Correlation between nutritional biomarkers and breast cancer: a case-control study.

A case-control study to explore associations between retinoids, tocopherols, total and beta-carotenes and breast cancer was conducted by analysing concentrations of these biomarkers in adipose tissue, cheek cells and plasma. A total of 414 French-Canadians in Montreal with new diagnoses of breast cancer were age-matched to 429 population-based controls. Subjects were interviewed using a questionnaire, and biological samples from 287 cases and 112 controls were collected within 3 months of the diagnosis. Mean beta carotene concentrations in cheek cells were significantly lower among controls. Odds ratios (ORs) from logistic regression analysis were used to compare higher and lower tercile concentrations. Significant positive associations were observed in adipose tissue for retinoid [OR=2.11; 95% CI (1.09-4.08)] and beta carotene [OR=3.18; 95% CI (1.70-5.93)]; in cheek cells for beta carotene [OR=2.22; 95% CI (1.21-4.50)] and for total carotenes [OR=2.94; 95% CI (1.59-5.42)] and in plasma for beta carotene [OR=1.53; 95% CI (0.80-2.93)] and total carotenes [OR=1.04; 95% CI (0.53-2.05)]. Among the control groups, significant Pearson correlations were observed between cheek cells and adipose tissue for total carotenes (r=0.27; p=0.01) and cheek cells and plasma (r=0.22; p=0.04). In contrast to previous works, this study shows that high concentrations of retinoids and carotenes in adipose tissue and cheek cells are associated with increased risk of breast cancer. However, all these studies are limited by small sample size. Although our study tested a limited number of controls, important associations were observed. These results suggest that the effect of disease on biomarkers is fundamental to the interpretation of epidemiological data. We suggest either that the high levels of these biomarkers found in cancer patients in this study may be due to the disease process that affects the pharmacokinetics of the biomarker or that the disease causes a change in dietary habits. In addition, in studies involving the application of biomarkers to cancer epidemiology it is imperative that a typical biomarker concentration is not associated with breast cancer risk before further examination of the methodological limitations of epidemiological studies investigating this relationship. Therefore, sample size, selection bias, information bias, and confounding should be considered in the design of studies investigating the aetiological relationship between biomarkers and breast cancer.