 Good afternoon, and welcome to today's workshop, as you know, co-sponsored by the McLean Center, the Global Health Initiative, the Urban Health Initiative, and the Robert Wood Johnson Disparities Program. I learned this morning that Dr. Olapati will not be able to give today's talk and has asked Jim Fackenthal, a research associate who works with Dr. Olapati in her program, to work from her slide set, as I understand it, and to present the talk that Dr. Olapati intended to give. Jim has very quickly reviewed the slides in the last 10 or 15 minutes, but knows the work pretty well. And so it's a pleasure to welcome you at short notice. Thank you very much for coming. OK, well, thank you for having me. Well, thank you very much for coming. Frumi apologizes for not being able to make it. I will do the best I can to improvise over the slides that she had prepared, and we'll see how it goes. I just want to give a brief overview of what I think are the most important points we want to make. First, we want to talk about the role of genetics in cancer risk, especially breast cancer risk, the role that ethnicity and historical genetic background plays in breast cancer risk, and how that's going to play different roles in different parts of the globe. Want to talk a little bit about how we detect cancers early on and the ways we can bring that to bear on global health disparities. And finally, want to talk a little bit about personalized medicine, because in a lot of ways those two subjects are very closely connected, interested not just in one-size-fits-all prediction methods or one-size-fits-all treatment methods, but in matching prediction methods and treatment methods to individuals. And that match often has to bear in mind personal background, including genetic ethnic background. OK, so there are lots of risk factors associated with the onset of breast cancer, like any other cancer. Being female is obviously an important risk factor, although male breast cancer does happen at about 1,000th the rate of female breast cancer age. And we'll go a little bit more into that later on. Genetic risk factors are what I want to talk about mostly, but also family history, which is tied to genetic risk factors. And then there are a lot of different terms we use to describe unopposed estrogen exposure over the course of a lifetime, including early monarchy, late menopause, late first birth, nulliparity. Previous history of breast or ovarian cancer, even in cases where we don't have necessarily a clear genetic link between a risk factor found in the genome and breast cancer risk, having a strong family history is still a strong predictor. Obesity and, again, some other terms, many of which are probably different ways of looking at unopposed estrogen or hormone exposure risk. OK, well, in 1994, there was a big breakthrough in the identification of a Mendelian risk factor. That is a single gene that, when inherited in the mutant form, caused an extremely high likelihood of developing either breast or breast ovarian cancer family syndrome, very high likelihood of breast cancer risk. And it was called, imaginatively enough, the breast cancer gene, BRCA1. And it made quite a lot of news. And a few years later, BRCA2, a different gene on a different chromosome, but also associated with a very high risk of breast cancer, was identified. And here you see the structures of those two genes. These are simply numbering the exons. Each one of them has a very large exon 11. BRCA1 is a very large gene. BRCA2 is an even larger gene. They both encode immense proteins. These are the kinds of things I think about in the laboratory, what these messages do, how they're regulated, and the different kinds of proteins they encode. Both of them play strong roles in DNA repair. And we'll say a little bit more about that later. But if you look at the literature, especially of BRCA1, you'll see that the number of different proteins that this gene encodes participates in just about every pathway and just about every metabolic schema that you can imagine. These are highly functionally diverse proteins. It's also important to remember that BRCA1 and BRCA2 don't just exist to prevent breast cancer. I mean, we call them tumor suppressor genes, but they play numerous roles during the development of an individual. They're both critically important for embryogenesis, especially early neurogenesis. And they play critically important roles in transcriptional regulation, chromatin modification, DNA repair, the kinds of functions that you would expect to be absolutely critical to every cell in the body throughout development. It's curious that mutations in these genes result in high risks of breast and ovarian cancer and not just, and not cancers of general metabolism in all potential tissue types. So this illustrates a little bit of the way BRCA1 and BRCA2 were discovered. This was done by a linkage analysis as opposed to an association study. We'll talk more about the difference between them. But in a linkage analysis, what you do is you look for a family history of affected individuals and you try to find genetic markers on different chromosomes that seem to travel with the disease in the pedigree that you see here. And that's how BRCA1 was isolated to chromosome 17. BRCA2 was isolated a few years later. And because these were so easily identified with strong family history, especially family history among first-degree relatives, it has become