Bladder cancer is one of the cancers associated with

Protective Effects of Plasma Carotenoids on the Risk of Bladder Cancer Rayjean J. Hung, Zuo-Feng Zhang, Jian Yu Rao, Allan Pantuck, Victor E. Reuter, David Heber and Qing-Yi Lu* From the Departments of Epidemiology (RJH, ZFZ), Pathology and Laboratory Medicine (JYR) and Urology (AP), and Center for Human Nutrition, Department of Medicine (DH, QYL), University of California-Los Angeles, Los Angeles, California, Department of Pathology, Memorial Sloan-Kettering Cancer Center (VER), New York, New York, and International Agency for Research on Cancer (RJH), Lyon, France Purpose: We examined the associations between plasma micronutrients and bladder cancer risk, and evaluated the combined effects of carotenoid and cigarette smoke. Materials and Methods: We performed a case-control study in 242 patients with bladder cancer and 204 healthy controls at Memorial Sloan-Kettering Cancer Center from 1993 to 1997. Epidemiological data and blood specimens were collected on 84 cases and 173 controls. Plasma micronutrients, including lutein, zeaxanthin, -cryptoxanthin, lycopene, -carotene, -carotene, retinol, -tocopherol and -tocopherol, were determined by high performance liquid chromatography. The logistic regression model was used to estimate the effects from carotenoid, tocopherol and retinol on the risk of bladder cancer. Results: Based on quartiles of plasma micronutrient levels and continuous variables, adjusted ORs were estimated for bladder cancer after controlling for potential confounders, including patient age, sex, education and pack-years of smoking. When using plasma levels of micronutrients as continuous variables, the adjusted OR was 0.22 (95% CI 0.05 to 0.92) for -carotene, 0.42 (95% CI 0.18 to 1.00) for lutein, 0.16 (95% CI 0.02 to 1.06) for zeaxanthin, 0.94 (95% CI 0.89 to 0.99) for lycopene and 0.90 (95% CI 0.81 to 1.00) for -cryptoxanthin. The adjusted OR for the joint effect of plasma carotenoids and tobacco smoking was 6.22 (95% CI 1.87 to 20.8) in smokers with lower lutein and 5.18 (95% CI 1.57 to 17.1) in smokers with lower zeaxanthin. Conclusions: Our results show protective effects of carotenoids on bladder cancer. They suggest that bladder cancer may be a preventable disease through nutritional intervention, especially in smokers. Key Words: bladder, bladder neoplasms, carotenoids, smoking, protective agents Submitted for publication September 26, 2005. Study received approval from the MSKCC Institutional Research Board on Human Subjects. Supported by National Institutes of Health, National Institute of Environmental Health Sciences, National Cancer Institute, Department of Health and Human Services Grants ES06718, ES011667, CA77954, CA09142, CA16042, CA 42710 and CA96116. * Correspondence and requests for reprints: University of California-Los Angeles Center for Human Nutrition, 900 Veteran Ave., Room 14-165, Box 951742, Los Angeles, California 90095-1742 (telephone: 310-825-3126; FAX: 310-206-5264; e-mail: qlu@mednet.ucla. edu). See Editorial on page 1192. Bladder cancer is one of the cancers associated with industrialization and tobacco smoking. It was estimated that 61,420 new cases would be diagnosed and 13,060 deaths would occur in the United States in 2006. 1 Two major risk factors of bladder cancer, that is tobacco smoking and occupational exposure, have been widely accepted for 4 decades. Tobacco smoking contributes to 25% to 60% of bladder cases. Chemicals related to occupational exposure and bladder cancer risk include the aromatic amines and amides, such as 2-naphthylamine, aniline and 4-aminobiphenyl. A number of epidemiological studies have shown variations in bladder cancer risk in different populations with a similar prevalence of exposure to these known causes. 