 All right. Well, welcome back for those of you who have been with us this morning. You'll know that our sessions have been tremendous and quickly has surfaced what is stake in creating a research agenda. I am looking forward to the talks and discussion this afternoon, and we will start with our second plenary talk by Dr. John Carpton. Dr. John Carpton is an internationally recognized leader in cancer genomics and precision oncology. He is the founding chair of translational genomics at the University of Southern California. His current work focuses on DNA and RNA sequences of tumors to identify biochemical vulnerabilities that can be targeted with new and existing therapies. He was a lead author on the first study to probe the entire genome for inherited prostate cancer genes, and on that study identified a novel mutation in a gene that plays a role in the development of breast, colorectal, and ovarian cancers. Dr. Carpton will speak for approximately 20 minutes and then again we'll open it up to questions from the workshop participants. So thank you very much for being here, Dr. Carpton, I'll hand it over to you now. Thank you. These are my disclosures. I'll probably go a little longer than 20 minutes I think we have a 40 minute session I'll try to get through this as quickly as I can. But just wanted to thank events and the organizers for the opportunities opportunity to share this morning. Hopefully everyone can hear me okay and see my screen. I was asked to talk about the current talk about current research in genomics and health equity and of course my worst nightmare came true this morning the entire session this morning essentially covered the vast majority of what I will talk about. I guess I'll just get the opportunity to use, you know this time to provide, you know some of my own perspective around some of the things that were discussed this morning. I think, you know, you know the time is now. You know, the, in terms of looking at the United States is predicted to be with minority majority by 2045. So as this as the country continues to become more and more diverse addressing these healthcare inequities will be critical towards improving overall outcomes in the US. Some states are already minority majority. And so this could actually happen even sooner than this. We've talked about some of the complexities this morning and of this topic as we think about health, health equity. And it being such a broad term and encompassing a variety of components including structural and social components as we think about social environmental exposures access to care. So in terms of genomics, consider financial toxicities payment and reimbursement I think LSEO brought some of that up, and these issues that play a significant role in health equity or inequities. And so how can every health equity if there's no real equity right so we, we have to stand on the fact that the long lasting detrimental impact of persistent and pervasive structure racism is at the core of healthcare inequities. We also have to consider financial imbalances, you know regardless of race. So consider persistent poverty and urban and or rural areas, regardless of race, think about Appalachia and other areas in the United States or urban urban areas, the structural and the impact of these environments on healthcare inequities in terms of the food desert stuff. The lack of quality healthcare in these in these areas. So we have to consider, you know, all of these issues when thinking about healthcare inequities I know number of individuals are participating in the workshop. We're part of a Congressional League Commissioned Progress Report on Cancer Health Disparities to the ACR by an in our title we did consider racial and ethnic minorities, and other underserved populations. We've also heard about, you know, other other groups such as this individuals with disabilities, adolescent and adults. There are a number of these groups that serve as underserved populations and we have to consider them in these discussions as well. And of course, you know, the at the at the core and center of a lot of these discussions are the ethnic and racial disparities that we see in many areas in healthcare. And so of course there's health equity or inequity and health disparities right, and we know that these things go hand in hand by where health inequity will likely lead to our exacerbate health disparities. So achieving health equity should help lead to the reduction in our limited elimination of health disparities. So they go hand in hand there was also some discussion in the chat I was perusing about, you know, why are we focusing on race it's all about wealth and well you know the strong relationship and correlation right as we look at family wealth, you know across several decades, looking at significant growth when we consider whites and and and flat rates considering blacks and Hispanic Latinos and a growing gap. And then even for some disparities contextual disparities, even when you correct for socioeconomics within groups, right we can see differences and outcomes across groups so it is a part of the issue but it's not the only issue. It's so genetics right where where does genetics come into play. And we know that genetics is sort of a buy is a biological context up, you know I don't want to digress too much but I was in a major meeting recently and you know someone sort of, you know, sort of challenged me, you know, disparities are not about biology, and then you know I know that there are differences in genetics, but genetics is not biology I think everybody's face kind of look weird but you know genetics is at the core of biology and if they're going to be genetic alterations that perturb phenotype, or lead to detrimental conditions than those genetics are going to tune the biology in such a way that leads to these diseases and these conditions. I know that this is also rooted largely in a degree of under land of the underlying genetics of various individuals and in large part related to geographic genetic ancestry, which is been generally tuned by by hundreds of thousands of years of natural selection. It could be continental or regional, whether we, we look at or think about sickle cell anemia as a disparity. Right, we know that that was, you know, based on natural selection and one environment those individuals were able to live to the age to reproduce. You take them out of that malaria infested environment and now they have a disease. And so a lot of the underlying genetics can be tied to natural selection. That's related to geographic ancestry again continental and or regional. So for the sake of this presentation, I'll be focusing somewhat on major underrepresented minority groups, African Americans Hispanic Latinos we know about native or American Indians as well. I also understand the broad perspective of heterogeneity across these populations as have been discussed by, you know, LSEO and and a stay bond that when you think of these populations spending Latinos and even African Americans that that there's broad heterogeneity across these groups. And so I fully appreciate that these broad categories are representative of very diverse, very diverse groups of people. Specifically these populations. I think there is both a racial context and an ancestral context. I think at the heart of this discussion is again population genetics, extracting the and categorizing specific genetic variation that can help to distinguish ancestral group particularly as pertains to trunk populations, African American Indian and or Asian and preaching to the choir because many of the individuals participating in this workshop we're among the pioneers that identified these variants and developed the tools that allow us to deduce genetic ancestral variation in individuals and across populations. After the some of that initial similar work in this area was done a number of large initiatives such as human genome project have map 1000 genomes. Just to name a few, I have have allowed us to begin to characterize the large degree of human variation in the context of continental and regional and or local ancestry. And I think many of us are now going to be somewhat interested in, you know what with the new lost 400 million bases that were recently revealed will provide in terms of our increased understanding of ancestry. We know that these ancestral informative markers or aims are now known and can be genotype to extract the sequencing data to provide to provide an accurate measurement of ancestral proportions for given individual or individuals from various populations. Using what are now industry standard tools and these data can be used in various ways. For instance, ancestry com just to understand your ancestry. And, and we know that these genetic variants are also used a lot and research to better understand relationships and correlations between individuals populations, and various conditions and diseases. This is some. This is sort of nostalgic, there are a lot of different ways I could show this principle components, graphs, but for for some of some of the older folks in the House and remember, remember the hat map and how some of the remember the hat map and how some of the genetic variants were identified and genotyped across different populations, the European, the Han Chinese, the Japanese from the Japanese populations from Tokyo, the Yoruba from Ibadan, and what we can see in these pie charts looking at the two different genetic alleles at any given marker, we can see the differences in allill frequencies across these populations. This is just an example of these ancestral and formative markers and how we can use these to better understand ancestral proportions in individuals. But as has been discussed we know that there's this race and ancestry and they do represent different but related factors and I think we've had the conversation about race and self-identification, the story of Rachel Dallazal is four white grandparents who essentially considers herself an African-American woman or individuals who might have 75% European ancestry and consider themselves European. So this whole concept of self-identification and race and then the social context around race where we do know that individuals like individuals will sometimes congregate together and build communities together so that you have individuals who may have more similar ancestral proportions living in similar environments or communities and conditions and being impacted by those conditions so that there will be a relationship between the ancestry as well as the racial and the social construct that's been developed around those communities. And so I think a number of us have various iterations of this slide as we think about social and societal versus biological and genetic and I used to have verses and I've modified this to have and here right so the