Good morning, everybody, to ENDO 2022, back in person. I think we're all excited to be here. We're also especially excited to be moderating and presenting this session to you about disparities in reproductive medicine. My name is Samantha Butts. I am a reproductive endocrinologist at Penn State Hershey Medical Center. My esteemed co-moderator is Dr. Nuju Dokris, who is a reproductive endocrinologist at the University of Pennsylvania. We'll be moderating the session. And we are incredibly fortunate to have this session be presented by four outstanding reproductive endocrine researchers, clinicians, and people who have a sincere interest in this content area and space. We're going to introduce each one separately. We'll keep that relatively short just to stay faithful to the time. They have a lot of outstanding content. We want them to be able to present it to you. We also will give maybe time for about one to two questions in between sessions, again, trying to stay on time. And then if we have time at the very end to allow for additional questions from the audience, and we also have questions coming in from our virtual participants, we will also introduce those as well. So it's my honor to bring up our first speaker, Dr. Erica Marsh, who's a reproductive endocrinologist and the division chief of REI at the University of Michigan. Thank you. Thank you so much to the Endocrine Society, to our esteemed moderators, and to my co-panelists for this symposia. Really excited to talk to you today about reproductive health disparities using uterine fibroids as a model and as a jumping point. Because I'm going first, I'm actually going to spend a lot of my time of my talk talking about key definitions and terms just to set the stage. These are my disclosures. Few learning objectives. Really, again, want to cover the definitions of health disparities, health inequities, how they've changed over time. Really understand the difference between the terms race, ancestry, and ethnicity. We currently use them interchangeably, and we need to stop that and be, I think, more impeccable with our language. And certainly talk about social and structural factors that influence health outcomes. And then really prompt reflection on some key questions. For whom was our health care system designed? And who is providing the care that is designed to deliver? Who's designing, performing, and interpreting our research? From what populations are our normative data derived? Is the research we do equitable? Is the care we provide equitable? And how can each of us in our current role work toward health justice? Because we all have a role to play, whether we are seeing patients, whether we're in the lab, whether we're working with cells, mice, human beings, and any other creatures on the planet. So just to jump in, while I'd like to pretend that everyone is as passionate about uterine fibroids or leiomyomas as I am and thinks about it all day, every day, I know that's probably not the case. So I'm going to just spend a few minutes giving an overview of what these tumors are, that I spend a good chunk of my time thinking about. So leiomyomas are smooth muscle tumors of the uterus. They're well-circumscribed, benign walls. Here's a picture of a couple of them here. A given uterus can have one fibroid, which is the late term for uterine leiomyomas, or it can have dozens of them. I think my record for removing fibroids is somewhere in the 50s. They grow in response to sex steroids, so estrogen and progesterone. And they vary in size from being microscopic to being very large, greater than 30 centimeters. They're largely extracellular matrix, and they account, but do have a cellular component. And they account for up to $34 billion annually, including direct and indirect cost, and continue to be the leading cause of hysterectomy in the United States. So we care about these tumors because they're highly prevalent. About 60% to 70% of individuals born with the uterus will have fibroids by the age of 50. And about up to half of those individuals will have symptoms from them. And what do those symptoms look like? The symptoms vary depending on the size and location of the fibroids, but the most common symptom is abnormal uterine bleeding of the heavy menstrual bleeding subtype, pelvic pressure, pain, infertility. I will say specifically recurrent pregnancy loss, not a single pregnancy loss. But in cases of recurrent pregnancy loss, fibroids are certainly implicated. And so therein lies the driver for the $34 billion and for the leading cause of hysterectomy. And just to show you that that drawing is not just a drawing, this is an example of a uterus, a 28-week size fibroid uterus. Normally, the uterus is about the size of your fist, and you can see that this uterus is markedly enlarged and distorted from fibroids. So who gets them? As I just mentioned, most individuals who are born with a uterus, most of those individuals self-identify as women, not all. I will use the term women in this talk frequently, but I want to make sure that everybody knows that I understand that not everybody with the uterus self-identifies as a woman or female. So what are the risk factors for getting these, given their high prevalence? Age is the most significant one. We don't really see these tumors until individuals reach the reproductive window. But the second most significant risk factor is certainly race. And we see a significantly increased prevalence of fibroids in individuals of African ancestry. And I want to be, I'm using the term race and ancestry very intentionally, and we'll get to that in a bit. Other risk factors include obesity, hypertension, pelvic infections, alcohol, specifically beer. And then parity and smoking have been shown to be actually protective against fibroids. We certainly are not going to go out and encourage either childbirth or smoking solely to protect against fibroids. But those associations have been noted in the epi literature. So what is our understanding about fibroids and race, and what is it based on? There have been anecdotal and case studies and case series in the literature about the association of fibroids and race, because that's currently how it's implicated in the literature or utilized in the literature. Actually, for 100 plus years, that association has been made. But one of the first well-done epi studies came out of the Nurses' Health Study II in the late 90s. And they actually looked at individuals who had had an ultrasound confirmation, who had an ultrasound or hysterectomy for fibroids. And what they found was that black individuals had a significantly higher incidence rate. So that incidence rate was kind of backed into. Obviously, this is largely cross-sectional data versus white women. They did not see a difference in Hispanics and Asians, but they quite frankly weren't powered to see a difference in those groups, given the cohort of nurses at that time. Now, remember, fibroids are not symptomatic in up to half of individuals who have them. So it's really important not to rely solely on people who have ultrasounds for symptoms or who have surgery for fibroids to determine the prevalence. We know the prevalence is much higher than those who have an intervention for them. And Donna Day-Baird, who's based out at NIHS in the Research Triangle down in North Carolina, did a phenomenal study in the early 2000s where she actually did a random ultrasound study of black and white individuals. And by age group showed the following data that there's a much higher prevalence of fibroids with black women being represented by black bars, white women being represented by the gray bars here, and that the prevalence increases with age from the mid-30s on, both in white and black women, but starts out pretty high in black women. Now we're going to take just a quick detour and really talk about some definitions before we talk more about the disparities we see in fibroids. And I think this quote by Martin Luther King, Jr. that most of us, if not all of us, are familiar with really sums up the challenges that we face being in the research space, being health care providers, and why we do what we do. So let's talk about some definitions. Ancestry. So ancestry is purely line of descent. And it's probably the cleanest and most objective of the terms that we're going to talk about. It is x begat y begat z, and can be traced back using SNPs, typically to a geographic area on the planet. So the 23andMe, the Ancestry.com is about ancestry specifically. Now ethnicity refers to really a cultural expression or identification, a common diet, a common language, a common geographic location or decision to live in a certain area, a common way to worship even is included in ethnicities. And it is something that we are born into or can opt into or out of. Whereas race is associated with physical characteristics, but is experienced socially and politically. So my 23andMe may say that I am 30% from Europe, 30% from West Africa, 30% Native American, and 10% Asian. But my lived experience is as a black woman. I don't wear my 23andMe or my Ancestry.com mapped out on my face. Our social interactions, regardless of what my ancestry, is driven by kind of what you see when you walk up to me. It's eye shape, eye color, skin color, hair texture, those things, those physical characteristics. So health disparities, the definition of health disparities has really evolved in a very interesting way. Over time, one of the earliest definitions we have from NIH is here, but from back from 1999. Differences in the incidence, prevalence, mortality, and burden of diseases and other adverse health conditions that exist among specific population groups in the United States. Whereas health inequities, and this is a definition from WHO, describes the things that we really want to prioritize to address, which are