Christine Metz, PhD & Peter Gregersen, MD - The Analysis of Menstrual Effluent: Emerging Insights into Endometriosis Pathogenesis

Christine Metz, PhD & Peter Gregersen, MD - The Analysis of Menstrual Effluent: Emerging Insights into Endometriosis Pathogenesis

Christine Metz, PhD & Peter Gregersen, MD - The Analysis of Menstrual Effluent: Emerging Insights into Endometriosis Pathogenesis

Endometriosis Foundation of America
Medical Conference 2019
Targeting Inflammation:
From Biomarkers to Precision Surgery
March 8-9, 2019 - Lenox Hill Hospital, NYC

Thank you very much for the invitation, it's really wonderful to be here and see so many people who are so enthusiastic about learning more ways to help women with endometriosis.

One of the most difficult problems that we've learned about through working on endometriosis and working with patients is that the diagnosis of endometriosis can take up to 7 to 10 years and that's been touched on several times today. I don't think there is a single reason why this occurs, I think it's a very complex group of things but clearly one thing that is not in the scholarly journals and not in high impact papers if that there is a pervasive dismissal of women's pain and we've learned a lot about that through women with endometriosis.

In addition, the condition, as we learned today, is a very complex very disease, sometimes with vague and non-specific symptoms. Many women present, for example, with GI symptoms and they end up seeing a gastroenterologist as part of their diagnostic odyssey. In many cases, as we learned today, many women are symptomatic and they end up confronting infertility later in life and receiving their diagnosis of endometriosis. Finally, there are many women and teens who actually delay having invasive laparoscopic surgery for their diagnosis because they would prefer not to undergo surgery.

Clearly, there is a need for an early non-invasive diagnostic for this condition. We've talked earlier, [Dr. Balun 00:01:58] actually did a beautiful presentation on how the eutopic endometrium is really altered in this disease. He's shown very elegantly that aromatase production in the eutopic endometrium of endometriosis patients leads to increased local estrogen and prostaglandins and inflammatory response. He specifically showed the role of stromal cells in this process.

In addition, there are increased inflammatory cytokines steering the menstrual phase and that's been shown beautifully in the non-human primate models and in women. There are differences in the endometrial stem cells in endometriosis particularly the role of these stem cells in a process known as decidualization that Linda Judice has shown to be defective in endometriosis.

Finally, and I think this is one of the most compelling, is that there is cross-talk between the ectopic endometriosis lesions and the eutopic endometrium. This work was done in collaboration with Aski, who's here today, in which the changes in gene expression in the human condition between cases and controls versus in the baboon and you can see that there is clustering of up and down-regulated genes in the endometrium. That led us to reason that menstrual effluent or menstrual blood may be a reasonable thing to be studying if there is such a difference in the eutopic endometrium of women with and without endometriosis.

We began collecting and studying menstrual effluent about four or five years ago with the help of a wonderful nurse coordinator, Margaret DeFranco. We collect menstrual effluent during the first two days of the cycle when menstruation is the heaviest, most of earlier work was done using the Diva Cup which can be inserted for up to 12 hours and we recruited subjects with endometriosis through our Research OutSmarts Endometriosis ROSE study. Our control study of subjects come from our genotype and phenotype registry and both of these studies we have complete genotype on the subjects.

What's interesting is we learned a lot about collecting menstrual effluent. Women can do it in the privacy of their own home and they can send us their menstrual blood overnight where it comes to the lab and we process it. Most of our work in the recent years has been focused on the stromal cells, but we've also explored novel ways to collect menstrual effluent because women with pelvic pain have real difficulty inserting the Diva Cup or a menstrual cup to collect. So we've been using soft tampons that come out of Europe which is these gynotech sponges but we've also a developed an external pad which is adhered to a feminine hygiene product and we're able to collect the menstrual cells right from the pad once the patients can send them to us.

We've also been collecting endometriosis lesions on some of these patients who undergo surgery, and I just want to point out that those lesions actually contain the stromal cells that we're most interested in.