2 Dietary intake of fruits and vegetables has been associated with a decreased risk of cancer at several sites in laboratory, animal and epidemiological studies. However, results in laboratory and observational epidemiological studies of the association between vegetable/fruit intake and bladder cancer risk have been mixed. 3 7 Interventional studies and clinical trials have also shown conflicting results. 8 10 Despite the conflicting reports, multiple studies, including laboratory and animal, 11 epidemiological, 3,6 and clinical and intervention 10 studies, have demonstrated an inverse association of vitamin A derivatives and bladder cancer. A few groups have assessed the risk of bladder cancer and serum carotenoids, retinol and vitamin E, of which most had a relatively small sample size. 12,13 One group reported lower serum lycopene and -tocopherol in cases compared to controls. 13 A nested case-control study showed statistically significant inverse linear trends in risk for serum -carotene, -carotene, lutein plus zeaxanthin, -cryptoxanthin and total carotenoids. However, after adjusting for cigarette smoking none of the inverse trends remained significant. 14 In this hospital based case-control study we determined plasma micronutrients in 84 patients with bladder cancer 0022-5347/06/1763-1192/0 1192 Vol. 176, 1192-1197, September 2006 THE JOURNAL OF UROLOGY Printed in U.S.A. Copyright 2006 by AMERICAN UROLOGICAL ASSOCIATION DOI:10.1016/j.juro.2006.04.030

PROTECTIVE EFFECTS OF PLASMA CAROTENOIDS ON BLADDER CANCER RISK 1193 and 173 healthy controls. We evaluated the possible role of carotenoids, retinol and tocopherols on the risk of bladder cancer when controlling for potential confounding factors, including patient age, sex, education and pack-years of smoking. We also explored potential joint effects of plasma carotenoid and tobacco smoking on bladder cancer risk. MATERIALS AND METHODS Study Population A hospital based case-control study was done at MSKCC from July 1, 1993 to June 30, 1996. The study was approved by the Institutional Research Board on Human Subjects at MSKCC and all participants provided informed consent. Eligible cases were patients seen at MSKCC with a pathologically confirmed diagnosis of bladder cancer. Cases were newly diagnosed or undergoing bladder surgery. We approached 242 consecutive patients with bladder cancer and 229 consented to be interviewed. Because the study was originally focused on tissue markers, only 90 cases donated blood samples and 84 had adequate plasma samples available for nutritional analyses. Approximately 20 ml blood samples were collected at pre-admission from patients to be admitted to the MSKCC urology service. A research nurse interviewed patients after they were admitted to the medical/surgical ward. The process from obtaining informed consent to face-to-face interview required between 2 and 4 days. Pathology reports were obtained after surgery or at the time subjects were discharged from the hospital. Controls were recruited from patients without cancer at MSKCC, including blood donors at MSKCC Blood Center, who had resided in the United States for at least 1 year and were in stable medical condition, as determined by their physician. Controls were not restricted by age, sex and race. During the study period 222 controls were approached and 204 were recruited and interviewed, of whom 173 provided blood samples. Epidemiological Data Collection Cases and controls were interviewed by