Assessment of a Dietary Questionnaire in Cancer Patients Receiving Cytotoxic Chemotherapy
http://www.100md.com
《临床肿瘤学》
the Department of Medical Oncology, Dana-Farber Cancer Institute
Department of Nutrition and Epidemiology, Harvard School of Public Health
Channing Laboratory, Department of Medicine, Brigham and Women's Hospital, Boston, MA.
ABSTRACT
PURPOSE: Few studies have examined the influence of diet on survival and chemotherapy-associated toxicities in patients with cancer. Although several comprehensive dietary questionnaires have been validated and calibrated in healthy populations, similar studies have not been performed among cancer patients.
METHODS: Two hundred patients with colorectal, breast, or neuroendocrine cancer undergoing treatment with cytotoxic chemotherapy completed a self-administered, 131-item, semiquantitative food frequency questionnaire. Using the questionnaire, we calculated dietary intakes of carotenoids, tocopherols, and fatty acids, and correlated these values with relevant biomarkers measured in simultaneously collected plasma specimens.
RESULTS: The Pearson correlation coefficients for various carotenoids as measured by the questionnaire, with the corresponding measurements in plasma specimens, ranged from 0.33 to 0.44 (all P < .001), adjusted for total energy intake, body mass index, age, sex, smoking status, and total plasma cholesterol. Similarly, the adjusted correlation between self-reported total vitamin E intake and plasma alpha-tocopherol was 0.34 (P < .001). Correlations between questionnaire and plasma measurements of trans-fat, eicosapentaenoic acid, and docosahexaenoic acid were 0.55, 0.29, and 0.42 (all P < .001), respectively. These levels of correlation are consistent with those reported in similar studies of self-reported diet in otherwise healthy populations.
CONCLUSION: Among patients with cancer receiving cytotoxic chemotherapy, questionnaire-based measurements of various micronutrients and dietary factors appeared to predict meaningful differences in the corresponding measurements in plasma specimens. This dietary questionnaire could offer an informative and practical means for assessing the influence of diet in cancer patients receiving chemotherapy.
INTRODUCTION
Multiple epidemiology studies have shown a relationship between diet and the risk of developing a variety of cancers, including colorectal, breast, upper GI, and prostate.[1-6] However, data on the influence of diet on the outcomes of patients with established cancer are considerably more limited.[7] A recent panel convened by the American Cancer Society reported that there is insufficient evidence to provide specific dietary recommendations for patients with breast, colorectal, lung, and prostate cancer.[7] The panel concluded that properly conducted studies of the effect of nutrition on the prognosis of cancer survivors are needed urgently, and should be a high priority for all academic and research funding agencies.
To assess associations between diet and cancer risk in healthy populations, large prospective studies principally have used semiquantitative food frequency questionnaires to ascertain estimated average long-term dietary habits.[8-12] Participants report their typical intake of commonly consumed foods from which average intake of macronutrients and micronutrients can be calculated. To identify dietary risk factors for various chronic diseases in adults, including cancer and heart disease, Willett et al[13] designed a self-administered, semiquantitative food frequency questionnaire that consists of 131 food items plus vitamin and mineral supplement use that collectively account for more than 90% of the intake of the nutrients assessed. In validation and calibration studies among healthy participants of various races and ethnicities,[13-16] mean calorie-adjusted nutrients as measured by the questionnaire were well correlated with nutrients measured from weekly diet records.[13,17,18] In addition, nutrients measured by the questionnaire correlated with corresponding biochemical indicators measured from plasma and adipose tissue specimens.[19-21] Moreover, evidence that the dietary questionnaire can detect important biologic relationships is provided by studies demonstrating associations between diet and the risk of developing various diseases, including heart disease and cancer.[22]
Although the semiquantitative food frequency questionnaire developed by Willett et al has been validated and calibrated in various healthy populations,[13-16] it remains uncertain whether the instrument is a valid measure of dietary habits in patients with cancer undergoing chemotherapy. Assessment of the utility of a semiquantitative food frequency questionnaire for a given population can be done using various techniques, including 24-hour recalls, diet records, or measurement of corresponding biochemical markers. Studies with biochemical markers, such as plasma carotenoids, tocopherols, and nonendogenous fatty acids,[20,21,23-25] have the distinct advantage of being objective; moreover, such biochemical assessments are subject to measurement errors that are not correlated with those inherent to a semiquantitative food frequency questionnaire.[22]
Cancer patients receiving chemotherapy may experience side effects that influence diet choices, food absorption, and energy balance. To support the future investigation of diet and cancer survival, studies that assess the utility of using semiquantitative food frequency questionnaires among cancer patients are clearly needed. We therefore administered a widely used semiquantitative food frequency questionnaire to a population of patients receiving cytotoxic chemotherapy and compared the results with corresponding measurements in simultaneously collected plasma specimens.
METHODS
Study Population
Patients for this study were enrolled at the Dana-Farber Cancer Institute (Boston, MA) between October 2003 and October 2004. Patients with colorectal, neuroendocrine, or breast cancer receiving cytotoxic chemotherapy were eligible for participation. Patients treated only with hormonal therapy or single-agent targeted therapy (eg, gefitinib, erlotinib, bevacizumab, cetuximab, or trastuzumab) were not eligible, unless they were given a cytotoxic therapy concurrently. Patients had to be receiving chemotherapy for at least 4 weeks and could not have had any major surgery within the 4 weeks before completion of the semiquantitative food frequency questionnaire and plasma collection. All patients were required to have an Eastern Cooperative Oncology Group performance status of 2 or less (ambulatory, up and about at least 50% of waking hours, and able to do self-care) and have an adequate caloric intake (as determined by the treating physician). Patients lacking physical integrity of the upper GI tract or who had a malabsorption syndrome were not eligible. In addition, patients receiving total parental nutrition or tube feedings were not eligible, although oral supplements were allowed. Patients receiving a cholesterol-lowering agent (including niacin, gemfibrozil, and 5-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors) were excluded. Participating physicians were provided with their clinic schedule for a given month and indicated which patients met eligibility for the study. Of the patients deemed eligible, 92% agreed to participate and signed informed consent; of those patients, 90% completed all components of the study. The study protocol and informed consent form were approved by the institutional review board of the Office for the Protection of Research Subjects for the Dana-Farber Harvard Cancer Center.
Semiquantitative Food Frequency Questionnaire
All patients completed a single semiquantitative food frequency questionnaire developed, tested, and refined by Willett et al.[1,8,9,13] Most patients completed the questionnaire while in the clinic, during the administration of chemotherapy. The questionnaire included 131 food items, vitamin and mineral supplements, and open-ended sections for other supplements and foods not specifically listed. Our questionnaire was modified to inquire about average food intake during the immediate last 3 months. We inquired about the exact breakfast cereal, multivitamin supplement, margarine, and vegetable oil used for frying or baking. For each food, a commonly used unit or portion size (eg, one egg or slice of bread) was specified, and participants were asked how often, on average during the last 3 months, they consumed that amount of each food. There were up to nine possible responses, which ranged from never to six or more times per day. We computed nutrient intakes by multiplying the frequency of consumption of each food by the nutrient content of the specified portions, using composition values from US Department of Agriculture sources supplemented with other data, including the components of specific vitamins and breakfast cereals.[26] We included supplements to total intake of specific nutrients. All values were energy adjusted using the residuals methods.[27]
Collection and Analyses of Biologic Samples
Patients underwent venipuncture at the time of a clinic visit. When possible, patients were requested to fast before the blood draw and 39% of patients reported no oral intake in the 8 hours before plasma collection. For carotenoid, tocopherol, and fatty acid measurements, blood was collected in an ethylenediaminetetra-acetic acid–containing tube, and plasma was separated from the cellular component by centrifugation, aliquoted, and frozen at –70°C until analysis (which was completed within 1 year of collection). Samples underwent a single thaw for analysis. Total cholesterol, high-density lipoprotein, and triglycerides were measured according to institutional standards at the Dana-Farber Cancer Institute.
