Michael S. Rogers, PhD - Toward New Approaches to Endometriosis-associated Pain

Um, he is, um, assistant professor of surgery at Harvard Medical School. Um, he has postdoctoral training here after receiving his PhD from the Mayo Graduate School, um, with the DAmato Lab and the Vascular Biology Program, and, and, um, he has been, uh, teaching here at HMS since, um, and has been a well-respected teacher and won many teaching awards with, um, his role there and has been a very productive researcher, multiple NIH, um, funded grants, RO1 grants, um. Uh, working in, uh, various fields, um, mainly around the area of, uh, angiogenesis and, um, novel, uh, sort of opportunities in the field of angiogenesis, and he's here to speak to us about one exciting area which is, um, um, new approaches for endometriosis associated pain. So, um, thank you, Doctor Rogers. Thank you. Um, and I have nothing to disclose. So if somebody would like to arrange that, I'm happy to think about it. Um, so, I'm gonna talk today about endometriosis, and I think it's useful to, when you, um, start, uh, talking about the disease, um, to get a little flavor for what it's like, um, from a patient's perspective. So I'm gonna give you a brief vignette um that describes uh, a fairly typical uh experience. This isn't any one patient, but it's a combination of several. So shortly after menarchy, uh, 13 year old Emma begins to experience increasingly intense pain every month. Her mother gives her ibuprofen and tells her that menstrual cramps are a normal part of being a woman. Initially, this helps, but as the months go by, it becomes clear that ibuprofen isn't helping enough and Emily starts, or, and Emma starts missing school. A visit to Emma's, uh, pediatrician results in a prescription for oral contraceptives, and when taken faithfully, these dramatically improve her symptoms. However, occasionally, she still experiences pain, the fact that she and her mother attribute to missed doses. In her junior year of high school, this becomes more common and she starts missing school again, even though she's careful to take her pills every day and is taking substantial doses of ibuprofen. By late in her senior year, she's skipping a few days of school every month and winds up missing her high school graduation because she isn't feeling well. During her freshman year of college, her roommate encourages her to visit student health services after an especially painful bout, and during that visit, she describes her newest symptom, dyspepsia. And she's referred to a local gastroenterologist. When she finds nothing wrong, uh, Emily returns to student health services where she describes her symptoms more fully and is referred to a gynecologist who makes a presumptive diagnosis of endometriosis. After considering other options, Emily decides on surgery, and this is, um, what her surgeon saw. So, um, this is, uh, underwater and you can see, um, Uh, these lesions here that are slightly attached, um, to the peritoneum. Um, such lesions, uh, tend to be quite small, um, but, uh, they can have other appearances as well. Um, you can see, uh, red lesions, clear lesions, these, these brown lesions or black lesions. All of these, uh, are, uh, examples of, of what the disease, uh, can look like surgically. And it's very common for a situation, uh, uh, like, uh, uh, Emma's to result in, uh, a diagnosis of endometriosis. And as optics have gotten better, it has become more common to see, um, lesions in patients with, um, with, with, uh, pelvic pain, uh, that, uh, isn't, uh, effectively treated by, by, um, uh, ibuprofen or, or other meds. Um, the disease can, uh, occur, uh, in a variety of different places. Most common in the cul-de-sac, but also anywhere, uh, throughout the peritoneum. Uh, when it's on the ovary, it tends to grow quite large and results in endometriomas. It can also appear on the uterus, in which case it's adenomyosis. Um, other common sites are, uh, intestine, bladder, and, uh, uh, the uterosacral ligament. Um, but distal organs can also be affected. Um, uh, endometriosis on the lung is common. Uh, it also appears on the skin and in other places. Um, it can have a variety of appearances as I showed you previously. Um, it can be quite large, uh, several centimeters in the case of endometriomas, or it can be, um, very small. So, um, uh, the smallest lesions that, that, uh, are commonly observed. a few 100 microns across. Um, it can range throughout various colors, and there is a staging system that has been developed by the ASRM that is moderately useful in predicting fertility. It has no correlation to pain. Um, but, uh, it, it, it does, uh, just, it, it is used to describe, um, different stages of endometriosis, though it is not clear whether patients progress through these stages or not. Um, that staging system takes into account, uh, the location and size, uh, of the lesions, whether the ovary is involved, um, whether the cul-de-sac remains intact. Um, and the extent of, of adhesions and whether those, uh, adhesions, uh, include, uh, or, or, or affect the ovary or, or the, um, tubes. Um, the disease itself is proliferative and invasive. As I indicated, it is also metastatic, so it is a disease growing in a location other than it, than it should be. It is, however, not dysplastic, and that differentiates it importantly from cancers. Um, the disease can take on a variety of, uh, of appearances, but most typically you will see, um, a dense stroma, um, uh, with glands embedded in it. Um, it is also very, very common to see hemosiderin, and, uh, two of those three, features, so, uh, endometriotic, uh, glands and, uh, endometriosis or endometriotic stroma and hemosiderin, two of those three is sufficient, uh, to, to generate a pathologic diagnosis of endometriosis. Um, so how big is the problem? Uh, endometriotic lesions, um, are an incidental finding, uh, in a, in, uh, a wide variety of abdominal and pelvic surgeries. They are, um, the, the, the fraction of women whose lesions cause pain is currently unknown. Endometriosis is the most common cause of infertility, uh, in the United States, and, and lesions are probably present in about 10% of women. Um, they can occur, as I said, anywhere in the body. There are rare case reports