Childhood Cancer 2016
Risk factors for childhood cancer: what are the challenges in research?
Dr Kurt Straif - IARC, Lyon, France
I will be talking about the established and suspected risk factors for childhood cancers and particularly for childhood leukaemias and also brain cancers.
Which ones are the most established?
This is based on evaluations of the IARC Monographs Programme and there is clear evidence that environmental tobacco smoke, in fact that is smoking exposure from the mother or the father, can lead to an increased risk of a rare childhood cancer of the liver, that is hepatoblastoma. There’s also some evidence but not yet sufficient at the last evaluation for certain types of childhood leukaemias, particularly acute lymphoblastic leukaemias.
There is also other evidence, particularly on ionising radiation that is linked to childhood leukaemias and then there is a series of other factors where there is, in the IARC terminology, limited evidence. That means there is a credible association but in the end chance bias confounding could not be ruled out with reasonable confidence, such as exposure from painting which has some biological plausibility due to the solvents and benzene, for example, is a well-known human carcinogen for adult leukaemias. Then there is non-ionising radiation - the extremely low-frequency electromagnetic fields that have been linked in epidemiological studies to childhood leukaemias. However, at the time the mechanism wasn’t understood at all, in fact there was no idea how this could be caused and therefore it was classified as possibly carcinogenic to humans.
What are some of the challenges in ongoing research?
This talk was specifically about the challenges in ongoing research for childhood cancers and there we have to consider, first of all, that childhood cancer is a very rare disease as compared, for example, to cancers in adults. This imposes particular challenges - typically with very rare outcomes you would want to do a case control study but then with a case control study and traditional assessment of the exposure you would always have some concern about recall bias. Therefore if you want to study that in cohort studies with a prospective assessment of the exposure that would not be prone to recall bias you need large cohorts or, even better, pooling of cohorts and that is what is going on at the moment, for example the I4C, the International Childhood Cohort Consortium, to study certain questions including environmental exposures in these pooled cohorts.
What else ought we to consider?
Typically the time from the first exposure to the cancer outcome, as we know for cancers in the adults, can be very long, in the extreme up to forty years, for example, for exposure to asbestos in mesothelioma. So in terms of other considerations we also need to take into account the latency, that is the time from the first exposure to a cancer outcome. We know from studies of cancer in adults that this can sometimes be very long, up to forty years. That, of course, has two consequences for studies of childhood cancer: it’s perhaps then not a cancer in children but it could be the exposure in the children that is relevant for a cancer later in life and we should perhaps not only consider childhood cancers as such but also critical exposures during childhood. On the other hand it is also a possibility due to other exposure effects, meaning that the same exposure in children could have a stronger effect in children as compared to adults or, perhaps, shorter latency periods because of increased susceptibility – they are more prone to a cancer due to development of organ systems and other factors. So perhaps with a shorter latency period there would still be possibilities to find things that you would otherwise not expect in children.