International Medical Conference Endometriosis 2025:
Endometriosis 2025: Your Mother Should Know, Your Doctor Should Know Better!
The Familial and Hereditary Aspects of Endometriosis - Krina Zondervan, MSc, DPhil, FRCOG
Hi everyone. My name's Karina Zan. I am so sorry not to be able to be with you today, but I will be with you virtually and look forward to joining the discussion a bit later before I start. Huge thanks to the organizers, particularly Dr. Serkin and Dr. Martin for inviting me to speak. And I will take you through some of the latest findings on understanding the familial and hereditary aspects of endometriosis. I think there's a really good realization now that that endometriosis is a condition we have to view across the life course right from its first beginnings. Obviously genetics and susceptibility based genetics and genetic variance is very much part of that. But to really understand the causes of endometriosis and its progression and manifestations across the life course, how it causes symptoms like pain, infertility, and fatigue, and all the various symptoms that women can experience.
We've said this for many years. I mean, we really don't quite understand in depth the causes of endometriosis. We are fairly confident that quite a considerable part of the origins of endometriosis can be explained by retrograde menstruation, where cells that are viable at present in menstrual debris and they enter the pelvic cavity at three of fallopian tubes during menstruation. But we know that's essentially a physiological process that happens in the majority of women. So what we don't know is to what extent there are innate or acquire differences in the growth of any mutual cells or the way they adhere to pelvic surfaces. Are there innate or acquired altered differences in hormone regulation or in altered immune system or inflammatory responses, or in fact what the environmental factors are that impact on this? And genetics can really be a tool, an unbiased way in trying to understand the biological causes of endometriosis where that causation isn't clear.
So why do we study the effects of genes and genetic variation in endometriosis, but in any complex condition for sort of main reasons? One, as I mentioned, to uncover biological pathways involved in causing disease, providing potential targets for biomarker as well as treatment development to really understand the disease better from a subtype point of view. So identify disease subtypes that could lead us to a much better targeted biomarker identification and treatment development, weak term stratified medicine to provide measures to predict risk and ultimately be able to add all these effects of genetic variants provide so-called polygenic risk scores. And that's an area that's certainly developing across many different conditions now, and one which perhaps is still a little bit farfetched, but perhaps in futures to provide information on environmental factors that could interact with these genetic differences that are actually modifiable than these environmental factors that could reduce disease risk or prevented from occurring.
So before I move on to telling you a bit about the advances in genetics, understanding genetics for endometriosis, a bit of a genetics 1 0 1. So Genes basic unit of genetics, part of our DNA, that code for specific proteins as humans have around 20,000 of such genes. And we call that the genome. But these protein coding bits of the DNA only really count for about 3% of our DNA. The remaining 97% about 10 years ago still was referred to as essentially junk DNA. We now know that's certainly not junk DNA. It's incredibly important part of the DNA that regulates how genes are being expressed. How do we study the effect of genes on disease risk? Well, just as we study any ethological exposures really look at variation. So DNA has many variants, places where the coding difference between people in a population when that variation is common, we call it polymorphisms, where it's rare.
We call them mutations, but they're essentially different flavors of the same concept. And for kind of studies that we do to understand complex disease genetics, we look at what's called single nucleotide polymorphisms or SNPs, which are single base pair changes across the genome, which are highly abundant. We have millions of them and they act as a sort of reference, a map across the genome that we can look at in terms of associations with disease risk. When we talk about genetic variation, we have two categories, flavors of them, heritable and somatic. The heritable version, it is mainly from parents. There's a small proportion that arises, of course, de nova conception, but most of it comes from parents. They're typically presented in every cell. And that type of variation is what constitutes heritability of traits and diseases. Somatic mutations, mutation events in the DNA of a single cell, they're present in selected cells only.
They expand in tissues through clonal expansion, and they're the basis for most cancers. But I'm in interest of time going to focus this talk entirely on the heritable variation and it's linked to endometriosis. So we've known for a long time that endometriosis can run in families. We know that from studies that looked at first degree relatives of women with and without the disease from large geological studies and from twin registry studies both in Australia and Sweden. Heritability is essentially the proportion of disease in a population to genetic factors. And we know from the twin study data that proportion's quite sizable. It's about 50% of endometriosis risk in our populations due to genetics. And we also know that about half of that, so around 25, 20 6% is due to common variants, these common SNPs. So it's that proportion that we hope to detect through our genetic studies.
A question that always arises and is whether we can identify a single gene heritable subtype of endometriosis where the a particular single gene or some variants in a particular gene are responsible for most of the familial transmission of endometriosis. And in that sense, we often link our evidence to that, what we see for breast cancer. So for breast cancer, for example, the BRCA one gene that was originally discovered through the analysis of families with many breast cancer sufferers, and I won't go into too much detail, but essentially the familial study showed a lot score, which is a type of significance value of just over 21, which is a huge signal. And that was derived from just over 200 families with breast cancer sufferers and that gene and the variants within that accounts for about 60 to 70% for familial breast cancer in a population. Now, we did similar studies in endometriosis and found a significant lot score, but not nearly as high as observed breast cancer. We also looked at comorbidity. This is genetic correlation analysis, trying to see to what extent the genetics underlying endometriosis is shared with other conditions. And we group these into autoimmune inflammatory conditions, pain related conditions, reproductive traits and conditions and metabolic conditions. And this analysis was led by Peter Lo Muer and meta noga group.