One of the most important jobs of the stromal fibroblast cells from the menstrual effluent or from the endometrium is to prepare the uterus for decidualization. This is a unique process, spontaneous decidualization is a unique process in women who menstruate, or in animals who menstruate. All other animals do decidualization after the fertilized egg comes along, non-human primates and animals that menstruate actually undergo spontaneous decidualization every single month in which these cells differentiate from a spindle-shaped stromal cell in the proliferative phase and they go to, under the direction of hormones which causes an increase of cyclocane MP inside the cells, they go into these nice plump stromal cells and they become protein factories. They make a lot of growth factors and nutritive factors that will prepare the endometrial lining for implantation. Some of those molecules are prolactin and insulin-like growth factor, binding protein 1 or IGFBP1.

If you look at the histological slides you can actually see these very small spindle-shaped cells become very enlarged protein factories to prepare the uterus every single month. This can recapitulate invetro using stromal cells isolated from very painful uterine biopsies or through stromal cells collected from menstrual effluent. We use a bromal cyclocane MP which can permeabilize the cell, plus or minus hormones, and we measure decidualization index by measuring IGFBP1.

Previous studies by Linda Judice's group, as I've said before, had reported that with an NF3 endometriosis subjects, their cells decidualized less when assayed in this assay. We developed a method to measure the conversion of the cells into decidualized cells and it took about, I don't want to say, but almost two months and we realized that this would not be a way to improve the diagnostic, it would be taking too long. With the help with of a post-doc in the lab who's here today, we've collected menstrual effluent, cultured the stromal cells from subjects with and without endometriosis and we performed decidualization assays in a micro-assay using a 96 weld plate and measured IGFBP1 levels. With approximately 30 subjects who have pathologically confirmed endometriosis versus control subjects, we find a very significant difference between IGFBP1 production or their decidualization capacity.

In addition, we've profiled these stromal cells to compare them to endometrial stromal cells that are isolated through uterine biopsies and most of the markers are almost identical to what's observed in the biopsies, but we also decided to do was to start looking at some of these markers that are associated with inflammation and aggressive pathogenic stromal cells. The first one here is fibroblast activation protein alpha or FAP alpha, it's a serine protease that cleaves after proline and it can digest matrix and it promotes cell migration and invasion. You see a lot about FAP in the cancer literature and it's associated with metastases. It's also up-regulated by cytokines such as IL1 and what we see is a significant increase in the expression of FAP1 in patients with endometriosis in their stromal cells.

Similarly, another marker podoplanin which is also associated with invasion and migration, also upregulated in cancer-associated fibroblasts and induced by TNF is also significantly up-regulated in the stromal cells isolated from menstrual blood of women with endometriosis.

We have three ... I guess we missed one here. We also have another factor shown here in this slide, ALDH1A1, we've done complete gene expression of cells from patients with and without endometriosis. One of the most highly down-regulated genes in endometriosis is ALDH1A1. That encodes a protein that converts retinol to retinoic acid and retinoic acid is known to play a very important role in regulating cytokine production. So we have three ways that we can discriminate between patients with endometriosis and without endometriosis and we're hoping to use these to develop a screening diagnostic.

Now I'll hand over the podium to Peter to talk about this role of inflammation that we've identified.

So it's really great to come to this meeting because we have a learned about a lot more things that we could be measuring these cells that might be useful for a diagnostic. I think we're getting close to having something that really can have pretty high specificity and sensitivity.

We're thinking about inflammation and wondered what TNF and IL1 would do decidualization in and actually in the literature there's plenty of evidence that TNF, well, particularly TNF and IL1 do in fact inhibit decidualization by stromal cells, this is just an experiment showing that it really takes extremely low quantities of TNF to inhibit decidualization of these cells. To do this acutely over a 24 hour period you, interestingly, IL6 is immediately produced in response to relatively modest levels of TNF and, interestingly, the cells also increase their proliferative capacity, they're not decidualizing, they're continuing to grow, they're producing cytokines.

