[Epidemiological link between eating habits and cancer: the example of colorectal cancer]

One million people worldwide are affected by colorectal cancer (CRC) every year, resulting in 500,000 deaths. Incidence rate varies widely according to areas. It has been much more common in high-income countries. This trend begins to be reversed, with a stabilization of incidence rates in high-income countries and a dramatic increase in developing regions. CRC increase seems to overlap with the increasing level of urbanization in the latter. In the eighties, Doll and Peto already reported that 80% of cancer mortality was avoidable, particularly for CRC for which environmental influence is dominating. Observational studies have reported that migrants tended to reach similar CRC incidence levels that those of natives in the host country. An estimated 60% drop of all CRC cases could be obtained provided eating habits are modified, since these seem to be the main cause of incidence variations. Over the last ten years, many epidemiological studies and meta-analyses have been conducted. However, this abundant literature gave contradictory results. Therefore, in 1997, two groups of experts: the World Cancer Research Fund (WCRF) and the American Institute of Cancer Research (AICR), considered drafting a report on the correlation between diet and cancer a necessity. This systematic literature review, updated in 2007, established that red meat, processed meat and alcohol increase the risk of CRC. Conversely fibers, milk and calcium probably protect against CRC. Considering the most important prospective studies, randomised trials and the 2nd WCRF's report, our article aims at reviewing the evidences about diet and CRC.