standard practice now for patients who have a strong family history to have their BRCA1 and BRCA2 genes sequenced. That is, it's a good predictive measure of who is going to have a mutation in a family. So it is often the case that if you're from an Ashkenazi Jewish ancestry and you're seeking genetic counseling, a genetic counselor will see that you have a strong family history. They'll suggest that you have BRCA1 and 2 sequencing, and they might suggest, if you're of Ashkenazi background, that you only have those three mutations screened for. Saves time and money. From other populations, though, that have not undergone a recent bottleneck, there aren't going to be any very high likelihoods that the mutation will be just one of a number of a small number of alleles you have the entire gene sequenced. So this availability of the sequencing, availability of knowing whether your family is a carrier of a mutation in the BRCA1 or 2 gene, some people believe may be contributing to health disparities. That is, awareness of genetic testing, access to genetic testing, and the capacity to act upon that information. Some people believe is an important aspect of health disparities. If you look at the prevalence of BRCA1 and BRCA2 mutations in families that are associated with a strong family history, that is, or at high likelihood of risk, that is more than two affected individuals within the family. According to a paper by Rita Nanda from this institution, you can see that 27% of caucasians that were described by this, I'm gonna lose it, okay. 27% of caucasians carried a BRCA1 mutation and 14% carried a BRCA2 mutation, whereas the numbers were considerably smaller in African-American populations. Only 15% carried a BRCA1 mutation, only 13% carried a BRCA2 mutation. Well, BRCA1 and BRCA2 are probably it for high penetrance Mendelian mutations. That is, we're probably not going to discover a BRCA3 or a third mutation that when mutated or inherited in the mutated form is going to be associated with a very high risk of breast cancer. Chances are the rest of the genetic components to breast cancer risk are going to be in genes or factors that have individually much smaller individual contributions but collectively contribute to significant risk. And we describe these statistically by being variants in genes associated with very low odds ratios. That is, we've pretty much taken a relative risk of 10, that is a very high likelihood of being associated with the mutation. We've seen in BRCA1 and BRCA2, there are a couple other genes that are associated with high breast cancer risk in the context of other multi tissue syndromes like Lee-Frowman syndrome or Calden syndrome associated with high breast cancer risk. These are very high mutations and these genes are very high penetrants. But what we're looking at now is instead of very rare mutations associated with high risk, we're now looking more and more using genome-wide association studies at much more common variants, each of which is associated with a comparatively low risk and it's combinations of those low risk factors which you will not see in a family history like a Mendelian factor that is contributing to the rest of the genetic component of breast cancer risk. We've also come a long way in stratifying what we used to call breast cancer, the single disease. And some of this has come from expression studies using tumors from lots of different breast cancer patients done by our collaborator, Chuck Peru, now in North Carolina who moved there after he did this work with Brown in Southern California. And what he noticed was that breast cancer, if you look at the expression of lots of different characteristic genes that are either up-regulated or down-regulated that distinguish tumor from normal or one tumor from another tumor can be broken down by molecular signature into at least, probably more now, five different groups. There's the Claude and Lowe, which is sort of basal-like but it's associated with a reduction of a couple of other markets including the Claude and Marker. There's the basal-like, that is these are tumors that have a gene expression pattern that is very similar to what normal basal epithelial cells look like. There's the HER2-like, one of the strong predictors of outcome in tumors is amplification of the HER2 gene. And these are tumors that have a strong signature that's similar to HER2-amplified. These are normal-like and that's sort of a vague description when you look at the gene expression pattern of adjacent normal tissue and these tumors, you don't really see a big difference, so they're called normal-like tumors. And the luminals, A and B, that's two slightly different gene expression patterns that look very similar to the normal expression pattern of luminal epithelial cells. It is Claude and Lowe and basal-like that we typically call triple negative by immunohistochemistry. And by triple negative, what we mean is when you stain for antibodies against estrogen receptor, the progesterone receptor, or HER2, they're negative for all three. And that is a very aggressive behavior. If you look at mortalities associated with these different tumor subtypes, what you see is that the basals are typically the most lethal. What we call by molecular signature basal-like, what we call by more convenient IHC assays that you could get from a typical pathology assessment, triple negative, those aren't exactly the same thing, but they overlap