I expect you won't be able to see the x axis here. All we've done here is grouped the various different conditions that we could look at on the X axis, and I'll talk you through that with the genetic correlation. So essentially the extent of genetic sharing on the Y axis. So we saw that for both asthma and osteoarthritis, there was significant genetic sharing with endometriosis. Almost all the pain related conditions we looked at showed significant genetic sharing, and that included things like migraine, lower back pain, hip and other joint pain, et cetera. And that really surprised us. There was quite a sizable genetic sharing there. Now we delved into that a little bit more to see to what extent there were certain specific genes and genetic loci that were shared looking at two pain conditions that showed the highest correlation with endometriosis, multi-site, chronic pain in migraine.
And we identified a number of SNPs that were essentially the same signal as the risk variance for endometriosis. And four genetic loci were entirely shared between endometriosis, multi sych, chronic pain and migraine. We also did this for osteoarthritis and rheumatoid arthritis, and this publication has just been accepted in human reproduction. And again, that led to three genetic loci shared between endometriosis and osteoarthritis, and one shared between all three conditions. Now, how can we use those associated variants? What's next? So we need to understand the biological effects of genetic risk factors on different tissues to identify new treatment targets. It's not as simple as saying this variant sits near a particular gene, because quite often we then realize that it's not actually that gene that's being perturbed in terms of expression, but a different gene in another location. And to do that, we need to do that in specific data sets.
This is just a slide to highlight that understanding and having genetic validation of potential new drug targets really increases the success rate of bringing that to market. And if you haven't read this particular paper by airing Erik Medical, I really recommend it. So it's not just the fact that, so quite often people argue that genetic effect sizes of a particular variant tends to be very small, and that is indeed the case. But that doesn't account for the fact that a lot of these different effect sizes within a pathway, a biological pathway, soon add up and really then are able to deliver new targets that are genetically meaningful. So we can link these genetic variants to gene expression datasets to DNA methylation datasets to see what happens at a tissue level, what these genetic variants actually do. We, we've done that for a number of different endometrium and blood data sets.
So doing that and linking this back to those results, again for osteo osteoarthritis and rheumatoid arthritis, we ended up with a number of biological pathways that are enriched in terms of their genetic drivers across all three conditions. And I won't take you through that in the interest of time, but there are certainly three key genes that came out of this analysis, particularly between endometriosis and osteoarthritis. One, a really interesting one, the hyaluronic acid pathway, which is a pathway that's currently being considered in of treatment for osteoarthritis and has also been suggested as a potential target for endometriosis. Now, another area we really want to look further in is this identification of subtypes of endometriosis, because we're really only just scratched the surface of this, and that's the aim of the next phase of the consortium analysis using deeply phenotyped data sets. So it's really this question of stage one, stage three, two, what does that really mean?
What does that designate in terms of the ultimate subtypes that mirrors? Essentially, we are looking at peritoneal versus ovarian, we're looking at inflammatory subtypes, et cetera, or are we looking at a whole range of subtypes that are distinct and could be treated differentially? And also looking at these comorbidities will be really informative and we need large scale data, detailed clinical and phenotypic data collected from many women to tease these effects at. And many biobanks just don't have that type of data. So over 10 years ago, as a large collaboration, Stacey Miser and I started a harmonization project, the Worth Effect Project, to harmonize and standardize the data and samples that are being collected for endometriosis research. We published a number of papers on this detailing standardized questionnaires for surgical sub phenotyping, clinical sub phenotyping, and also SOPs for fluid and tissue collection, and is actually just over 60 centers in 24 countries that are adopting these standards.
Some of these data sets are included in the next phase of the consortium. That's very much ongoing led by Stacey Misma, and we're hoping to delve much deeper into this phenotyping and understanding the underlying genetics of these, including the different pang sub phenotypes. Now, almost to the end, one really important part of what we do next is really to increase the ancestral nest ethnic diversity in our global studies to ensure that findings are applicable to all, and we really need to do better at that. Currently, we've obviously just used data that are available globally, but we're really trying to increase in the number of data sets and studies that consider endometriosis at a global scale. We just got funding from the Bilin Gates Foundation to do a pilot study working with some fantastic collaborators in Ghana, Kenya, and South Africa to do the first genome-wide genetic studies on endometriosis there.
And then lastly, the question often asked, what is the current utility for genetic screening on the basis of these variants that we found? Well, remember that the variants only really explain about 5% of disease risk currently, and that's limiting. So if you, and this is work done by PA Rodda and Peter Lo muer with Met Newgard, if you look at these GWAS variants and create polygenic risk score with an increasing exposure towards certain genetic risk factors, then the area under the curve, the A UC, which is a metric for how well a marker of disease does is only just over 60%. That's not really good enough for screening, let alone diagnosis. So there's quite a lot of work to be done there, and perhaps in future combinatory tools are the way forward. So just to finish up on the conclusions, no single gene for endometriosis explaining heritability.
It's much more complex than that. Stage three four disease is certainly different in genetic heritable origin from stage one two, but that's a very, very crude classification, slightly driven by distinct origins of verian versus peritoneal disease. But we need to delve into those different sub phenotypes much in a much deeper fashion. We know there's a common genetic basis between endometriosis and other pain conditions and other inflammatory conditions, and we can leverage that for drug repurposing avenues and possible subtype identification. We need, and we are doing much more detailed functional studies in relevant cells and tissues that are required to understand the impact of these variants on biology and pathogenesis. And there's real promise there, I think, for translating those findings into novel treatment targets. And lastly, as I mentioned, the utility of genetic variants for risk prediction currently is limited, but may improve in future. Lastly, a huge thanks to the many, many collaborators that have worked with us on these studies, but most importantly, to all the women who participate in the endometriosis studies across the world. Thank you very much for your efforts. And finally, a little plug for those of you, which I know all of you are interested in endometriosis for the 16th floor Congress of an endometriosis, which is happening in May in Sydney. Thank you very much for your attention.