of men with endometriosis, but that all of those men, um, have been, uh, uh, taking drugs that are estrogenic. Um. Symptoms at, uh, children's have been observed, uh, in, in, uh, girls as young as 7 years old, uh, with surgical diagnoses as young as 8 years old. Uh, but it's very uncommon for that, for, for a diagnosis to be made so young. In fact, uh, a diagnostic delay of between 7 and 11 years, um, is common with, uh, delays, uh, in the, in the Northeast of the United States, uh, being around 8.5 years right now. Um, a woman will typically see 7 to 9 healthcare practitioners before diagnosis, indicating that, that this is not something that, that is at the top of physicians' minds, um, as they're looking at, at, uh, pain symptoms. And the cost of endometriosis, the economic cost is about $160 billion in the US. So that includes, um, lost work, that includes medical costs, um, uh, and other costs as well. So to put that into perspective, um, that's, uh, uh, just below the cost of type 2 diabetes in the United States. Um, so what causes, uh, endometriosis? Uh, the most accepted theory, uh, is Samson's theory of retrograde menstruation. Uh, retrograde menstruation occurs in about 90% of women, but only a fraction of those women go on to get, uh, uh, endometriosis. Um, shorter cycle time, uh, which results in increased exposure to, uh, uh, uh, menstrual fluid increases the risk supporting that. I'll also show you some, some data in a few minutes that also supports that. Um, but that doesn't, that can't explain all cases of endometriosis, in particular, endometriosis, uh, in men can't be explained that way. And so there are probably some fraction of cases, um, that are explained by, um, uh, the, by incomplete, uh, regression of the mallerian duct during embryonic development. Um, there's also a suggestion that there may be some cases that result from local metaplasia. Um, and this is used, uh, um, or, or this is used primarily, um, to explain, uh, premenstrual endometriosis, although there are other explanations, uh, uh, for that observation. Um, endometriosis is at least partially a genetic disease, so twin studies indicate, uh, about 50%, uh, heritability with about half of that heritability, uh, uh, being, uh, explicable by common, uh, single nucleotide polymorphisms. Um, so the most recent, uh, uh, data is a meta-analysis of 11, uh, genome-wide association data sets. Um, with 17,000 cases and not quite 200,000 controls. Um, this only explains a minor fraction of, of the variants that could be explained, um, so about 5.2% of the variants. And, uh, the adjacent, um, genes, uh, fall into, uh, two major categories, one of them having to do with hormonal effects. So, um, uh, for example, the estrogen receptor, um, uh, is, is, uh, one of those, uh, uh. Low side that's affected. Uh, also genes having to do with the microenvironment, so win, fiberectins, uh, VEGFR2, which I'll talk about again in just, just a little while. Um, and then, uh, uh, genes where we really don't have a good clue as to, uh, what, what might be going on. In addition to germline, um, uh, susceptibility, there's also some indication that somatic mutations may affect, uh, endometriosis. Uh, this was first suggested by the fortuitous finding, uh, of, uh, Daniela, uh, Dillescu, um, in the JAXS lab in 2000 in the early 2000s, where she She was actually trying to generate an ovarian cancer model, um, and, uh, wound up with her genetic manipulations also affecting, um, uh, other parts of the reproductive tract and, and found that, uh, activating KRAS, um, can result in lesions that look very much like endometriosis. Um, that was followed up, uh, by, uh, Stephen Charnock Jones, um, uh, and, and, and they observed that activated KRAS in donor, uh, epithelium and stroma will promote lesion growth. It will allow the growth of lesions that otherwise could not grow, um, in a, in a mouse model. And finally, um, there are recent studies, um, uh, demonstrating, uh, that in fact KRAS PI3 kinase. And a number of other mutations that don't appear here, um, are present in, uh, in, um, the, the lesions of, um, uh, endometriosis patients. So, uh, KRAS, PI3 kinase, uh, they're both present in deep infiltrating, uh, endometriosis and in endometriomas. Uh, peritoneal, uh, endometriosis has not been assessed for, uh, for, um, For mutation and that's something that's, that, that, that we're currently working on in my lab. Um, other genes, uh, involved in this include, um, uh, PPP2R1A, um, and AID1A, which is a transcription factor that's associated with, uh, with ovarian cancer. So, we have a sense for what the disease is and what causes it. What are the options? So in the case of Emma, um, we, uh, we, we went through medical therapy, um, and then, uh, drugs that target estrogen. So in addition to oral contraceptives, uh, aromatase inhibitors, um, and progestins in higher doses are also used. Um, androgens, uh, have been used historically. The side effects make them, uh, disfavored currently. Um, uh, particularly with, uh, increased use of GNRH receptor agonists and antagonists, uh, just a couple of months ago, um, the first, uh, antagonist, uh, uh, GNR GNRH receptor antagonist was approved for use in endometriosis. Surgery is, uh, also, uh, a common therapeutic option for, um, for endometriosis patients. Um, for patients who, um, are, are diagnosed, uh, for endometriosis because of infertility, um, surgery often enables pregnancy to, to be established and proceed normally. Um, it can also provide pain relief, although this relief is often only temporary. Um, and a combination of, of surgery followed by some of these other therapies can be more effective than, than, uh, surgery alone. Um, and so, so, so those are options, but it is very common for patients to require multiple surgeries, um, over their reproductive years in order to achieve pain relief, and, uh, there are patients for whom surgery does not, uh, result in, in relief of pain. So, from the perspective of, of, uh, drug, uh, discovery, um, there are currently two targets, uh, really for, um, for, uh, endometriosis therapy. One of them is COX-2, is targeted by NSAIDS, and the other is estrogen production. Um, and we've asked the question, uh, is there more? Um, and one of