Plasma carotenoids as biomarkers of intake of fruits and vegetables: individual-level correlations in the European Prospective Investigation into Cancer and Nutrition (EPIC).
Al-Delaimy WK., Ferrari P., Slimani N., Pala V., Johansson I., Nilsson S., Mattisson I., Wirfalt E., Galasso R., Palli D., Vineis P., Tumino R., Dorronsoro M., Pera G., Ocké MC., Bueno-de-Mesquita HB., Overvad K., Chirlaque M., Trichopoulou A., Naska A., Tjønneland A., Olsen A., Lund E., Alsaker EH., Barricarte A., Kesse E., Boutron-Ruault MC., Clavel-Chapelon F., Key TJ., Spencer E., Bingham S., Welch AA., Sanchez-Perez MJ., Nagel G., Linseisen J., Quirós JR., Peeters PH., van Gils CH., Boeing H., van Kappel AL., Steghens JP., Riboli E.
OBJECTIVE: The aim in this study was to assess the association between individual plasma carotenoid levels (alpha-carotene, beta-carotene, lycopene, beta-cryptoxanthin, lutein, zeaxanthin) and fruit and vegetable intakes recorded by a calibrated food questionnaire (FQ) and 24-h dietary recall records (24HDR) in nine different European countries with diverse populations and widely varying intakes of plant foods. DESIGN: A stratified random subsample of 3089 men and women from nine countries participating in the European Prospective Investigation into Cancer and Nutrition (EPIC), who had provided blood samples and dietary and other lifestyle information between 1992 and 2000, were included. RESULTS: beta-Cryptoxanthin was most strongly correlated with total fruits (FQ r = 0.52, 24HDR r = 0.39), lycopene with tomato and tomato products (FQ r = 0.38, 24HDR r = 0.25), and alpha-carotene with intake of root vegetables (r = 0.39) and of total carrots (r = 0.38) for FQ only. Based on diet measured by FQ and adjusting for possible confounding by body mass index (BMI), age, gender, smoking status, alcohol intake, and energy intake, the strongest predictors of individual plasma carotenoid levels were fruits (R(partial)(2) = 17.2%) for beta-cryptoxanthin, total carrots ((partial)(2) = 13.4%) and root vegetables (R(partial)(2) = 13.3%) for alpha-carotene, and tomato products (R(partial)(2) = 13.8%) for lycopene. For 24HDR, the highest R(partial)(2) was for fruits in relation to beta-cryptoxanthin (7.9%). CONCLUSIONS: Intakes of specific fruits and vegetables as measured by food questionnaires are good predictors of certain individual plasma carotenoid levels in our multicentre European study. At individual subject levels, FQ measurements of fruits, root vegetables and carrots, and tomato products are, respectively, good predictors of beta-cryptoxanthin, alpha-carotene, and lycopene in plasma.