social and the biological. So as we think about race and ethnicity and the social determinants of health the social construct that impact health inequities and health disparities and then the genetic ancestry and how natural selection and variation at the genetic level can also impact disease incidents and or outcomes but we know that these two things interact as they impact host biology or conditions. So we know that disparities and incidents and outcomes exist for a number of diseases and conditions of chronic diseases diabetes cardiovascular pulmonary disorders and cancer. We can't ignore these differences think about diabetes and higher rates among American Indians Hispanics and Blacks. We consider heart disease cardiovascular disease and the higher rates among Blacks and we look at various forms of cancer men and women and the significant disparities particularly among Black men across various tumor types are not so much in women as an interesting sex related issue and could lead to some interesting observations and questions here as well. But we know that there various factors are influencing these disparities again including those extrinsic factors societal diet lifestyle and possible interest intrinsic factors such as genetics and ancestry and many of us have been exploring these disparities particularly among African Americans and Hispanic Latinos. So we consider these common diseases and the age of the GWAS and now the age of PRS in the hopes of creating these precision prevention and precision detection approaches utilizing polygenic risk scores which is touched upon this morning by Nancy and part of the other discussions this morning but also you know the context around the the dearth of representation as we think about the cohorts that have been utilized for these studies where the vast majority of individuals and many of these large GWAS studies are of European ancestry or European descent and very small percentages when you consider individuals of Black or African ancestry or Hispanic Latino or Native heritage. Just want to walk through a few vignettes here and you know I think this is a really cool study I think Charles O. Teemings at NHGRI was a part of the study there are a lot of large PRS studies that have been published but I really like this one which was looking at polygenic risk prediction and type 2 diabetes in Africa and they were able to utilize a really large cohort of meta-analysis of over 200,000 cases and a million controls again by where a subset were white or European and a subset were Black or African American and they generated PRS from the entire multi-ethnic cohort from just the African American individuals and just the European individuals and we're able to show that although the discriminatory ability was similar for the African American and a multi-ethnic generated PRS the African American derived PRS was more transferable and generalizable across multiple sub-saharan African populations and at the 10th Dessal where you have fewer SNPs but a greater increased risk you could detect a 3.6 fold increase in risk and disease was detected at almost three years earlier when using the African American PRS when you compare it to the first Dessal so again utilizing a multi-ethnic cohort deriving PRS from very specific populations and then using those and showing that when you tune these PRS appropriately you can get more power and that's opposed to this really interesting study by some of the practical but they created what they call a polygenic risk in a multi-ethnic and multi-ethnic population so the practical consortium again largely European data set and when they validated this PRS and about 80,000 individuals 71 72,000 which were European right you can see that although you could predict risk in all populations European Asian and African you can see that the performance was much lower in the in the African individuals and when trying to use these PRS this 46 variant PRS to predict aggressive disease there was again a positive hazard ratio but two fold less compared so this was K1 being the European 2 being African and then the admixt you could see that the performance was much lower when you looked at the prediction of risk in in African American or African black prostate cancer cases so again by not tuning the PRS appropriately you see poor overall performance I think another area that's been there's been a lot of work will you ask me to present right I do a lot of work in cancer it's probably my primary area of investigation and there's been a lot of progress in this area NHGRI had a huge role in the cancer genome atlas study of the TCGA where over roughly 12,000 individuals were profiled across I think 33 different tumor types and the compendium of somatic alterations were revealed across many of these tumor types for the first time but what we know is that there was also limited diversity in these data sets despite many of these tumor types demonstrating significant disparities in incidents and outcomes prostate triple negative breast co rectal endometrial leukemia multiple myeloma and the important thing here is that when you try to to generate statistically significant results all of these most of these groups are under significantly underpowered to identify mutation differences at 5 or 10 percent an MD PhD student at USC McKinsey postal has been working with us and looking at two large genomic studies TCGA and our Orion data set which also has about 12,000 somatic profiled cancer cases and looking at TCGA and looking specifically at race seeing that the vast majority of individuals were white you have 9% black or African American 6% Asian when we look at the Native American or Hispanic it's almost nil and then when you of course when you look at ethnicity considering Hispanic Latina or Hispanic ethnicity you can see that you know again only 4% the vast majority were non-Hispanic white so in spite of that investigators did take that data set the TCGA data set and deduced genetic ancestry from all of the individuals and then looked at somatic mutation profiles in the context of self-identified race and genetic ancestry and there was some really interesting outcomes from that study that showed that we looked at certain tumor types such as breasthead neck endometrial cancer tumors derived from African Americans exhibited higher levels of chromosomal instability they had higher frequencies of p53 mutation and cyclone e amplification and lower absurd observed frequencies and genomic alterations within the PI three counties AKT pathway and so so despite the lack of diversity you know we still see differences right when we look at the the composition of somatic alterations and tumors derived from individuals across different groups you know our group performed a really large study one of the largest single cancer studies we looked at over 700 multiple myeloma tumors generating somatic coexome normal tumor sequencing whole exome and RNA sequencing data I'm 700 myeloma patients of which about 125 were African American and over 500 were European and I bring this up because multiple myeloma is a relatively rare tumor type but one of the most significant cancer health disparities both in terms of incidence where it ranks second and in terms of mortality rates where it's fourth among men and second among women so multiple myeloma is rubber has represented one of the most significant cancer disparities and I think that the outcomes or the the mortality difference is one that's really interesting I think we were able to provide some clarity as to the factors likely influencing those outcomes we were able to deduce genetic ancestry from all of our cases from the germline data to reveal the somatic the companion of somatic mutations across myelomas in African Americans and European cases and then identifying that one of the most well-known cancer genes TP50 TP53 mutation frequencies were significantly higher in tumors from whites and when we looked at genetic ancestry wild type versus mutant we can see that the vast majority of of individuals whose tumors that harbored p53 alterations had over 90 percent European at ancestry which sort of bucked the dogma in terms of outcomes as p53 loss is associated with poor outcome in myeloma but it's enriched among European cases so these data could suggest that African American patients may have tumors that have molecular features associated with more favorable outcome so if if African Americans were to receive equal treatment we might see equal or better outcomes and that's exactly what we're starting to see clinically by where studies looking at outcomes using these new drug classes of drugs the immunomodulatory agents and that proteasome inhibitors were seeing significant increase in overall of five year survival rates among African Americans that succeeding that of whites and so in multiple myeloma we could be you know seeing that the difference in outcomes or or or mortality rate is socioeconomics and access to to therapeutics but it doesn't explain the difference in incidents right so we still have other opportunities to explore and there could be more of a biological context to the disparity in incidents it could be the rate of the pre-malignant lesions and the conversion of full-blown myeloma immunological factors germ lines and that risk environmental exposures or some combination of the buck so we think we've figured out the incidence I mean the the mortality rate disparity is probably socioeconomics and access to care for myeloma so we can probably fix that right hopefully but we still need to explore the the the disparities in incidents in this disease and over the last 20 years it's been up an explosion of studies in this area where we're consistently identifying differences in tumors derived from individuals across populations with real therapeutic and translational consequences we consider ALL pediatric ALL called prostate cancer lung cancer breast cancer and prostate cancer and so just a few additional thoughts I think some of the other important implications when considering genomics and disparities some of this has been touched upon some maybe not but the therapeutic toxicities and drug metabolism right and differences across groups that might be associated with with biology right or race and the social construct or genetics the under underlying genetics also clinical genetics and the false discovery rates around v us and and the difference in incidence penetrance of sequela so we think about for instance retinopathy and diabetes or in-stage renal disease associated with disparities in hypertension so understanding the population genetics mostly defined by aims will be critically important and not just understanding disparities but how to best approach disease prevention detection and treatment as such if an aim or series of aims is associated with a phenotype at least to