Epidemiology of CLL: Interlymph consortium and genome-wide sequencing studies

Prof Claire Vajdic - University of New South Wales, Sydney, Australia

You’ve been doing research on the epidemiology of CLL and I was wondering what led you to do research in this area.

That’s a very good question. In the 1980s and 1990s we saw a 50% increase in the incidence of non-Hodgkin’s lymphoma in Australia and other Western countries and a lot of us set about to do case control studies to try and identify the risk factors behind that increase. None of our studies identified an explanation for that pattern of increase and so we understood that if we were able to pool our data altogether, because we’d all studied non-Hodgkin’s lymphoma and it consists of at least forty different subtypes, if we pooled it together we would have the statistical power to examine associations for individual subtypes and that might reveal more than if we’re analysing them all together because it’s not a single disease.

So specifically what kind of results have you been finding?

Very novel results with respect to a couple of risk factors. The first was ATP, so we saw an inverse risk in association with personal history of atopic disease. That might indicate an association with IgE but we do have a question mark about the possibility of reverse causation with that particular exposure even though we’re seeing some evidence of an inverse association also in cohort studies where we have that longitudinal data. Secondly we saw an inverse association with sun exposure. So this is self-reported outdoor hours, a reduction in risk for those people that had high time outdoors. This obviously implicates ultraviolet light, possibly vitamin D, and there we’ve seen in cohort studies there’s some evidence of also an inverse association with high levels of serum vitamin D. So that’s an interesting lead to follow up.

Another area we’ve seen significant doubling in risk for people who have a first degree family member with a history of a haematological malignancy. So a much higher risk in those with a family member who has leukaemia or a male affected relative. We saw some suggestive evidence with exposure to working or living on a farm and that might implicate exposure to pesticides and herbicides.

What are the next steps in your research? What plans have you got coming?

I should also tell you about our genetic findings. So what we’ve done is we’ve pooled data from case controls and cohort studies and run genome wide association studies. There we’ve found a number of novel single nucleotide polymorphisms or SNPs that significantly predict risk of CLL. They point to a number of biological mechanisms and most prominently a role for apoptosis and telomere length, so that’s really interesting. That’s being followed up in the laboratory with functional studies, we’re also going to be doing family studies and when we have enough data looking at the interaction between genetic mutations and environmental exposures because that may reveal some really interesting associations there and lead us to better understand the biology of this condition.

Can you tell me more about your international consortium, is it Interlymph?

Yes, Interlymph was established in 2001 so that was the coming together of all those individual case control studies that could not individually solve this problem. We came together, grouped our data so basically we share all our epidemiological data and our DNA data at the individual level. So we’ve shipped the data to a central location, the Mayo Clinic in the United States, where all our risk factor information has been harmonised and is available for any of the consortium members to use to test their hypotheses.

What do you think the future holds for those sorts of research studies?

They’re really fascinating areas of work. So the genome wide studies at the moment can identify common mutations and the next steps really are looking at the rarer mutations. That’s where I think we’re going to be filling in a lot of the gaps because at the moment we can only explain about 17% of the hereditability of CLL. So we’ve got a lot of missing information there and the whole genome work will probably identify some strong leads, as will the epigenetics and methylation studies that are underway.