Concentrations of carotenoids and tocopherols were measured as described by El-Sohemy et al.[28] Samples were quantitated by high-performance liquid chromatography. The system manager software (D-7000, Version 3.0) was used for peak integration and data acquisition (Hitachi, San Jose, CA). Because lutein and zeaxanthin coelute on the chromatogram, the two are grouped and provided as lutein + zeaxanthin. Overall (between-run) coefficients of variation were all less than 5%.
Plasma fatty acids were determined by extracting and transmethylating the lipids with methanol and sulfuric acid as described by Zock et al.[29,30] After esterification, the fatty acid methyl esters were redissolved in isooctane and quantitated by chromatography. Peak retention times and area percent of total fatty acids were identified by injecting known standards, as described previously.[31] Overall coefficients of variation for the major peaks were approximately 5%.
Ascertainment of Clinical Data
Computerized medical records were reviewed to extract the cancer diagnosis, current chemotherapy regimen, height, weight at time of study participation, weight approximately 4 to 6 weeks earlier, time the patient received chemotherapy, and age. Height and weight were obtained by standard clinical practice guidelines at the Dana-Farber Cancer Institute. Body mass index was calculated as weight in kilograms divided height in square meters. Smoking status was obtained from the self-administered questionnaire that included diet and classified as current, past, or never smoker.
Statistical Analysis
Pearson correlation coefficients were used to assess the relationship between plasma concentrations and dietary intake. To better approximate a normal distribution, all variables, except age and body mass index, underwent natural logarithmic transformation. Consistent with prior analyses, we excluded patients with implausibly high (> 3,500 calories for women; > 4,200 calories for men) or low intakes (< 600 calories for women; < 800 calories for men).[32] We adjusted for potential confounders using the PROC CORR PARTIAL procedure in SAS version 9.1 (SAS Institute, Cary, NC). For fatty acids, we used the percent of total fat intake made up by a particular fatty acid. Previous studies have estimated that 100 to 200 participants can provide sufficiently stable correlation coefficients comparing the food frequency questionnaire with a comparison method.[22] Two hundred twenty-seven patients were enrolled onto the study. Four patients were withdrawn due to ineligibility (all were not receiving a cytotoxic chemotherapy), 23 did not complete all components of the study, and eight patients were excluded because of implausible caloric intake. Thus the final analysis cohort was 192 patients.
RESULTS
Baseline Characteristics
One hundred ninety-two patients were included in our analyses ([Table 1]). Median age for the population was 55 years, and the majority of patients had either breast cancer or colorectal cancer treated with a variety of cytotoxic chemotherapy regimens. Given that rapid changes in energy and nitrogen balance may influence the correlation between self-reported dietary intake during a 3-month period and single biomarker measurements, we determined weight changes during a 4- to 6-week period. Notably, the majority of participants had maintained a relatively stable weight; 95% of patients experienced a weight change of less than 3 kg in the 4 to 6 weeks before study enrollment.
Correlations of Carotenoids and Tocopherols
We calculated crude and multivariate Pearson correlation coefficients between energy-adjusted dietary micronutrients as measured by the semiquantitative food frequency questionnaire and corresponding biomarkers measured in plasma specimens. Consistent with prior similar studies, we adjusted correlations for age, body mass index, smoking status, and total cholesterol.[21] In addition, we included sex and weight change in the models. Because the majority of patients were not able to fast before sample collection, we did not adjust for plasma triglyceride levels. However, when we limited analyses to the 74 patients who fasted at least 8 hours, the correlations were not substantially different (data not shown). Results for various carotenoids and tocopherols as measured by the questionnaire were significantly correlated with plasma determinations ( [Table 2]). These correlations are comparable with those obtained from other studies of healthy populations ([Table 3]). For example, we found a correlation of 0.33 for beta-carotene (?-carotene), compared with a range of 0.22 to 0.39 in other studies.[21,23-25,33] Similarly, for alpha-tocopherol (-tocopherol), our correlation of 0.34 is within the range of 0.25 to 0.51 demonstrated by others.[21,24,33,34] Of note, the negative correlation between plasma gamma-tocopherol and total vitamin E intake (r = –0.40) has been observed by Ascherio et al[21] (r = –0.51), and is consistent with the experimental observation that -tocopherol supplements decrease plasma -tocopherol.[37]
Correlations of Fatty Acids
We similarly calculated crude and multivariate Pearson correlation coefficients ([Table 4]) between dietary fatty acids (as percentage of total fat intake) and the corresponding plasma fatty acids (as percentage of total fatty acids in plasma). These correlations for various fatty acids are comparable with those obtained from other studies of healthy populations ([Table 3]). For example, our correlations of 0.44 for total omega-3 and 0.42 for docosahexaenoic acid compare with those reported by others who measured plasma fatty acids (0.51[35] and 0.42,[20] respectively). Restriction to patients who were fasting did not appreciably change these results (data not shown).
Influence of Time on Chemotherapy
We modified the existing semiquantitative food frequency questionnaire by assessing average food intake during the previous 3 months (versus the typical 12 months). Patients in this trial were eligible if they had been receiving chemotherapy for at least 4 weeks. Given that this was a cross-sectional study among patients who received chemotherapy for various lengths of time, we confirmed that such variability did not influence our results. The correlations for carotenoids, tocopherols, and fatty acids do not vary substantially when we stratified the analyses by patients receiving chemotherapy for less than 3 (33% of cohort) or more than 3 months (67% of cohort). For example, adjusted correlations for -carotene were 0.39 (P = .003) for those receiving chemotherapy less than 3 months compared with 0.31 (P = .0005) for those treated for more than 3 months. Similarly, adjusted correlations for trans-fat were 0.54 and 0.56 (P < .0001 for both), respectively.
DISCUSSION
In a population of patients who were actively receiving cytotoxic chemotherapy for various solid tumor malignancies, we found that levels of nutrients as measured by a self-administered semiquantitative food frequency questionnaire were significantly correlated with the corresponding biomarkers measured from plasma. The correlations found for carotenoids, tocopherols, and fatty acids are comparable to those demonstrated using the same instrument in healthy populations without cancer. Our results suggest that this semiquantitative food frequency questionnaire may be an accurate instrument to assess dietary intake in cancer patients undergoing chemotherapy.
Within our study population, we considered other methods for validation of our semiquantitative food frequency questionnaire, such as diet records that would be maintained by each subject or 24-hour food recall surveys. These methods have the ability to correlate more nutrients because there are only a limited number of nutrients that have corresponding biochemical parameters that are sensitive to dietary intake. However, in a population of patients with cancer, diet records or recall surveys may be impractical and potentially less valuable. Because many of the chemotherapy regimens are given either on a weekly or every-other-week schedule, we would anticipate that daily intakes of patients would vary widely depending on toxicities experienced from the therapy. Moreover, diet records and 24-hour diet recall surveys are subject to sources of measurement error that are similar to those for a semiquantitative food frequency questionnaire and thereby may create artifactually positive correlations. In contrast, sources of measurement error for plasma biomarkers are not typically correlated with those found in a semiquantitative food frequency questionnaire.[22]
In our study, we prospectively collected information on known and potential confounders of the studied biomarkers, including smoking status, total cholesterol, body mass index, and age. Fasting triglycerides may also be a useful covariate in analyses of tocopherols. However, because all of our patients were actively receiving chemotherapy, less than one half were able to fast for more than 8 hours before plasma collection. Chemotherapy may affect plasma triglyceride levels in particular, leading to less accurate measurements, whereas cholesterol levels appear to be less affected.[38-42] Similarly, few studies have suggested that cancer therapies (including surgery, chemotherapy, and radiotherapy) may influence plasma tocopherols and carotenoids levels.[43-45] However, the comparable correlations between our study and prior reports suggests these biomarkers are still reasonable to use in this study of patients receiving chemotherapy.