the differences in health, which are not only unnecessary and avoidable, but in addition, are considered unfair and unjust. Now, the newer definition coming from the federal government that was printed in Healthy People 2020 is really expanded in many ways, but also much more focused in many ways. And it talks about a particular type of health difference that is closely linked with social, economic, and or environmental disadvantage. And then goes through and talks about the different basis on which a group can be identified to experience health disparities. And it also has added health equity, which is the attainment of the highest level of health for all people. And that achieving health equity requires valuing everyone equally. Now, let's talk about inequality and disparities and equality. This is a figure that may be familiar to many of you, but I think it's very basic. And it's very helpful in understanding the work that we need to do. So starting from a place of inequality, two little kids, one on one side, a tree leaning toward them with a lot of fruit on that side, another kid with the tree leaning away from them and less fruit. So inequality speaks to unequal access to opportunities. So when we talk about equality and wanting equality, equality is not the answer necessarily to address inequality. Equality means giving the same resources to everyone, which doesn't really solve the problem as we see here. So evenly distributed tools and assistance in a system that is uneven, if you will, to start. So we want to move next to a system of equity where we have custom tools that identify and address the inequality and really give people a more equitable opportunity. And ultimately, we want to get to a place of justice where we can actually go back to equality and have a system based on equality because the system itself is just. The tree is upright. There's equal fruit on either side. And this is a context where we can talk about equality and where we can talk about really meritocracy and all those other things. So what are the bins in the tree trunk that we deal with as scientists and providers and researchers? They are largely with the newest definition of health disparities that we have in Healthy People 2020, the social structural determinants of health. So the things that we have not historically dealt with, but that are key and very significant drivers. So food, education, physical environment, social environment, and of course, health care and economic stability. And so what are the fruit? Why are we trying to straighten up the tree? Because we want people to have access to these things in the context of their health. These are components of patient-centered care that are identified by the Picker Institute years ago. And we all know patient-centered care is one of the columns or foundations of quality, health care quality, that was identified by what was formerly the IOM, now the National Academy of Medicine. We want people to be able to access this fruit so that they can have an optimal health care outcome. So back to fibroids. When does the disparity begin? We have to make sure we're asking the right questions. We can't jump to the 30s and talk about, well, what's the right treatment? What's the right intervention? Assuming these tumors present at 35 or in the late 40s. Because as we can see here in this study, looking at very young women between 18 and 30, that disparity exists much earlier in individuals of African ancestry who are represented by black bars. Here, individuals of European ancestry are represented by white bars. And we see a very significant difference in prevalence in these early ages. Are we telling? And what is the impact of that? Younger African-American women are having myomectomies early or having multiple myomectomies and are being counseled to have hysterectomies in many cases before they start their families. And so we need to think about, well, how does a playing ground different? Now, are there hormonal differences that contribute to the disparity? We did a study about a decade ago now, it's hard to believe, looking at that question. And what we found was that African-American women have higher estrogen levels than compared to white women. And particularly mid-cycle and the luteal phase. And that was despite having no difference in FSH. Follow-up work showed that there are differences in aromatase that drive that. And so as we transition from the pathophysiology, where there have been a number of areas where there have been shown differences between races or people of different ancestry, I think is really what we've actually been looking at. Why we do what we do is also for us to understand what is being experienced by the intact patient. There's a lot of theoretical models that we have. But one is the Anderson model of total patient delay. But if you think about this in whatever you study, in the case of fibroids, we can walk through. But really, who knows when to raise a concern with a health care provider? Which patients know that they're having a problem? Which patients get a diagnosis? Who gets a referral for a specialist or subspecialist? Who can get an appointment? Which treatments are they offered? Who engages in treatment? And which individuals are those treatments successful? And who ultimately obtains the right outcome or the optimal outcome? And we looked at treatment delay and some qualitative work that we published a few years ago and found that there is a difference in people's understanding of normal. This is one of my favorite quotes, saddest, but also one of my favorite from one of the participants. My period lasted for 30 days. And it was heavy. And it was horrible. These are all individuals who have symptomatic fibroids. I was wearing pads like the size you get when you're in the hospital after you have a baby. And I was so used to that happening that at the time, I didn't call anybody because it was like, this is normal. If this is the experience of your mother, of your sisters, of your aunties, then when the doctor says, are your periods normal, you say, yeah, they're normal. And so it's not enough for us to say, is X normal? Are your cycles normal? Are your periods normal? Is your sex drive normal? Is your hair growth normal? Tell me about your periods. Tell me about your hair growth. Another example, uncomfortable symptoms that people don't realize that not everybody has. Treatment for fibroid options range from conservative medical treatment to more minimally invasive and open surgical options with some interventional radiological procedures in the middle. And the drivers of that treatment choice for patients are ultimately all of these things. And what often gets left out is the star in the middle, which is patient preference as well as access. So just wrapping up, how do we think? We know that fibroids are the leading cause of hysterectomy in the United States. This is a paper done by Whitney Robinson or her team out at UNC. And she does some incredible work looking at benign hysterectomy using a cohort in North Carolina. And what she found was that for benign hysterectomy indications, most of which are fibroids, but not all, more, and this is county data, so representing about 28,000 individuals over, I think, a four-year period, that black women more often than white women have hysterectomies. And that when we look by economic status of a given county, low being the poorer counties, high being the wealthier counties, that we see, interestingly, even more racial disparity, where many more black women, the vast majority of black women are having hysterectomies versus their white counterparts. So understanding what's driving that, how we can understand the intersectionality of things like race and income status and the importance of that. My screen is going in and out, so I'm going to look this way now. So I was part of a group that looked at fibroid outcomes across the country. And you can see the teams and the locations. And we published this work. And this is work that I felt really challenged by, but ultimately thought that it gave an important message. Because the title was Black Women are More Likely than White Women to Schedule a Uterine Sparing Treatment for Liomyomas. This study was not designed to speak to what black women prefer versus white women prefer. It was part of a larger study looking at outcomes. But we collected racial data and had this finding. But if you look at where the study sites were, they are in no way representative of where most people get their fibroid care. They are places that are either solely UAE sites or they're big academic medical centers, for the most part, with one exception. So this is a reminder that we got to be careful what we say, how we message our findings, what the implications are. In summary, fibroids, highly prevalent, racial difference, likely a difference in the biology of the tumors, but also key differences in the social and structural determinants of health across people of different ancestries. Ultimate goal is equality for all, but we need to ensure that equity and justice are currently present and want to focus on ending health inequities and the fact that it requires intentional, thoughtful, and well-funded, yes, well-funded research that is inclusive of all populations at all stages. With that, I want to just acknowledge my incredible clinical and research teams and funders and collaborators. And again, thank you to the society and to the moderators. Thank you. Just to keep us on time, we're going to move to the next speaker, Dr. Marsh. And then if anybody has questions for Dr. Marsh, we will come to those towards the end of the session. So thank you, Dr. Marsh. Yes, I will. OK, next speaker is Dr. Kieran O'Donnell. And he is an investigator at Yale speaking to us about perinatal mental health and the epigenome. Good