Contenuto
COS Study - Rationale and methodology
Main scientific objective
To investigate the interaction between nutritional factors and genetic susceptibility in the occurrence of breast cancer in young women.
Main expected achievements
- Identification of dietary and other life-style changes that may prevent or postpone breast cancer in women belonging to high-risk families,
- development of primary preventive recommendations for high risk families and of strategies for long term dietary intervention trials, and
- development and validation of a software for classifying subjects (and families) as for the opportunity of genetic testing for high penetrance mutations of BRCA genes.
Rationale of the study and main hypotheses
Several environmental factors have been hypothesised to affect the penetrance of hereditary breast cancer genes, including nutritional factors, ionising radiation, tobacco smoking, reproductive factors, oral contraceptives, and treatment with tamoxifen. Nutritional hypotheses refer mainly to the antioxidant effect of fruit and vegetables consumption, to the dietary regulation of the availability of sex steroid hormones and insulin-like growth factors and related peptides, and to dietary components that may affect the expressiono of BRCA proteins, such as soy phytoestrogens and cruciferous indoles. If confirmed, these hypotheses would have a great impact on understanding the mechanisms of carcinogenesis and on the perspective of preventing both genetic and sporadic breast cancers.
Heritable breast cancer is thought to account for 5 to 7% of all breast cancers, and about half of these will be related to BRCA1 or 2 gene mutations. Such mutations confer very high lifetime risks of developing breast cancer, and most of the cases already occur at young ages. Lifetime penetrance estimates have ranged from 36% in population-based series to 85% in women from affected families. Ten to twenty percent of patients who develop breast cancer before the age of 40 are likely to have inherited a deleterious mutation of BRCA1, BRCA2 or other high penetrance genes. Some of these mutations also increase the risk of ovarian cancer, of male breast cancer and, to a lesser extent, prostate, colon and larynx cancer. Ovarian cancer penetrance estimates vary from 12 to 63%.
A sizeable proportion of mutation carriers, however, does not develop breast cancer at all or develop it only late in life. It is reasonable to hypothesise, therefore, that the penetrance of the genetic trait may be regulated through other genetic or non-genetic factors.
The main gene-environment hypotheses that it is intended to test include the protective interaction with the consumption of various vegetables and fruit, dietary sources of phytoestrogens (soy products, other legumes, whole grain cereals and cruciferous vegetables), dietary factors affecting insulin sensitivity (low glycaemic index food, fibres, mono and polyunsaturated fats), and with low body mass index (BMI) and waist to hip ratio (WHR). Secondary aims include testing the interaction with oral contraceptive and radiation exposure in childhood, which have been hypothesised to increase the risk by a greater magnitude in mutation carriers than in non carriers.
Methodology
The C.O.S. study will be based on the collection of a few key information from several thousands European women who got breast cancer before the age of 40. They will be classified according to their probability of carrying a high penetrance mutation on the base of their family history of breast and ovarian cancer. The odds of a given exposure among breast cancer cases who most likely carry an high penetrance mutation (because of a highly predictive family history) will be compared with the odds computed for breast cancer cases who most likely are not mutation carriers (because no other case occurred in an otherwise informative family).
Whenever the odds ratio will come out significantly different from one, the study will indicate a gene-environment interaction, i.e. that the factor under study affects the incidence of genetic cancer by a greater (or lesser) magnitude than the incidence of sporadic cancer. Analyses will also be carried out separately for patients positive and, respectively, negative for oestrogen receptors.
The proposal has three major scientific and technological innovative aspects:
- the case-only methodology, which is well established in genetic epidemiology but has never been used to address the effect of dietary factors on cancer gene penetrancea reliable methodology of classifying cases as BRCA mutation carriers or non carriers
- a validated methodology for the assessment of dietary habits in the past (before the diagnosis of breast cancer).
1) The logic of case-only studies can be summarised as follows:
| Exposed | Unexposed | |
|---|---|---|
| Genetic Cases | A | B |
| Sporadic Cases | C | D |
| Source Population | E | F |
The case-only odds ratio (AD/BC) can be viewed as the ratio of two risk ratios, i.e. [(A/E)/(B/F)] / [(C/E)/(D/F)]. It expresses how greater is the association of the environmental exposure with genetic cancers than with sporadic cancers, that is the statistical interaction between the exposure and the genetic trait. Such a study design, therefore, cannot identify risk factors that affect the occurrence of both genetic and sporadic cases by the same magnitude (on the Relative Risk scale), but only departures from the multiplicative joint effect of the genotype and the exposure. Such departures, however, are predicted by the major dietary hypotheses of the study. The method assumes that there is independence between environmental exposure and genotype, i.e. that genotype does not determine whether a woman will be more or less likely to have any of the exposures under study.