The really striking thing is that if you give them TNF for a couple of days and then remove the TNF, grow the cells, split them in the absence of any other oxygenous cytokines and look at them a month later, they still have these phenotypes. So there's kind of a switch going on it seems in response to an inflammatory cytokine, we've done some of this with IL1, the proliferation of these cells continues out to a month beyond what cells that have not been previously treated with TNF have. We've seen some elevation of some of these markers but we haven't looked out distantly yet to see if they're maintained but it does suggest that the reason that we can see this decidualization phenotype in menstrual blood-derived stromal cells, even after multiple passages, is because that abnormality has been induced in them by exposure invevo to TNF, IL1 or other inflammatory cytokines.

We think that there's a switch going on with these cells and that intrauterine inflammation is a big part of what this abnormality is about. This is sort of a model invoking inflammation, we think that relatively high levels of TNF and IL1 in the intrauterine environment that could be there because of chronic endometritis, it could be there because a person has perineal endometriosis lesions that are making these cytokines and that's what you see in baboon models. There may be ongoing systemic inflammation from elsewhere, such as inflammatory bowel disease which we think is an unexplored association between endometriosis and we would be a source of TNF. Actually, IBD has high circulating levels of TNF in the serum typically.

Obviously genetic predisposition. We've heard a lot about genetic predisposition, it might be that only some people are responsive to this or are making this stuff or that their cells are not as responsive to these inflammatory cytokines as normal. Again, one could imagine then inducing all of these things, including retinoic acid deficiency and the progesterone resistance. So which of these come first in the cascade, we don't know, we'd like to use these cells to ask that. But the result of this is once they get in the perineum, they're gonna have increased proliferative capacity, they've got overexpression of these molecules that are known to be invasive in terms of fibroblast function.

Many of you may know that I've spent most of my life doing the genetics of rheumatoid arthritis, rheumatoid arthritis is really interesting now because there are four or five different fibroblast subsets that are found in rheumatoid synovium and those that are in invasive and pro-inflammatory both of these molecules are highly expressed. We think that fibroblasts are in fact driving the inflammation in rheumatoid arthritis joint and may well be driving some of this inflammation.

Interesting, podoplanin is one of the only known ligands for CLEC2 which is on platelets. These things are bleeding, they're platelets in there and the result of that interaction is the production of cheap TGF data which probably also promotes the differentiation of these cells.

We think that inflammation really may be a very primary thing and it may be asymptomatic inflammation within the uterine lining that's generating a difference in these cells that get into the endometrium. I'll just finish with, again, the early diagnosis is what's been driving our interests in getting menstrual blood and analyzing it. We think a good place to look for this and try to incorporate this pad. We just have developed this pad in the last 6-12, 6-10 weeks. A company outside of New Haven has designed this for us, we've used it, a person can put this on for three-four hours, we get tons of cells, they grow like crazy, stromal cells really grow very well, we can demonstrate the decidualization deficit. This avoids the use of the cup which obviously is an issue for many people and we'd like to know if we screen adolescent populations that have symptoms, that have missed schools, that needs to be on OCPs have non-cyclic abdominal pain which seems to be a typical symptom of endometriosis as discovered by our friend [ Noyami Aladodu 00:17:09] has developed an app for this.

We're developing another app with her where we're going to integrate assessment of symptomatology and serial assessment of stromal cell phenotypes in a population of adolescent girls. To see if we can see a correlation between symptomatology, differentiate just primary dismenorrea from something that looks more like endometriosis.

Many people, obviously, have contributed to this and I'll point out Laura Warren who just got her PHD thesis on this. Margaret DeFranco has mentioned again who's our master recruiter. We can tell you more about the ROSE study afterward. We are very grateful to the Endometriosis Foundation of America and to Diva Cup for giving us free Diva Cups initially. Northwell just gave me and Christine a half a million dollar Innovation Award last year to develop this is a diagnostic and I think we are on the way to doing so. We're really looking forward to integrating it into population studies.

So thank you very much.