morning, everyone. It's such a pleasure to be here. This is my first Indo meeting, and I'm delighted to be part of this panel with my esteemed colleagues. And I must say, this is an area that I feel very passionate about and an area that my education is ongoing. So today, I'd like to tell you a little bit about the research that my group is involved in looking at the origins of perinatal mental health and its impact on child development. So I lead the Healthomics and Perinatal Epidemiology Research Group at Yale, and we're interested in trying to integrate measures of complex biological systems and prediction models of perinatal mental health, but also trying to understand the biological mechanisms that link impaired maternal mental health and pregnancy to altered child neurodevelopment and mental health across the lifespan. So why should we care about this problem? Well, this is a common major public health concern. So if I was talking to you just a couple of years ago, we would talk about a prevalence of around one in seven pregnant individuals struggling with their mental health. We're seeing a generational increase in individuals that are struggling with their mental health in and around pregnancy, with a recent report from a large UK epidemiological cohort suggesting that that number has reached one in four. That was published just a couple of years ago, and now if we look at a systematic review in the times of the ongoing COVID pandemic, we see that approximately around one in three pregnant individuals are struggling with their mental health. This is a common problem. How can we address it? Well, of course, we can screen for perinatal mood and anxiety disorders, and there's a number of well-validated screening tools that, of course, many of you be aware of. Of course, you're probably all aware of the American College for Obstetricians and Gynecologists' recommendation for screening, a comprehensive screening of perinatal mental health in the postpartum visit. I would argue that we need to follow the lead of Australia and the UK to actually have screening in pregnancy, where it's recommended to be screened both in the antenatal period and in the postnatal period. But we need to think about this in the context of equity and racial disparities in health outcomes. So even in the United Kingdom, where there is a clinical recommendation for the screening of mood and anxiety disorders, both in pregnancy and the postpartum, we see that this isn't done universally. There are groups, subgroups, that don't receive screening and don't receive treatment for further subsequent follow-up treatment. And of course, the group that's least likely to receive screening and follow-up treatment, even in the context of universal screening practices, are women of color. And then if we move to thinking about the tools that we use to assess perinatal mood and anxiety disorders, well, of course, one of the most well-established tools is the Edinburgh Postnatal Depression Scale, which is validated for use in pregnancy, developed by John Cox. We need to think about how sensitive this is in certain subgroups. There's some idiomatic questions, such as things getting on top of you, I've been able to see the funny side of things. And sometimes these questions and these items don't necessarily translate well to individuals and subgroups, where English may not be their first language. And in fact, Deborah Perry, in a large study, comparing the clinical cutoffs used to identify pregnant individuals at risk of postpartum depression, suggested that there should actually be a lower threshold used for women who self-identify as African-American or black women, so really suggesting that we can't use a one-size-fits-all approach to screening. And I think we also just need to think about the broader context of whether or not someone feels safe to disclose symptoms of anxiety or depression, which can be a barrier to actually identifying women that have the greatest need. Anecdotally, I'm based at Yale. I've moved there for the last two years and have had some fascinating conversations with social workers there. But they've been fascinating, but also quite depressing, really, because historically, with these social workers that have been working in underserved communities, one of the games that the children used to play in that community was, the social worker is coming, and they would try and run away and hide. So that was their experience of interactions with the Department of Children and Families and social work. So I think we need to think about the contextual factors that can influence whether or not an individual feels safe about disclosing their mental health symptoms. And just in terms of a historical review of screening practices in the Yale New Haven hospital system, we're now approaching around 85% to 86% of pregnant individuals that are being screened, mostly in the postpartum period. But when we look at these subgroups, we do see that since 2016, there is between a 4% and a 10% difference between white women, or women who identify as white, being screened with the EPDS, and women that identify as women of color being screened. So basically, we see that there's around 48% of women that historically weren't screened, whereas that number jumps to 54% and higher for women of color. So some of the strategies that we're trying to implement in my group now to try and develop new or more personalized approaches to screening, including the integration of different measures of biology. So we know that there's not going to be a one size fits all. So we're, through support from the Brain Behavior Research Foundation, and also from the Canadian Institute for Advanced Research, we're investigating developing a molecular screen for perinatal mental health, which I don't have time to talk to you about today. But it really seeks to identify subgroups of individuals that are sensitive to the hormones of pregnancy, particularly estradiol and progesterone, to identify a transcriptional signature that can predict individual differences in trajectories of mental health symptoms across pregnancy. And in partnership with Kasana Health and the Canadian Institute for Advanced Research, we're also implementing a passive data capture approach to try and reduce the burden associated with these comprehensive screening approaches that we use in our epidemiological studies. Where are you using passive data capture, using smartphone-based application, to try and create a digital index of social support, and with the idea that social support is a critical determinant of perinatal mental health. And why is this important? Well, of course, we know that there are effective treatments out there that can actually prevent perinatal depression. So if we can implement these improved screening practices, then we stand to be able to intervene and prevent perinatal depression. And of course, in addition to the human costs, this comes at a huge economic cost. So this is the latest cost analysis from Kara Zivan and colleagues in the Commonwealth Fund, suggesting that the annual cost of untreated perinatal mood and anxiety disorders is approximately $18 billion US dollars. That's per year. Now, similar cost analyses from the UK and Australia indicate that between 40% or as much as 72% of these costs could be derived from the adverse effects of untreated perinatal mood and anxiety disorders on child outcomes. And you may ask, why is that such a large kind of confidence interval, the 40% to 72%? Well, the 40% estimate for the costs associated with adverse effects on child outcomes comes from cost analyses that only look at outcomes up until the child is aged five years of age. In the United Kingdom, they looked at the effects or the costs associated with untreated perinatal mood and anxiety disorders up until 18 years of age. And that's where they arrived at the 72% cost estimate. So we know that this is a common problem. This is a costly problem. So what are the effects on the child? Well, work that I've been involved in with the Avon Longitudinal Study of Parents and Children, a cohort of 15,000 pregnancies based in the UK, recruited from a single geographic area, Avon close to Bristol, if you're familiar with it, found that children born to mothers that experienced high levels of anxiety or depression in pregnancy had approximately double the risk for clinically relevant symptoms of socio-emotional and behavioral problems. These effects were evident at age four and they persisted until at least age 13. In fact, in more recent analyses, we see these effects persisting into early adulthood. Now, of course, a question that we get with these kind of analyses as well, you know, is it not all just genetic? Is there not just simply a genetic transmission of risk? An anxious pregnant individual begets an anxious child. And so we've been trying to address this directly by integrating measures of genetic variation and genetic propensity for psychiatric disorders. So many of you will be aware of the fantastic work of the Psychiatric Genomics Consortium now allows us to create summary measures or propensity measures for psychiatric disorders, ADHD, schizophrenia, cross disorders, where we look at genetic risk factors across multiple psychiatric disorders. And we've integrated these summary measures of genetic risk in our analyses, looking at trajectories of symptoms across childhood and into adolescence. You're looking at symptoms going from four years of age until 16 years of age in the ALSPAC longitudinal study of parents and children. And what we found is that maternal prenatal anxiety and with very similar effects for prenatal depression remains significant risk factor for socio-emotional behavioral problems, even when we adjust for these genetic risk factors for ADHD, schizophrenia, depression, or a model