Genomic trends across the African Diaspora for breast cancer

Prof Melissa B Davis - Weill Cornell Medical College, New York, USA

For this particular meeting I spoke about our international breast cancer research where it’s actually an organisation called the International Centre for the Study of Breast Cancer Subtypes. I serve as a scientific director. We conduct breast cancer research across the African diaspora specifically with the intention of identifying genetic and molecular phenotypes that are specific to the West African diaspora and hopefully to explain some of the disparities in cancer mortality that we see in breast cancer.

So today I talked about the organisational structure, some of our infrastructure in collecting samples, the partnerships we have with our African countries, the capacity building we do there. Then I also presented some emerging findings from the work that we do with the cohort.

What collaborations do you have in Africa?

I would be remiss not to mention that I work with the Chief of Breast Surgery, Lisa Newman, we were both recruited there this past year. She, over the last 17 years, has been working with various teaching hospitals throughout Ghana. She also works with a teaching hospital in Ethiopia. So the sites in Ghana include Kumasi Teaching Hospital and some of its satellite clinics. In Ethiopia it’s St Paul’s Millennium Institution or medical college and also some of their satellite hospitals and clinics as well. We have site coordinators and site principal investigators; typically the site principal investigators will be the Chief Medical Officers and the Chief of Surgery for those institutions. Then we also work with the departments of pathology to provide reagents, to upscale or augment the quality of their tissue preservation and subtyping for tumours as well.

Can you outline some of the emerging data?

As a geneticist, one of the things that I’m very interested in is how the shared West African genetic ancestry of the African-Americans in the United States, because the large majority of the African-Americans in the United States are descendants of African slaves that were brought over from the transatlantic slave trade. So I’m interested in identifying some of the shared ancestry predispositions that come hand in hand with our West African ancestry.

So Dr Newman previously observed and published prolifically on different phenotypes that she saw from surgeries she performed in patients who had breast cancer throughout West Africa. Then when she compared those phenotypes to East Africa she saw that there was a distinct pattern of tumour biology in women in West Africa currently that also paralleled the tumour biology that we see in African-Americans in the United States. So, for instance, triple negative breast cancer is the most prevalent subtype of breast cancer that’s diagnosed in African-American women in the United States. If you look across the board it’s probably about 40% chance that if a black woman has breast cancer it will be triple negative. In West Africa, in Ghana and Nigeria for instance, it’s upwards of 80% so the predominant type of breast cancer that’s diagnosed in West Africa is triple negative.

So what we started to uncover are what we call population specific alleles in certain genes. These are mutations in genes that exist only in that population in sub-Saharan West Africa that are linked to predisposition for triple negative breast cancer. One of those genes is called DARC, so the title of my talk today was ‘The DARC Side of Breast Cancer’ because I’m a sci-fi fanatic. What we study about the DARC gene is the fact that there is a very famous mutation called Duffy null in the gene and it confers a resistance to certain species of malaria pathogens. So it’s fixed in sub-Saharan Africa. The result of the mutation is that it’s removed from red blood cells, which is why it’s a blood group antigen, but what we identified in terms of cancer context is that tumours that express the gene have a better or just higher prevalence of immune cell infiltration. So we identified that on the RNA level using some transcriptomics and so now what we’re starting to do is look within the tumours on a spatial proteomic level to actually see that tumours that express DARC have better infiltration in the tumour space whereas tumours that don’t express DARC, although there may be tumour associated leukocytes in the region of the tumour, in the microenvironment of the tumour, they are sequestered or compartmentalised into the stromal area of the tumour and they don’t infiltrate into the epithelium.

So what we’re trying to determine is if that is because of the Duffy null mutation, for instance, in the gene. Does that track with whether or not the tumour is expressing in the gene? It appears it probably does but we’re still working on a mouse model, for instance, to definitively determine that.

Why is it important to compare populations?

Just generally as a scientist who is interested in disparities across an admixed population it’s very important to deconvolve the genetic diversity that’s in that admixed population. So to date most GWA studies that look for genetic associations with breast cancer predominantly only looked at European ancestry populations. More recently there have been some consortia that have brought together smaller cohorts to create a larger cohort to replicate some of those GWA studies of African-Americans for instance. They, by far, have identified some alleles that appear to be associated with breast cancer in those populations. But mechanistically they haven’t necessarily been linked to a why.

It’s important, some of the presumption behind that is it’s linked to their West African ancestry, that the reason you could find these genetic associations in African-Americans is because they have a shared ancestry. But that hasn’t actually been connected directly to the ancestral population. So, as a geneticist, I think it’s very important that we don’t presume because we have the tools now to actually go to the ancestral population, identify the risks in that population and, if the theory is right, that the shared ancestry in African-Americans is what’s causing triple negative breast cancer incidence to be higher in that population then we should be able to identify in the ancestral population and then see it imprinted in the admixed population.

So I think it’s important in that regard but in terms of the clinical impact immunotherapy is one of the most hottest therapies pan-cancer. Regardless of where your tumour is, everyone wants this super-immune therapy treatment that looks like… it’s amazing when it works. But in some people it doesn’t work at all and so while we stratify the population to say it works for you, sorry, it doesn’t work for you, I think it’s important to identify the why. So beyond the fact that we know that certain tumours are more immunogenic than others, is there a way to identify why tumours are more immunogenic than others? If I can identify the genetic or a gene that’s driving tumour infiltration then that gives a platform for creating or recreating that type of phenotype in a tumour. So we’re a long way from doing that but I think that eventually it will be feasible.