the approaches that we're trying to take is, is, is to try to, uh, to, to learn from, um, the, the surgical experience, and there are, um, endometriosis patients who present with widespread, uh, lesions and widespread disease, but no pain. They present because, uh, they can't get pregnant. And so that tells us that the lesions need not be accompanied by pain and so, um, uh, essentially, um, our, our, what we're trying to do is go from a place where a patient, um, has, uh, pain to no pain, even, even though the lesions might remain intact. Um, so basically, can we shift the patients towards, uh, a, a no pain phenotype. So, um, as we began this work just a couple of years ago, uh, our first thought was, well, let's go and let's look in the literature and let's find the best models that we can of, of endometriosis-associated pain. And when we looked, uh, in the literature, we found exactly 0 models of endometriosis-associated pain. And so, uh, we have set out over the last 2 years or so to develop a, a mouse model of endometriosis-associated pain. So, what we want, uh, from this model is, or, or, or the, the benchmarks that we set out for ourselves are, are that it reproduced the important, uh, features of the cellular and tissue pathology of the disease so that it looked like endometriosis, and that the animals behave as though they are, um, uh, experiencing pain. Um, and that it recapitulate, uh, the human drug response. And one of the main reasons or, or the main motives for developing this is that a successful model will enable us to rapidly test candidate molecules, um, first with the, with the first goal of repurposing approved drugs and with later goals of, of developing our own drugs. So, there were, uh, at the outset of this, many protocols, um, that produced glands, stroma, and, and hemosiderin, um, but almost all of the reported lesions from those models were clear, indicating that, that, um, that, uh, they don't recapitulate the, the spectrum of disease. Um, we had no models of sensory testing. Almost all of the models out there involve surgery, um, which can confound measures of pain. The pain from the surgical incision and manipulation, particularly with human-sized fingers on a mouse, um, uh, is likely to, uh, confound any disease-associated pain. Um, and almost all of the existing models, uh, included estrogen supplementation, which meant that we would be unable to analyze estrogen modulators, which is I, which as I said, are a major target of, of medical therapy currently. So the model that we developed, um, is to pre-treat a donor uh animal, um, with estrogen, uh, take, uh, uh, each of its ovarian or, or each of its uterine horns, um, uh, chop them up and inject them into, uh, a recipient. Um, it sounds crude, but it works. And, uh, and, uh, that results in implantation, uh, of lesions as early as day 8 and, and well established lesions by day 14. Um, these lesions, uh, by about 2 months later, um, show a wide variety, um, of different colors, um, and, uh, uh, histology that is typical of the disease. So you see here, um, glands, uh, you see blood vessels, you see, um, uh, you, you, you see muscle tissue that is commonly found, um, and with special stains, we also see nerve infiltration, uh, into the lesion. These lesions express estrogen receptor, um, and we also see, um, uh, infiltration of immune cells, um, just as we see in human patients, and, uh, that, that is not just local to, to the lesions, but we see, um, a general involvement of, of the peritoneum as well. Uh, so, we have now, uh, a, a model that, uh, doesn't supplement with estrogen, meaning that we'll be able to test, um, estrogen modulators, um, and, uh, it, it, uh, reflects the, the histology of the disease. Um, but the next, uh, and tougher question is, do these animals feel pain? Um, so to, to get at that, we first have to start with what is pain. We all kind of have an intuitive knowledge of it. Um, but as you start to move into animals, you have to break that down so that you can assess whether the animals are experiencing, um, what we would call pain. So pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage or is described, uh, in terms of such damage. So importantly, there are sensory, in other words, peripheral components to pain. And also, uh, emotional or central or brain-related, uh, uh, components of pain. And both of those can be important in, in, uh, humans and are therefore important to be able to assess in animals as well. So how do we measure pain in rodents? Um, to measure the peripheral part of pain, we use arc reflexes where essentially receptors, um, uh, uh, in, in the skin, uh, are activated, send signals through the, through, um, the afferent neuron, um, to inner neurons, um, and then result in, in, uh, reflex, uh, responses, uh, in, in the animal. So, uh, for those, there are two main things that we look at. One of them is mechanical hyperalgesia, so increased sensitivity to touch, and the other is thermal, thermal hyperalgesia, increased sensitivity, um, to thermal stimuli. I can tell you that, well, early on we had indications. that we were seeing, uh, changes, uh, in, in, uh, thermal response in the paw, even though we have, uh, the lesions in the abdomen that didn't hold up, but we do see, uh, substantial and reproducible changes, uh, in, in mechanical sensation. So this is what the assay looks like. Uh, essentially, you see, um, a calibrated fiber that we go in and touch the animal with, and you'll see the animal respond. This one gives a nice horse kick. Um, uh, this one you can see, um, uh, just, uh, a flinch and a, and a, and a rapid movement. Um, it takes a fair bit of training for someone to learn to recognize the difference between normal mouse movement and a response, but, but we're fortunate to have in our lab, uh, someone with about 1010 years' experience in these types of assays, uh, a man by the name of Victor Batori, and he has been very helpful in getting these assays, uh, up and running and be reproducible. So, um, with that said, this is what we saw, as I said, with the thermal, uh, measures of thermal hyperalgesia. So we don't see substantial changes, uh, in the adjacent paw in, in its response to heat. But we do see, uh, these are sham animals, animals