that considers all three of these polygenic risk scores for psychiatric disorders. Now, if anyone's a polygenic risk score export in the audience, you'll be saying, but these are terrible predictors. These are weak predictors. And I would agree with you. They account for a small proportion of the variance and outcome, but they're currently the best approach that we have at the moment to directly assess genetic risk at the level of the individual and integrate it into a prediction model. And so we really want to just test the robustness of the association between maternal mental health and pregnancy and child outcome. And even when we adjust for child genetic risk factors, we still see these effects of maternal mental health in pregnancy. And we find additive effects of maternal mental health, both in the prenatal period and the postnatal period. And I think that really speaks to this idea of early intervention so that we can at least reduce exposure at one developmental period. So one of the things that I find fascinating about this area is that while maternal mental health is a salient risk factor for adverse mental health outcomes in the next generation, not all children are affected. And that's a really important message to get out there for anyone who may be pregnant or know a pregnant individual that's worried about their child or worried about their mental health. Of course, it's one of the great ironies that you're telling someone, don't be stressed about being stressed. So I think we want to better understand why is it that some children are affected and others are not? And that's really what's laid the foundation for some of the work that my group has been involved in, trying to develop different biomarkers for assessing the impact of prenatal stress or maternal mental health on child outcome. And so for anyone who's in the audience that's involved in epigenetic epidemiology, you'll probably all be familiar with the idea of epigenetic clocks. So this is a relatively new area of epigenetic epidemiology where we can take any biological sample. Steve Horvath from UCLA has developed a nice multi-tissue predictor. I could take a skin cell, I can take blood, we can take whatever biological tissue we would like to take and we can estimate someone's epigenetic age. Some would say that that's a representation of your biological age and some individuals age more quickly than others. So you can actually see epigenetic age acceleration where someone has a greater deviation or difference from their chronological age and their predicted epigenetic age. And indeed what you find is that those individuals that have higher levels of epigenetic age relative to their chronological age have increased risk for early mortality, have reduced cognitive function and quicker cognitive decline in older life. And the workflow is pretty simple. You take a blood sample or a biological sample, you extract the genomic DNA and then you use a methadone profiling technology such as the Epic array from Illumina and you get your measure of epigenetic age acceleration. Now, one of the challenges with applying this to pediatric studies and to developmental studies is that there is a prediction error of approximately 3.6 years. Now, of course, for anyone who's worked with children, you'll know that 3.6 years is a long time in the life of a child. So we've been involved in developing epigenetic clock suitable for pediatric cohorts. We've used the same technology or methodology in elastic net regression to identify DNA methylation sites that are predictive of chronological age in a cohort of over 1,500 DNA methylomes. So we read DNA methylation from a large group of kids range in age from zero to 19 years of age. And we were able to identify 94 DNA methylation sites, CPGs, that associated with chronological age. And so this is just showing you longitudinal DNA methylation from a cohort of around 150 kids. And you can see that each line represents a child with two time points of DNA methylation. And what we should see here is our positive slopes because we're looking at epigenetic age at time point one and epigenetic age at time point two. They're separated at a minimum of six months. But you can see when we apply these previous or older generation epigenetic clocks, the slopes are all over the place for a technical term. And when we apply the pediatric epigenetic clock, we can see that those slopes become more positive, suggesting reduced error in the prediction of age. But of course, you just say, well, why would you just not ask someone's date of birth, record the current date? You can just predict their age. Well, what we find is that age acceleration in the pediatric cohort is associated with neurodevelopmental outcomes. This particular publicly available data set, we found the pediatric epigenetic age was associated with autism spectrum disorder. And I won't spend too long on this paper. This is a paper from my group that came out earlier this year, where we found that maternal prenatal anxiety was associated with accelerated epigenetic age. We found these effects in a predominantly white cohort from the Netherlands, and also in a multi-ethnic cohort from Singapore. So suggesting robust associations between prenatal anxiety and child epigenetic aging in diverse sociocultural contexts. But just to highlight just one quick study in the last 30 seconds from some unpublished data, there's real intense interest in whether or not these epigenetic clocks can be used as biomarkers to quantify the impact of racism and discrimination. So Arlene Geronimus's weathering hypothesis, whether or not we can use these epigenetic clocks to understand the biological embedding of racism and discrimination. And so this is work that I'm carrying out in collaboration with Dr. Alison Hipwell and Kate Keenan, who lead the Pittsburgh Girls Study. So this is a large prospective cohort that has followed a group of young women from the age of six until early adulthood. And so through support from the Office of Research and the National Heart and Lung Institute, we have assessed epigenetic age in approximately 200 young women from the Pittsburgh Girls Study. And we found that experience of discrimination through the everyday discrimination scale in early adulthood is associated with accelerated epigenetic aging using a novel second generation epigenetic clock. And we have hypotheses about what could be the potential mediating mechanisms, but we do think that there's a critical role for inflammation in driving some of these effects. So just to wrap up there, I am kind of over time, I apologize. I think that epigenetic variation is a logical candidate for the biological embedding of experience. And the question is, how can we mobilize this information to inform clinical care? We need to think about genotype and cell type. These are critical determinants of variation in DNA methylation. But we also need to think about the developmental timing of these exposures, which is emerging as an important factor for determining or specifying the biological embedding of experience. But really, if I can leave you with one take home message, I would say that perinatal mental health is a wise societal investment. Irrespective of the causal association between child outcome, the human suffering represented with perinatal mental health indicates that we need to be doing a better job of screening and treating perinatal mental health disorders. So we need to ask early, ask often, and ensure access and investment. And I will just stop with that. Thank you so much. Thank you so much. Thank you, Dr. O'Donnell. I think in the interest of keeping us on time, we'll move to the next speaker. But we won't forget about getting your questions in because these are wonderful presentations and we'll get to them at the end. We're gonna introduce our next speaker now. So thank you so much. This is all mine. So I'm gonna introduce our next speaker. I can introduce myself. Is it working? I just wasn't sure if my microphone was on. It is. It is on, right? Yeah, it's not working. No, it's not. I can just... I'm Jessica Spencer. I work about a mile from here at Emory University. And I'm gonna talk a little bit about racial disparities in assisted reproduction. I have some external activities. I'm unpaid for them, so I have no financial disclosures for this talk. So I wanna talk briefly about some of the inequities in reproductive health, specifically within IVF, egg and sperm donation, and for male fertility. And talk a little bit about some potential solutions to these. I recognize there are many health disparities, many subpopulations that are marginalized that are affected by this. But I'm going to focus for this talk, because it's about 15 minutes, specifically on black women, mostly in the southeast and in Georgia. So there's a great ASRM Ethics Committee statement from last year that I wanna highlight here. And I bolded a few of the important points on here. It's important for people to know that infertility is common. About one in eight couples experience this. It's a basic human right to build a family. And it is a disease. This is not an elective thing. This is something that is absolutely a disease, and there are medical treatments for it. I probably don't have to remind you and this audience of that fact, but it is important to say that over and over again to everybody else. We are not meeting the need in the United States for infertility support. There are twice as many IVF cycles per capita in Europe as in the United States. And that's not because they like IVF. It's because they have better access to it. And they have similar prevalence of infertility. There are significant differences in both the utilization and the outcomes for black, Asian, and Hispanic women in the United States. And these are due to economic factors, social