Previous studies have conducted repeated administrations of the questionnaire as well as repeated plasma collections to reduce measurement error. Although we did not perform similar corrections in measurement error, our findings are nonetheless consistent with those studies that used repeated determinations of diet and biochemical parameters.[20,21,25,34-36]
Specific carotenoids and tocopherols may be associated with the development of a variety of malignancies,[46-52] and preclinical data suggest that tocopherols may influence treatment-related toxicities.[53-55] -Tocopherol is likely the predominant component of vitamin E supplementation, and our correlation for -tocopherol (r = 0.34) was consistent with correlations observed in other validation studies within healthy populations (r = 0.25 to 0.51). Moreover, prospective studies in previously healthy populations using this semiquantitative food frequency questionnaire have found significant associations between vitamin E intake and the risks of coronary heart disease[56] and possible risk of certain cancers.[57]
The correlations using plasma biomarkers (r = 0.3 to 0.6) may be considered modest.[58] However, other factors beyond dietary intake, such as variations in absorption and host metabolic parameters, will influence the levels of plasma biomarkers.[22] Thus, our findings do not allow a quantitative estimate of the correlation between our food frequency instrument and true intake. A prior study used the semiquantitative food frequency questionnaire, multiple 24-hour recalls, and biochemical markers (including certain carotenoids and fatty acids) to estimate the correlation of each with true intake using a triangulation method. Correlations between the semiquantitative food frequency questionnaire and the biochemical indicators were similar to those that we report in this study, and the triangulation analysis suggested that the correlations of the semiquantitative food frequency questionnaire intake with true intake were generally stronger than the correlations between the biochemical measurements and true intake.
We conducted this study in a population of patients with three different cancer types on a variety of chemotherapy regimens. Although we believe that our findings are generalizable to other cancer populations, the utility of a semiquantitative food frequency questionnaire in measuring between-subject variations in patients with cancers that may affect food absorption (eg, esophageal and gastric) is not addressed. Furthermore, all patients in this study were deemed to have adequate nutritional intake by their physicians, primarily on the basis of relative weight stability. Although this clearly is a subjective and inaccurate measure, only two patients had implausibly low total calorie counts and 95% of patients had relatively stable weight in the 4 to 6 weeks before enrollment. Nonetheless, use of a semiquantitative food frequency questionnaire to describe between-person variations in cancer patients with inadequate caloric intake cannot be addressed in this study.
Nutritional therapies are the most common form of complementary therapy among cancer patients.[7] After a diagnosis of cancer, patients are highly motivated to seek information about diet, physical activity, dietary supplement use, and nutritional complementary therapies. Studies among patients with breast cancer indicate that a majority are interested in making healthful changes in their diets.[59-62] In one study, 52% of cancer patients wanted nutritional guidance at the time of diagnosis or soon after, although few participants reported ever receiving dietary recommendation from their physicians.[59] Unfortunately, few data exist in the literature to enable firm conclusions to be drawn.[7] Confirmation of the utility of a semiquantitative food frequency questionnaire in this population has important implications. Dietary intake may affect treatment-related toxicities, effectiveness of adjuvant therapy for patients with curable disease, and progression-free and overall survival in patients with metastatic disease on long-term therapy. These results confirm that such a questionnaire can offer an informative assessment of dietary intake for the purpose of such future studies in such patients. Ongoing and completed cooperative group trials sponsored by the National Cancer Institute use this questionnaire to assess the influence of dietary intake on treatment-related toxicities and survival in patients with regional and advanced cancers. The data provided in the current analysis support the ability of measuring dietary intake in patients receiving chemotherapy and will allow additional investigation into the relationship between nutrition and outcomes in patients with cancer.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Jeremy Furtado for assistance in biomarker analyses and Laura Sampson for assistance in nutritional analyses.
NOTES
Supported in part by a K07 award from the National Cancer Institute (K07CA097992) and an American Society of Clinical Oncology career development award (J.A.M.).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Willett WC: Diet and cancer. Oncologist 5:393-404, 2000
La Vecchia C, Altieri A, Tavani A: Vegetables, fruit, antioxidants and cancer: A review of Italian studies. Eur J Nutr 40:261-267, 2001
Dagnelie PC, Schuurman AG, Goldbohm RA, et al: Diet, anthropometric measures and prostate cancer risk: A review of prospective cohort and intervention studies. BJU Int 93:1139-1150, 2004
Giovannucci E: Nutritional factors in human cancers. Adv Exp Med Biol 472:29-42, 1999
Blackburn GL, Copeland T, Khaodhiar L, et al: Diet and breast cancer. J Womens Health (Larchmt) 12:183-192, 2003
Thomson CA, LeWinn K, Newton TR, et al: Nutrition and diet in the development of gastrointestinal cancer. Curr Oncol Rep 5:192-202, 2003
Brown JK, Byers T, Doyle C, et al: Nutrition and physical activity during and after cancer treatment: An American Cancer Society guide for informed choices. CA Cancer J Clin 53:268-291, 2003
Willett WC, Reynolds RD, Cottrell-Hoehner S, et al: Validation of a semi-quantitative food frequency questionnaire: Comparison with a 1-year diet record. J Am Diet Assoc 87:43-47, 1987
Willett WC: Future directions in the development of food-frequency questionnaires. Am J Clin Nutr 59:171S-174S, 1994
Slattery ML, Dyer A, Jacobs DR Jr, et al: A comparison of two methods to ascertain dietary intake: The CARDIA Study. J Clin Epidemiol 47:701-711, 1994
Block G, Hartman AM, Dresser CM, et al: A data-based approach to diet questionnaire design and testing. Am J Epidemiol 124:453-469, 1986
Caan BJ, Slattery ML, Potter J, et al: Comparison of the Block and the Willett self-administered semiquantitative food frequency questionnaires with an interviewer-administered dietary history. Am J Epidemiol 148:1137-1147, 1998
Willett WC, Sampson L, Stampfer MJ, et al: Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 122:51-65, 1985
Subar AF, Thompson FE, Kipnis V, et al: Comparative validation of the Block, Willett, and National Cancer Institute food frequency questionnaires: The Eating at America's Table Study. Am J Epidemiol 154:1089-1099, 2001
Longnecker MP, Lissner L, Holden JM, et al: The reproducibility and validity of a self-administered semiquantitative food frequency questionnaire in subjects from South Dakota and Wyoming. Epidemiology 4:356-365, 1993
Hernandez-Avila M, Romieu I, Parra S, et al: Validity and reproducibility of a food frequency questionnaire to assess dietary intake of women living in Mexico City. Salud Publica Mex 40:133-140, 1998
Willett WC, Sampson L, Browne ML, et al: The use of a self-administered questionnaire to assess diet four years in the past. Am J Epidemiol 127:188-199, 1988
Rimm EB, Giovannucci EL, Stampfer MJ, et al: Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am J Epidemiol 135:1114-1136, 1992