that receive saline injections, uh, both the red. And the blue lines, uh, here indicate, um, uh, mice that received, uh, uh, uh, tissue injections, and, uh, you see, uh, a drop in the sensitivity threshold indicating that the animal is more sensitive, so it's sensitive, uh, to a smaller stimulus. Um, and that is, uh, in the, um, in the placebo-treated animal, uh, true throughout, um, the, the, the duration of the experiment, so throughout all eight weeks. Uh, we start treatment, uh, with ibuprofen at, uh, at 4 weeks here and we see, uh, a relatively rapid, uh, reversal, but it's not a complete reversal of, of the, this behavioral change in the animals. Um, it's not just ibuprofen that works. Uh, letrozole, uh, so an aromatase inhibitor also results in a reversal of, of hypersensitivity, um, as does the androgen danazole. Um, and, uh, we also see in the case of, of letrozole and danazole. We see a, a, a regression of about 60% of the lesions. Um, we do not see, but, but, but the remaining lesions are not of substantially different size, uh, um, indicating that, um, that, that, that these, uh, drugs are working as expected. Um, but that's just part of it. So that's just, uh, the peripheral part. So, so this, this, this, uh, arc reflex, we also wanted to know whether there's, uh, uh, a central aspect, um, to, to the changes in the animals, uh, and for that we have to look for spontaneous behaviors. Um, one of the big problems in the pain field in general has been the use of, of these peripheral measures of pain. And it turns out that they have not been predictive of drug efficacy in humans. So there are a number of compounds that have, uh, gone through using, uh, reflex measures, uh, made it all the way into humans, and, and the results have been disappointing. And so one of the The directions that the pain field has been going is towards, uh, spontaneous measures that are believed, uh, to, to, um, reflect, um, uh, not only, um, reflex, but also central processing of the stimuli. So the first, uh, attempt at this was using something called a palm reader. Um, this is, uh, a device that's in, uh, development in the wolf lab to try to, um, get at where the mouse is putting weight on its paws. Um. And this mouse right here has a lesion on this side right here. So, uh, on the right side of the animal, and you can see that it's putting less pressure on that paw than it does, uh, on the other hind paw, indicating that it is, uh, uh, uncomfortable with putting substantial pressure there. Um, another measure, so that's mechanical sensitivity. Another non-evoked measure of pain is thermal place preference. So essentially how warm the animal likes the room. Um, and to measure that, we, we allow the mice to run back and forth, uh, along, um, a, a strip, one end of which is at 50 degrees and the other, uh, end of which is at 7 degrees. Um, with the thought being that, that a change in where that animal rests, uh, will indicate a change in its thermal sensation. Um, and so this is what that looks like. Um, so this is, um, a sham mouse, and this is a mouse, uh, that has endometriosis. This is, um, about 15 minutes, uh, total, uh, real time. And you can see that the endometriosis mouse, um, uh, actually, uh, has a hard time finding a place where it's comfortable. It moves back and forth, um, it will settle, uh, in a, in a temperature that's a little bit different from that of the sham-treated mouse, but the biggest difference that we saw is that the endometriosis mice, uh, can't get comfortable. They move back and forth, uh, looking for a place, um, but never find a place where they're actually, uh, comfortable. Um, quantitating that. Um, Uh, we, we look at, um, at, uh, uh, this is a sham-treated mouse here, so it prefers a temperature, um, of about, uh, 34 degrees, and endometriosis, uh, mouse, you can see it spreads out, uh, the time that, that it spends in different zones, uh, more than does, uh, the sham treated animal. Um, and, uh, it prefers, um, the, the, this mode here is a little bit, at a little bit higher temperature. Um, the best measure that we found, uh, for this, uh, instability of, of thermal place, uh, is the absolute sum of squares, um, of, of the difference from the mean, um, and that increases fairly substantially in the endometriosis mouse. When we treat with letrozole or danazole, that is reduced, uh, indicating that the animal, that, that, that the symptoms that the animal is experiencing are experiencing are reduced. Um, we also look at the number of times the animal spends in a particular zone. So basically the number of times it moves, uh, back and forth. And again, we see, um, a wider spread with the endometriosis mice, um, and versus the sham-treated mice, and that is again reduced with, uh, letrozole and danazole. Um, but importantly, uh, so that's, that's thermal place. Um, one of the things that this experienced graduate student, uh, noted also when he was looking at the animals to, to do, um, other assays, um, was that they were exhibiting a behavior that he had only previously seen, uh, when mice are injected, uh, with, uh, acetic acid or carrageenan, both strong irritants, uh, in the abdomen. Uh, that's a behavior that, that is commonly called writhing or abdominal contortions. Um, and so you can see this animal, for example, essentially stretching out its abdomen. Um, and the idea there is that it's, it's, um, activating stress, stretch receptors to try to reduce the pain. Um, you see that in this animal, uh, and this animal as well. Um, so, uh, this is exciting for people in the pain field because this is the first time that abdominal contortions. have been observed in a disease model. Previously, as I said, uh, strong irritants, uh, have to be injected in order to observe this behavior. And, uh, so, um, we're excited about that. Uh, another thing that, that was observed is called direct abdominal licking. So, this is, uh, what an animal looks like when it's normally grooming itself. It will move, uh, in, in a systematic way from place to place across, uh, its body. Um, that's, uh, that, that, that's a typical behavior. You see this animal, uh, goes straight to a position on the abdomen adjacent to where the lesion in this animal is, indicating that