and cultural factors. And we are all responsible for fixing that. So I want to talk a little bit about a few examples of this. So in the context of health disparity in reproductive medicine, how would I define it? Well, we had some definitions earlier, which were great and gave us that historical perspective. I think the way I define it is it's a higher prevalence of reproductive disease or treatment failure among a specific racial or ethnic population, the cause of which is rooted in systemic bias and discrimination in both society and the health care system. And I want to give a couple of examples for this talk. One is higher exposure to factors that cause infertility, so for example, reproductive toxicants, but also STIs, with poor screening, referral, and treatment, which leads to higher prevalence. And then lower adjusted pregnancy rates with fertility treatment. So even when they have access to that same fertility treatment, their outcomes are lower. Their chances of success are lower. This was a slide that Dr. Marsh shared, and just as a reminder that health disparities do not exist in a vacuum. They are driven by social and economic inequities. And when your patient walks in the door in a fertility clinic, they bring with them the burden of inequity and racial discrimination from every aspect of life, from their access to good food, to their transportation to the clinic, the safety they have at home. And so that carries with them in the clinic. And it's important as providers to remember that. This is an old study, but this is from about seven years ago. At the time, I was working on a large cohort study looking at reproductive outcomes in cancer survivors. And one of the graduate students said, I want to look at our comparison women. We had 1,000 comparison women in the study, and they were recruited all across the state of Georgia. And they underwent a one-hour computer-assisted telephone interview. And she said, I want to look at racial disparities and access to care for help getting pregnant. It was a great idea. We had these women. They had already answered tons of questions about reproductive problems in their past, their prevalence of different diagnoses, their periods of infertility and subfertility. What we found is that the black women in our cohort were more likely to have a lower income, lower education. They live in a large metro area and be unpartnered. They had similar subfertility, which we defined as six months, and infertility as one year of trying to conceive without pregnancy as white women. But they were more likely to report a history of gonorrhea, chlamydia, PID, and fibroids. Black women were less likely to visit a doctor for help getting pregnant, and the time to the appointment was twice as long for white women. So if you look at the graph here, the lower graph at the solid line are black women, and the top dotted line are white women. And you can see that the probability of a visit over time, this is survival analysis, was twice as fast for white women as black women in our cohort. And this has been seen in many, many other access to care analyses, but this is in the context of seeking help for fertility. Sadly, these women were also asked if they reached their desired family size, and black women were more likely to be childless. And they stated that they had fewer children that they desired in their lifetime at the time that they were interviewed. So I want to talk a little bit about STI prevalence. I had a student once ask me, well, how is tubal disease a disparity? Isn't it something you acquire because of something you do? And it's way more complex than that. So gonorrhea and chlamydia are major causes of tubal disease in the United States. And if you think about in the context of years of distrust and fear in the health care system from generations of experimentation without consent for sterilization and hysterectomies, coupled with a provider and an institutional bias, what you get is a system that has less screening and education in black women, and they become more prevalent, and they're more likely to be left untreated, and that results in higher rates of tubal factor infertility. And this has been shown. There's two times higher rates of tubal factor infertility. So then you would say, as an IVF doctor, well, if the tubal factor infertility is so much higher, then we should see a lot more IVF cycles in black women than white women. But actually, it's the opposite. So you could say, well, maybe it's because IVF is so expensive. And that is true. It is expensive. And in many states, like in Georgia here, we don't have a mandate that requires insurances to cover IVF. So this is a study that interviewed over 1,000 women and asked them what their barriers to IVF care were. And what you're seeing in this graph is that when they had coverage, they didn't report it as a barrier. When they didn't have coverage or when they had inadequate coverage, they did report it as a major barrier. But this was true for both black and white women. What's different about black women is they also reported that in addition to this, I also noticed that my relationship status, my weight, a lot of centers have BMI cutoffs, and my race and ethnicity were reported as barriers. You may say, well, IVF centers are far away from where people live. And that's also true. This is a map. This is kind of old data. But they're both 2020 data on both sides. On the left side, you see the CDC graph of all the IVF centers in the United States. It looks very similar today. There's a few more dots. And on the right, you see the population of black Americans and really kind of density by geography. And if you overlay these two together, you can see there is a paucity of IVF centers in high-density black populations. And I have patients in my own clinic that drive five, six hours just for an ultrasound for routine treatment. So something like an IVF cycle is really insurmountable for many people. OK. So we actually took that same data that I just talked to you about, the comparison women, and we looked at, well, maybe the distance and your level of urbanization is what affects your access to care. Like maybe if you live far away, you're less likely to actually go see care. Well, we also asked them about maybe because you see such political and ethnic and religious differences in urban versus suburban and rural areas, maybe their feelings about IVF are very different depending on where they live. And that is true. The graph on the right is showing you that as you move from small town rural to urbanized, you get more comfortable with IVF. But what's interesting is that women living in rural and urban areas were least likely to seek infertility care. It was actually the women in the suburbs that were most likely to seek care. So this is not just a issue of distance, right? This is much more complex than that. I want to talk a little bit about Dr. Butts and her colleagues did a nice study last year that looked at racial disparities in a single large clinic. This is important because when you look at large databases, sometimes it's very hard to control for having different experiences in different places. So this is a very large academic center. And they looked at frozen embryo transfer, live birth rates. And they controlled for BMI, age at the time of egg retrieval, and the number of embryos they transferred. And they did find that despite controlling for these things, black women had lower success rates per embryo transfer, even when you control for all these other things. It was 45% per transfer in white women and 33% in black women. This was looked at on a large scale with SART data. This is the Society for Assisted Reproductive Technology Database, which included over 7,000 cycles. They saw the same thing. They adjusted for age, BMI, parity, smoking, uterine factor, tubal factor, prior transfer. So trying to control for all the things that may be different in these two populations. And even with that, you see that the live birth rate in black women was 37% versus 45% in white women. So some possible mechanisms that these authors suggested is that, well, maybe it has to do with residual effect of surgery on fibroids. Or maybe because they were less likely to pay for more expensive or additional things, like pre-implantation genetic testing, that that affected their live birth rate. Maybe it was low vitamin D that was unscreened. Maybe it's due to weathering and chronic stress, as we just heard. And maybe also because of delay in treatments, that they arrived to the center with a more severe, significant type of infertility than they would be if they had been referred earlier. Our group partnered with a large private practice in Atlanta to look at a donor egg bank. And we thought the donor egg bank was a nice model for trying to tease out the difference between egg quality and uterine receptivity and the health of the person carrying the pregnancy. So we looked at 327 oocyte donors, 900 recipients, and 1,600 embryo transfers over a period of seven years. And what we found is that even though black women had more usable embryos after you thawed and fertilized and grew blastocysts, they had the lowest live birth rate per transfer. And this was true whether or not they used a black egg donor or a white egg donor of their race. So it was pretty impressive. I think that we don't entirely know why this is true. And I think it's important to see that it is not some genetic difference in the egg. The other barrier that these women have is access to eggs. For those of you that do reproductive endocrine, you'll know that it's very hard to find donors of color. This is a study that looked at 12 donor egg banks. And they looked at how many donors they had to offer. And not surprisingly, black and Asian donors were far underrepresented compared to the recipients who were buying these eggs. And white donors were overrepresented. And this is true in sperm banks, too. It's very difficult, the time to actually purchasing the egg or the sperm and getting into treatment cycle, I would hypothesize, is much longer in part due to this. We see the same in males. Referral for male evaluation is lower. Just as a reminder, up to 40% of the time in couples with infertility, there is a male factor. 