Willett WC, Stampfer MJ, Underwood BA, et al: Validation of a dietary questionnaire with plasma carotenoid and alpha-tocopherol levels. Am J Clin Nutr 38:631-639, 1983
Ma J, Folsom AR, Eckfeldt JH, et al: Short- and long-term repeatability of fatty acid composition of human plasma phospholipids and cholesterol esters: The Atherosclerosis Risk in Communities (ARIC) Study Investigators. Am J Clin Nutr 62:572-578, 1995
Ascherio A, Stampfer MJ, Colditz GA, et al: Correlations of vitamin A and E intakes with the plasma concentrations of carotenoids and tocopherols among American men and women. J Nutr 122:1792-1801, 1992
Willett WC: Nutritional Epidemiology (ed 2). New York, NY, Oxford University Press, 1998, pp 101-147
McNaughton SA, Marks GC, Gaffney P, et al: Validation of a food-frequency questionnaire assessment of carotenoid and vitamin E intake using weighed food records and plasma biomarkers: The method of triads model. Eur J Clin Nutr 59:211-218, 2005
Newby PK, Hu FB, Rimm EB, et al: Reproducibility and validity of the Diet Quality Index Revised as assessed by use of a food-frequency questionnaire. Am J Clin Nutr 78:941-949, 2003
Resnicow K, Odom E, Wang T, et al: Validation of three food frequency questionnaires and 24-hour recalls with serum carotenoid levels in a sample of African-American adults. Am J Epidemiol 152:1072-1080, 2000
U.S. Department of Agriculture ARS: USDA Nutrient Database for Standard Reference, Release 10: Nutrient Data Laboratory, 1993. http://www.nal.usda.gov/fnic/foodcomp]
Willett W, Stampfer MJ: Total energy intake: Implications for epidemiologic analyses. Am J Epidemiol 124:17-27, 1986
El-Sohemy A, Baylin A, Kabagambe E, et al: Individual carotenoid concentrations in adipose tissue and plasma as biomarkers of dietary intake. Am J Clin Nutr 76:172-179, 2002
Zock PL, Gerritsen J, Katan MB: Partial conservation of the sn-2 position of dietary triglycerides in fasting plasma lipids in humans. Eur J Clin Invest 26:141-150, 1996
Zock PL, Mensink RP, Harryvan J, et al: Fatty acids in serum cholesteryl esters as quantitative biomarkers of dietary intake in humans. Am J Epidemiol 145:1114-1122, 1997
Kabagambe EK, Baylin A, Allan DA, et al: Application of the method of triads to evaluate the performance of food frequency questionnaires and biomarkers as indicators of long-term dietary intake. Am J Epidemiol 154:1126-1135, 2001
Wu K, Hu FB, Fuchs C, et al: Dietary patterns and risk of colon cancer and adenoma in a cohort of men (United States). Cancer Causes Control 15:853-862, 2004
Romieu I, Stampfer MJ, Stryker WS, et al: Food predictors of plasma beta-carotene and alpha-tocopherol: Validation of a food frequency questionnaire. Am J Epidemiol 131:864-876, 1990
Parrish LA, Marshall JA, Krebs NF, et al: Validation of a food frequency questionnaire in preschool children. Epidemiology 14:213-217, 2003
Sasaki S, Ushio F, Amano K, et al: Serum biomarker-based validation of a self-administered diet history questionnaire for Japanese subjects. J Nutr Sci Vitaminol (Tokyo) 46:285-296, 2000
Hyman MD, Insull W Jr, Palmer RH, et al: Assessing methods for measuring compliance with a fat-controlled diet. Am J Public Health 72:152-160, 1982
Baker H, Handelman GJ, Short S, et al: Comparison of plasma alpha and gamma tocopherol levels following chronic oral administration of either all-rac-alpha-tocopheryl acetate or RRR-alpha-tocopheryl acetate in normal adult male subjects. Am J Clin Nutr 43:382-387, 1986
Kailajarvi M, Ahokoski O, Virtanen A, et al: Early effects of 5-fluorouracil, epirubicin and cyclophosphamide therapy on common laboratory tests. Anticancer Res 21:2867-2872, 2001
Morimoto K, Koh M: Postoperative changes in serum lipid levels of breast cancer and adjuvant chemotherapy [in Japanese]. Gan To Kagaku Ryoho 31:1517-1524, 2004
Moschovi M, Trimis G, Apostolakou F, et al: Serum lipid alterations in acute lymphoblastic leukemia of childhood. J Pediatr Hematol Oncol 26:289-293, 2004
Ellis PA, Fitzharris BM, George PM, et al: Fasting plasma lipid measurements following cisplatin chemotherapy in patients with germ cell tumors. J Clin Oncol 10:1609-1614, 1992
Alexopoulos CG, Pournaras S, Vaslamatzis M, et al: Changes in serum lipids and lipoproteins in cancer patients during chemotherapy. Cancer Chemother Pharmacol 30:412-416, 1992
Potischman N, Byers T, Houghton L, et al: Effects of breast cancer treatments on plasma nutrient levels: Implications for epidemiological studies. Cancer Epidemiol Biomarkers Prev 1:555-559, 1992
Clemens MR, Ladner C, Ehninger G, et al: Plasma vitamin E and beta-carotene concentrations during radiochemotherapy preceding bone marrow transplantation. Am J Clin Nutr 51:216-219, 1990
Filer LJ Jr, Kibler CE, Mason KE, et al: The influence of surgery of serum tocopherol (vitamin E) and carotenoid levels. N Y State J Med 53:2836-2839, 1953
Gann PH, Ma J, Giovannucci E, et al: Lower prostate cancer risk in men with elevated plasma lycopene levels: Results of a prospective analysis. Cancer Res 59:1225-1230, 1999
Tamimi RM, Hankinson SE, Campos H, et al: Plasma carotenoids, retinol, and tocopherols and risk of breast cancer. Am J Epidemiol 161:153-160, 2005
Sato R, Helzlsouer KJ, Alberg AJ, et al: Prospective study of carotenoids, tocopherols, and retinoid concentrations and the risk of breast cancer. Cancer Epidemiol Biomarkers Prev 11:451-457, 2002
Giovannucci E: A review of epidemiologic studies of tomatoes, lycopene, and prostate cancer. Exp Biol Med (Maywood) 227:852-859, 2002
Epstein KR: The role of carotenoids on the risk of lung cancer. Semin Oncol 30:86-93, 2003
Helzlsouer KJ, Huang HY, Alberg AJ, et al: Association between alpha-tocopherol, gamma-tocopherol, selenium, and subsequent prostate cancer. J Natl Cancer Inst 92:2018-2023, 2000
Comstock GW, Alberg AJ, Huang HY, et al: The risk of developing lung cancer associated with antioxidants in the blood: Ascorbic acid, carotenoids, alpha-tocopherol, selenium, and total peroxyl radical absorbing capacity. Cancer Epidemiol Biomarkers Prev 6:907-916, 1997
Leonetti C, Biroccio A, Gabellini C, et al: Alpha-tocopherol protects against cisplatin-induced toxicity without interfering with antitumor efficacy. Int J Cancer 104:243-250, 2003
Legha SS, Wang YM, Mackay B, et al: Clinical and pharmacologic investigation of the effects of alpha-tocopherol on adriamycin cardiotoxicity. Ann N Y Acad Sci 393:411-418, 1982
Pace A, Savarese A, Picardo M, et al: Neuroprotective effect of vitamin E supplementation in patients treated with cisplatin chemotherapy. J Clin Oncol 21:927-931, 2003
Stampfer MJ, Hennekens CH, Manson JE, et al: Vitamin E consumption and the risk of coronary disease in women. N Engl J Med 328:1444-1449, 1993
Wu K, Willett WC, Chan JM, et al: A prospective study on supplemental vitamin E intake and risk of colon cancer in women and men. Cancer Epidemiol Biomarkers Prev 11:1298-1304, 2002
Cohen J: Statistical Power Analysis for the Behavioral Sciences. Mahwah, NJ, Lawrence Erlbaum Associates, 1988
Demark-Wahnefried W, Peterson B, McBride C, et al: Current health behaviors and readiness to pursue life-style changes among men and women diagnosed with early stage prostate and breast carcinomas. Cancer 88:674-684, 2000
Lee MM, Lin SS, Wrensch MR, et al: Alternative therapies used by women with breast cancer in four ethnic populations. J Natl Cancer Inst 92:42-47, 2000
Monnin S, Schiller MR, Sachs L, et al: Nutritional concerns of women with breast cancer. J Cancer Educ 8:63-69, 1993
Thomson CA, Flatt SW, Rock CL, et al: Increased fruit, vegetable and fiber intake and lower fat intake reported among women previously treated for invasive breast cancer. J Am Diet Assoc 102:801-808, 2002(Jeffrey A. Meyerhardt, De)
Department of Nutrition and Epidemiology, Harvard School of Public Health
Channing Laboratory, Department of Medicine, Brigham and Women's Hospital, Boston, MA.