me I would no longer consider it an aim but I would consider it a biomarker right and I think this type of nomenclature is far more amenable clinically right then trying to use these differences and race or ancestry if something is associated with a phenotype or difference let's call it a biomarker and so that being said we know that these two things continue to play together I'm really excited about these new study designs and hoping Shanita Hughes-Halbert talks about some of the amazing work she's doing we're really trying to understand the impact of structural racism and social determinants and the on the environmental exposures the social stressors and conditions and a built environment and how they impact physiology right the the epigenetic changes transcriptional changes the work that Nancy talked about this morning a protein modifications lipid modifications and other other physiology such as chronic inflammation right and how these drive towards the increase in pathways that impact disease incidents and progression in this case I'm talking about the hallmarks of cancer but this could be replaced by pathways associated with various chronic diseases other considerations touched on workforce and that's broadly the health care and research and training as well international collaborations biorepositories and access to diverse patient cohorts and data sets diverse model systems for validations and the application of advanced genomic technology so moving beyond exomes and RNAseq and moving into single cell and spatial approaches to assess macro and micro environment as we think about the heterogeneity that exists in these diseases how can we utilize a tool such as single cell RNA sequencing to look at the composition of white cells or the composition of pbmcs across individuals at different stages of development across different populations and how that the genetic ancestry can impact those differences and also spatial differences and looking at cancer identifying these hallmark pathways that are spatially related to a very specific physiological feature of cancer cells and how we can show that that type of physiological cancer cell is enriched in tumors derived from african-american patients versus Caucasian patients I'll be presenting some of this at AECR in the coming days so finally significant progress has been made I think all things being considered we know that there's a dearth in diverse cohorts available for study and we can't just think about getting hitting numbers that that meet the census proportions right we need to think about oversampling in these populations especially for these diseases that disproportionately impact underrepresented minority populations there have been numerous studies published in high impact journals in spite of these limitations 20 years ago there were few or none so I look back and I remember Rick and I started the AHVC study back in 1996 and there was literally nothing happening in that space and now we're seeing papers and and and cell nature you know cancer cell cancer discovery high impact journals showing that these differences do exist do the appropriately powered studies we need to perform these studies we need to build the models generate the evidence and then we can change policy because policy will only change when we generate the appropriate evidence and then finally my hope is that we can tackle these big issues right and as we do it we not only make sure that everyone has access to standard care but that as we do as we better define the relationship between genomics and disease disparities that everyone has access to the most precision care right possible such that disease management is most tailored to each individual appropriately and just this final slide we had a workshop as we're considering a business business model to approach cancer health equity and there were individuals from academia industry government and the interesting thing probably 150 people we had a very active chat going and I suggested that we create a word cloud from the chat and this is the result of that word cloud the most significant word utilized was community I think for us to really move this forward we have to continue to engage the community right if we're really going to address health equity and utilize genomics in the appropriate way as one of the means to move towards health health care equity we have to do a better job of engaging the community so I'll stop there and I don't know if there's time if there's time for questions but thanks there is time for questions your timing is great and thank you for that really interesting presentation I want to encourage folks to ask their questions in the chat and we'll make sure that we have time for Dr. Carpton to answer I did want to start with a question my own question to you Dr. Carpton you know you you pointed out that the performance of PRS scores is variable across populations but yet you showed pretty robust evidence that comparisons between groups at least in your your line of research is possible and then you mentioned using aims as a as a classification or as a descriptor as opposed to race and ancestry I just wanted to ask you a little bit more about that like how first of all the rationale and then how would that be operationalized in research well yeah I think and I'm assuming you're you're you're mentioning sort of my comment that sort of we have we have ancestral informative markers right aims right and and the context of that of that genetic variant is ancestry informative correct um I'm just saying that if that ancestry ancestry informative marker or aim becomes associated with a significant with significantly with a phenotype right let's say it's associated with increased risk uh or it's associated with drug metabolism differences I would say that at that point it's a biomarker right it yes it's enriched in a specific population and that's why it's considered an ancestry informative marker but an individual could self identify as Hispanic Latino race but have a high proportion of the African ancestry if they're from the Dominican Republic right so instead of calling it an ancestral informative marker at that point let's call it a biomarker of drug metabolism right because it will be impactful across all all populations but yes it will be enriched in certain populations so I just think that we talk about and I can't remember I think it was estabonic used the the term you know ancestry I mean a race defined versus you know race informed right I think that we we can get away from from in using genetics and ancestry we can get away from race and we can possibly get away from ancestry right and just focus on the fact that that variant is associated with a phenotype and now it's a biomarker I see so um that's very helpful and I guess my question would be then um ancestry you're trying to get at the admixture is that right for for the for the aims using aims instead of ants genetic ancestry you're trying to get at the ad admixture but there will be there could be certain let's let's take prostate cancer for instance um there are genetic variants of the 8q24 region right that are strongly associated with the increased risk of prostate cancer across groups the interesting fact though that as many of those variants are enriched in sub-Saharan African populations and I use the word enriched right meaning they're more common they're more frequent there but does that mean that they're that they're only in individuals who might consider themselves African American or Black or right I mean there could be an individual who has 25 African ancestry right 75 European ancestry right and and contain that genetic variant at 8q24 so that person may self identify as white but that genetic variant is informative for their risk of developing prostate cancer so my point those markers become biomarkers of increased risk of prostate cancer and not just aims right or ancestral and informative markers because again we continue we we continue to throw in race we continue to throw in ancestry which are terms that can kind of separate right at the end of the day if something is associated with a phenotype right a biological contextual feature right it's a biomarker at that point interesting yeah I would love for folks in the in the room to weigh in on on this approach it's very provocative we do have a question from Dave Kaufman and I'm going to read it but I also would invite him to add as he likes so he asked or he states you know PRS would seem to represent multiple pathways leading to disease and many of those pathways represent the effect of exposures to multiple known risk factors where they're known so some of these non-genetic factors are social determinants of health related to inequities in the same phenotypes when evaluating the power of a PRS is it important to adjust for often easily measurable known non-genetic risk factors including social determinants of health if we don't is the PRS partially a proxy measure for a proxy measuring the effects of known exposures so that was a long question I no no it's a it's a great question and I'll extend it I think I think I think absolutely right but and I also think there there are other clinical there's additional clinical context that could be added as well you think about PRS or prostate cancer right you might also want to consider family history you might also want to consider PSA trends and and values right but absolutely I think there's undoubtedly an interplay between the social construct and ancestry absolutely and and and I think you know individuals who may have the same underlying genetic ancestral risk factors right living in different environments actually may have different you know levels of risk risk of disease initiation and and risk of disease progression and and poor and poor overall outcomes so I tend to agree I know that the there were the two New England journal papers one saying throw race out and then others coming in saying let's not throw the baby out with the bathwater I hate that phrase just like I hate devil devil's advocate but lack of better phrase we have to keep all of this information in context and my belief is that as we build the appropriate models and do perform the appropriate studies generate the data right we can then develop the models that will help us best determine risk in a more precision way right I think this whole term precision medicine came on the heels of what was originally genomic medicine by where we would sequence or utilize information from the genome right to help understand you know clinical management or help you know better perform clinical management but at the end of the day you know there there's a lot of additional context in play to ensure that we that we apply the most precision approach to health care and I do believe that social construct the social determinants of health