a lot. Those are typically the most lethal and the most refractory to most kinds of treatments. Interesting thing is when you look at tumors that are associated with BRCA1 mutations, you see that their molecular signature is very basal-like. BRCA1 mutation associated tumors are also lacking in estrogen receptor, lacking in progesterone receptor, and lacking in HER2. So that suggests they may have a common etiology, but more clinically relevant is the idea that a lot of the targeted therapies that we're using now, therapies directed against the presence of an estrogen receptor, for example, the serums, for example, they're not gonna work. And herceptin, which is a monoclonal antibody, it's a humanized mouse monoclonal antibody directed against amplified the HER2 receptor, which is indicated when you see HER2 amplification. That's not gonna work, because it's not amplified. So the conventional targeted therapies we have for treating ER positive and HER2 positive tumors are not gonna work on these very aggressive, fast proceeding tumor types. So this raises the questions, are there population differences in the incidence of triple negative or triple threat breast cancer? Is that nature or nurture? And does it explain health disparities? Okay, breast cancer is a global problem. It's not always perceived as a global problem, though. There are some people in some places who believe, and not without reason, that cancer is a disease that older people get. In fact, the older you are, the more likely it is you're gonna get it, because the principal cause of cancer is bad luck. Genetics influences it a lot, environmental factors influence it a lot, but what it's influencing is the underlying bad luck. It's gonna happen. There are a lot of countries that are devoting a lot of resource to infectious diseases that are killing people at a very early age and are wondering why should we worry about cancer when our population isn't always getting old enough to get cancer? Infectious diseases are taking out a lot more people. And it is true that you're going to see a higher incidence rate in places where you have higher incomes. And that is something we're working to address, especially in some of the African nations, some of the emerging developing nations, where we're seeing increases in cancer incidence as a lot of the other types of diseases are on the wane. But there are other considerations, and a lot of it has to do with what's being offered and who's taking advantage of it. When you look at people who have cancer and see at what rates they die, between 81 and 83 and 1992, the rates of mortality for blacks and whites were similar. But then something happened. The rates of mortality went down for white population in Chicago, stayed roughly the same for black populations. Not entirely clear why that is. But one thing that we're gonna try to impress on you is that the proportion of these triple negative breast cancers, of these basal-like breast cancers is higher in black populations than it is in white populations. These are the cancers that are more refractory to treatment. It may also be influenced by access to treatment, but right now I'm just gonna focus on the genetic characteristics. Oops. Okay, that shows that the study was done in different cities with different outcomes. Why do we think about the African diaspora? There are two important points I wanna make. First, a lot of what we study in breast cancer genetics has to do with what the background genetic diversity is in different human populations. And the first thing you have to remember is half, if not more, of the genetic diversity that occurs in people occurred in Africa before anybody left Africa to populate another nation, okay? So half, if not more of the genetic diversity that exists on Earth existed before there was a diaspora from Africa to populate other continents. In fact, I've heard some African scholars say that there is more genetic diversity between East and West Sudan than there is in the rest of the world, and it's plausible. I don't know if it's true, but I like how it sounds, and it's plausible. The other important thing to remember is a slightly technical one. A lot of the genetic diversity we look at comes from samples that go into what we call the International HapMap Project, where they take samples from different populations, Han Chinese, Japanese, the Ceph, which is a Caucasian population, and a Yoruba population, which comes from Nigeria. And these are taken to be the representative examples of peoples from throughout the world, and the starting point for looking at genetic diversity, and sort of the landscape on which we project our risk factor hunting. Note, though, that slave trade did not just come from this one small spot in Nigeria, which gave rise to African American populations in North and South America, came from all over the West Coast of Africa and even a little bit into Ethiopia. So all that genetic diversity that was taking place during the first half of human evolution, that is the source of the African population during the diaspora, not just what's going on in the Yoruba as in the HapMap. So that's an important consideration to keep remembering. Okay, so many of those samples that came from the Yoruba population, that is an ethnicity within Nigeria, came from a collaboration between the University of Chicago and the University Hospital in Ibadan in Nigeria. And many of those samples that