it's trying to rectify something with its grooming behavior. And finally, um, another spontaneous behavior indicative of mechanical, uh, of differences in mechanical sensation is squashing. So essentially, uh, normally the animal will keep its abdomen off of the cage floor. Um, and when, uh, when you look at animals with endometriotic lesions, you see it's most clear in this animal, where you can see the hair poking, poking out through the bottom of the cage floor. It's, uh, it's pressing its abdomen, uh, against the cage floor. Again, uh, our interpretation of that is it's trying to relieve an abnormal, uh, sensation. So, um, uh, as was true with thermal place preference, we also see with these spontaneous behaviors, uh, that, that, um, drugs that relieve, uh, symptoms in humans also relieve these symptoms in animals. So letrozole here with abdominal contortions and danazole, um, via the direct abdominal licking, uh, again, letrozole and danazole, um, relieve those, those, uh, uh. Um, behaviors, um, and, uh, we also see less abdominal squashing. Um, so, uh, with the strongest effects being on contortions, uh, and, and squashing. So again, this is an indication that drugs that we know work in women are also relieving these behaviors, uh, in animals, suggesting that, that both these animals are experiencing pain and that the drugs that relieve pain in humans also relieve them in animals, demonstrating that these, that, that these assays, uh, are likely to predict the effect of, of future drugs. So, um, one of the reasons that these animals become more sensitive, um, is because of changes in the peripheral neurons. Um, and, uh, so to assess whether there are, uh, changes in peripheral neurons, uh, in, in, in these animals, um, we, uh, isolated dorsal root ganglia from the animals, um, 8 weeks after, uh, initiation of, of the lesions. Um, we then looked to see whether there are changes in, in calcium signaling. Um, so we look, uh, at, at sham and, uh, endometriosis animals. Um, we, we, we record for 2 minute, uh, 2 minutes to get a baseline, then we add either capsaicin or AITC which activate two different, uh, sensory receptors, uh, so calcium channel sensory receptors, uh, in these neurons. And then we depolarize the neurons with potassium chloride in order to get a sense for what the maximum response, uh, possible in, in the, the neurons is. Um, and so you can see here, um, that, uh, so black line is the sham. Um, red line is the endometriosis, uh, model, and we see, uh, an increase in response to AITC, so the tripA one, agonist, that's, uh, often called an itch receptor, um, but it, it, it responds to a number of different stimuli. Um, uh, and there is not a significant difference in the potassium chloride response indicating that, that, uh, the big difference is probably upregulation of tripA1, uh, responsiveness in these neurons. We also see a statistically significant increase in the fraction. Of dorsal root gang ganglion neurons that are responsive to AITC, uh, indicating that not only has the sensitivity of individual neurons been increased, but also additional neurons, uh, have been increased to be responsive to these stimuli. We see the same thing with TRV1. So again, an increase in responsiveness to capsaicin, so this is a burning response. Um, uh, and, uh, and, uh, so we see increased neurons and increased responsiveness to capsaicin. Um, with a non-significant, with no significant difference in, in the response to, uh, potassium chloride. Again, indicating that, uh, both for mechanical and for thermal stimuli, and increasing, uh, number of neurons and an increasing sensitivity of the neurons themselves. So, uh, we're definitely seeing peripheral sensitization in these animals, um, as demonstrated by these neuronal readings. What's causing that, um, Uh, a very early experiment designed to try to assess this is to take extracts from the lesions and put them onto naive, uh, DRG neurons and look to see, uh, uh, what happens and in fact, uh, lesion extracts do, uh, stimulate, um, uh, a fraction of both, uh, capsaicin responsive. And, uh, AITC responsive, uh, neurons from naive animals indicating that there are actually, um, uh, chemicals, either proteins or small molecules in the lesions themselves that are, that are able to stimulate, uh, uh, the neurons. So, what we've got now, um, uh, after a couple of years of work, um, is an animal model that we think accurately reflects, uh, what it is that, that, uh, women with endometriosis experience. Uh, the animals exhibit, um, mechanical hyperalgesia. They may, um, alter their weight distribution. They're showing, uh, um, central responses, so, so, uh, planned, uh, or, or behaviors that are consistent. Um, with, with, uh, central sensitization, uh, as well, or at least a central response. Um, the dorsal root ganglia from, from neurons from these animals, um, showed changes in calc in, in, uh, calcium responsiveness that are consistent with peripheral sensitization as well. And, um, we've been able to show that drugs that relieve, uh, symptoms in Women also, um, reverse some of these changes, uh, in the animals themselves. So we think that, that we have a model that will enable us to, uh, to begin to test approved drugs to see whether they might be repurposed for, um, for therapy and endometriosis. Um, and, uh, this is our current list. So we have some matrix metalloprotease inhibitors, um, uh, that, uh, may inhibit the tissue remodeling that's associated with, with, uh, establishment and progression of the disease. Um, we have some atypical, uh, uh, VEGFR2 antagonists, and I'll describe why it is that we're looking at those, uh, in just a minute. We have, um, uh, a couple of anti-inflammatory agents. We have metformin because metformin seems to do everything. Um, we have sulfasalazine as a recommendation from our, uh, pain collaborators, uh, in the wolf lab, um, and then a number of other, um, uh, drugs, uh, that we hypothesize, uh, may, uh, affect, uh, endometriosis pathology, uh, for various reasons. Gabapentin, for example, um, has, uh, recently been used more commonly in, in endometriosis that's, that's refractor. to other therapies and, and seems to um uh show some activity there. But I am open and I would appreciate