55075) from the NIH.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Mathew A, Devesa SS, Fraumeni JF, Jr., Chow WH. Global increases in kidney cancer incidence, 1973–1992.
Eur J Cancer Prev
2002
;
11
:
171
–8.
2
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.
3
Bergstrom A, Hsieh CC, Lindblad P, et al. Obesity and renal cell cancer—a quantitative review.
Br J Cancer
2001
;
85
:
984
–90.
4
IARC. Fruit and Vegetables. In: Vainio H, Bianchini F, editors. IARC handbooks of cancer prevention. Lyon (France): IARC Press; 2003. p. 98–237.
5
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.
6
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.
7
Weikert S, Boeing H, Pischon T, et al. Fruits and vegetables and renal cell carcinoma: findings from the European Prospective Investigation into Cancer and Nutrition (EPIC).
Int J Cancer
2006
;
118
:
3133
–9.
8
Rashidkhani B, Lindblad P, Wolk A. Fruits, vegetables and risk of renal cell carcinoma: a prospective study of Swedish women.
Int J Cancer
2005
;
113
:
451
–5.
9
McLaughlin JK, Mandel JS, Blot WJ, et al. A population-based case-control study of renal cell carcinoma.
J Natl Cancer Inst
1984
;
72
:
275
–84.
10
Wolk A, Gridley G, Niwa S, et al. International renal cell cancer study. VII. Role of diet.
Int J Cancer
1996
;
65
:
67
–73.
11
Yuan JM, Gago-Dominguez M, Castelao JE, et al. Cruciferous vegetables in relation to renal cell carcinoma.
Int J Cancer
1998
;
77
:
211
–6.
12
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.
13
Maclure M, Willett W. A case-control study of diet and risk of renal adenocarcinoma.
Epidemiology
1990
;
1
:
430
–40.
14
Prineas RJ, Folsom AR, Zhang ZM, Sellers TA, Potter J. Nutrition and other risk factors for renal cell carcinoma in postmenopausal women.
Epidemiology
1997
;
8
:
31
–6.
15
Smith SA, Campbell DR, Elmer PJ, et al. The University of Minnesota Cancer Prevention Research Unit vegetable and fruit classification scheme (United States).
Cancer Causes Control
1995
;
6
:
292
–302.
16
Joshipura KJ, Ascherio A, Manson JE, et al. Fruit and vegetable intake in relation to risk of ischemic stroke.
J Am Med Assoc
1999
;
282
:
1233
–9.
17
Lee JE, Giovannucci E, Smith-Warner SA, et al. 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.
18
Pennington JAT. Bowes and Church's food values of portions commonly used. New York: Lippincott-Raven; 1998.
19
U.S. Department of Agriculture, Agriculture Research Service. USDA Nutrient Data Laboratory. USDA Nutrient Database for Standard Reference, release 13. Washington (DC): Department of Agriculture; 1999.
20
Salvini S, Hunter DJ, Sampson L, et al. Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption.
Int J Epidemiol
1989
;
18
:
858
–67.
21
Feskanich D, Rimm EB, Giovannucci EL, et al. Reproducibility and validity of food intake measurements from a semiquantitative food frequency questionnaire.
J Am Diet Assoc
1993
;
93
:
790
–6.
22
Rosner B, Willett WC. Interval estimates for correlation coefficients corrected for within-person variation: implications for study design and hypothesis testing.
Am J Epidemiol
1988
;
127
:
377
–86.
23
Willett WC, Sampson L, Stampfer MJ, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire.
Am J Epidemiol
1985
;
122
:
51
–65.
24
Rimm EB, Giovannucci EL, Stampfer MJ, et al. Reproducibility and validity of a expanded self-administered semiquantitative food frequency questionnaire among male health professionals.
Am J Epidemiol
1992
;
135
:
1114
–26.
25
Michaud DS, Giovannucci EL, Ascherio A, et al. Associations of plasma carotenoid concentrations and dietary intake of specific carotenoids in samples of two prospective cohort studies using a new carotenoid database.
Cancer Epidemiol Biomarkers Prev
1998
;
7
:
283
–90.
26
Stampfer MJ, Willett WC, Speizer FE, et al. Test of the National Death Index.
Am J Epidemiol
1984
;
119
:
837
–9.
27
Storkel S, Eble JN, Adlakha K, et al. Classification of renal cell carcinoma: Workgroup no. 1.
Cancer
1997
;
80
:
987
–9.
28
Willett W, Stampfer M. Implications of total energy intake for epidemiologic analysis. In: Willett W, editor. Nutritional epidemiology. New York: Oxford University Press; 1998. p. 273301.
29
Cox DR. Regression models and life-tables.
J R Stat Soc
1972
;
34
:
187
–220.
30
SAS Institute, Inc., SAS/STAT Software: The PHREG Procedure. Changes and enhancements, release 8.1. Cary (NC): SAS Institute Inc.; 2000.
31
Laird NM, Ware JH. Random-effects models for longitudinal data.
Biometrics
1982
;
38
:
963
–74.
32
Harville DA. Maximum likelihood approaches to variance component estimation and to related problems.
J Am Stat Assoc
1977
;
72
:
320
–40.
33
DerSimonian R, Laird N. Meta-analysis in clinical trials.
Control Clin Trials
1986
;
7
:
177
–88.
34
Cochran WG. The combination of estimates from different experiments.
Biometrics
1954
;
10
:
101
–29.
35
McLaughlin JK, Gao YT, Gao RN, et al. Risk factors for renal-cell cancer in Shanghai, China.
Int J Cancer
1992
;
52
:
562
–5.
36
Negri E, La Vecchia C, Franceschi S, D'Avanzo B, Parazzini F. Vegetable and fruit consumption and cancer risk.
Int J Cancer
1991
;
48
:
350
–4.
37
De Stefani E, Fierro L, Mendilaharsu M, et al. Meat intake, ‘mate’ drinking and renal cell cancer in Uruguay: a case-control study.
Br J Cancer
1998
;
78
:
1239
–43.
38
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.
39
Talamini R, Baron AE, Barra S, et al. A case-control study of risk factor for renal cell cancer in northern Italy.
Cancer Causes Control
1990
;
1
:
125
–31.
40
Fraser GE, Phillips RL, Beeson WL. Hypertension, antihypertensive medication and risk of renal carcinoma in California Seventh-Day Adventists.
Int J Epidemiol
1990
;
19
:
832
–8.
41
Hu FB, Stampfer MJ, Rimm E, et al. Dietary fat and coronary heart disease: a comparison of approaches for adjusting total energy intake and modeling repeated dietary measurements.
Am J Epidemiol
1999
;
149
:
531
–40.
42
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.