trained research nurses. Epidemiological data were collected with a detailed questionnaire on 1) demography, including sex, race, date of birth and age, 2) life-style risk factors, such as cigarette smoking, alcohol drinking, coffee and tea consumption, 3) job titles and occupational exposure, 4) family history of cancer, 5) dietary history, 6) physical activity, 7) sexual behavior and history of sexually transmitted disease, and 8) medical history. To ensure interview quality 5% of subjects were re-interviewed randomly for selective questions. Blood Sample Collections and Laboratory Analyses After cases and controls were identified 5 ml blood were collected into tubes, transported immediately to the laboratory and stored at 4C. The blood was centrifuged at 200 gravity for 30 minutes to separate plasma and blots. The plasma was then divided into several 1 ml aliquots and stored in a 70C freezer. Samples were packed in Styrofoam containers with dry ice, shipped from MSKCC to UCLA in 1997 and stored at 70C. Individual plasma tocopherols, retinol and carotenoids were measured by HPLC at the UCLA Center for Human Nutrition. Case and control samples were prepared simultaneously. Extraction and HPLC analysis were done in subdued light using previously described procedures. 15 The system was validated against standard reference serum from the National Institute of Standards and Technology Micronutrients Measurement Quality Assurance Program 2 to 3 times annually. Coefficients of variation for intra-assay pool plasma sample were 7.4 for lutein, 8.7 for retinol, 8.8 for -tocopherol, 10.3 for -carotene, 10.6 for -tocopherol, 11.5 for -cryptoxanthin, 12.2 for lycopene and 14.2 for -carotene. Quality control procedures included periodic analysis of a pooled plasma sample daily and the National Institute of Standards and Technology Reference Material. Of the samples 5% were randomly selected and remeasured. Statistical Analysis Associations between plasma carotenoids, retinol and tocopherols, and bladder cancer risk were estimated with the adjusted OR and 95% CI using logistic regression. Variables were first analyzed as continuous variables and then as ordered variables based on the quartile distribution among controls. When the data were analyzed as continuous variables, the logistic models of retinol, -tocopherol, -tocopherol, lutein, -carotene and -carotene were based on 10 U micronutrient concentration, while those of lycopene, zeaxanthin and -cryptoxanthin were based on 100 U micronutrient concentration because the concentrations were low. Trend tests were performed by assigning the score j to the jth exposure level of an ordered variable, where j 0, 1, 2..., and treating the ordered variable as an interval predictor in the unconditional logistic regression. We adjusted for potential confounding factors, including patient age, sex, education and pack-years of smoking (continuous variable). RESULTS Table 1 shows the comparison of demographic variables between subjects with and without a blood sample. There were no meaningful differences between the 2 groups in terms of age, sex, years of education, smoking status and cumulative tobacco consumptions. Table 2 lists the mean, median, 25th and 75th percentiles of micronutrient concentration in cases and controls. Median levels of carotenoids, retinol and tocopherols in controls were higher than those in cases. Table 3 shows crude and adjusted ORs, and corresponding 95% CIs for retinol and tocopherols. No obvious associations were found between plasma -tocopherol, -tocopherol and retinol, and bladder cancer risk after adjusting for potential confounding