will be critically important in understanding disease risk in terms of risk of initiation and progression okay great so we we have a question from Anandi Krishnan who asks do you feel that we still have gaps in engagement between investigators from disparate disciplines and here in parentheses cancer clinicians immunologists social scientists genomicists to address these equity disparity questions and if so what are some strategies you suggest for the next generation of investigators to come together that's a great question you know I've been at this for about 30 years now and I'll tell you if there's a stark difference from where things were 30 30 years ago from where they are today where the social scientists would hold their meetings and say it's all social science and the biologists and geneticists would hold our meetings and we'd say it's all genomics and geneticists but there were always those individuals who understood that it was multifactorial and we've seen a growing cadre of of individuals playing in that space again you're going to hear from a number of them over the course of the next couple of days where the gap has definitely begun to close how do we get there policy funding mechanisms right funding mechanisms that that incentivize right the integration across disciplines the creation of multidisciplinary interdisciplinary teams to to to ask the right questions design the right studies so that we can most effectively find the solutions to these health disparities and towards towards achieving health health care equity so it's a great question I think we're a lot better off today than we were 30 years ago but I would love to see continued funding in this area and I really do believe that we're we're heading in that direction we now have the the National Institute for minority health and health disparities under LSEO's leadership which started off as an office and I watched it go from an office to a center to an institute and between the you know that I see and other ICs across campus NCI and others in HGRI I really would love to see the continuation of RFAs and funding announcements and mechanisms that focus on incentivizing integration across disciplines great so we do have a question from Ebony Madden and she's asking you know can you expand on on your comment to that we we should just not not only achieve cohorts that reflect the US population but there's a need to oversample when investigating diseases that might affect underrepresented populations and and I'll add how do we how do we go about that given what we know about who is actually signing up for research yeah no that's that's a great question I you know it's it's funny I think you know we we have gotten to the place to where you know we feel good right if we think we we can get to 12 percent or 10 percent African Americans in a study hey we're close to the the census proportions we we did it but if a disease disproportionately affects that particular group I really firmly believe that we should oversample in order to ensure that we have appropriate statistical power right when we're when we're you know when we're describing our results and when we're general generating those results so I really do believe that we have to oversample because there could be additional heterogeneity hidden in that information right and and and again it goes to power right when you think about identifying differences at 1 percent versus 5 percent versus 10 percent we will need to oversample in some cases I was on a call the other day with Doug Lowey I think it was a day before he knew he was going to become the interim NCI director but I was just on a call with Doug and and he agreed totally actually he was the one who said it and he agreed totally right that we need to think about oversampling just just getting to those proportions is not good enough understanding where we are with those proportions and and especially as we think about Hispanic Latinos given the broad heterogeneity across that group as has been mentioned from from from European mestizo all the way through native various right American Indian components all the way to the the Afro Caribbean where you we have Puerto Rico Dominican where individuals are have you know upwards of you know 40 50 60 African ancestry so I think that we have to consider oversampling in order to ask the right questions and answer them appropriately and then finally to your question Sandra how do we get there we get there through funding right I mean let's let's just be real I don't I don't even know why we're even right it's money it's funding right it's resources the people who want to do this are here the people who know how to do this are here I remember when there was a time when cohorts were being created and there was there was a shutdown when individuals were trying to come together to say let's generate a large cohort of African-American prostate cancer cases was shut down right and and so so so it's it's in you know it's it's funding right we we can write the proposals we can generate the studies we can design the studies we can perform the research right but we can't do it without funding we can identify the cases both both nationally and internationally right we can do all of the work but it requires funding to get there yeah I think that's an excellent note for us to end on thank you so much Dr. Carpton for for a great talk and a wonderful discussion now we're going to move to our panel our final panel of the day this is entitled identifying research gaps and opportunities and I'm going to go ahead and introduce that our three panelists as well as our moderator so we will first hear from Dr. Rick Kittles Dr. Kittles is a professor and founding director of the direct division of health equities within the department of population sciences at city of hope he is also associate director of health equities in the comprehensive cancer center his research has focused on understanding the complex issues surrounding race genetic ancestry and health disparities and over the last 20 years he has been at the forefront of the development of ancestry informative genetic markers and how genetic ancestry has been quantified and utilized in genomic studies on disease risk and outcomes next we will hear from Dr. Lauren Salisbury Dr. Salisbury is an assistant professor in health policy and health services research in the department of public health sciences at the university of chicago her research studies pharmacogenomics and how to guide its implementation in a manner that advances health equity within genomic medicine she is currently pursuing this work as part of an hgri career development award she is the assistant program leader of the cancer prevention and control program within the university of chicago medicine comprehensive cancer center and the assistant director of diversity studies within the university of chicago center for personalized therapeutics our third panelist is Dr. Michael Inouye Dr. Inouye is a computational biologist who has been analyzing human genome data for more than 20 years he is the director of research in the department of public health and primary care at the university of Cambridge the months chair of cardiovascular prediction and prevention at the baker heart and diabetes institute in australia and director of the cambridge baker systems genomics initiative and then finally i'd like to introduce dr. shinita hughes halbert uh from the university of southern california dr. hughes halbert is the associate director for cancer equity at the usc norris comprehensive cancer center and a professor and vice chair of research in the department of population and public health sciences at the kech school of medicine at usc her research is aimed at reducing the disparities in cancer outcomes that affect patients from underrepresented communities with primary focus on african-american communities she identifies social cultural psychological genetic and environmental determinants of cancer health disparities and translates this information into interventions to improve health equity among racially and ethnically diverse populations i'm going to hand it over to uh dr. kittles all right thanks thanks andra um seems like uh we're we're always at the same meeting huh so um uh let me see what do i start uh i have uh several um well let me let me just rephrase it i have strong sort of affinity with uh um the stavon's comments earlier today um and being a um uh a researcher of color um we sort of have had um an interesting journey and insight that has brought us to um where we are today so so my my comments today are just going to really just talk about some of the fears and hopes as we as we enter into this space of genomics and health equity um and uh and i'll tell a couple of interesting stories that you might find uh insightful as we as we talk about this uh this journey um you know uh early on you know this whole discussion around disparities and and health equity but i think we all have to be honest and and and real and realize that disparities and and health overall in the united states is really due to two major contributors poverty and and and racism and um while you know many of these diseases are are genetic um those differences that we see in incidence and mortality for the majority of these disparities have a very little to do with with with genetics um but i will tell you i study one of those diseases that where there is a strong genetic underpinning and and and i've been very vocal about it trying to understand the role genetics is playing uh in in prostate cancer i i loved nancy nancy cox's um volcano plots earlier today where where um you know she most of those phenotypes were cardiovascular phenotypes but if we were to put prostate cancer on on that um genetic ancestry um associated um where they could she control for race prostate cancer would be high on that on that on that plot um and you know it's an area where we we do know that there's some strong genetic um underpinnings um and and i think that's where um a lot of the um uh attention and action when we talk about genetics and genomics in in health equity should be i mean we should we should i try to understand race and and genetic ancestry in order to tease apart the um the risk factors for these for these disparities and if we find out that there is a strong genetic component to it then we um we we put some money and effort to try to understand that um but if it's not then we have to be honest and say look you know maybe we need some behavioral interventions here um uh and and i say that because historically you know geneticists and i and i hope you know we don't go in this direction in health equity space with with genetics is that you know geneticists have been very trendy and and you know we jump on new technology you know we go from um r you know um um uh r of l p's to to