22 US and Canadian andrology centers coupled with infertility clinics were surveyed. And black men were found to be less likely to be seen by a urologist for infertility. They were older. They were referred later. They were more likely to have severe infertility like azoospermia and have longer duration of infertility. So just kind of a summary, a little bit of what we see in reproductive medicine is on the patient side, they have a long distance for travel across the US. There is a huge cost barrier. Many people are underinsured, especially in the state. We have added layers that add more barriers to people like BMI cutoffs, marriage and contract requirements, which are common in the state. And in the context of systemic racism, which has created this distrust and fear and hesitancy to seek care. On the provider side, we absolutely all have implicit biases. We're less likely to refer patients, longer time to refer, less aggressive treatment, and inadequate screening and treatment of STIs and fibroids. And importantly, a lack of diverse providers. This is from the ACGME, the providers in training from a couple of years ago. Only 6% were black in reproductive endocrinology. We can do better than this. As an educator, I look at this. And it's not just an issue of getting them into fellowship. It's an issue of starting much earlier and helping put systems in place to get people excited about the field and involved so they can then take care of their community and our community. So solutions, educate providers about referral. What is infertility? What are the questions you ask to pick up on it sooner when people don't say things? Improving screening and treatment for STIs and uterine fibroids. Improve patient education for underserved populations about infertility and options. Educate REIs about structural racism and existing barriers to care. And recruit a diverse workforce at all provider levels. Recruit diverse trainees, mentor, and retain them. And lobby for improved insurance coverage, please, at your employer level and at the state level. All right. Thank you. Thank you so much, Jessica. For those who came in late, we are going to do our Q&A at the very end. So invite our last speaker, Dr. Jessica Chin, who is a reproductive endocrinologist at Cedars-Sinai in California. Thank you so much, Dr. Dokerson. Thank you so much to the Endocrine Society. I'm so happy to be connected with my former mentors and a lot of the folks in this room. So thank you very much. So I'm going to be talking a little bit about a big topic, racial and ethnic differences in polycystic ovary syndrome. All right, so for those of you who, again, don't eat, breathe, sleep PCOS, polycystic ovary syndrome, it is one of the most common endocrine disorders in women and impacts one in 10 women. It is typically characterized by irregular cycles, symptoms of hyperandrogenism, like acne and hirsutism, and as well as these polycystic ovaries that we can often see on ultrasound. In addition to some of those more visible symptoms, it's also associated with a plethora of long-term issues, including infertility, mood disorders, endometrial cancer associated with obesity, dyslipidemia, insulin resistance, glucose intolerance, so several severe metabolic issues as well. So as most of the people in this room know, there is not one potentially accepted diagnostic criteria for PCOS. There have been several that have been used, including the NIH, Rotterdam, and AEPCOS Society have different definitions, which then puts prevalence of this disorder at different percentages. There's also a proposed phenotype approach to PCOS. So there's ones where there are women who have all three of those symptoms. So hyperandrogenism, oligo-ion ovulation, as well as this morphology on ultrasound. So as you can see, there's not one size fits all. So even though a lot of women fall under this umbrella, the disorder itself can look very, very different. So I came up with this schema, or just this picture. Everybody knows this. So the combination or the contribution of the pathophysiology to PCOS is likely genetics, environment, and what I'd like to focus on a little bit is how race and ethnicity play into the phenotypic expression of PCOS and its long-term metabolic outcomes. So let's start on the genetics portion first. So through multiple family and genome-wide association studies, we can see that there are several recognized loci that are associated with PCOS. However, the proportion of heritability is still less than 10%, which is very common for very complex heterogeneous disorders, such as PCOS. And so then, of course, we look towards epigenetic variations, including microRNAs, which are non-coding small RNAs that can potentially affect post-transcriptional expression of several genes. So this is work that came out of our lab with Dr. Pizarska. So we looked at women with kind of the more severe phenotypes of PCOS, so phenotype A, which is where women have all three of those classic symptoms, compared with phenotype D, which are women who are non-hyperandrogenic. And this is our PCA plot of these 15 women in each group that show that there is a separation in the phenotypic expression on the microRNA level. And what we identified was 820 differentially expressed circulating microRNAs that were different between phenotypes A versus D. This is just a short list of some of the ones that were upregulated in that phenotype A. So there is some sort of genetic contribution. And eventually, through this study, as we look at more samples, we will be able to pick out whether or not race slash ethnicity have some role in this. And this is a volcano plot of those differentially expressed microRNAs. So you can see that they do separate in A versus D. So there's something on that microRNA level that is contributing. So to focus more on this race ethnicity piece, to bring up the ESHRA slash ASRM goals back 10 years at this point. This was published 10 years ago. And these were the goals of the societies at that time. They said, we need to look at effects of migration and rapid economic development for different ethnic groups for long-term cardiovascular metabolic risk. We need to look at population-based prevalence of PCOS in all ethnicities, best management for manifestations by ethnicity and the role of genetic and environmental factors to explain ethnic variations, and to look at the effect of insulin sensitizers in different ethnic groups. And at that point, they were suggesting that we look for optimal specific recommendations in various races or ethnicities. So it's now been 10 years. Where are we at with all of these goals? So I argue that we have barely scratched the surface. So this is a summary, very, very broad. And again, research in this area is very challenging. Because like Dr. Marsh mentioned, race, ethnicity are interchangeable in the literature. It's very difficult to really parse this out. But in general, and again, Asian women, this encompasses a large and wide variety of Asian people. But in general, are shorter, have lower BMIs, have the milder hyperandrogenism type symptoms. South Asians have a higher prevalence of metabolic syndrome, type 2 diabetes, and central obesity. East Asians have milder symptoms of hirsutism, but worse acne. So this is how we see some of these differences in that phenotype. In African-American women, it's been shown that there's a greater risk for hypertension and obesity. In Hispanic women, higher risk for type 2 diabetes and metabolic syndrome. What are the challenges in research in this area? So again, we lack this standard definition of PCOS. And this will set us up for that endocrine debate later on. What is driving the bus of PCOS? There are conflicting studies. Again, the definitions for race, ethnicity are very challenging. Sometimes race is self-identified. There's very few ancestry studies that can really point out that line of ancestry. And there's really limited data comparing both within race, ethnic differences across geographic regions, and then between race, ethnic differences in similar geographic regions. So this is a study I did in fellowship with Dr. Dirkers and in collaboration with several international collaborators, where we looked at the prevalence of metabolic syndrome and its components in women with PCOS in multiple regions. So we all know what metabolic syndrome looks like, but it's greater than or equal to three of the following, obesity, elevated triglycerides, low HDL, elevated blood pressure, elevated fasting glucose. And this metabolic syndrome puts women at risk for cardiovascular disease, diabetes, stroke, and atherosclerosis later in life. So prior studies have shown that there is this association between PCOS and a higher prevalence of metabolic syndrome. However, there are very few population-based studies that have directly compared the prevalence of metabolic syndrome and its components in women of different races, ethnicities, and women with PCOS. So this was a cross-sectional study in over 1,000 women comparing, again, the metabolic syndrome and its components in the US, Brazil, Finland, Norway, and India. And what we found was, and I'm just going to this very busy, but just to point out, even using Rotterdam criteria or NIH criteria, the phenotype of PCOS looked very different across the site. So again, this just goes to show that PCOS in and of itself is very variable in its