factors, including patient age, sex, education and pack-years of smoking (table 3). Inverse associations were found for plasma lutein, zeaxanthin, -cryptoxanthin, -carotene and lycopene after adjusting for potential confounders (table 4). The adjusted OR for plasma lutein was 0.11 (95% CI 0.01 to 0.91) for the highest vs the lowest quartile (p for trend 0.03). When plasma lutein was treated as a continuous variable, the adjusted OR was 0.42 (95% CI 0.18 to 1.00, p 0.05). The adjusted OR for -carotene was 0.12 (95% CI 0.02 to 0.95) for the highest vs the lowest quartile with a clear dose-response relationship (p for trend 0.03). The adjusted OR for lycopene was 0.05 (95% CI 0.002 to 1.14) for the highest vs the lowest quartile (p for trend 0.06). The adjusted OR for zeaxanthin was 0.16 (95% CI 0.02 to 1.06) for the highest

1194 PROTECTIVE EFFECTS OF PLASMA CAROTENOIDS ON BLADDER CANCER RISK TABLE 1. Demographic characteristics of bladder cancer cases and controls with and without plasma available No. With Plasma (%) No. Without Plasma (%) Cases Controls Case Controls Total No. Totals 84 173 158 31 189 Age: 60 or older 61 (72.62) 19 (10.98) 114 (72.15) 6 (19.35) 120 Younger than 60 23 (27.38) 154 (89.02) 44 (27.85) 25 (80.65) 69 p Value with vs without plasma 0.94 0.19 Sex: M 72 (85.71) 132 (76.74) 130 (82.28) 24 (80.00) 154 F 12 (14.29) 40 (23.26) 28 (17.72) 6 (20.00) 34 p Value with vs without plasma 0.49 0.69 Smoking: Yes 66 (79.52) 76 (44.44) 131 (84.52) 16 (55.17) 147 No 17 (20.48) 95 (55.56) 24 (15.48) 13 (44.83) 37 p Value with vs without plasma 0.33 0.28 Pack-yrs: 20 or Greater 55 (65.48) 28 (16.18) 97 (61.39) 5 (16.13) 102 Less than 20 29 (34.52) 145 (83.82) 61 (38.61) 26 (83.87) 87 p Value with vs without plasma 0.53 0.99 Education (yrs): 12 or Greater 75 (89.29) 169 (97.69) 140 (88.61) 28 (90.32) 168 Less than 12 9 (10.71) 4 (2.31) 18 (11.39) 3 (9.68) 21 p Value with vs without plasma 0.87 0.07 (Fisher s exact test) quartile (p for trend 0.03). The adjusted OR was 0.90 (95% CI 0.81 to 1.00, p 0.05) when -cryptoxanthin was treated as a continuous variable. Table 5 shows the estimated joint effects of smoking, and plasma lutein, zeaxanthin, -carotene and -cryptoxanthin. ORs for the joint effect were 6.22 (95% CI 1.87 to 20.76) in smokers with lower lutein and 5.18 (95% CI 1.57 to 17.08) in smokers with lower zeaxanthin. This indicates that the effect of smoking on bladder cancer risk might be modified by lutein and zeaxanthin. DISCUSSION Most studies of the association of micronutrients and bladder cancer risk have been based on estimated intake from food frequency questionnaires. Results are often inconsistent due to the inherent methodological limitations of such a study design. By measuring micronutrients in plasma we were able to estimate the internal doses of nutrients. However, such measurements can only reflect micronutrient levels at the time of blood sample collection and not long-term dietary patterns. Plasma carotenoids only reveal a shortterm dietary intake because the half-life of carotenoids in blood is only a few days. In addition, measured levels may represent a changed diet in cases after cancer diagnosis. If patients changed their life-styles and dietary habits by increasing the intake of dietary antioxidants after cancer diagnosis, plasma concentrations of these carotenoids should be higher than before the diagnosis. However, this would lead to an underestimation of the associations of carotenoids with bladder cancer and bias the association toward null. Despite this potential limitation we found significant inverse associations between some plasma carotenoids and