um microsatellites to snips to um sequencing um whole genome exome uh RNA seek and and we're we're leveraging new technology but we're we're asking the same questions and and and that's that's the problem that i that i have in this space because we are embracing technology that's really you know state of the art and i you know i don't have any problem with the technology but we're asking the same questions and why is that i think a lot of it has to do with the fact that um there's homogeneity in the room uh in the room when questions emerge in the room where decisions are made about research proposals in the room where where r f a's are written there's not a lot of diversity there and and so we end up asking the same questions with this new technology and coming up with um very little progress in this space and so what does that progress look like well let me just give you an example of some progress um historically that that i've seen i remember when i was at Howard University and and um uh we were trying to do a um we were trying to get involved we had an asthma center asthma research center we were trying to get involved with this um in this um clinical trial with this drug company which i'm not going to mention they had this inhalant um uh and and there was some data showing that african-americans behaved differently they didn't respond um similarly so we wanted to be involved in this and so i remembered calling some of the scientists at um at this company uh this drug company and saying you know we we can help recruit african-americans in this trial and very quickly they said oh well we're unsure about that we don't we're not maybe you know we're not really open to increasing the number of african-americans because uh you know i said why they said because african-americans are dirty the data is dirty and there's too much home with heterogeneity we want clean we want a clean trial going through the FDA now that would tell you something about when this how long ago this was but that was the current state at the time and and i remembered myself others other young investigators on the east coast young investigators on the west coast we were all actively trying to understand what this dirtiness in the data was what was this heterogeneity that everybody was so scared to to touch because they wanted clean data and so folks like myself and and they stayed on and others at UCSF we started working on developing ancestry informative markers and leveraging that in these studies that was a bubbling up of of of individuals who wanted to study disparities who didn't have the the the the tools and the methodologies to really effectively do it so we had to do it ourselves but if we were the same people in the room it wouldn't have gone as far as it did i think and so that's why i think this this this workforce diversity piece is critical because bringing more attention and more minds in the room will create more diversity and ideas and i think more progress in this field but what what what are some of the questions we have to answer the the role of race and racism in disparities and how when we talk about racism in america how is that mediated by skin color we still don't know much about that that's an area that i think we really need some attention on because race and genetic ancestry as as many have said before there's a relationship there in some cases they have independent their independent risk factors when i say race i'm talking about the social construct of race not the not a biological race but there's this information there that it carries is a biomarker as or it's a it's a it's social social marker as a risk factor and then there's ancestry which is itself a a risk factor and then there's this interaction of the two so so that's an area that i think we really have to to to focus on and then we also have to stop this opportunistic sampling and i think john mentioned that in his talk i mean you know historically not historically but there was a period of time where there was no additional cohorts or or recruitment uh in african america and and latino populations because n i h just yes thank you rick those are great comments but we would i would like to hear from our other panelists and you know we i i think these are really setting an important stage for us to um to have our panel discussion so lauren did you want to introduce yourself and give us sort of your reactions to the talk and um to continue our conversation sure thank you just want to make sure uh everyone can hear me all right all right seeing some shaking heads okay great um well it's truly a pleasure to be here today and uh as a health policy and health services researcher studying pharmacogenomics i focus largely on some of the later stages of the translational cycle um namely how genetically based prescribing is equitably or inequitably implemented into clinical practice and its implications for underrepresented populations uh dr carpton and others have eloquently discussed today the significance of the dearth of diverse cohorts and genomic studies and as a part of that kind of given where my research is situated i want to emphasize that this dearth of information and lack of representation applies to implementation studies as well in genomic medicine and this is important because if the goal is to avoid perpetuating and exacerbating health disparities or even potentially introducing new health disparities uh knowledge of the patient experiences of underrepresented populations that are receiving genetically directed healthcare is especially critical patient engagement in genomic medicine is critical um in our especially in our decentralized us health system where patients will be the keepers of their genetic information as they travel from one clinical encounter to another and we've seen in numerous instances within healthcare where you know lack of that ability or knowledge and proper communication tools um to be able to truly be that keeper and carry it across a decentralized system can truly affect health outcomes and communication has been a topic that's come up throughout today's sessions i can say in the studies that we've completed to date we consistently see results that reiterate opportunities in pharmacogenomics clinical trials for improved communications about the role that genomic medicine and personalized care can have for patients especially during clinical visits where genetic testing might be important for medication decisions and importantly those patients across all backgrounds that indicate enthusiasm for the promises of personalized care but we're still seeing differences in the views and the preferences within these underrepresented patient populations that could require tailor approaches and that those aren't necessarily fully always incorporated into care with this in mind i think there are three kind of primary considerations uh that i would propose for going forward and first i'd say you know the communication between patients and providers about genomic medicine um and the type of care that they're receiving that's genetically guided um definitely merits additional focus um already we've heard a call for more empirical evidence to better understand the question of what is actually being communicated in the context of these care visits where genetics can be clinically relevant uh to therapeutic decisions and secondly um as has been noted and i think can't be understated uh appropriately delivering genomic medicine such as pharmacogenomics to underrepresented and underserved patient populations especially those who may be at risk for experiencing health disparities is going to require a better understanding of their life experiences both within and beyond the health system uh in other words we're going to need a more comprehensive picture of their whole lives that account for both the medical determinants as well as the health related social economic political and the full range of other factors that influence patient health outcomes an additional consideration i'll add to that is that germline genetic tests last a lifetime and they can guide care across therapeutic areas which may not only have ramifications for the individuals undergoing testing but it can also be meaningful for their family's health and finally i'll conclude with the need to ensure our systems and institutions have the resources and infrastructure to promote and support equitable access to and delivery of genomic medicine this will involve building levers and tools such as appropriate guidelines directing clinical practice for example it's come up earlier today about the appropriate use of race and medicine and clinical algorithms that direct genomic that direct genomically guided care laws that evolve alongside our medical advancements and health policies that regulate the reimbursement for genetic testing and really discovering the types of payment models that will facilitate personalized medicine thank you great comments thank you and now michael great thanks shanita um so uh i i don't really have a whole lot to introduce about myself so i'm kind of an unusual uh person here in that i am a japanese american um who has not lived in the us since 2005 um so you know i i am i'm finding myself sitting here listening to the conversation that's happening and just incredibly impressed i have to say with the level of of awareness of consciousness level of debate the insight that we're seeing right across the the entire table of people from the us who are leading this debate on how we use genomics to promote health equity and and indeed you know at least not exacerbate inequity so i i i see i'd like to say as as someone sort of a little bit from the outside looking in but previously inside looking out um that that you're you're you're you're actually leading i think the world and the debate that you're having here and the the actions that n h g r i in particular takes from here we'll set an example for the international community and um and i would say you know don't lose sight of that uh you'll you'll tailor to solutions that address the the inequities in the us and that's entirely appropriate you know given the use of us tax dollars but i i think don't don't lose sight of of the enormous impact that can be had uh for you know the uk uh for australia for europe for africa for you know uh asia i think how we how we sort of take the insights and learnings here and and engage uh you know of course very diverse communities and health systems right around the world uh let's let's also keep that front of mind as well that's that's all i've got thank you uh thank you so much and thanks everyone for staying for this really important panel discussion i just like to summarize a couple of key points that i heard um both in john's talk and then through the comments made by the panelists one is um community michael was just