an email or, or, or your recommendation of anything else that we should be testing uh in this model um because uh we're now in a position for the first time, uh, to actually be able to at reasonable cost, uh, assess whether an existing approved drug might be, uh, effective in the disease. So, with the few minutes that I have left, um, uh, I'll talk about another approach that we're taking, uh, to identify new therapeutic targets. Um, so as I said, current targets are just, uh, cyclooxygenase and, uh, estrogen receptors. Um, but, uh, we, we, we noted, uh, an old friend in the list of, of, uh, loci that are linked to, uh, endometriosis. This is VEGFR2. Um, and so we thought, uh, very quickly, uh, about what might be done, um, to, to test to see whether VEGF, uh, our two antagonists might be effective. Um, uh, there are a number of things that indicate that, that VEGFF plays an important role and angiogenesis plays an important role in the disease. Uh, in addition to kind of the obvious, uh, surgical, um, uh, observation where we see blood vessels, um, growing into the lesion, um, it's also been observed that increased serum-soluble VEGFR1, a natural, uh, VEGF antagonist, is correlated with, uh, less severe endometriosis. Um, VEGFE is also a survival factor for endometrial epithelial cells, uh, which express both VEGFR1 and VEGFR2. Uh, danazole therapy decreases serum VEGF levels, uh, in patients, uh, with endometriosis. closer to that of patients without endometriosis and GNRH analogs reduce VEGGF production, cultured endometrial cells, and in, in, uh, uh, and they also reduce uh peritoneal fluid VEGF in patients uh with endometriosis. So we said, all right, we've got a list, a long list of, of VEGF, uh, R2 antagonists that are currently being used for cancer. There's just one problem. All of them are teratogens, and so would be inappropriate for use in a, in, in, in a young female, uh, uh, fertile population, uh, especially in women who may be trying to get pregnant. And so, um, we turn to, uh, another, uh, class of VEGFR2 antagonists. These are the dopamine D2 receptor agonists. So essentially what they do, um, is they down regulate VEGFR2 protein. When they bind to, uh, dopamine receptor D2, um, not only D2 receptor gets internalized, but VEGFR2 gets internalized together as part of a complex, um, and as a result, uh, you have less VEGFR2 on the surface, uh, of cells. Um, both cabergoline and quinagolide inhibit mouse models of endometriosis, so they inhibit, uh, lesion establishment and growth. Um, uh, in human and mouse endometrial xenographs, it inhibits, um, inhibits growth, it reduces lesion angiogenesis and reduces VEGFR2 phosphorylation, and quinagolli in a small proof of concept clinical trial, um, reduced, uh, the, the number of lesions that persisted in second-look surgeries. Um, and, uh, so, uh, as a result of these observations, uh, Amy DeVosta, um, is, uh, working to try to test the ability of, of cabergoline, another D2, uh, agonist, um, to, uh, relieve pain in, in some of, uh, the patients here at Children's Hospital. But there's a second reason for us to be interested, uh, in, in, uh, VEGF, and that is, uh, a recent discovery, um, that there is a functional role for VEGGFR1. So not VEGGFR2, which is the, the receptor involved in angiogenesis, but VEGGFR1, which acts as a decoy receptor in the context of angiogenesis, um, in the, in the context of peripheral neurons, appears to sensitize, uh, to pain. Um, and, uh, uh, as demonstrated by this, this relatively recent paper that really got us, uh, interested in this, uh, as a whole. So essentially what this group found is that, uh, VEGF induces, um, uh, no susceptive sensitization via VEGGFR1 in peripheral sensory neurons, so it induces the same changes that you saw, uh, being induced by endometriosis lesions. Um, there are, there are more VEGFR1 molecules on the sensory neurons in, in a human, uh, in, in a, in a mouse model of human cancer. Um, sensory neuronspecific loss of VEGFR1 attenuates, uh, tumor-induced nerve remodeling. And, uh, local um or systemic blockade of VEGFR1 or VEGF family ligands, uh, attenuates cancer pain. So basically, if you give Avastin or if you give an anti-VEGFR1 antibody, but not if you give an anti-VEGFR2 antibody. Um, the, this pain in a, this animal model of cancer pain, um, shows reduction, uh, in, in, in pain, much like what I showed you, um, with danazol and, and the other, uh, compounds, uh, in the endometriosis model. So, is this likely to be active in the case of endometriosis? So we looked in the literature, we find that microvessel density correlates with pain even though VEGF, uh, expression does not, which suggests that, uh, although VEGGF is important for endometriosis, there are additional factors, uh, share responsibility that, that may share responsibility. Um, TRIPV1, uh, is upregulated in the foci of women with chronic pelvic pain. Um, women, uh, with endometriosis have a higher nerve fiber density versus controls, and in a rat model, both human, uh, endometriosis tissue and, and rat uterine tissue induced sensory nerve, uh, invasion. All of these, uh, um, uh, features were observed in the cancer pain model, um, and that suggests that, that, uh, something similar is likely going on in endometriosis. So, uh, to do that, we've got 3 things that we're working towards and I can tell you we're only a little bit of a ways towards this. So, um, the first question that we ask is, is there enough V VEGFEA, VEGFEB, and PLGF? So the 3 known, uh, ligands of VEGFR1 in endometriosis lesions to activate VEGFR1. Um, and then the questions that we don't know the answers to yet, is VEGFR1 activated in human endometriosis lesions and can VEGFR1 blockade prevent endometriosis associated pain in mice? So, how much VEGFR1 ligand is present. Um, we gathered, uh, peritoneal fluid from, uh, a number of, of, um, uh, endometriosis patients. Uh, much of this peritoneal fluid, um, we got as a result of collaboration with the Boston Center for Endometriosis. Um, and, uh, what we find is we find that a substantial fraction of patients, so 31 out of 42 patients have sufficient VEGFR1 ligand in their peritoneal fluid to result in 50% receptor occupancy. We anticipate that the concentration in the