2) A major requirement of a case-only study is an unbiased classification of genetic and sporadic cancers. Several methods have been used to estimate the probability of carrying a BRCA mutation, based on the number of breast and ovarian cancer among 1st and 2nd degree relatives, the age of occurrence, the presence of bilateral breast cancers and, when relevant, ethnicity. Most methods, however, do not take into account the size of the family (namely the number and ages of healthy relatives), the structure of the pedigree, nor the change of cancer risk over subsequent generations. These methods are suitable for selecting patients for genetic testing, a clinical use that does not require high specificity, but would not be suitable for classifying patients in a C.O.S. study. G. Parmigiani developed a method that properly incorporates the whole information available from each family member. The probability that a given family member is a carrier is obtained from Bayes's theorem, using the mutation prevalence in the population as the prior distribution and the family history as the evidence, taking into account mendelian transmission pathways. The weight of the maternal grandmother or aunt with breast cancer, for instance, will be higher if the mother had breast cancer too (or if she died when still too young to have breast cancer) than if the mother reached adult or old age without developing the disease. The results of the Parmigiani method are quite sensitive to varying assumptions about the age specific incidence function of breast and ovarian cancer in mutation carriers and in the general population. In the C.O.S. project the method will be improved using generation specific incidence functions estimated from age-period-cohort models, i.e. a different function for each family member. Cumulative BC risk of a European woman born in the fifties, in fact, is about three times higher than the risk of her grandmother born at the beginning of the 20th century. A similar cohort effect has also been shown for mutation carriers (Such a finding corroborates the study hypothesis that environmental factors modify the penetrance). With respect to the Parmigiani software, which incorporates information on first and second degree relatives, a further improvement of the C.O.S. software will be the possibility of taking into account more distant relatives such as cousins (third degree) and parent cousins (fourth degree). The present version of the C.O.S. software, which provides also the probability of mutation for each family member, is freely available for family clinics that want to test it. Further updating will be available for users that want to share with the C.O.S. working group their family history information to improve the penetrance estimates. The basic output of the software is the probability that a C.O.S. participant has inherited a BRCA mutation given her family history (90% in the following figure). For use in the family clinic, however, a specific probability is computed for each relative (indicated by different colors in the figure)
3) A further critical aspect of the C.O.S. study is that the dietary investigation must classify patients as for their usual diet before the diagnosis of breast cancer. For most cases the interval between diagnosis and recruitment in the C.O.S. study will be a few months to a few years; in a minority of cases, however, this interval will be many years. A major limitation of questionnaires aiming at describing dietary habits in the past is that memory may be distorted by more recent habits. Studies on past diet of women that were already investigated on their current diet in the past showed that the results of the present questionnaires on past diet are more correlated to present diet than to the past one as originally recorded. Such misclassification of past diet would bias the results of epidemiological studies towards no association (in the C.O.S. study towards no interaction) but is unlikely to generate any spurious association (because it will equally effect women with genetic and sporadic cancer). As the C.O.S. questionnaire will refer to a precisely defined period in the past (the year before diagnosis), the recall is likely to be easier then just enquiring on the diet many years ago as evaluated in previous studies. To validate this hypothesis 100 women who participated in a dietary intervention study and who had filled in a semiquantitative food questionnaire at baseline (before intervention) have been sent the same questionnaire 5 years later to enquire on the usual diet before the dietary intervention. Preliminary analyses show quite satisfactory results on nutrients (Spearman's rho usually above 0.7). Correlations for single food items range between 0.3 to 0.76 and are fairly good for many relevant food. Women seem to be able to recall their usual diet before a major event that effects dietary habits, such as the enrolment in a dietary intervention study. This suggests that C.O.S. participants will be able to recall their dietary habits before breast cancer diagnosis. The dietary questionnaire used in this study was developed and validated in the frame of the European Prospective Investigation into Cancer and Nutrition (the EPIC study). In the C.O.S. project it is planned to use the same questionnaire and to expand the validity study to a larger number of women.
EU project, Quality of Life, Key action 1-Health, Food and Environment,
QLK1-2000-466-COS
![Pagina validata XHTML 1.0 Strict! Verifica con la validazione automatica [link esterno, apre una nuova finestra]](/istituto/ico/valid-xhtml10.gif)
![CSS Validato! Verifica con la validazione automatica [link esterno, apre una nuova finestra]](/istituto/ico/w3c_css_ico.bmp)