bladder cancer, which indicates that our estimates are relatively conservative. Recent studies indicate that smokers had significantly lower serum/plasma -carotene, -carotene, lutein/zeaxanthin and -cryptoxanthin than nonsmokers, 16 indicating that smoking might be a potential confounder for the relationship between carotenoids and bladder cancer. Although we adjusted for cigarette smoking in our regression analysis, the possible residual confounding effect by cigarette smoking might still exist. Nevertheless, the protective effect of high plasma lutein, which persisted when analyses were restricted to those who never smoked, argues for a true association (table 5). As a hospital-based case-control study, potential selection bias might exist. However, using controls from the same clinic is a commonly accepted method for controlling the potential selection bias. The categorization of micronutrient concentrations into quartiles by the cutoff point in the control group might lead to the loss of some information and the artificial category might result in some misleading results. The effect of micronutrients might be diluted by quartile categorization, which might be biased toward the null. How- TABLE 2. Plasma carotenoids, retinol and tocopherols in cases and controls Cases (0.1 mol/l) Controls (0.1 mol/l) Mean SD Quartile 1 Median Quartile 3 Mean SD Quartile 1 Median Quartile 3 Retinol 13.04 5.75 9.75 11.88 15.77 14.73 7.27 9.42 12.94 18.51 Lutein 0.95 0.58 0.56 0.78 1.22 1.48 1.20 0.67 1.22 1.91 -Tocopherol 28.65 21.31 16.12 22.99 35.74 33.09 23.65 18.41 27.31 40.34 -Tocopherol 195.09 121.87 116.95 146.64 227.87 203.50 115.10 114.17 177.45 265.93 -Carotene 0.67 0.63 0.33 0.43 0.75 0.95 0.91 0.39 0.70 1.18 -Carotene 3.46 7.57 0.97 1.65 3.19 3.11 2.79 1.34 2.04 4.22 Lycopene 1.83 1.10 1.04 1.63 2.56 3.42 4.80 1.92 2.82 4.01 Zeaxanthin 0.51 0.47 0.25 0.39 0.58 0.68 0.51 0.34 0.54 0.83 -Cryptoxanthin 0.62 0.48 0.32 0.46 0.71 1.30 1.54 0.51 0.83 1.64

PROTECTIVE EFFECTS OF PLASMA CAROTENOIDS ON BLADDER CANCER RISK 1195 TABLE 3. Bladder cancer risk for plasma tocopherols and retinol by control quartiles No. Cases No. Controls Crude OR (95% CI) p Value Adjusted OR (95% CI) p Value -Tocopherol: 0 28 43 1.00 0.12 1.00 0.52 1 23 43 0.82 (0.41 1.65) 1.94 (0.34 11.2) 2 15 43 0.54 (0.25 1.14) 1.34 (0.28 6.50) 3 18 44 0.63 (0.30 1.30) 1.81 (0.38 8.64) Continuous 84 173 0.99 (0.98 1.00) 0.15 1.01 (0.99 1.04) 0.31 -Tocopherol: 0 18 43 1.00 0.61 1.00 0.26 1 28 43 1.56 (0.75 3.22) 4.53 (0.72 28.6) 2 22 44 1.19 (0.56 2.53) 0.44 (0.07 2.91) 3 16 43 0.89 (0.40 1.97) 0.91 (0.17 4.95) Continuous 84 173 1.00 (1.00 1.00) 0.59 1.00 (0.99 1.00) 0.17 Retinol: 0 20 43 1.00 0.18 1.00 0.24 1 31 43 1.55 (0.77 3.13) 0.67 (0.14 3.26) 2 20 44 0.98 (0.46 2.07) 0.70 (0.11 4.45) 3 13 43 0.65 (0.29 1.47) 0.32 (0.05 1.96) Continuous 84 173 0.96 (0.92 1.00) 0.07 0.98 (0.89 1.08) 0.67 *Adjusted for age, sex, pack-years and education. Continuous variable with 10 U of increment. ever, we found similar results using continuous variables in our analyses. Although the participation rate was 94% in cases and 92% in controls, a potential selection bias must be noted because the proportion of cases with blood samples was substantially low (39% vs 85% of controls). Patients with blood samples may be a select population, which may have affected the association under study. We analyzed our data to explore whether there was a potential selection bias for cases and controls with and without blood samples. No obvious differences were found in cases in terms of age, race and education. However, this still does not fully preclude potential selection