ABSTRACT
PURPOSE: Few studies have examined the influence of diet on survival and chemotherapy-associated toxicities in patients with cancer. Although several comprehensive dietary questionnaires have been validated and calibrated in healthy populations, similar studies have not been performed among cancer patients.
METHODS: Two hundred patients with colorectal, breast, or neuroendocrine cancer undergoing treatment with cytotoxic chemotherapy completed a self-administered, 131-item, semiquantitative food frequency questionnaire. Using the questionnaire, we calculated dietary intakes of carotenoids, tocopherols, and fatty acids, and correlated these values with relevant biomarkers measured in simultaneously collected plasma specimens.
RESULTS: The Pearson correlation coefficients for various carotenoids as measured by the questionnaire, with the corresponding measurements in plasma specimens, ranged from 0.33 to 0.44 (all P < .001), adjusted for total energy intake, body mass index, age, sex, smoking status, and total plasma cholesterol. Similarly, the adjusted correlation between self-reported total vitamin E intake and plasma alpha-tocopherol was 0.34 (P < .001). Correlations between questionnaire and plasma measurements of trans-fat, eicosapentaenoic acid, and docosahexaenoic acid were 0.55, 0.29, and 0.42 (all P < .001), respectively. These levels of correlation are consistent with those reported in similar studies of self-reported diet in otherwise healthy populations.
CONCLUSION: Among patients with cancer receiving cytotoxic chemotherapy, questionnaire-based measurements of various micronutrients and dietary factors appeared to predict meaningful differences in the corresponding measurements in plasma specimens. This dietary questionnaire could offer an informative and practical means for assessing the influence of diet in cancer patients receiving chemotherapy.
INTRODUCTION
Multiple epidemiology studies have shown a relationship between diet and the risk of developing a variety of cancers, including colorectal, breast, upper GI, and prostate.[1-6] However, data on the influence of diet on the outcomes of patients with established cancer are considerably more limited.[7] A recent panel convened by the American Cancer Society reported that there is insufficient evidence to provide specific dietary recommendations for patients with breast, colorectal, lung, and prostate cancer.[7] The panel concluded that properly conducted studies of the effect of nutrition on the prognosis of cancer survivors are needed urgently, and should be a high priority for all academic and research funding agencies.
To assess associations between diet and cancer risk in healthy populations, large prospective studies principally have used semiquantitative food frequency questionnaires to ascertain estimated average long-term dietary habits.[8-12] Participants report their typical intake of commonly consumed foods from which average intake of macronutrients and micronutrients can be calculated. To identify dietary risk factors for various chronic diseases in adults, including cancer and heart disease, Willett et al[13] designed a self-administered, semiquantitative food frequency questionnaire that consists of 131 food items plus vitamin and mineral supplement use that collectively account for more than 90% of the intake of the nutrients assessed. In validation and calibration studies among healthy participants of various races and ethnicities,[13-16] mean calorie-adjusted nutrients as measured by the questionnaire were well correlated with nutrients measured from weekly diet records.[13,17,18] In addition, nutrients measured by the questionnaire correlated with corresponding biochemical indicators measured from plasma and adipose tissue specimens.[19-21] Moreover, evidence that the dietary questionnaire can detect important biologic relationships is provided by studies demonstrating associations between diet and the risk of developing various diseases, including heart disease and cancer.[22]
Although the semiquantitative food frequency questionnaire developed by Willett et al has been validated and calibrated in various healthy populations,[13-16] it remains uncertain whether the instrument is a valid measure of dietary habits in patients with cancer undergoing chemotherapy. Assessment of the utility of a semiquantitative food frequency questionnaire for a given population can be done using various techniques, including 24-hour recalls, diet records, or measurement of corresponding biochemical markers. Studies with biochemical markers, such as plasma carotenoids, tocopherols, and nonendogenous fatty acids,[20,21,23-25] have the distinct advantage of being objective; moreover, such biochemical assessments are subject to measurement errors that are not correlated with those inherent to a semiquantitative food frequency questionnaire.[22]
Cancer patients receiving chemotherapy may experience side effects that influence diet choices, food absorption, and energy balance. To support the future investigation of diet and cancer survival, studies that assess the utility of using semiquantitative food frequency questionnaires among cancer patients are clearly needed. We therefore administered a widely used semiquantitative food frequency questionnaire to a population of patients receiving cytotoxic chemotherapy and compared the results with corresponding measurements in simultaneously collected plasma specimens.
METHODS
Study Population
Patients for this study were enrolled at the Dana-Farber Cancer Institute (Boston, MA) between October 2003 and October 2004. Patients with colorectal, neuroendocrine, or breast cancer receiving cytotoxic chemotherapy were eligible for participation. Patients treated only with hormonal therapy or single-agent targeted therapy (eg, gefitinib, erlotinib, bevacizumab, cetuximab, or trastuzumab) were not eligible, unless they were given a cytotoxic therapy concurrently. Patients had to be receiving chemotherapy for at least 4 weeks and could not have had any major surgery within the 4 weeks before completion of the semiquantitative food frequency questionnaire and plasma collection. All patients were required to have an Eastern Cooperative Oncology Group performance status of 2 or less (ambulatory, up and about at least 50% of waking hours, and able to do self-care) and have an adequate caloric intake (as determined by the treating physician). Patients lacking physical integrity of the upper GI tract or who had a malabsorption syndrome were not eligible. In addition, patients receiving total parental nutrition or tube feedings were not eligible, although oral supplements were allowed. Patients receiving a cholesterol-lowering agent (including niacin, gemfibrozil, and 5-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors) were excluded. Participating physicians were provided with their clinic schedule for a given month and indicated which patients met eligibility for the study. Of the patients deemed eligible, 92% agreed to participate and signed informed consent; of those patients, 90% completed all components of the study. The study protocol and informed consent form were approved by the institutional review board of the Office for the Protection of Research Subjects for the Dana-Farber Harvard Cancer Center.
Semiquantitative Food Frequency Questionnaire
All patients completed a single semiquantitative food frequency questionnaire developed, tested, and refined by Willett et al.[1,8,9,13] Most patients completed the questionnaire while in the clinic, during the administration of chemotherapy. The questionnaire included 131 food items, vitamin and mineral supplements, and open-ended sections for other supplements and foods not specifically listed. Our questionnaire was modified to inquire about average food intake during the immediate last 3 months. We inquired about the exact breakfast cereal, multivitamin supplement, margarine, and vegetable oil used for frying or baking. For each food, a commonly used unit or portion size (eg, one egg or slice of bread) was specified, and participants were asked how often, on average during the last 3 months, they consumed that amount of each food. There were up to nine possible responses, which ranged from never to six or more times per day. We computed nutrient intakes by multiplying the frequency of consumption of each food by the nutrient content of the specified portions, using composition values from US Department of Agriculture sources supplemented with other data, including the components of specific vitamins and breakfast cereals.[26] We included supplements to total intake of specific nutrients. All values were energy adjusted using the residuals methods.[27]
Collection and Analyses of Biologic Samples
Patients underwent venipuncture at the time of a clinic visit. When possible, patients were requested to fast before the blood draw and 39% of patients reported no oral intake in the 8 hours before plasma collection. For carotenoid, tocopherol, and fatty acid measurements, blood was collected in an ethylenediaminetetra-acetic acid–containing tube, and plasma was separated from the cellular component by centrifugation, aliquoted, and frozen at –70°C until analysis (which was completed within 1 year of collection). Samples underwent a single thaw for analysis. Total cholesterol, high-density lipoprotein, and triglycerides were measured according to institutional standards at the Dana-Farber Cancer Institute.
