Introduction

The hypothesis that the intrauterine environment is of importance to the aetiology of certain forms of cancer is not new. Studies of children exposed in utero to radiation in Japan at the end of the Second World War indicate a special vulnerability during the perinatal period for future cancer risks (Kato et al. 1989). Although the increased risk was most pronounced for haematopoietic cancers, an excess number of breast cancers has also been demonstrated. Similarly, strong associations have been shown between exposure to diethylstilbestrol (DES) given to expectant mothers during pregnancy and vaginal adenocarcinoma in young women (Greenwald et al. 1971). These two examples are of major importance as they show that there is a window of vulnerability already ‘open’ in utero.

Although the findings mentioned above prove that there are biological pathways which can act in a cancerogenic process at an early stage in life, the two exposures (DES and radiation on a level with Hiroshima and Nagasaki) are extremely rare. However, in the case of haematopoietic cancers, especially leukaemia in childhood, studies have shown that chromosomal translocations are already present at birth in children that later will manifest with leukaemia (Hjalgrim et al. 2002), indicating that factors other than those mentioned above are of importance in utero. However, in order to study other cancers that will occur in adult life, researchers are facing a formidable challenge. What sort of exposures are we interested in?