lesions themselves would be higher, um, and so we expect that many, if not most of these patients actually have enough ligands in their peritoneal fluid, um, to result in, uh, in, uh, activation of VEGFR1. Um, So, um, is VEGFR1 activated in human endometriosis lesions? Um, we're currently finalizing our protocols, uh, to, to be able to stain, uh, uh, lesions from, uh, endometriosis patients. These slides are relatively precious and so we're double-checking all of our staining procedures now, but we expect, uh, by, by the end of January or, or in early February to know the answer to that question. And then finally, can VEGFR1 blockade prevent endometriosis associated pain in mice? Um, that, that's, uh, that question is the subject of a two-pronged attack. Um, first, we've recently, um, been, been, uh, given permission to use, uh, an anti-VEGFR1 antibody developed, uh, by Eli Lilly to test to see whether that, uh, that might be useful, um, in, uh, endometriosis-associated pain. And we're also currently in the process of, of doing the breeding to generate tissue-specific VEGFR1 knockout mice, uh, in order to see whether knockout, uh, of VEGFR1 specifically, uh, in sensory neurons, um, will affect, uh, the response in, in our model. So to summarize, a substantial fraction of endometriosis patients are underserved by current therapeutic option. This is an important unmet, uh, need in medicine. Uh, mouse and mouse models of endometriosis can recapitulate key features of, of the disease without exogenous estrogen. Um, these models induce, uh, behaviors consistent with pain, um, and consistent with their effect in humans, ibuprofen, letrozole, and Danazole, um, uh, reverse, uh, pain in our mouse model. Uh, we currently have additional, uh, standard of care drugs that we're testing with results on those expected shortly after Christmas. Um, our mouse model then we think has, uh, strong promise as a tool in drug repurposing and also in target validation with VEGFR1 being the first target that we're working to validate. Um, and we think that although there's a lot of work left to go, VEGFR1 is a plausible target for, for the therapy of endometriosis-associated pain. With that, I'd like to thank the people who were involved in this work. Um, Phil and Christina have gone on to more lucrative pastures, uh, as has Noah. Um, uh, Erica and Chris are undergraduates that contributed substantially to the work that we've done here. Victor is the very talented graduate student that I talked to you about. Arum has helped a lot with, uh, molecular mechanisms, um, Nick Andrews in the Wolf Lab, um, helped us to design these pain-related experiments and, and, uh, help provide important guidance, uh, along the way. Um, uh, Brooke and Cindy have helped us, uh, with pathology specimens and helping us to determine whether our mouse model actually looks like, uh, endometriosis. And then, uh, a, a large number of people, um, at the Boston Center for Endometriosis, not least Mark Laufer have been extraordinarily helpful in helping uh, in, in, in letting us understand the disease and also, uh, in, in helping us to provide samples and design, uh, studies, uh, uh, with human tissue. And with that, I'll take questions. Well, Michael, I'd first like to thank you for bringing to us today the remarkable work that you've done in the last uh several years in the lab. I, I find the whole development of the pain analysis fascinating how, how you went, went through it. I have a couple of questions and I'm sure they're The additional ones from the audience. The first is you showed early on that the presence of endometriosis didn't necessarily correlate with the presence of pain. In other words, lots of women could be found to have endometriosis but not have pain. Is there any insight into the mechanism or what, what leads to that distinction? So, it, it's currently a mystery in the field. I would argue that, that, um, VEGF does correlate with pain. The presence of VEGF does correlate with pain. So I would suggest that there are some lesions that produce, uh, sufficient VEGF to induce, uh, neurite ingrowth. Other lesions established with lower levels of VEGF and as a result, um, uh, don't induce neurite ingrowth and as a result, there's nothing there to, to sense, uh, abnormality. So I would suggest that, that, that a big difference is the presence of sensory neurons in the lesion. And the second question is, obviously, the pain is what, you know, plagues these young, young women, uh, early on, but what ends up being the, the biggest morbidity for them is, is the issue with the, the lesions and the scarring that produces infertility. Do you know anything yet as far as the agents that you've looked at that have been successful at controlling pain? Does it have any, any effect on the, the growth or So scarring from these lesions. So, so, so, so, so we haven't looked at, look, looked for adhesion formation, uh, in this model, uh, at least we haven't quantitated it. That doesn't mean it's not present. It just means that, that, that we haven't paid close attention to it. Um, uh, some of the agents that we use, so the ones that affect, uh, estrogen production do in fact result in, in lesion regression. Um, in, in the model, and so, uh, we would expect that that might also, um, reduce, uh, the extent of, of adhesion formation and other mechanical disruption, uh, uh, in the animals, but that's not something that we're focused on right now. Additional questions for Doctor Rogers. Doctor Jacki. Uh, thank you very much for uh just an excellent talk on this very vexing disease for all of us. You know, it's very difficult for us to even diagnose this disease, uh, reproducibly, and as pediatric surgeons, we often look at the gynecologists and say they're making it up, uh, looking at what they actually see. Uh, having said that, there, this is a very real disease, and I was wondering in your animal model, what is the natural history of those implanted? Uh, uh, endometriomas, if you will, uh, in the, uh, mice. As the mice get more senescent, do they disappear, uh, or do they, uh, uh, stay forever? So, that, that, that, that's a good question, and, and, and that's actually something that has been a problem in the field, um, because, as