Concentrations of carotenoids and tocopherols were measured as described by El-Sohemy et al.[28] Samples were quantitated by high-performance liquid chromatography. The system manager software (D-7000, Version 3.0) was used for peak integration and data acquisition (Hitachi, San Jose, CA). Because lutein and zeaxanthin coelute on the chromatogram, the two are grouped and provided as lutein + zeaxanthin. Overall (between-run) coefficients of variation were all less than 5%.
Plasma fatty acids were determined by extracting and transmethylating the lipids with methanol and sulfuric acid as described by Zock et al.[29,30] After esterification, the fatty acid methyl esters were redissolved in isooctane and quantitated by chromatography. Peak retention times and area percent of total fatty acids were identified by injecting known standards, as described previously.[31] Overall coefficients of variation for the major peaks were approximately 5%.
Ascertainment of Clinical Data
Computerized medical records were reviewed to extract the cancer diagnosis, current chemotherapy regimen, height, weight at time of study participation, weight approximately 4 to 6 weeks earlier, time the patient received chemotherapy, and age. Height and weight were obtained by standard clinical practice guidelines at the Dana-Farber Cancer Institute. Body mass index was calculated as weight in kilograms divided height in square meters. Smoking status was obtained from the self-administered questionnaire that included diet and classified as current, past, or never smoker.
Statistical Analysis
Pearson correlation coefficients were used to assess the relationship between plasma concentrations and dietary intake. To better approximate a normal distribution, all variables, except age and body mass index, underwent natural logarithmic transformation. Consistent with prior analyses, we excluded patients with implausibly high (> 3,500 calories for women; > 4,200 calories for men) or low intakes (< 600 calories for women; < 800 calories for men).[32] We adjusted for potential confounders using the PROC CORR PARTIAL procedure in SAS version 9.1 (SAS Institute, Cary, NC). For fatty acids, we used the percent of total fat intake made up by a particular fatty acid. Previous studies have estimated that 100 to 200 participants can provide sufficiently stable correlation coefficients comparing the food frequency questionnaire with a comparison method.[22] Two hundred twenty-seven patients were enrolled onto the study. Four patients were withdrawn due to ineligibility (all were not receiving a cytotoxic chemotherapy), 23 did not complete all components of the study, and eight patients were excluded because of implausible caloric intake. Thus the final analysis cohort was 192 patients.
RESULTS
Baseline Characteristics
One hundred ninety-two patients were included in our analyses ([Table 1]). Median age for the population was 55 years, and the majority of patients had either breast cancer or colorectal cancer treated with a variety of cytotoxic chemotherapy regimens. Given that rapid changes in energy and nitrogen balance may influence the correlation between self-reported dietary intake during a 3-month period and single biomarker measurements, we determined weight changes during a 4- to 6-week period. Notably, the majority of participants had maintained a relatively stable weight; 95% of patients experienced a weight change of less than 3 kg in the 4 to 6 weeks before study enrollment.
Correlations of Carotenoids and Tocopherols
We calculated crude and multivariate Pearson correlation coefficients between energy-adjusted dietary micronutrients as measured by the semiquantitative food frequency questionnaire and corresponding biomarkers measured in plasma specimens. Consistent with prior similar studies, we adjusted correlations for age, body mass index, smoking status, and total cholesterol.[21] In addition, we included sex and weight change in the models. Because the majority of patients were not able to fast before sample collection, we did not adjust for plasma triglyceride levels. However, when we limited analyses to the 74 patients who fasted at least 8 hours, the correlations were not substantially different (data not shown). Results for various carotenoids and tocopherols as measured by the questionnaire were significantly correlated with plasma determinations ( [Table 2]). These correlations are comparable with those obtained from other studies of healthy populations ([Table 3]). For example, we found a correlation of 0.33 for beta-carotene (?-carotene), compared with a range of 0.22 to 0.39 in other studies.[21,23-25,33] Similarly, for alpha-tocopherol (-tocopherol), our correlation of 0.34 is within the range of 0.25 to 0.51 demonstrated by others.[21,24,33,34] Of note, the negative correlation between plasma gamma-tocopherol and total vitamin E intake (r = –0.40) has been observed by Ascherio et al[21] (r = –0.51), and is consistent with the experimental observation that -tocopherol supplements decrease plasma -tocopherol.[37]
Correlations of Fatty Acids
We similarly calculated crude and multivariate Pearson correlation coefficients ([Table 4]) between dietary fatty acids (as percentage of total fat intake) and the corresponding plasma fatty acids (as percentage of total fatty acids in plasma). These correlations for various fatty acids are comparable with those obtained from other studies of healthy populations ([Table 3]). For example, our correlations of 0.44 for total omega-3 and 0.42 for docosahexaenoic acid compare with those reported by others who measured plasma fatty acids (0.51[35] and 0.42,[20] respectively). Restriction to patients who were fasting did not appreciably change these results (data not shown).
Influence of Time on Chemotherapy
We modified the existing semiquantitative food frequency questionnaire by assessing average food intake during the previous 3 months (versus the typical 12 months). Patients in this trial were eligible if they had been receiving chemotherapy for at least 4 weeks. Given that this was a cross-sectional study among patients who received chemotherapy for various lengths of time, we confirmed that such variability did not influence our results. The correlations for carotenoids, tocopherols, and fatty acids do not vary substantially when we stratified the analyses by patients receiving chemotherapy for less than 3 (33% of cohort) or more than 3 months (67% of cohort). For example, adjusted correlations for -carotene were 0.39 (P = .003) for those receiving chemotherapy less than 3 months compared with 0.31 (P = .0005) for those treated for more than 3 months. Similarly, adjusted correlations for trans-fat were 0.54 and 0.56 (P < .0001 for both), respectively.
DISCUSSION
In a population of patients who were actively receiving cytotoxic chemotherapy for various solid tumor malignancies, we found that levels of nutrients as measured by a self-administered semiquantitative food frequency questionnaire were significantly correlated with the corresponding biomarkers measured from plasma. The correlations found for carotenoids, tocopherols, and fatty acids are comparable to those demonstrated using the same instrument in healthy populations without cancer. Our results suggest that this semiquantitative food frequency questionnaire may be an accurate instrument to assess dietary intake in cancer patients undergoing chemotherapy.
Within our study population, we considered other methods for validation of our semiquantitative food frequency questionnaire, such as diet records that would be maintained by each subject or 24-hour food recall surveys. These methods have the ability to correlate more nutrients because there are only a limited number of nutrients that have corresponding biochemical parameters that are sensitive to dietary intake. However, in a population of patients with cancer, diet records or recall surveys may be impractical and potentially less valuable. Because many of the chemotherapy regimens are given either on a weekly or every-other-week schedule, we would anticipate that daily intakes of patients would vary widely depending on toxicities experienced from the therapy. Moreover, diet records and 24-hour diet recall surveys are subject to sources of measurement error that are similar to those for a semiquantitative food frequency questionnaire and thereby may create artifactually positive correlations. In contrast, sources of measurement error for plasma biomarkers are not typically correlated with those found in a semiquantitative food frequency questionnaire.[22]
In our study, we prospectively collected information on known and potential confounders of the studied biomarkers, including smoking status, total cholesterol, body mass index, and age. Fasting triglycerides may also be a useful covariate in analyses of tocopherols. However, because all of our patients were actively receiving chemotherapy, less than one half were able to fast for more than 8 hours before plasma collection. Chemotherapy may affect plasma triglyceride levels in particular, leading to less accurate measurements, whereas cholesterol levels appear to be less affected.[38-42] Similarly, few studies have suggested that cancer therapies (including surgery, chemotherapy, and radiotherapy) may influence plasma tocopherols and carotenoids levels.[43-45] However, the comparable correlations between our study and prior reports suggests these biomarkers are still reasonable to use in this study of patients receiving chemotherapy.