bias, and so we should interpolate the results with great caution. In addition, the small sample size may have decreased the power of the study, resulting in imprecise measurements and limiting our ability to estimate the association precisely. Despite limited statistical power TABLE 4. Bladder cancer risk for plasma tocopherols and retinol based on control quartiles No. Cases No. Controls Crude OR (95% CI) p Value Adjusted OR (95% CI) p Value Lutein: 0 34 43 1.00 0.0001 1.00 0.0261 1 29 43 0.85 (0.45 1.64) 0.65 (0.16 2.67) 2 16 43 0.47 (0.23 0.98) 0.34 (0.06 1.95) 3 5 44 0.14 (0.05 0.40) 0.11 (0.01 0.91) Continuous 84 173 0.48 (0.33 0.71) 0.0002 0.42 (0.18 1.00) 0.0508 -Carotene: 0 32 43 1.00 0.0012 1.00 0.0321 1 28 43 0.77 (0.45 1.69) 0.49 (0.11 2.34) 2 14 43 0.32 (0.21 0.93) 0.30 (0.06 1.48) 3 10 44 1.69 (0.13 0.70) 0.12 (0.02 0.95) Continuous 84 173 0.56 (0.35 0.88) 0.1113 0.22 (0.05 0.92) 0.0378 -Carotene: 0 31 43 1.00 0.0313 1.00 0.047 1 21 44 0.66 (0.33 1.33) 0.15 (0.03 0.83) 2 18 42 0.59 (0.29 1.22) 0.30 (0.05 1.72) 3 14 44 0.44 (0.21 0.94) 0.11 (0.04 0.80) Continuous 84 173 1.01 (0.96 1.07) 0.593 1.01 (0.94 1.09) 0.8221 Lycopene: 0 49 43 1.00 0.0001 1.00 0.059 1 23 43 0.47 (0.25 0.90) 0.77 (0.18 3.24) 2 9 43 0.18 (0.08 0.42) 0.58 (0.13 2.63) 3 3 44 0.06 (0.02 0.21) 0.05 (0.002 1.14) Continuous 84 173 0.93 (0.91 0.96) 0.0001 0.94 (0.89 0.99) 0.0255 Zeaxanthin: 0 34 43 1.00 0.0013 1.00 0.0295 1 23 43 0.68 (0.34 1.33) 1.19 (0.24 5.82) 2 18 43 0.53 (0.26 1.08) 0.36 (0.07 1.77) 3 9 44 0.23 (0.11 0.60) 0.16 (0.02 1.06) Continuous 84 173 0.92 (0.86 0.98) 0.0124 0.87 (0.74 0.97) 0.0126 -Cryptoxanthin: 0 45 43 1.00 0.0001 1.00 0.031 1 21 43 0.47 (0.24 0.91) 1.43 (0.32 6.31) 2 13 43 0.29 (0.14 0.61) 0.31 (0.06 1.70) 3 5 44 0.11 (0.04 0.30) 0.11 (0.01 1.03) Continuous 84 173 0.89 (0.84 0.94) 0.0001 0.90 (0.81 1.00) 0.0523 *Adjusted for age, sex, pack-years and education. Continuous variable with 10 U of increment. Continuous variable with 100 U of increment.

1196 PROTECTIVE EFFECTS OF PLASMA CAROTENOIDS ON BLADDER CANCER RISK TABLE 5. Potential micronutrient and cigarette smoking joint effects on bladder cancer risk Smoking No. Cases No. Controls Adjusted* OR (95% CI) Lutein: High No 8 48 1.00 High Yes 12 38 1.30 (0.34 4.94) Low No 9 47 4.61 (0.99 21.58) Low Yes 54 38 6.22 (1.87 20.76) -Carotene: High No 9 45 1.00 High Yes 15 41 0.71 (0.19 2.69) Low No 8 50 0.86 (0.19 3.84) Low Yes 51 35 2.89 (0.85 9.81) Zeaxanthin: High No 8 51 1.00 High Yes 18 35 1.26 (0.36 4.48) Low No 9 44 3.22 (0.69 14.92) Low Yes 48 41 5.18 (1.57 17.08) -Cryptoxanthin: High No 7 49 1.00 High Yes 10 36 0.95 (0.21 4.43) Low No 10 46 3.90 (0.80 18.9) Low Yes 56 40 6.54 (1.71 24.97) High third and fourth quartiles based on control cutoff and low first and second quartiles based on control cutoff. * Adjusted for age, sex and education. we achieved statistical significance with several carotenoids. We believe that the limitation of the small sample size might not fully affect the conclusion of the study. In this study protective effects on bladder cancer risk were found for plasma lutein, -carotene, zeaxanthin and -cryptoxanthin with dose-response gradients. The protective effect of lycopene was of borderline significance. Our results are in agreement with 2 recent studies. A large cohort study indicated strong inverse associations between bladder cancer risk and the intake of dark green vegetables (p 0.01 for linear trend), yellow-orange vegetables (p 0.01), tomato products (p 0.03) and citrus fruits/juices (p 0.002) after adjusting for nondietary risk factors such as cigarette smoking. 