went to the HapMap were duplicates of the samples that came to the University of Chicago as part of an earlier collaboration. And part of what we wanted to look at was what's the genetic landscape difference between the populations that received most of the study in North America, the populations that gave rise to the mutation prediction models, the populations that were used for the clinical trials for various kinds of treatments. What is the genetic differences between those populations and a population that is ancestral to much of the North American African American population? So the first one of the things we wanted to do is look at the spectrum and frequency of BRCA1 and BRCA2 mutations. Remember I showed you an earlier study that showed that BRCA2 mutations were less frequent in African Americans than they were in Caucasian populations. Well, we looked at the frequency and spectrum of those mutations in this Nigerian population, which was primarily Yoruba, although there were some Igbo and some other ethnicities represented in here. And what we found was that a very high number, 7.1% carried BRCA1 mutations and 3.9% carried BRCA2 mutations. Why are those high numbers? Well, most of the other studies, they look at mutations in patients who have strong family history and are therefore very likely to carry a mutation. We didn't do that. We looked at all patients, regardless of family history, and found we had extremely high numbers. These would be considered high numbers, even if we had selected only for patients that had family history. So why is it that in this Nigerian population, we have BRCA1 and 2 mutation frequencies that are higher than any have been seen before, but in the African American population, those mutation frequencies are lower than have been seen before. Not entirely clear why that is. Part of it may have to do with admixture. Part of it may have to do, though, with the genetic diversity that gave rise to the African American population in North America. Not all North American African Americans are descended from the same population that gave us modern Nigerian Urabas. Now, this is kind of a cute diagram. It shows how the prediction models work. What we're interested in looking at is the models that allow us to predict whether somebody is a likely carrier of a mutation. And most of the information we use to try to make that determination is what is the strength of their family history. We bring into it other things like age of onset and personal cancer history, whether somebody has bilateral breast cancer, whether there's breast ovarian cancer in the same family, but mostly it's going to be family history. And what you see is that if you have zero affected family members, only one in 20 patients will carry a BRCA1 or 2 mutation. If there's one affected member, that drops to one in 16, still not a very strong predictor, but it's one in 16. If there are two affected family members, one in eight, and if they're more than two, half. Okay, so even when you have very strong family history, half of the time it's not going to be because you carry a BRCA1 or a BRCA2 mutation. It could come from other reasons. And we showed you before that the BRCA1, the tumors associated with BRCA1 mutations have a unique molecular signature that makes them look very like basal-like tumors. This shows that phenotypically, they also have a distinctive pattern. Medullary and atypical, medullary, well, all tumor types show those on occasion, but here in BRCA1 associated tumors, they're more frequent. High mitotic rate, that just means that cells are dividing a little faster or more of the cells are actually engaged in dividing. Higher degree of aneuploidy, that's an association with a genomic instability, possibly associated with a defect in the BRCA1 gene itself, which is so important for DNA repair. High proliferation fraction, I don't know why that's different from high mitotic rate, but I guess it is. And as I mentioned before, ER negative, PR negative, and NOHR2 gene amplification, these are the indicators of triple negative tumor status. Frequent TP53 mutations will, of course, most tumors have frequent TP53 mutations, and this is what's very interesting, and I alluded to it earlier before. This is a pattern described for basal-like tumors. It's a pattern described for triple negative tumors, and it's the pattern more commonly seen in young African-American women than in other populations. So when I mentioned before that part of the reason for the differences in decline in mortality after these different kinds of drugs were available, part of the reason could be because the frequency of triple negative breast tumors is higher in young African-American women than it is in other populations. So this, again, is simply illustrating a study that we're participating in where we're taking tumors from different sample centers, not just tumors, but epidemiological information. We're collecting blood samples, preparing DNA, RNA, looking at some of the not just genetic risk factors, but also calibrating the genetic landscape that we have to work with in identifying risk factors. And this is simply an illustration of what a pathologist can do to help stratify different tumor types because it's very difficult and very expensive to try to get RNA that you need to do the molecular characterizations. So luminal A's are gonna be characterized by staining positive for ER and PR. Luminal B are gonna stain