Previous studies have conducted repeated administrations of the questionnaire as well as repeated plasma collections to reduce measurement error. Although we did not perform similar corrections in measurement error, our findings are nonetheless consistent with those studies that used repeated determinations of diet and biochemical parameters.[20,21,25,34-36]
Specific carotenoids and tocopherols may be associated with the development of a variety of malignancies,[46-52] and preclinical data suggest that tocopherols may influence treatment-related toxicities.[53-55] -Tocopherol is likely the predominant component of vitamin E supplementation, and our correlation for -tocopherol (r = 0.34) was consistent with correlations observed in other validation studies within healthy populations (r = 0.25 to 0.51). Moreover, prospective studies in previously healthy populations using this semiquantitative food frequency questionnaire have found significant associations between vitamin E intake and the risks of coronary heart disease[56] and possible risk of certain cancers.[57]
The correlations using plasma biomarkers (r = 0.3 to 0.6) may be considered modest.[58] However, other factors beyond dietary intake, such as variations in absorption and host metabolic parameters, will influence the levels of plasma biomarkers.[22] Thus, our findings do not allow a quantitative estimate of the correlation between our food frequency instrument and true intake. A prior study used the semiquantitative food frequency questionnaire, multiple 24-hour recalls, and biochemical markers (including certain carotenoids and fatty acids) to estimate the correlation of each with true intake using a triangulation method. Correlations between the semiquantitative food frequency questionnaire and the biochemical indicators were similar to those that we report in this study, and the triangulation analysis suggested that the correlations of the semiquantitative food frequency questionnaire intake with true intake were generally stronger than the correlations between the biochemical measurements and true intake.
We conducted this study in a population of patients with three different cancer types on a variety of chemotherapy regimens. Although we believe that our findings are generalizable to other cancer populations, the utility of a semiquantitative food frequency questionnaire in measuring between-subject variations in patients with cancers that may affect food absorption (eg, esophageal and gastric) is not addressed. Furthermore, all patients in this study were deemed to have adequate nutritional intake by their physicians, primarily on the basis of relative weight stability. Although this clearly is a subjective and inaccurate measure, only two patients had implausibly low total calorie counts and 95% of patients had relatively stable weight in the 4 to 6 weeks before enrollment. Nonetheless, use of a semiquantitative food frequency questionnaire to describe between-person variations in cancer patients with inadequate caloric intake cannot be addressed in this study.
Nutritional therapies are the most common form of complementary therapy among cancer patients.[7] After a diagnosis of cancer, patients are highly motivated to seek information about diet, physical activity, dietary supplement use, and nutritional complementary therapies. Studies among patients with breast cancer indicate that a majority are interested in making healthful changes in their diets.[59-62] In one study, 52% of cancer patients wanted nutritional guidance at the time of diagnosis or soon after, although few participants reported ever receiving dietary recommendation from their physicians.[59] Unfortunately, few data exist in the literature to enable firm conclusions to be drawn.[7] Confirmation of the utility of a semiquantitative food frequency questionnaire in this population has important implications. Dietary intake may affect treatment-related toxicities, effectiveness of adjuvant therapy for patients with curable disease, and progression-free and overall survival in patients with metastatic disease on long-term therapy. These results confirm that such a questionnaire can offer an informative assessment of dietary intake for the purpose of such future studies in such patients. Ongoing and completed cooperative group trials sponsored by the National Cancer Institute use this questionnaire to assess the influence of dietary intake on treatment-related toxicities and survival in patients with regional and advanced cancers. The data provided in the current analysis support the ability of measuring dietary intake in patients receiving chemotherapy and will allow additional investigation into the relationship between nutrition and outcomes in patients with cancer.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Jeremy Furtado for assistance in biomarker analyses and Laura Sampson for assistance in nutritional analyses.
NOTES
Supported in part by a K07 award from the National Cancer Institute (K07CA097992) and an American Society of Clinical Oncology career development award (J.A.M.).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Willett WC: Diet and cancer. Oncologist 5:393-404, 2000
La Vecchia C, Altieri A, Tavani A: Vegetables, fruit, antioxidants and cancer: A review of Italian studies. Eur J Nutr 40:261-267, 2001
Dagnelie PC, Schuurman AG, Goldbohm RA, et al: Diet, anthropometric measures and prostate cancer risk: A review of prospective cohort and intervention studies. BJU Int 93:1139-1150, 2004
Giovannucci E: Nutritional factors in human cancers. Adv Exp Med Biol 472:29-42, 1999
Blackburn GL, Copeland T, Khaodhiar L, et al: Diet and breast cancer. J Womens Health (Larchmt) 12:183-192, 2003
Thomson CA, LeWinn K, Newton TR, et al: Nutrition and diet in the development of gastrointestinal cancer. Curr Oncol Rep 5:192-202, 2003
Brown JK, Byers T, Doyle C, et al: Nutrition and physical activity during and after cancer treatment: An American Cancer Society guide for informed choices. CA Cancer J Clin 53:268-291, 2003
Willett WC, Reynolds RD, Cottrell-Hoehner S, et al: Validation of a semi-quantitative food frequency questionnaire: Comparison with a 1-year diet record. J Am Diet Assoc 87:43-47, 1987
Willett WC: Future directions in the development of food-frequency questionnaires. Am J Clin Nutr 59:171S-174S, 1994
Slattery ML, Dyer A, Jacobs DR Jr, et al: A comparison of two methods to ascertain dietary intake: The CARDIA Study. J Clin Epidemiol 47:701-711, 1994
Block G, Hartman AM, Dresser CM, et al: A data-based approach to diet questionnaire design and testing. Am J Epidemiol 124:453-469, 1986
Caan BJ, Slattery ML, Potter J, et al: Comparison of the Block and the Willett self-administered semiquantitative food frequency questionnaires with an interviewer-administered dietary history. Am J Epidemiol 148:1137-1147, 1998
Willett WC, Sampson L, Stampfer MJ, et al: Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 122:51-65, 1985
Subar AF, Thompson FE, Kipnis V, et al: Comparative validation of the Block, Willett, and National Cancer Institute food frequency questionnaires: The Eating at America's Table Study. Am J Epidemiol 154:1089-1099, 2001
Longnecker MP, Lissner L, Holden JM, et al: The reproducibility and validity of a self-administered semiquantitative food frequency questionnaire in subjects from South Dakota and Wyoming. Epidemiology 4:356-365, 1993
Hernandez-Avila M, Romieu I, Parra S, et al: Validity and reproducibility of a food frequency questionnaire to assess dietary intake of women living in Mexico City. Salud Publica Mex 40:133-140, 1998
Willett WC, Sampson L, Browne ML, et al: The use of a self-administered questionnaire to assess diet four years in the past. Am J Epidemiol 127:188-199, 1988