3 Another case-control study showed that high dietary intake of total carotenoid and nonprovitamin A carotenoid (lutein, zeaxanthin and lycopene) were associated with a 44% and 43% decrease in bladder cancer risk, respectively. 6 Inverse associations between serum carotenoids and bladder cancer were found in this study. However, it is difficult to infer a true causal relationship based on the limitations discussed. It has been suggested that anticancer effects of carotenoids are at least in part related to their antioxidant activity. Antioxidant activity of carotenoids in multilamellar liposomes assayed by inhibiting the formation of thiobarbituric acid reactive substances was highest for lycopene and lowest for lutein. Mixtures of carotenoids were more effective than the single compounds and this synergistic effect was most pronounced when lycopene or lutein was present. This superior protection of mixtures may be related to the specific positioning of different carotenoids in membranes and lipoproteins. 17 In addition, anticancer activities of carotenoids have been reported to be associated with the induction and stimulation of intercellular communication via gap junctions, which have a role in the regulation of cell growth, differentiation and apoptosis with the regulation of enzymes important for metabolizing xenobiotics and carcinogens, and the modulation of nuclear receptors and cellular signaling of pathways. 18 Animal studies provide further evidence in support of carotenoids as protective factors in bladder cancer development. Tanaka et al found that the incidence of bladder cancer in mice treated with the carotenoid astaxanthin was significantly lower than in control mice. 19 Results also show strong combined effects of cigarette smoking and certain carotenoids, namely lutein, zeaxanthin, -carotene and -cryptoxanthin. Cigarette smoke contains oxidizing agents, among more than 4,000 identified constituents, 20 in which polycyclic aromatic hydrocarbons, aromatic amines and other chemicals are carcinogenic, mutagenic and toxic. In addition, free radicals in cigarette smoke may cause oxidative damage to macromolecules. A recent study demonstrated main effects for smoking and carotenoid intake as well as for effect modification in joint effect analyses. Analysis of the independent effects of smoking revealed that current smokers were at 4.8-fold increased risk for bladder cancer, whereas former smokers were at 2-fold increased risk. Furthermore, current and heavy smokers were at a greater risk than former smokers and those who never smoked, and there was a decreased risk for high vs low carotenoid intake for each smoking stratum. 6 Another casecontrol study also showed that currently smoking healthy controls had lower serum levels of carotenoids than nonsmokers. 14 Taken together these results indicate that carotenoids may counteract the adverse effects caused by cigarette smoking. CONCLUSIONS The increase in plasma lutein, -carotene, zeaxanthin, -cryptoxanthin and lycopene appeared to be associated with a decreased risk of bladder cancer with dose-response gradients. Our results suggest the potential protective effect of these carotenoids on bladder cancer, especially in smokers. Abbreviations and Acronyms HPLC high performance liquid chromatography MSKCC Memorial Sloan-Kettering Cancer Center REFERENCES 1. Cancer Facts and Figures, 2006. Atlanta: American Cancer Society, 2006 2. Zhang, Z. F. and Steineck, G.: The epidemiology and etiology of bladder cancer. In: Principles and Practice of Genitourinary Oncology. Edited by D. Raghavan and H. Scher. New York: J. B. Lippincott, pp. 215 222, 1997 