positive for ER, PR, and HER2. The HER2s, which are gonna be typically ER negative, will stain only for HER2. The basal likes will stain for cytokeratins five and six and for the epidermal growth factor receptor and the unclassifieds just don't stain with anything. So this is one of the assays we use for determining who has what kinds of tumor. And using these kinds of assays, we see that African-American women, premenopausal and postmenopausal, African overall, are gonna have a higher frequency of basal-like tumors than other populations. And remember, the basal-like tumors, the triple negative tumors are the ones most refractory to standard treatments and are also usually typically the fastest progressing. And we mentioned these before, Fumi inserted this slide. Again, there are different breast cancer risk factors. Again, we're talking mostly about the genetic risk factors, but we've mentioned estrogen receptor, ER a couple of times. Now you can see why all these different terms that really describe a lifetime exposure to estrogen like early monarchy, late menopause, first birth, null parity, exogenous hormone use. That's why these are considered risk factors. Now, in addition to genetic factors and environmental factors, we're considering a realm that sort of falls in between. That is epigenetic factors, risk factors. These are covalent changes to gene structure that can be inherited like a mutation, but can also be environmentally induced. So it's almost like a return to a Lamarckian model of evolution. Some of these factors that can induce some of the covalent modifications, by which we mean methylation of certain C groups in DNA, we'll get into that in a minute, are stresses which can come from social interactions or failed social actions, lacking interactions, and there's an active research program going on here to address all of those things. This is a map associated with that Chicago wide study and I really don't know what more Fumi would have said about that, but this is a collaboration we're doing in conjunction with Suzanne Kansen who's very interested in glucocorticoid receptors and other hormonally inducible responses to stress, one of which could be things like social isolation or a certain other neighborhood associated environmental influences. We also mentioned previously looking at personalized medicine that is trying to apply therapies that are especially well suited to somebody's personal and more specifically genetic and molecular issues. We said, for example, that you can't apply a hormone-based therapy to somebody who's got an ER negative tumor because there's no target for it, so that is, but on the other hand, then that's somebody you can spare from side effects when they wouldn't have benefited from the therapy. So that's an example of the kind of personalized medicine we're looking at. Here is another approach that we're taking and it draws from the observation that one of our recent graduate students in the lab was working with, she was looking at one of these epigenetic modifications I mentioned before. I said it was a methyl, it was a covalent modification of gene structure. It's actually the addition of a methyl group to a cytosine base in the context of a CPG island. That is, if near the promoter of a gene of interest, you have C, G, C, G, C, G, some of those C's will acquire a methyl group. When that methyl group is attached to the C, then certain modifications are going to, it'll attract methylated DNA binding proteins and those are going to attract other modifiers that modify the histones in the regions and promote an inactive chromatin configuration. So a methylated C is an indicator of transcriptional inactivation or a promoter of chromatin configuration that inhibits transcription. We expect less expression, therefore from a gene whose promoter is associated with a methyl C. And what she noticed was that there are CPG islands in the promoter region of BRCA1. And in addition to BRCA1 genes that are mutated, being associated with these basal-like phenotypes, sometimes when you see not mutated but promoter methylated BRCA1s, those tumors will also be associated with BRCA1-like phenotypes. It is as if the gene was inactivated by methylation rather than by being mutated. We also mentioned that BRCA1 is intimately involved in DNA repair and it's believed that part of the reason that cells start accumulating the kinds of mutations that ultimately can lead to a cancer is because of the failure in the DNA repair because of the failure in BRCA1 gene function. There are lots of genes though that are required for DNA repair. Another one of them is the PARP gene. And what this student noticed is that in tumors that had methylated BRCA1s, BRCA1 gene expression level was much lower as you predict in basal-like tumors that were unmethylated at the BRCA1 site, BRCA1 expression levels were much higher. If you look at those same tumors, again, where BRCA1 promoters are methylated, PARP1 expression is much higher but in unmethylated tumors it's a little bit lower. So from this and similar observations, the idea is that if a tumor is associated with a defect in DNA repair because it's lacking in BRCA1 function, either because it's mutated or because the BRCA1's inactivated by methylation, you might disrupt DNA repair functions even further if you inhibit PARP. That is, to become a tumor, you need to have a little bit of genomic instability but not too much. A little