talking about the community of scholars in the united states and those who are international lauren talked about um and rick talked about the communities um that in which we are we need to be engaged with to advance equity and genomic medicine um i also heard a lot of discussion about the whole person and social determinants of health and then related to that the importance of ensuring that we have a diverse workforce that there's equity in terms of access to the novel um strategies and the advancements that are made as a result of applying um genomic and genetic technology to um diseases in different ways different ways um i think rick made really important points about how and where genetics and genomics plays a role um as it relates to um the broader systemic issues of poverty and racism and so you know one of the questions or one of the points that i that i think was really important from john's presentation was that you know there's really this intersectionality that i think has in my mind has not really been um examined with sufficient depth of breath to really inform um how we think about um health equity research and genomics so with those kind of integrative comments and that framework i'd like to open it up for the panel discussion and and also um as as in our previous session if you can submit your questions um in the chat what i'll do is is um is read them and and invite our panelists to uh respond i'll start uh with the question with you know ask any panelists to think through and and share with us your views about if an equity framework were applied to all dimensions of genetic research and clinical translation what um mechanisms practices and expertise are needed um rick i'll start with you because you were talking about um workforce diversity and i think you've done a lot of thinking and work in the space of equity frameworks yeah so i think depending on the actually depends on the research question right in terms of all the expertise that you'd want um in the in the room or at the table but um i think we need to move beyond just the uh the um the genetic or uh technology and the statistical analysis and um and really try to dig deep in terms of uh uh that health equity uh framework which encompasses you know the the social determinants and um uh the environment uh the the physical and the social environment and and and how that uh how some of these health behaviors interact uh not just with the with the environments but also with with the genes that you carry so so i mean you know you have to have that level of that diversity or that interdisciplinary um team uh to encompass effectively all of those dimensions in that framework um because disparities isn't isn't isn't just driven by genetics or or just the environment and and so we we need to understand and and model that uh accordingly thank you lauren what are your thoughts about that yes thank you i i think you know i couldn't reiterate more uh the value um at least that i've seen of team science approaches to these topics you know i think dr cox was talking earlier about bringing in social scientists uh into this discussion and into directing um you know everything from the research question to the study design to the interpretation of the results and and what the implications then might be for um different populations um and i think that's going to be critically important um and from a kind of broader systematic standpoint i think that you know interdisciplinary and multi-sector collaborations are also going to be you know critically important there um and i think i just add to that that i think being very clear uh about for popular for studies that involve multiple populations you know what what are the population labels that we're using uh within these investigational studies um particularly those that are evaluating health disparities um is it race ethnicity is it genetic ancestry is it some combined factor of the two and i think um you know across uh different uh types of studies in the research and and biomedical enterprise we've seen um that those population labels are not employed um consistently and so i think you know defining them and being very clear about what you're studying um and and evaluating whether it's appropriate to the research question thank you and so i'm related to that you know when we talk about international health equity efforts if the us is becoming a leader in the space um how can individuals companies universities continue to contribute to the advancement globally and michael i'll like to start with you to to start our conversation about that it's a it's a great question um and you know i mean there's there's so much that that can be done i'll i'll maybe tackle it from an area where where i'm relatively expert in uh and and that's actually uh in in terms of data so we obviously know that you know there's a major sort of data uh gap as in like an under representation of various you know protected uh characteristics various groups uh whether it's you know ethnicity or socioeconomic or you know those sorts of things so that area uh in terms of you know how we're missing you know data on those individuals and then not able to kind of you know train predictive models or or leverage other sort of genetic insights into you know those groups is is something that you know those sort of people can work together to to do better and things like the you know the all of us study uh which is you know oversampling um minority uh ethnicities is uh is a pretty good example of of how you know people can come together and vary from various sort of you know individuals institutions companies that sort of thing can come together and do something really good um i'll copy at that with uh making it international is another step and obviously uh being based in the uk uh you know uk biobank is is a i believe a international exemplar uh in terms of making data available and and and usable right around the world um with very little restrictions actually apart from doing you know approved research um for health so you know i i think that's something that all of us can can really take on board and ensure that the international community uh can also contribute to you know the the equity that research that's going on using all of us um and and indeed that you know that all of us can help contribute uh to the uh you know to helping create uh or to helping promote you know equity research in other countries uh as well and and to do that you know making that data accessible to international researchers will be very important but that's that's just one angle of a very complicated uh question i mean there's all sorts of things people can do and i think the one thing that you know we we can really sort of i think you know take on board is raising awareness keeping it front of the agenda we've talked a lot about uh how equity and and research has really taken off in the past sort of five 10 years or so and the the challenge will be keeping it there and if we keep it there front of mind for people then you know various uh good things will happen connections will happen amongst these groups and yeah i'll leave it there so there's a clarifying question about intersectionality and you know this is related to um sort of how Crenshaw focused on race and gender emphasizing that they must be considered together because the two create more discrimination um than they do separately but i think that the question being raised by this uh in this the issue being raised in this question is you know what um you know are we really focusing on identity and you know how are we using it within this context and i think you know intersectionality is a really important concept um as we think about race ethnicity and in my mind i think there are multiple ways to think about and use an intersection use an intersection the framework of intersectionality to help us think through these issues so in my mind and i'd love to hear the panelists talk about this as well you know they're sort of in my and i'll agree with with what i think i heard john say um is that you know we often kind of had a very siloed approach to thinking about genetics separate from social issues and i think intersectionality in my mind in this space from means that you know the two um domains really work together and they they intersect with one another so the expression of a genomic marker um is through a social context and i think that's you know one of the intriguing areas um in my mind and one of the ways where i think there's an absence of really thought about um you know how in which you know there's this sort of intersectionality among uh risk factors previously i think it's been depicted in frameworks where it's shown to show this multi-level approach but you know those if you think the classics are the depiction of the social ecological framework because we're like these circles that kind of are nested within each other and i guess what i would like to what i really think is important is to have sort of the the Venn diagram of understanding where and how different types of determinants intersect but panelists what do you guys think about that i agree okay does anyone disagree i mean it's kind of i think it's a very interesting focus for us because you know i you know as john was talking and as rick was talking you know i we we all started our careers at relatively the same time and we've seen sort of the progression so you know i can think back to the first sort of multi-level model developed through the centers of population health and health disparities and now there's you know the nimhd research framework which really illustrates sort of the multi-level nature of variables that are important to minority health and health disparities and now i think we're sort of moving towards this more intersectional approach where we can think of these sort of determinants and are interacting with each other there's another question in the chat which is about how can students entering this field help to incorporate lessons from those who have studied and promoted equity and genomics during their future career or educational pursuits so what are the lessons learned and i might start with rick again because rick has been a pioneer in this area you have a lot of lessons that you might want to pearls of wisdom that you might want to share yeah i guess the first thing i want to share is for them to finish pass it's all null and void after that but but um all right so i i think i think um having having good um mentorship or at least um engaging with folks who have been um around uh and have seen certain things is important um you may not necessarily agree with everything that people say but having that that dialogue and that discussion and an understanding of that history um plays a big role and and it's gonna it's actually actually it's gonna play a big