3. Castelao, J. E., Yuan, J. M., Gago-Dominguez, M., Skipper, P. L., Tannenbaum, S. R., Chan, K. K. et al: Carotenoids/ vitamin C and smoking-related bladder cancer. Int J Cancer, 110: 417, 2004 4. Michaud, D. S., Pietinen, P., Taylor, P. R., Virtanen, M., Virtamo, J. and Albanes, D.: Intakes of fruits and vegetables, carotenoids and vitamins A, E, C in relation to the risk of bladder cancer in the ATBC cohort study. Br J Cancer, 87: 960, 2002 5. Zeegers, M. P., Goldbohm, R. A. and van den Brandt, P. A.: Are retinol, vitamin C, vitamin E, folate and carotenoids intake associated with bladder cancer risk? Results from the Netherlands Cohort Study. Br J Cancer, 85: 977, 2001

PROTECTIVE EFFECTS OF PLASMA CAROTENOIDS ON BLADDER CANCER RISK 1197 6. Schabath, M. B., Grossman, H. B., Delclos, G. L., Hernandez, L. M., Day, R. S., Davis, B. R. et al: Dietary carotenoids and genetic instability modify bladder cancer risk. J Nutr, 134: 3362, 2004 7. Steinmaus, C. M., Nunez, S. and Smith, A. H.: Diet and bladder cancer: a meta-analysis of six dietary variables. Am J Epidemiol, 151: 693, 2000 8. Lamm, D. L., Riggs, D. R., Shriver, J. S., vangilder, P. F., Rach, J. F. and DeHaven, J. I.: Megadose vitamins in bladder cancer: a double-blind clinical trial. J Urol, 151: 21, 1994 9. Hennekens, C. H., Buring, J. E., Manson, J. E., Stampfer, M., Rosner, B., Cook, N. R. et al: Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J Med, 334: 1145, 1996 10. Studer, U. E., Jenzer, S., Biedermann, C., Chollet, D., Kraft, R., von Toggenburg, H. et al: Adjuvant treatment with a vitamin A analogue (etretinate) after transurethral resection of superficial bladder tumors. Final analysis of a prospective, randomized multicenter trial in Switzerland. Eur Urol, 28: 284, 1995 11. Hicks, R. M., Turton, J. A., Chowaniec, J., Tomlinson, C. N., Gwynne, J., Nandra, K. et al: Modulation of carcinogenesis in the urinary bladder by retinoids. Ciba Found Symp, 113: 168, 1985 12. Tyler, H. A., Notley, R. G., Schweitzer, F. A. and Dickerson, J. W.: Vitamin A status and bladder cancer. Eur J Surg Oncol, 12: 35, 1986 13. Helzlsouer, K. J., Comstock, G. W. and Morris, J. S.: Selenium, lycopene, alpha-tocopherol, beta-carotene, retinol, and subsequent bladder cancer. Cancer Res, 49: 6144, 1989 14. Nomura, A. M., Lee, J., Stemmermann, G. N. and Franke, A. A.: Serum vitamins and the subsequent risk of bladder cancer. J Urol, 170: 1146, 2003 15. Lu, Q. Y., Hung, J. C., Heber, D., Go, V. L., Reuter, V. E., Cordon-Cardo, C. et al: Inverse associations between plasma lycopene and other carotenoids and prostate cancer. Cancer Epidemiol Biomarkers Prev, 10: 749, 2001 16. Dietrich, M., Block, G., Norkus, E. P., Hudes, M., Traber, M. G., Cross, C. E. et al: Smoking and exposure to environmental tobacco smoke decrease some plasma antioxidants and increase gamma-tocopherol in vivo after adjustment for dietary antioxidant intakes. Am J Clin Nutr, 77: 160, 2003 17. Stahl, W., Junghans, A., de Boer, B., Driomina, E. S., Briviba, K. and Sies, H.: Carotenoid mixtures protect multilamellar liposomes against oxidative damage: synergistic effects of lycopene and lutein. FEBS Lett, 427: 305, 1998 18. Sharoni, Y., Danilenko, M., Dubi, N., Ben Dor, A. and Levy, J.: Carotenoids and transcription. Arch Biochem Biophys, 430: 89, 2004 19. Tanaka, T., Morishita, Y., Suzui, M., Kojima, T., Okumura, A. and Mori, H.: Chemoprevention of mouse urinary bladder carcinogenesis by the naturally occurring carotenoid astaxanthin. Carcinogenesis, 15: 15, 1994 20. Pryor, W. A.: Cigarette smoke radicals and the role of free radicals in chemical carcinogenicity. Environ Health Perspect, suppl., 105: 875, 1997