Comprehensive assessment, validation and causal identification of dietary exposures and cancer risk

  • Topic: Combination of Cancers
  • Institution: University of Ioannina School of Medicine
  • Country: Greece
  • Status: Ongoing

Scientific abstract

(View plain language abstract)

Background

There is ample evidence suggesting that cancer is largely a set of preventable diseases. Cancer incident rates differ remarkably worldwide, suggesting an important role of lifestyle factors on cancer risk. It is widely accepted that nutrition-related factors such as obesity and physical activity play an important role in the occurrence of several cancers. However, the evidence for specific foods or nutrients affecting cancer risk is largely inconsistent with few exceptions according to the World Cancer Research Fund. A potential reason for these inconsistencies is that foods and nutrients are strongly correlated, and it is difficult to decipher their independent effects. In addition, current efforts to associate single nutrients with cancer risk are susceptible to selection effects and may create spurious claims of association. There is still an urgent need to identify modifiable dietary factors for the primary prevention of cancer. Therefore, we propose to systematically evaluate and validate the association between a wide range of foods and nutrients and cancer risk.

Hypothesis and Objectives

We will perform nutrient-wide association studies (NWAS) for several food and nutrient intakes in relation to risk of lung, breast, colorectal and prostate cancer. We will evaluate the bias and credibility in the existing observational literature of dietary intakes and cancer risk, and compare the statistically significant findings from the NWAS to the results of this evaluation. To further explain a substantial component of cancer risk, we will evaluate gene-environment interactions between the latter NWAS findings and known genome-wide association study (GWAS) hits in relation to risk of the latter four cancers. Finally, to identify potential causal effects of dietary intakes on risk of cancer, we will perform mendelian randomization analyses for the significant NWAS findings.

Setting and Methods

In the NWAS, which follows the GWAS paradigm, we will evaluate associations for approximately 100 nutrients and foods in relation to risk of the four most common cancers using Cox proportional hazards models adjusting for known or suspected risk factors for each of the cancers while controlling for multiple comparisons in the European Prospective Investigation into Cancer and Nutrition (EPIC). Significant findings will be then validated in the independent Netherlands Cohort Study. In the second aim, we will identify all published studies on dietary intakes and cancer risk, and apply two novel and state-of-the-art approaches (excess significance test, sensitivity analyses with credibility ceilings) to evaluate bias and credibility in this literature. The gene-environment interaction analyses and the mendelian randomization will be performed in EPIC and in three genetic consortia (International Lung Cancer Consortium [ILCCO], United States National Cancer Institute Breast and Prostate Cancer Cohort Consortium [BPC3] and Genetics and Epidemiology of Colorectal Cancer Consortium [GECCO]), which combine resources from many large and well-established cohort studies.

Impact

These studies will critically assess the current literature of dietary intakes and cancer risk, provide complementary and unbiased evidence for the potential role of a range of foods and nutrients in relation to lung, breast, colorectal and prostate cancer risk, and maybe assist in identifying novel dietary strategies for cancer prevention.


Plain language abstract

Background

There is ample evidence suggesting that cancer is a preventable disease. Lifestyle factors have been shown to affect cancer development and among those, obesity and lack of physical activity play an important role in the occurrence of several cancers. However, the evidence for specific foods or nutrients affecting cancer risk is largely inconsistent in the scientific literature. A potential reason for these inconsistencies is that foods and nutrients are strongly inter-related, and it is difficult to decipher their independent effects. However, there is still an urgent societal need to identify specific modifiable dietary factors for the prevention of cancer. Therefore, we propose to systematically evaluate and validate the association between a wide range of foods and nutrients and cancer risk.

Aims and Objectives

Our aim is first to identify the potential reasons behind the inconsistencies of the scientific literature between diet and cancer risk. We hypothesize that part of this literature is biased. Therefore, we will systematically collect and critically study this literature. In addition, we will perform new association studies using bias-free techniques for several food and nutrient intakes in relation to risk of the four most common cancers i.e., lung, breast, colorectal and prostate cancer. We will also evaluate whether certain foods or nutrients act together with variations in an individual's genome to further increase or reduce cancer risk.

How It Will Be Done

First, we will search and identify all published meta-analyses of dietary intakes and cancer risk, and apply two novel and state-of-the-art statistical approaches to evaluate bias and credibility in this literature. Second, we will test associations for approximately 100 nutrients and foods in relation to risk of lung, breast, colorectal and prostate cancer using bias-free epidemiological and statistical techniques in a large European study, and validate the findings in another large European study. Finally, the interactions between the diet and the genome will be studied in three genetic study consortia, which combine resources from many and well-established epidemiological studies.

Potential Impact

This project will critically assess the current literature of dietary intakes and cancer risk, provide complementary and unbiased evidence for the potential role of a range of foods and nutrients in relation to lung, breast, colorectal and prostate cancer risk, and maybe assist in identifying novel dietary strategies for cancer prevention.