presentation. When you look at the prevalence of metabolic syndrome in the study population, compared with the US white population, the US black women had a much higher percentage or prevalence of metabolic syndrome. They were more likely to have a higher BMI and also a higher rate of hypertension. When you look at the women from India, their presentation of metabolic syndrome was different. So they had a much higher rate of low HDL. They had higher rates of hypertriglyceridemia. So again, all of these women fell under the PCOS umbrella, but in and of itself, its metabolic syndrome components looked very different. And even looking at two adjacent Scandinavian countries that are right next to each other, you still see this difference in the presentation of metabolic syndrome and its clustering. So we can see here that there has to be more granularity in this research. When you look at the odds ratios of metabolic syndrome by race, you can see that U.S. black women, even after adjusting for age, have a higher risk of metabolic syndrome compared with their white counterparts. This went away after you adjusted for BMI. So this suggests that BMI may be what is driving this relationship. So the conclusions from this is that there's a significant difference in the prevalence of metabolic syndrome and the clustering of its components despite this unifying diagnosis of PCOS. And there is a need for complete metabolic screening in women with PCOS to potentially identify specific targets for intervention to be more precise in what we are offering for our patients. So looking in the U.S. alone, so again, another study from Dr. Dockers' group, this was a study of black women with PCOS compared with white women with PCOS and their controls in looking at the prevalence of, again, metabolic syndrome and cardiovascular risk. And what they found was that black women with PCOS had a higher adjusted risk for metabolic syndrome in both the adolescents, so the under 20 years of age population, as well as the adults. And this was a study, a secondary analysis out of the PPCOS-2 trial. So this looked at over 700 women with PCOS. And again, they wanted to compare the risk of insulin resistance and metabolic syndrome amongst non-Hispanic white women, non-Hispanic black women, and Hispanic patients. And what they found was across the three groups, there was no difference in BMI or waist circumference. However, Hispanic women had a higher prevalence of hirsutism, free androgen index, an abnormal HOMA score, and hyperglycemia. Black women had a lower prevalence of metabolic syndrome and triglyceridemia. So even across different studies, we see conflicting results. And what I found was that there are very few studies that include few, that include Hispanic or Latinx patients. There's, again, this conflicting evidence regarding the difference in the prevalence of insulin resistance between these groups. And we do need to look at the role of BMI. And finally, this is a study that a resident in our group just submitted. We are looking at a regional variation. So this is within the U.S. in two completely different regional areas or geographic locations in the U.S. So we looked at 1,600 women with PCOS, half from the University of Alabama at Birmingham, 700 from Cedars-Sinai in Los Angeles. So very different geoepidemiologic locations. And what we wanted to do was compare the hormonal and metabolic traits of the women with PCOS in these two geographic locations and also looked at it by race. So what we found was that overall, we saw a stronger pull in the geographic location variable. So meaning the women in Birmingham, Alabama had higher rates of hirsutism. They had also higher rates of obesity and a elevated mean HOMA-IR score. So from a hirsutism and metabolic perspective, there seemed to be a difference in the women in Alabama. When we looked at this by race, there actually was no difference. So we looked at both black and white women in these locations and it did not fall out. But when we looked specifically at the black women, so we compared black women in Alabama compared with those in California, what we saw was that interestingly, there was no difference in their BMI. There was no real difference in their hirsutism scores, but we did see higher rates of hyperandrogenism in the Cedars-Sinai Los Angeles cohort. So we looked at both location as well as race differences and there seems to be some differences there as well. So what this suggests is that there is a regional difference in the presentation of the hormonal and metabolic parameters. So there is this geo-epidemiologic contribution as well. And we do need to do more research in order to identify some of these modifiable environmental risk factors for PCOS, which leads me to this final piece and probably the most difficult to study. So as several of my co-panelists have mentioned, there are several environmental determinants of reproductive health. We have environmental toxins that could disrupt reproductive health. Obesity can exacerbate PCOS phenotype. And in terms of socioeconomic status, so what goes into that? Adverse health behaviors, geographic disparities, exposure to these toxins, access to healthcare. And some of the studies that look at SES in PCOS show that it's not just low SES that determines what PCOS looks like in the future. It may vary across a lifetime. So one of the studies saw that childhood low SES, but higher adult SES is associated with higher PCOS. So it's not clearly low socioeconomic status alone. And what is the role of race and ethnicity in all of these environmental determinants? Is there a higher exposure to toxins, higher exposure to obesity, certain race, ethnicity, associations, socioeconomic status that also play into this. So it's a very complex relationship for sure. And we also have to consider the cultural and social practices. So there is often this culture of silence in several race, ethnicity groups where some folks do not go to see doctors. They don't seek healthcare. A lot of women live with PCOS in silence and don't have the health literacy to go seek treatment. And the impact on their quality of life also differs. So geographic location, ethnic origin, cultural social practices are all likely contributors to the differing manifestations of PCOS. In summary, so there are variations that exist in the presentation of PCOS depending on the race or ethnicity that you identify with. There's not yet an established optimal or specific recommendation in various race ethnicities, so I believe that there should be at some point. And we do need to recognize that these differences do exist and they have to be considered when we're approaching our clinical care. And this is a critical public health issue. And we need to focus our research and our energy in bringing some of the more precision medicine recommendations to different groups because we can see that PCOS is certainly not one beast and it has to be addressed on an individual basis. Thank you very much. Thank you. to come up, and we can open up the session for questions. And we've had a few people who've logged in online, so maybe I could start with questions from our virtual audience. Dr. Marsh, as you're sitting, I can start with a question. There's a question for you from our online audience. Do we know why African-American women have higher estrogen levels? And is this independent of weight, or does it seem to correlate with levels of obesity? So in the work that we did, the participants were BMI matched and the requirement for participation was actually having a normal BMI. So we know that even in the setting of a normal BMI, that there are differences in levels of estradiol. Now to the larger question, does BMI contribute to estrogen levels? Yes, it does. And is there a higher prevalence of obesity in women of African descent? Yes, there is. So I think both statements are true. Yes, there are higher estradiol levels independent of BMI, but we also know that BMI likely contributes disproportionately to the elevated estradiol levels in the general population of women of African ancestry. Thank you. We have some questions from the audience. Thank you very much for this. This is wonderful. I'm trying to phrase this question, but a few years ago, this society was very involved in trying to discuss endocrine disruptors as they applied to this type of reproductive issues along with all the other things that endocrinology affects. And it seems to be very much something that could explain what several of you are studying. I just, it's always been very difficult to figure out how to study endocrine disruption itself. And I think that one of the things that I wanted to ask Dr. O'Donnell is whether your discussion of methylomes and descriptions of the epigenetic modifications and weathering, which is a new term for me, could be correlated with these exposures because it's very clear that your societal status and the types of exposures that you have with your food and with your environment can differ drastically and might explain some of these things. It's just been very difficult to study, but does this technology that you're describing help us learn more about that? Thanks so much. That's a great question. And I think the answer is yes. So one of the leading mechanisms for trying to understand how the environment influences variation in DNA methylation or histone modifications is altered transcription factor binding. So basically, if an environmental toxicant or an endocrine disruptor influences the activity or the binding of a transcription factor to the DNA, then that can alter the amount of DNA methylation either by protecting that site from DNA demethylating enzymes like the tet enzymes or by actually promoting increased access to the underlying DNA so you can actually have DNA demethylation. So in terms of some environmental