bit of genomic instability allows you to accumulate the kinds of mutations that allow tumor genesis to progress but too much just leads to metabolic meltdown. You don't have enough genome to support life of the cell. So if the tumor is already lacking in some DNA repair functions for lacking BRCA1 function, you can push it over the edge into lacking too much DNA repair and simply kill the cell if you give it a second hit, an inhibitor of the PARP1 gene. This illustrates part of the stem cell model. A lot of people are wondering why is it that BRCA1 associated tumors are so often ER negative, PR negative, HER2 negative, that is basal like? It came from an important, one model came from an important observation that showed that the BRCA1 gene function is necessary for tumors to differentiate into basal like epithelial, or for normal breast ductal and lobular tissues to differentiate into basal like epithelial cells and luminal epithelial cells. That in the absence of BRCA1 function, a progenitor cell, which would have differentiated into those more mature epithelial cell types will stay basal like. It will stain negative for ER, PR and HER2 new. The idea then is that the reason that BRCA1 inactivation leads to all these triple negative tumor types is it because lack of BRCA1 function traps cells into that progenitor luminal cell type. It doesn't allow cells to progress into the luminal cell fate. And so those cells never had an opportunity to come into existence, so they're not the ones that are able to give rise to the tumors. And this just illustrates what I was talking about a second ago. Imagine a two-hit model. If DNA repair is allowed to progress with normal base excision repair, which is a process governed by PARP1, and by homologous recombination, which is a process governed by BRCA1, you have full levels of DNA repair. If, however, you inactivate homologous recombination pathway because you've inactivated BRCA1, you still have base excision repair operating because PARP1 is still there. You still have a viable cell. Some defects in DNA repair, but there's enough DNA repair by alternate pathways to allow cell viability. Same thing, if you have a PARP deficiency because you've hit it with a drug, for example, that's a PARP inhibitor, you have some defects in DNA repair but the cells are still viable. And this is important because when you apply a therapy, you don't want it to poison all the cells in the body. You just want it to poison the cancer cells. Well, the rest of the cells in the body are not lacking in BRCA1 because they have an undergone complete loss of BRCA1 gene. They still have one viable copy of the BRCA1 gene left. So by this mechanism, a PARP inhibitor isn't going to be as toxic to other cells in the body, just the tumor cells that have lost all BRCA1 gene function there where they have lost all capacity to do homologous recombination because of having lost BRCA1 gene function and lost the capacity to do base excision repair because you've inhibited the PARP functions. Then that is a cell which isn't just defective for DNA repair, it's a failure in DNA repair and so the cell just collapses from inability to support itself with a viable genome. Now, I mentioned before that there are other targeted therapies. We mentioned estrogen receptor and I also mentioned HER2. That is, there is a portion of chromosome 17 that can become amplified. Many, many copies of this tiny portion of chromosome 17 that carries the HER2 gene. When that occurs, you have too much of the HER2 gene product which is a growth factor receptor and that results in tumors that's staying positive for HER2 or are detected HER2 by fluorescence and C2 hybridization and those patients and only those patients will receive the drug Herceptin which is target against the HER2 gene product. Herceptin is not without side effects. You don't wanna give it to everybody, only the people who would benefit from it. Okay, now when Fumi shows these slides, she points to points of density and says very clever doctor sounding things that I'm not gonna try to replicate here because I don't understand any of them. She says clever sounding doctor things here too. Ulcerated is one of them but what we see in a lot of African cases is a very late presentation of breast tumors. There are a couple different ways of thinking about late presentation and this confuses a lot of people so it's important point to make clear. Sometimes late presentation means a tumor has gone ignored for a very long time and is only presented in a clinic when it's something other than a worry when it's actually causing extreme pain or limited mobility. Late presentation can mean something else though when we're talking especially about triple negatives or basal like breast cancers. These are tumors that upon their earliest detection look like late stage tumors. Possibly because they proceed through early stages much, much faster. Possibly because they don't even bother with those early stages. They just jump right into what we call a late stage tumor. So that's an important distinction to make when you hear people talking about late stage presentation. It's not just a delay between detection and presenting in the clinic. It's also talks about the speed of different tumor development rates and they are very particular to different tumor subtypes. Okay, so this is actually too vague to describe so I'm not gonna try it. So this is an idea which isn't