 This research has been conducted at the Lady Davis Institute for Medical Research in the Siegel Cancer Center, which is part of the Jewish General Hospital. Our research in our lab focuses on the genetics of breast cancer. We are particularly interested in the role of the angioreceptor in breast cancer. Now while this may at first seem counter-intuitive, if you consider that the vast majority of breast cancer cases occur post-menopausal, when there is a change in the ratio of anvirgins to estrogens, you can see that it has a certain degree of logic, hopefully. In addition, there are now instances where different tumors from the same person have a different genetic profile, in other words they exhibit genetic heterogeneity, and even more recently, within the same tumor, you can have, again, what we call intra-tumor genetic heterogeneity. To try and get a handle on this problem, we decided to analyze variants in the angioreceptor gene. What we did is we took six advanced stage tumors and then subjected them to laser capture microdissection to separate out normal tissue within the tumor and the actual cancerous tissue, and in addition, we also sequenced matching blood samples from each of the patients. We then sequenced them using next-generation sequencing in which we used 20,000 reads rather than the normal number of reads, which can vary usually from about 30 to 50. The reason for this was because we were dealing with genetic heterogeneity, we decided to oversequence in order to be able to just examine the sequence, the angioreceptor gene sequence, in a few cells. Now the factor that we looked at in the angioreceptor gene was a functional polymorphism in the first exon of the angioreceptor gene, which is known to differentiate the activity of the angioreceptor gene. Now when we looked at our results, we indeed found a high degree of intra-tumor genetic heterogeneity of this CIG repeat. In normal tissues, the CIG repeat can vary on length from 18 to 25, and normally, of course, in normal individual, it's an excellent gene, you only have two variants. In our case, we had multiple variants, sometimes over 30 variants within each of the samples. What was interesting was that each of the samples had a different frequency of distribution of this CIG repeat. In normal tissue, what was fascinating was is that by far the dominant CIG repeat length was zero, something that never normally occurs within tissues. The breast cancer tissue, on the other hand, had many less zero repeats, and most of the repeats were in the 18 to 25 normal range, and the matching blood had no zero repeats. In fact, all its repeats were within the 18 to 25 CIG repeat length. Now these results illustrated something quite interesting. First of all, it showed that the CIG repeat lengths pre-existed in normal tissue, and that what appeared to happen was that there was a selection process taking place, whereby in menopause the longer repeats were selected for. Now, to understand the biology behind this, we have to understand that the shorter CIG repeat length is the most active, and therefore we can hypothesize that the shorter CIG repeat essentially was selected for pre-menopausal because there was a relatively low ratio of androgens to estrogens, but post-menopausal, this selection pressure had changed because now the androgen to estrogens ratio had changed radically so that longer repeats were now selected for. And the hypothesis is, is that the shorter CIG repeats are in fact somehow protective to prevent the cells from becoming breast cancer. The most important message, though, is that it is important now not just to consider the presence of a particular gene variant, but we have to start to analyze the frequency of a gene variant, and hopefully this will enable us to really pick out for the first time those genes and those variants that really are of critical importance in breast cancer.