toxicants that we would all be familiar with, per- and polyfluoroalkyl substances, so PFAS, are associated with DNA methylation, changes in DNA methylation. Really nice paper from Joseph Braun showing effects in cord blood and then showing that being sustained up until 12 years of age. Similarly, Francis Champagne and others have looked at BPAs and showing differences in DNA methylation as a function of exposure to BPA. I think what the field is recognizing at the moment is the need to account for the potential synergy between prenatal stress, so psychosocial stress, or perhaps racism and discrimination, and these environmental toxicants. So there's some really interesting work expanding mixture models to incorporate not just environmental toxicants but mixtures of prenatal stress or psychosocial stress, and that's work that we're involved in as well. Yeah, thank you. So great talk for all the presenters. I'm Stan Andres, an assistant professor at Howard University College of Medicine. My question is to the last presentation, and I recently had a publication, Falzerano et al., from 2021, that was looking at prevalence of NAFLD and PCOS, and it was kind of a worldwide look of different aspects and looked at race and ethnicity, and we actually, I think, reviewed probably everyone on the panel here, so love your work. I had a question about, I believe it was from Dr. Dukras's group, and it was looking at metabolic syndrome and PCOS, and it kind of was taking this worldwide look as well, but looking at the map, you can see that there's a big gap in that there was no representation from Africa, so I was wondering what was the reason for that and if you could explain a little bit as to why that was not presented in that paper. Thank you so much for that comment, and yes, absolutely, I completely agree. We have these collaborators at these five sites and happen to have their database. I'm actually working now with another collaborator. We're going to be looking at some West African data potentially, so I definitely think that is absolutely a crucial and critical piece that needs to be addressed as well and potentially go back and do the comparison with those sites as well, but congrats on your work as well. Yes, thank you, and to that very point, I think we need to be very intentional as to be inclusive in those aspects. Sometimes I feel that, I don't know if it's, I know when, for instance, when we were looking, some of the challenges that we were having was that some of the articles that were coming out of researchers from Africa didn't have, weren't translated into English or just had various ways of not being included, and I think we just need to be very intentional as researchers when we're thinking about this to be inclusive of those groups, so thank you. Absolutely, thank you so much. Dan DeMessig, UCLA, wonderful talks, and I appreciate everybody's work. Congratulations, outstanding. I have two kind of independent questions. The first may be for Drs. Marsh and Spencer. When I'm looking at the implantation rates for the black individuals, which seem to be lower independent of donor status, potentially raising an issue of the uterus, is there any evidence that the problem is at a myometrial level from, let's say, fibroids, et cetera, versus this idea that there might be an alteration of implantation time? If there's a difference in BMI versus steroid interactions, I can see where the endometrium might be pre-receptive. Any thoughts on the mechanisms for the decreased pregnancy outcome for FETs? Sure. No, I'll take a cut. I mean, great question, Dan. I think big pictures we don't know when it's been understudied, but I think the assumption has been fibroids, fibroids, fibroids, fibroids, fibroids, and that we actually haven't looked at things at the endometrial level sufficiently. In the work looking at fibroid prevalence in very young women that we published a few years ago, we actually looked at endometrial thickness independent of fibroids, independent of OCP status or hormonal exposure status, and found differences in endometrial thickness between white women and black participants. And there's been a small, some small studies that have started to look at endometrial differences amongst racial groups. So I think if anything, it's that we've been making assumptions that it's fibroids, and certainly some of it is fibroids, but I think some of it is potentially differences in endometrial physiology, response to the hormonal milieu, which we know is different, and response to how we kind of integrate the intersectionality of what we've been talking about of all of these different existences that we have, the environmental exposure, how we integrate racism into our physiology, how we integrate our diets, I mean, all of those things kind of separately come together and I think the endometrium is probably gonna tell us more than the myometrium will, or tell us as much as the myometrium will. Jessica. Yeah, I don't have a whole lot to add to that other than it's a very hard answer. To my knowledge, there aren't specific studies that have looked at that on the cellular level, but most of the data I showed are just kind of large database studies, like in the SART data, you just know if you have a uterine factor or not in an individual. We don't know how many fibroids were removed and what was the residual effect of scarring and was their transfer difficult, and all the kinds of things in IVF go into a higher life birth rate and a transfer. But I would caution that in general, we do keep looking for really granular reasons for why we see these disparities, but I wanna kind of go back to what Dr. Marsh was saying is that we also have to accept that there may be something about the context of that individual's experiences and their exposures, and when they get to that cycle, how they are managed and everything that results in these disparities other than just cellular differences. Great, thank you. The other question I have for Dr. O'Donnell, which is totally different, you know, when you talk about the Horvath model for epigenetics, particularly looking at blood, and I know the last slide, in terms of future directions, you're talking potentially about target tissues per se, recognizing that the epigenome can differ in various target tissues, and the idea that that model in the blood has recently been applied to IVF offspring, trying to figure out the long-term effects of IVF on offspring health. It looks like there isn't a big difference in the blood epigenetics with the Horvath model, but yet there are some long-term concerns about heart issues, hypertension, some other variables. How do we get at the epigenome of the target tissues per se in the offspring, because we need to go beyond the blood to get into the target tissues. Do you have any thoughts on how we would do that in the future? Yeah, I think that's a really good point. I think that tissue, so different tissues age at different rates, so for example, the cerebellum ages at a much slower rate than, say, breast tissue, which would age at a more accelerated rate, so I do think that there are tissue-specific effects. I think one of the most exciting developments in the context of epigenetic clocks that's happening more recently is realizing that these clocks are actually composed of subunits that can be, so CPGs that can be clustered into distinct networks that may be sensitive to different biological processes. So for example, so certain subunits may be being more sensitive to, say, inflammation, others being more associated with, say, for example, smoking or environmental toxicant exposure. This is work that's being done by Morgan Levine, which is very exciting, trying to collapse across multiple epigenetic clocks to actually get at more biological function. What's the biological pathways that are driving acceleration? And it may be that some of these subunits are gonna be more sensitive to those pathways than others. In terms of trying to do that in the target tissue of interest, in terms of my research interests, obviously the placenta is one tissue that I think is really fertile ground for looking at these aging measures because of the kind of critical role that it plays in fetal development. But I do think that it is a challenge trying to understand what is the correspondence between different tissues and the aging. And I think that's where these subunits or subnetworks will get us closer to trying to understand which are the subunits that are shared across tissues that tick at the same rate. Great, thank you very much. Since we're slightly over time, we're gonna leave the last word with Dr. Santoro. We would invite any other questions to come up and engage with the speakers. But Dr. Santoro, take us out. Okay, I'll be quick. Thanks, everybody. I did wanna just point out in some of the conversations we're having that estradiol is lower in obese women when you measure serum estradiol. So if there's the independent effect of obesity is not to give you more estradiol, it may give you more estrone. So we need to be a little careful when we talk about obesity lowering estrogen because estrone is about 1 100th as potent as estradiol. And my second point was just that the exosome world may also have something to teach us about epigenetics. Tracy Bale has some very compelling data in men showing that men with early life stress have different exosome cargo in their seminal fluid that may translate into reproductive changes. So we've got plenty of work to do. But thanks for kicking this off. Great. Thank you to all our panelists and the audience. Thank you.

disparities

reproductive medicine

uterine fibroids

perinatal mental health

PCOS

racial and ethnic groups

systemic biases

screening practices

epigenetic clocks

racial disparities

assisted reproduction

infertility

equitable care