just something we're thinking about in the lab. It's something that is a big part of the attention in the budgets going on in the National Institutes of Health. Fostering innovation, research to identify the most promising technology, improved delivery of evidence based interventions, changed genomics for a developing country to genomics by developing countries to build local research capacity, develop ELSI expertise, leadership development programs to form mutually enhancing partnerships. And again, this describes a lot of what we're doing in our collaborative efforts with University Hospital in Ibadan in Nigeria and with other places where we're working in Senegal and South Africa, other places where we're collecting samples. Here's a picture of Fumi. Again, she apologizes for not being here. I know it's a disappointment. She's a whole lot prettier than I am. This is Clement Adibamo. He's our surgery contact at the University Hospital in Ibadan and this is some of his crew there. This is De Jong Ho. I don't know if he's here but he's with the Center of Health Studies here at the University of Chicago and let me stop there. Thank you for your attention and take questions. Yeah. Is there currently treatment for triple negative breast cancer? And if so, does, or if not, would an early detection by a mammogram help? Yeah, early detection always helps. There are plenty of, yeah. Okay. Is there currently a treatment for triple negative breast cancer? If not, would early detection by mammograms help? Yes, the treatment is not personalized treatment but there are standard chemotherapy treatments that are used and early detection always helps. I mean, it almost seems to go without saying but people went and studied it anyway that people with more aggressive tumor types are going to be more easily treated with early detection than if they present too late. More promising targeted therapy though that is in a lot of development that are these PARP inhibitors or triple negatives. Other questions? I want to go back to that Chicago slide you showed of the difference in survival after 92. And I'm trying to get the take home lesson that you and Dr. Olapati are offering. Are the differences that we see, is part of your claim that the differences that we're seeing in populations like this primarily accounted for by the genetic variation or that the genetic variation is one of the elements to which you would add as you suggested in your talk lack of access to care, lack of early screening, lack of adequate treatment but I couldn't quite get the bottom line. Yeah, what can't say primarily we can only say what the different contributors are and right now I don't think we have a good feel for what proportion each contributes. One of the contributors is that African Americans have a higher frequency of these treatment refractory tumors than other populations but we don't know if that's the predominant or if that's more than just a slight contributor to the difference. I see, so you're not making that claim. No. Only that it ought to be included among all the other things that may be contributing. You said that the methylated genes get passed down so how can that be if the Cs are methylated and then when you make some more DNA you pull some Cs out of the environment but those aren't gonna be methylated. Okay, so remember the reverse complement of CPG in the other direction is CPG and there are enzymes that recognize hemimethylated DNA which is what you get after one round of replication. That is this C is methylated but this newly added C on the other side is not. There are enzymes that recognize that hemimethylated DNA, recognize the methylated strand as being the old one because it's methylated. They decide that's the right one that anything that's not methylated must be new and if it's different must be a mistake and it methylates it. Can you talk a little bit about early onset breast cancer and sort of what is going on in that area currently what populations are being affected the most? Early onset breast cancer, yeah. Okay, yeah, good point. Sorry, I didn't mention that before. So one of the characteristics of BRCA1 and BRCA2 associated tumor risk is earlier onset breast cancer. That is an earlier age of getting breast cancer than you would see among sporadic populations. In fact, there are studies that show that if you have a BRCA1 or two mutation, your chances of getting breast cancer by the time you're 40 are pretty high but if you haven't gotten it by the time you're 45 it's a little lower and if you haven't gotten it by the time you're 50 it's lower still and eventually it gets down to the level of which you'd expect to have population-based sporadic breast cancer. One of the reasons we thought that early onset breast cancer which is what you would predict from having a strong genetic component might be part of the same etiology as what we see in African-American patients is because in addition to being high frequency triple negative, high frequency basal like tumors in African-American populations and African populations despite very, very different environmental influences have a higher frequency of early onset breast cancer than in other populations which is again, one of the reasons why we thought there would be such a strong genetic component to breast cancer risk in patients of African ancestry. But I want to thank Jim back on Paul very much for stepping in before me. We really appreciate it.