ASCO 2010 Annual Meeting, 4—8 June 2010, Chicago

Interview with Professor Kathy Albain (Loyola University, Chicago, USA)

Potential biological causes of the racial survival disparity seen in breast cancer

The research I presented was done on behalf of a North American cancer co-operative group called the Southwest Oncology Group and a network of other groups called the Breast Cancer Inter-group of North America. It’s based on an analysis of a treatment-based trial that has already been published and the analysis that we presented at the meetings yesterday have to do with outcome of African Americans on early stage breast cancer trials.

What were the results of this study?

What we did was follow up a previous analysis of ours that we published in The Journal of the National Cancer Institute last year, and in that older analysis we looked at the entire population of patients with breast cancer enrolled on our early stage clinical trials over the years. And in that analysis we found that African Americans did worse than all other races, even when you adjusted for various factors such as number of positive lymph nodes or the body mass index, socio-economic status and other clinical features of the tumour. 

So the analysis we presented now was a follow-up of that to see if we could learn more about the reason for the disparity in outcome for African Americans. So these were all women who were treated by National Cancer Institute approved physicians and were all given the same treatment. So many of the barriers had already been passed that can cause disparities. In other words, these women were all at centres that had clinical trials available and were treated on these trials. 

So in the analysis we just presented yesterday, we took one trial that was very homogeneous and enrolled postmenopausal women who had positive lymph nodes and they were all treated with chemotherapy followed by Tamoxifen or Tamoxifen alone. And in the overall trial it was also true that African Americans did worse and when adjusting for other factors they still did worse. So the meat of what we presented has to do with looking at a subset of the trial whose tumours had been previously analysed for the 21 gene recurrence score assay, which in that analysis showed that women who have tumours that have a very high recurrence score benefitted from chemotherapy added to Tamoxifen whereas patients with very low scores did not, even though their lymph nodes were positive.

So that’s the background, and what we did was we looked first of all at the levels of oestrogen receptor by two ways. We looked at it in the standard way the pathologist scores the level of receptor positivity. All these women had oestrogen receptor positive tumours. So another possible cause of the disparity was not there, that African American women tend to have more triple-negative breast cancer but not in our study; these were all oestrogen receptor positive. So we looked at level of positivity of the oestrogen receptor and that was not different between African Americans and others. 

Then we looked at this 21 gene recurrence score assay which appeared to be a similar distribution in African Americans versus others although there is a trend that the African Americans had higher scores. And then we delved down to the individual genes in this assay and the genes are grouped into different axes: one’s called proliferation; one’s called oestrogen receptor; one’s called HER2; one’s called invasion and then there’s three miscellaneous genes. So the bottom line is for each of those axes and the individual genes that comprise the axes, there was no major increased expression of those genes in African Americans versus others but where there was a striking difference in the five genes that constitute the proliferation axis. So that’s the set of genes that drive the tumour’s aggressiveness, make the tumour divide and grow to a greater degree. So all of those genes had a higher level of expression in African Americans.

So that is a new finding which now needs to be explored – why is that the case? The discussant to my paper is postulating that it may be due to a greater level of body mass index in African Americans and Caucasians leading to different growth factor pathways that interact through the insulin pathway in overweight and obese people to cause tumour proliferation. So that’s a theory and now our next step is going to be to see if we can look at that further, learn more through other grants that we have to explore the biology in African Americans versus others and ultimately probably determine that it’s not an issue of race per se but it’s an issue of what may track with race and there may very well be Caucasians, for example, that will have the same profile once we learn more what the molecular underpinning of this proliferation elevation is.

What further research will you be doing to address this problem?

We have some grants to look at host factors; what genes you inherit that metabolise different aspects of the proliferation pathway. We can do a little bit more with the database itself but other groups have similar databases that could look to see if they see the same thing to validate this. But down the road we may come up with a profile that should be treated a little bit differently in whatever race has the profile. But we’re not there yet; that’s the next series of steps to take.