I said, most models rely on exogenous estrogen, and eventually the pellets that you implant run out of estrogen and, and lesions will regress in that case. So in our case, um, we're using natural, um, uh, uh, estrogen. Uh, we haven't taken the mice out to an age that you would call senescent. Um, but they're out, we, we've taken them out to about 9 months. Um, uh, and, uh, at 9 months, we still see, uh, the presence of, of lesions in the animals. Um, by that age, they're typically black in color, um, but, but they still have the, the typical, um, glands and stroma, um, and I shouldn't say they because we've done it twice, so. Um, as you might imagine, we want to use the, the animals as much as we can, but we had some early ones, uh, as we were doing an animal cleanup, and look to see whether they still had, had lesions, and in fact, they do. So, the lesion is present for an extended period of time in these animals. Uh, it undergoes color changes that are consistent with what we think but don't know is going on, uh, uh, in women. I guess, uh, just two questions. Uh, it's a fascinating disease and can't figure out, you know, what's biology, what's going on there, uh, is there any, so two questions. One is the simple one about, uh, why we see it at, at distant locales like in the lungs or elsewhere in the body. And the second is, what is it about. Endometrial tissue that, that, uh, that behaves this way. I mean, can you conjecture at all about other tissues don't, uh, don't implant in distant locations or invade or that sort of thing. It's, it's fascinating. Right, right. So, so, so I would say in answer the second question is actually easier for me to answer conceptually. So, um, I can't think of another epithelial tissue, uh, that, that sloughs off every month, right? So you've got living tissue, essentially the blood supply is cut off and it's shed, right? You do have shedding on the, on the surface of the skin, but that's dead cells. So you've got living cells that are shed, and that's relatively, uh, uncommon. Uh, Furthermore, that tissue has a pathway at least into the abdomen and peritoneum. Now, as to um the, the, the, the more distal locations. The, the, the, the best thing I can do, um, is guess. Um, if I had to guess, um, I would say that it may travel either through the blood or through the lymphatics to other locations. And a small cluster of cells or even um uh a, a few uh stem-like epithelial cells may be enough to, to cause the lesion to implant other, other places. So, I, I use the, the example of cancer metastasis there where it doesn't. Take very many cells to implant an existing uh, uh, uh, epithelial lesion someplace else. It's just, uh, cancer sheds its cells and, and the endometrium is shed, but I can't think of another place where, where you have, um, Uh, an, an epithelial tissue that gets shed, uh, the way that endometrium, uh, does. It's interesting, only, so, so there are two different ways, um, that, that the, uh, endometrium cycles in animals. One is an estra cycle, which is what mice do, where The endometrium is resorbed by the animal, um, and the other is a menstrual cycle, um, which is, uh, what humans do, and other animals that also have menstrual cycles. So for example, baboons also get spontaneous endometriosis. And so I think the shedding is a key reason, uh, why that happens. Doctor Moses and then Doctor Laufer. Thank you. Um, Mike, can you tell us what you know about the endothelium in this disease, vis a vis, say the tumor endothelium or the normal endothelium? I can tell you nothing about the endothelium. Why is that? That's because we don't have good access, uh, uh, to the endothelium at this point in time. Uh, also because nobody has looked. You have to look, obviously, and you could use some of the tissue that you get for your, from your mouse model, right? Yep, yep, and, and, and, and in fact, um, uh, before Tina got a better offer, she was gearing up to do exactly that. Mhm. So you're moving that to the top. We're moving in that direction, yes. Thank you for the presentation. Your accomplishments in a short period of time are amazing and pivotal to the field. Um, uh, and I thank you and the vascular biology Program for collaborating. I do have a question. What's your opinion of one of the theories that endometriosis really are multiple diseases? That there is superficial perineal disease, there's deeply infiltrative disease. And then there's disease that results in infertility and the disease of distant disease. Do you have an idea of what could be different from those standpoints? So, first of all, I will say, um, there is little to no evidence, uh, for that one way or the other. Um, uh, so anything that I'm going to be saying, uh, now is going to be rank speculation, but We're in the Folkman Auditorium and, and among friends, so I'll speculate. Um, I think that most endometriosis is approximately, though potentially with, with somatic genetic differences, approximately the same disease that lands in different locations. So I think for example, that, that the large size of endometrioma is a result of landing uh on a tissue that is basically the estrogen production factory um for the body. Uh, and so it is a well-fed, well-fed with estrogen, and, and so that accounts for its, uh, different, um, uh, presentation versus, for example, peritoneal disease. Now, whether there, whether there are differences among the, the lesions that, that result in their ability to invade or not, um, I think that's a harder question to answer and I'm sure that there will be some, some phenotypic differences between more invasive and less invasive lesions. Um, but I think that Samson's theory, um, explains enough of the disease that one, that, that, that if you take that as your baseline, you're going to be, to be able to explain much of what you see. Now, that's, uh, again, naive and without much evidence and I'm happy to be corrected. But I think you sit in a position where you can help us to find that, yeah, based on approaching all of the different types of superficial and deeply infiltrative disease. Yeah. Thank you. Time for one final quick question. Now, Michael, thank you so much for being with us today and presenting this information. Thank you. I