role as we move forward in this in this health equity space around genomics um because you know we we have to um you know we i like i like sort of what um uh stave on and and several others on this this paper that that was um written in newland journal the reckoning of racism you know and and and that's where i think things ultimately are going to have to uh uh go thank you other panelists do you have a response that you'd like to share well i can you know speak from uh you know both uh having a a probably more proximal recollection of what it is to be a student and a trainee and and now uh from a faculty person and i can and definitely say that um you know being you know a part of rooms and discussions such as this one as you are um are key you know components um to learning about that and uh seeking in and identifying different types of mentors that can guide different aspects of your you know scientific inquiry um but also your career um have also been really valuable strategies thank you so one of the biggest challenges facing research generally um raised today is and more generally overall is how to enhance engagement and accountability um with both individuals but it would to enhance engagement and accountability but it'd be helpful to have some discussion about more specificity about how to achieve this so how do we achieve um accountability along with engagement holding who accountable accountability with who for i mean i think there are multiple ways to think about accountability um certainly from you know my my first response is accountability among researchers accountability to the groups that we represent you know one of the things that strikes me is you know as we talk about community engagement and um and the lack of diversity in cohorts you know the the question is well what how can we um and one of the reasons why that there are barriers and lack of diversity is is due to um there not being a clear sense of where and how the results might benefit the communities at risk and the communities that we're working to that we're serving so if we think about communities as like one of the our major stakeholders to whom we would be and accountable i mean how do we and how do we ensure that i think i think you know what's been missing in a lot of this uh discussion has been the um the historic value of of minority serving institutions like HBCUs and and the role that they play in our communities and um uh in some in many cases in particular with the science and the technology area you know the many of these institutions have been have been left out of that out of that equation and i think um that could play uh you know those institutions can play a big role too in in the development uh in the training of more um scientists of color but then also in the community engagement uh around these issues and the advancement of the science and of disparities because most of them have a commitment part of their mission is health equity and and disparities and so and understanding disparity so so i i see that as um an area that definitely should be um engaged more okay and sort of a follow-up point michael did you were going to say something i i guess i i just wanted to add for for accountability it's it's i wanted to actually mention something from the data science perspective which i think is something that the community is struggling with uh and and it's how do we actually measure ourselves and hold ourselves to account um and and one of the issues in in equity research that that we run into time and again is that there are so many ways of measuring equity and fairness uh you know how do you actually like what you consider fair in terms of model performance for example whether it's a polygenic score or or or risk prediction model that sort of thing so so i i actually think that we're there there's a lot of there's a lot of good will i think in a lot of especially on the part of you know students and and postdocs that that they highly highly value the series research and rightly so you know and but but they don't actually have the tools that precede accountability um so one of the things that i think we need to do is is the genomics community uh and you know uh genomic data science community is to establish you know metrics and and standards by which we measure ourselves things that are akin to like you know AUC uh per standard deviation odds ratios etc there's a there's a fair literature on this but i think the community does need to come together and uh and and lay down a little bit you know how we should be measuring ourselves and therefore how do we know whether we're doing well or not thank you and just to reiterate and there's a lot of agreement with rick's point about um you know more resources and and funding being um focused towards um hpc us which i think is and then just to kind of bring a finer point to that um there is a significant role for hpc used to play and workforce diversity and community engagement relative to genomics and health equity one of the questions that came up in the in some of the previous chats is you know about the overall social system and here in the united states and how we need to think about achieving health equity within that context um so certainly they you know we live in a society where um you know there's a very strong discrepancy between the research funding that goes to majority academic institutions versus hpc use um there's also you know as you know we've learned through the covet 19 pandemic about um some of the perceptions that are that are prevalent about science and research um throughout the country and uh the issue one of the questions is in and you know we we don't have um and it seems like there's limited support for universal health care so how do we address you know the the broader challenges of equity um overall and then how can we you know really focus that on uh achieving equity and genomic medicine lauren would you like to start sure um so you know i think you know my approach to this you know being a a student of health policy has always been um you know we definitely have to generate a larger and more extensive evidence base um and that is uh you know sometimes challenging work to do especially with hard um complex questions such as the this one um you know but i i think that uh policymakers are uh you know maybe you know given current political climate and knowing that this wax and wanes they have distant receptivity um to evidence and what that means even kind of the definition of what are um you know facts or what is evidence but i think that as a researcher i've at least always taken the approach that if you continue to study topics that are timely that have meaningful implications for people's lives um and you continue to disseminate and promote that work um that i i've certainly seen that you know across uh different types of circumstances that windows will open for that information to make a meaningful change and shift um and how uh practice actually occurs and can uh you know effectively impact what policies are on the agenda and what's being considered um so i think that that you know highlights how important this discussion and um you know thinking you know strategically and intentionally about how uh you know everyone on this call um can think through equity and genomic medicine um is is incredibly important um and having kind of some of that evidence there uh that can inform um where we go from here when those windows open up thank you any other comments or thoughts from the panel i you know i i think about this that question um and i think about in the context of you know like precision medicine or genomic medicine or what historically was you know um initially was was termed individualized medicine and and and i see the value as a geneticist i see the value and the promise um but you know the the inequities and the structural uh racism that um hampers it hampers the ability of it to be effective um and and and create broad benefit across across all communities um you know just just looking at health care overall and i think there was a question in here about um uh healthcare and health equity you know studying um we really need to study that i think that that is a an area that definitely needs to be explored more because just the way that the healthcare system is set up you know there are these barriers for for um disparate communities to to actively engage to be actively be engaged and to participate in the building of better health for themselves in their community and so how do how do we how do we jump from there to this issue of genomics and precision medicine and think that you know things are going to be um uh better um and so that's that's the challenge that's the frustration that is because i see the promise but i don't see once it's implemented it at being done well because of these structural barriers that have been long-standing um in the in the healthcare system so i just i just wanted to echo very strongly what what rick just said about structural inequities in the american healthcare system it is it is apart from research that is the challenge that it's just completely uh uh distorting every research-based advance that comes through the pipe uh this is not a research question uh but it's something that surely is something that's you know needs to happen you know in dc and and you know scientists need to be a part of that conversation so thinking about some of the priorities that have been that i i as i see them and and i'd love to hear from the panelists about this but you know there's been uh funding opportunities for that that that have been designed to support research to address structural racism there's been um and this is one that was uh came out from that was supported by nimhd and really i think some innovative um proposals applications were funded as a result of that similarly there are um you know there's work now being you know developed support to really understand structural racism and how it can be many the manifestations it has within healthcare within the day to day lived experiences of people um in different uh different settings both community clinical work workforce and workplace i guess one of the things that i um that i'm thinking through is you know structural racism is really the big like fundamental issue um what do we need to do to to address it and you know how does that relate to the research framework with equity and genomics it's a challenging question i don't know that i have the answer um anyone want to offer some thoughts so what's the question again so the question is you know if the the main issue the fundamental issue that's like a barrier for all things and i and is structural racism how do we fix it how do we address it there's research being done now to testing some really innovative strategies um but what what do we need to do next