Rimm EB, Giovannucci EL, Stampfer MJ, et al: Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am J Epidemiol 135:1114-1136, 1992
Willett WC, Stampfer MJ, Underwood BA, et al: Validation of a dietary questionnaire with plasma carotenoid and alpha-tocopherol levels. Am J Clin Nutr 38:631-639, 1983
Ma J, Folsom AR, Eckfeldt JH, et al: Short- and long-term repeatability of fatty acid composition of human plasma phospholipids and cholesterol esters: The Atherosclerosis Risk in Communities (ARIC) Study Investigators. Am J Clin Nutr 62:572-578, 1995
Ascherio A, Stampfer MJ, Colditz GA, et al: Correlations of vitamin A and E intakes with the plasma concentrations of carotenoids and tocopherols among American men and women. J Nutr 122:1792-1801, 1992
Willett WC: Nutritional Epidemiology (ed 2). New York, NY, Oxford University Press, 1998, pp 101-147
McNaughton SA, Marks GC, Gaffney P, et al: Validation of a food-frequency questionnaire assessment of carotenoid and vitamin E intake using weighed food records and plasma biomarkers: The method of triads model. Eur J Clin Nutr 59:211-218, 2005
Newby PK, Hu FB, Rimm EB, et al: Reproducibility and validity of the Diet Quality Index Revised as assessed by use of a food-frequency questionnaire. Am J Clin Nutr 78:941-949, 2003
Resnicow K, Odom E, Wang T, et al: Validation of three food frequency questionnaires and 24-hour recalls with serum carotenoid levels in a sample of African-American adults. Am J Epidemiol 152:1072-1080, 2000
U.S. Department of Agriculture ARS: USDA Nutrient Database for Standard Reference, Release 10: Nutrient Data Laboratory, 1993. http://www.nal.usda.gov/fnic/foodcomp]
Willett W, Stampfer MJ: Total energy intake: Implications for epidemiologic analyses. Am J Epidemiol 124:17-27, 1986
El-Sohemy A, Baylin A, Kabagambe E, et al: Individual carotenoid concentrations in adipose tissue and plasma as biomarkers of dietary intake. Am J Clin Nutr 76:172-179, 2002
Zock PL, Gerritsen J, Katan MB: Partial conservation of the sn-2 position of dietary triglycerides in fasting plasma lipids in humans. Eur J Clin Invest 26:141-150, 1996
Zock PL, Mensink RP, Harryvan J, et al: Fatty acids in serum cholesteryl esters as quantitative biomarkers of dietary intake in humans. Am J Epidemiol 145:1114-1122, 1997
Kabagambe EK, Baylin A, Allan DA, et al: Application of the method of triads to evaluate the performance of food frequency questionnaires and biomarkers as indicators of long-term dietary intake. Am J Epidemiol 154:1126-1135, 2001
Wu K, Hu FB, Fuchs C, et al: Dietary patterns and risk of colon cancer and adenoma in a cohort of men (United States). Cancer Causes Control 15:853-862, 2004
Romieu I, Stampfer MJ, Stryker WS, et al: Food predictors of plasma beta-carotene and alpha-tocopherol: Validation of a food frequency questionnaire. Am J Epidemiol 131:864-876, 1990
Parrish LA, Marshall JA, Krebs NF, et al: Validation of a food frequency questionnaire in preschool children. Epidemiology 14:213-217, 2003
Sasaki S, Ushio F, Amano K, et al: Serum biomarker-based validation of a self-administered diet history questionnaire for Japanese subjects. J Nutr Sci Vitaminol (Tokyo) 46:285-296, 2000
Hyman MD, Insull W Jr, Palmer RH, et al: Assessing methods for measuring compliance with a fat-controlled diet. Am J Public Health 72:152-160, 1982
Baker H, Handelman GJ, Short S, et al: Comparison of plasma alpha and gamma tocopherol levels following chronic oral administration of either all-rac-alpha-tocopheryl acetate or RRR-alpha-tocopheryl acetate in normal adult male subjects. Am J Clin Nutr 43:382-387, 1986
Kailajarvi M, Ahokoski O, Virtanen A, et al: Early effects of 5-fluorouracil, epirubicin and cyclophosphamide therapy on common laboratory tests. Anticancer Res 21:2867-2872, 2001
Morimoto K, Koh M: Postoperative changes in serum lipid levels of breast cancer and adjuvant chemotherapy [in Japanese]. Gan To Kagaku Ryoho 31:1517-1524, 2004
Moschovi M, Trimis G, Apostolakou F, et al: Serum lipid alterations in acute lymphoblastic leukemia of childhood. J Pediatr Hematol Oncol 26:289-293, 2004
Ellis PA, Fitzharris BM, George PM, et al: Fasting plasma lipid measurements following cisplatin chemotherapy in patients with germ cell tumors. J Clin Oncol 10:1609-1614, 1992
Alexopoulos CG, Pournaras S, Vaslamatzis M, et al: Changes in serum lipids and lipoproteins in cancer patients during chemotherapy. Cancer Chemother Pharmacol 30:412-416, 1992
Potischman N, Byers T, Houghton L, et al: Effects of breast cancer treatments on plasma nutrient levels: Implications for epidemiological studies. Cancer Epidemiol Biomarkers Prev 1:555-559, 1992
Clemens MR, Ladner C, Ehninger G, et al: Plasma vitamin E and beta-carotene concentrations during radiochemotherapy preceding bone marrow transplantation. Am J Clin Nutr 51:216-219, 1990
Filer LJ Jr, Kibler CE, Mason KE, et al: The influence of surgery of serum tocopherol (vitamin E) and carotenoid levels. N Y State J Med 53:2836-2839, 1953
Gann PH, Ma J, Giovannucci E, et al: Lower prostate cancer risk in men with elevated plasma lycopene levels: Results of a prospective analysis. Cancer Res 59:1225-1230, 1999
Tamimi RM, Hankinson SE, Campos H, et al: Plasma carotenoids, retinol, and tocopherols and risk of breast cancer. Am J Epidemiol 161:153-160, 2005
Sato R, Helzlsouer KJ, Alberg AJ, et al: Prospective study of carotenoids, tocopherols, and retinoid concentrations and the risk of breast cancer. Cancer Epidemiol Biomarkers Prev 11:451-457, 2002
Giovannucci E: A review of epidemiologic studies of tomatoes, lycopene, and prostate cancer. Exp Biol Med (Maywood) 227:852-859, 2002
Epstein KR: The role of carotenoids on the risk of lung cancer. Semin Oncol 30:86-93, 2003
Helzlsouer KJ, Huang HY, Alberg AJ, et al: Association between alpha-tocopherol, gamma-tocopherol, selenium, and subsequent prostate cancer. J Natl Cancer Inst 92:2018-2023, 2000
Comstock GW, Alberg AJ, Huang HY, et al: The risk of developing lung cancer associated with antioxidants in the blood: Ascorbic acid, carotenoids, alpha-tocopherol, selenium, and total peroxyl radical absorbing capacity. Cancer Epidemiol Biomarkers Prev 6:907-916, 1997
Leonetti C, Biroccio A, Gabellini C, et al: Alpha-tocopherol protects against cisplatin-induced toxicity without interfering with antitumor efficacy. Int J Cancer 104:243-250, 2003
Legha SS, Wang YM, Mackay B, et al: Clinical and pharmacologic investigation of the effects of alpha-tocopherol on adriamycin cardiotoxicity. Ann N Y Acad Sci 393:411-418, 1982
Pace A, Savarese A, Picardo M, et al: Neuroprotective effect of vitamin E supplementation in patients treated with cisplatin chemotherapy. J Clin Oncol 21:927-931, 2003
Stampfer MJ, Hennekens CH, Manson JE, et al: Vitamin E consumption and the risk of coronary disease in women. N Engl J Med 328:1444-1449, 1993
Wu K, Willett WC, Chan JM, et al: A prospective study on supplemental vitamin E intake and risk of colon cancer in women and men. Cancer Epidemiol Biomarkers Prev 11:1298-1304, 2002
Cohen J: Statistical Power Analysis for the Behavioral Sciences. Mahwah, NJ, Lawrence Erlbaum Associates, 1988
Demark-Wahnefried W, Peterson B, McBride C, et al: Current health behaviors and readiness to pursue life-style changes among men and women diagnosed with early stage prostate and breast carcinomas. Cancer 88:674-684, 2000
Lee MM, Lin SS, Wrensch MR, et al: Alternative therapies used by women with breast cancer in four ethnic populations. J Natl Cancer Inst 92:42-47, 2000
Monnin S, Schiller MR, Sachs L, et al: Nutritional concerns of women with breast cancer. J Cancer Educ 8:63-69, 1993
Thomson CA, Flatt SW, Rock CL, et al: Increased fruit, vegetable and fiber intake and lower fat intake reported among women previously treated for invasive breast cancer. J Am Diet Assoc 102:801-808, 2002(Jeffrey A. Meyerhardt, De)