that's a great question i wonder about some of that that research though because you know especially the the funding recently in the direction of these sort of these transformative health equity uh warrants where you're out there measuring structural racism and then you're putting in you're doing implement um the interventions or into what am i trying to say you're implant you're implementing interventions on the community that has suffered from the racism without doing anything to break down the structural racism and so you know i i wonder sort of you know is it going to just continue and we'll just end up with better ways to cope hmm that's provocative so what do others think my my lived experience uh overseas is that um america is not more racist than any other country um shock horror uh and yes other countries that's hard to start it's just it's it's just it's it's just you know that it has i i think you just see it more is the thing it gets highlighted more you know um whereas in other places it's it's not for various reasons but other places have more equal um societies and healthcare systems so i i think structural racism is an extraordinarily important limitation uh to um the advances that we can make as you know it is medical researchers and genomicists um but it it's it's not something that's going to stop us from from creating a at least a more equal um healthcare system society that sort of thing i just wanted to make that slightly subtle point thank you lauren you get the last word oh gosh that's a big responsibility um but especially on such a such a question but you know i think i want to harken back to um you know the first session that we had today and some of the comments there that i think really pointed to uh you know the threat of being kind of in an echo chamber of people who already know and believe firmly um that structural racism is the source and the cause of a lot of the ills that we kind of see um in health and well-being and within our healthcare system uh here in the united states uh and i i think that you know again i i would kind of double down on this you know but we need to develop more and miracle research and studies that can really show how structural racism ties directly and is linked um to specific health outcomes and specific health disparities and that work you know that work has been happening a lot of the pioneers who've been doing that work are you know are here today um it's continuing um and it needs to continue you know we cannot let i think our foot off of the gas of doing that work um and developing new methodological tools and statistical methods and things in data science whether it's machine learning or what using the most advanced state-of-the-art technologies that we have today um to continue to demonstrate the impact that structural racism has on you know communities that are underrepresented and underserved with that i'd like to bring the panel discussion to a close um by thanking um all of the panelists for um their insightful comments um and all of you for joining in the discussion and i'd like to turn it back over to sandra um to continue the the meeting thank you great thank you so much shinita and to all the panelists um for for really a great discussion um all right well we are near the end um of our day one and um let me say it's been a tremendous day um i've been tasked to identify some highlights from what we heard and then if there's time i'd like to open it up for comments and questions from our audience um given the richness and breadth of today's discussion uh providing a summary is challenging but but i what i will say um is that as an anthropologist it's clear to me that this workshop confirms what social scientists have long taught us that science and scientific work is an inherently a social process it's a set of activities engaged by a set of actors to accomplish a set of goals that involve choices and trade-offs that reflect our values and commitments and this workshop is part of that process in our case it's the aspiration to use genomic research to relieve human suffering equitably so a major impetus for convening all of us to discuss a research agenda is a sense of urgency to intervene on what many of our speakers um refer to as the problem of the genomic diversity gap that stubborn gap in genomic data that disproportionately represents those of a european ancestry so addressing the gap in genomic data must begin with situating the current data landscape in the context of scientific practices and how human groups have been brought into genetic research this requires critically examining who has been sampled by whom and for what purpose and to recognize data gaps as a product of sociopolitical relationships over time these data gaps cannot be disentangled from structural inequities that directly contribute to health disparities that unfairly burden the most marginalized disenfranchised unrecognized in our society we heard this over and over again from our speakers there's been robust debate today over the use of categories race ethnicity and genetic ancestry and genomic research and how these terms have been used in varying ways to index a broad spectrum of variables lived experience racism in our society and human biology and the speakers throughout the day have from what I heard bemoaned how these categories are used and yet as noted by several of our speakers these are increasingly sedimented into our healthcare system and into our research data in uncritical ways we have heard that we need research to empirically demonstrate how the use or non use of specific categories impact human health so that's one recommendation that I heard quite clearly but we also it seems need future research on health equity and genomics that relies not only focusing on categories and missing data but really goes beyond issues around recruitment but on how do we nurture long-term relationships with communities that will engender trustworthiness of the research as such equity and genomics will require building infrastructure for community engagement critically important we've heard this throughout the day will be diversifying the field of genomics it requires eliminating barriers to not only entry into the field of genomics but supporting researchers to flourish and lead we know empirically that investigators from underrepresented groups produce novel research generate more innovative solutions to problems and publish more influential scientific papers a diverse scientific workforce that promotes the work of black indigenous people and other people of color women sexual and gender minorities individuals with disabilities and those from socioeconomically disadvantaged groups among others precisely the study populations now targeted for genomic and precision medicine research to fill the diversity gap we need to encourage their entry into the field in order to achieve equity so this challenge goes beyond goals of representation but gets to the heart of how a science can be co-produced by asking the questions that matter to the communities it seeks to include this also includes diversifying the entire genomic research ecosystem and critical junctures that narrow the field and we heard this about questioning who is on study sections who are on the journal editorial boards and review bodies who are in our training and funding programs all of this must reflect our commitment to equity we heard from several speakers that the clinical utility of genomic medicine for underserved communities must be studied this relates to the limitations of comparing populations how robust and accurate our standards and references and can these be improved we need more research of the psychosocial and behavioral outcomes of providing genomic information to underserved populations and this requires taking into account not only the research ecosystem but the clinical handoff addressing equity and genomic medicine thus means investigating what benefit actually means for communities that have limited access to care and are under resourced and addressing these barriers for example how do we ensure that benefits of genomic information and interventions are not only realized by communities that have the health healthcare infrastructure creating a research agenda that can achieve equity and genomic medicine will require creating mechanisms that encourage multidisciplinary research we heard this actually from the beginning and I'm thinking here about Nancy Cox talk in which she described her partnership with social scientists and research of electronic health records and race adjusted for genetic ancestry to do identify the role of the environment the lived experience of racialization to understand disparities and hypertension how do we leverage expertise to augment explanatory power of the environment along with genetics to intervene on disease how should teams be formed to fully integrate needed expertise these seem to be first crucial steps to address racism ableism sexism and other isms that undermine the quality of our science and its intended goal of relieving human suffering this will mean creating research that situates genomics within the larger ecosystem and love and our ability to leverage relevant expertise needed to equitably provide benefit this will mean not only creating partnerships but empowering community led research and investing in research and expertise that assesses whether equitable outcomes have actually been achieved and finally I just wanted to go back to Latrice Landry's suggestion that we get comfortable with uncomfortable words she and many others today asked how do we begin to address power and structural racism and other systemic inequities as core commitments and genomics should this be a core competency not only the ability to explain what that is but to recognize and integrate it into research questions and design so I think we're off to a good start and before we end I just wanted to go to any comments or questions that folks in our audience have and let's see here if not I think we can end a few minutes early it's been a long day I do have a couple of announcements well first and foremost I would like to thank our speakers and moderators for giving so generously of their expertise experience and time to this workshop and to the NHGRI leadership and staffer their vision and implementation of this important discussion as I mentioned this is the first day it set us up set us up nicely for day two which is tomorrow and tomorrow there will be more substantive time for all of you to engage with many of the issues that have surfaced today in small discussion groups please note that we will begin at 11 a.m which is one hour later than we started today 11 a.m eastern time and with that I wanted to thank all of you for participating today in this workshop and I look forward to engaging with each other tomorrow