EFA Medical Conference 2017
"Breast, Ovary and Endometriosis"
October 28, 2017 - Lotte New York Palace Hotel
Hormone and Cytokine Sensitivity in Endometriosis
Robert Taylor, MD, PhD
All right, well thank you very much. It's really an honor to be here. It's actually a great honor to follow Doctor Rich. Unfortunately, I'm gonna be talking a little bit about medical therapy for endometriosis, a topic that he doesn't really believe in. Not all of us were as good surgeons as Doctor Rich, so several of us have to kind of go down this pathway just because of our skill set. What I would like to talk you about today a little bit is how we've been informed over the years, and this topic of ovaries, breasts, and endometriosis, as Doctor Seckin has pointed out, really a unique one. I think we've really been informed about new drug development, and that's where I'm gonna be sort of taking this presentation, based on the information that we've learned a little bit from breast cancer, a little bit from ovarian hormone action. So I'm gonna talk about hormone oxytocine sensitivity, and try to understand really how these lesions of endometriosis might be approached in a medical fashion.
So, we know that the two dominant steroid hormones that are responsible for the growth of endometrium and endometriosis are estradiol and progesterone. I've listed here some of the things that we kind of all commonly know that these do. Much of this work actually was informed by the breast cancer literature. Estradiol is a mitogenic hormone. It stimulates the proliferation of cells. It causes nerves and blood vessels to kind of grow into the endometriosis lesions. It promotes inflammation. These are things that estradiol tends to do.
Progesterone on the other hand has some of the opposite effects. It tends to have an anti-inflammatory effect. It prepares the normal endometrium for apoptosis at the end of the cycle. It causes the stromal decidualization that we're kind of familiar with in early pregnancy. As reproductive endocrinologists, the way we kind of view the system is depicted in the figure on the right, which you're all quite familiar with, the growth of endometrium under the influence of starting at kind of the top of the axis. GNHR release, pituitary hormone synthesis, effects on the ovaries, production of estradiol, and then progesterone in the second half of the cycle, and the effects that it has on the endometrial lining. Frankly, this has really sort of directed a lot of our thinking about endometriosis.
The way we kind of approach medical therapies to date, and I will mention, so non-steroidal, anti-inflammatory drugs are kind of outside of this axis. Opiates, which you've heard a little bit about from Doctor Rich, and our president recently, also are important modulators of these pathways working outside of this hypothalamic, pituitary gonadal axis. Most of the drugs that we've used for the treatment of endometriosis symptoms and pain are really targeted toward this axis. All contraceptive pills, GnRH analogs, I'll talk briefly about those. Selective progesterone modulators, these are gonna become kind of interesting drugs in the future to look at. The [inaudible 00:03:25] releasing IUD has actually been fairly effective at reduction of pain symptoms associated with endometriosis. Aromatase inhibitors, maybe we'll hear more about that from Serdar Bulun, who has really kind of pioneered that area. Then, danazol, a little bit more of a historical perspective. Progestin use, and I'm gonna talk a little bit at the end about selective estrogen receptor modulators, and they're potential development in the future.
Now, I just sort of bring this up. It's been really sort of 40 years now since Andrew Schally and Roger Guillemin won the Nobel Prize for the discovery of GnRH, and it's been kind of an interesting process figuring out that this ten amino acid peptide could drive the entire axis, and the fact that between peptides five and six was where that protein is degraded, allowed us to, over the years, develop super agonists by substituting those amino acids, so that we had drugs that would be much more effective than natural GnHR. Ultimately, antagonist and now, there's a new class of drugs, these non-peptidergic, orally active, small molecule medications, of which, Elagolix will be kind of the first one of those, probably, that comes to market, which actually have the potential of really being able to suppress the hypothalamic, pituitary gonadal axis very effectively with an oral medication, and we've seen from the recent New England journal article from Hugh Taylor and colleagues, that this could be really quite effective in the treatment of endometriosis pain in the future.
So again, just to kind of briefly review because as Doctor Seckin said in his opening remarks, part of my role here is to try to give you people a little bit of an understanding about how some of these pathways that we're thinking about targeting for the treatment of endometriosis pain might operate. Just to kind of remind you a little bit about how steroid hormone action, estrogen action now through classical estrogen receptors, estradiol represented by these little triangles, diffuse into the cells, would bind to the estrogen receptors, which come in sort of classically in two flavors: the nuclear receptor, ER alpha and ER beta, and I'll talk a little bit more about those. They activate gene transcription and turn on new genes that are important for cell growth and cell proliferation, etc. that's kind of what estrogen does.
Now, estrogen receptor beta was discovered and cloned in 1996, and a few years following that, we asked questions about whether estrogen receptor beta was differentially expressed in endometriosis and made the observation in endometriotic cells that the ratio of estrogen receptor beta to estrogen receptor alpha, these two sort of similar, but different forms of the estrogen receptor might be different that they are actually increased in these isolated cells. Serdar Bulun and his group went on to show, actually, that in tissues, it was even more dramatic. If you actually take whole tissues, that the amount of estrogen receptor beta that's expressed in those cells is really much, much higher in endometriosis. This is really quite an interesting observation and kind of gives us some insight into new drugs potentially in the future that might actually target these receptors.
This is a study from [Lozan 00:06:59], from a group in Lozan, looking at estrogen receptors in general, and what it shows is that in normal endometrium, the estrogen receptor alpha levels are certainly present, but not nearly as high as they are in ectopic lesions. This is estrogen receptor alpha. This is estrogen receptor beta. I've already shown you a little bit of data that would support that. Interestingly, and you saw those nice pictures, really both from Harry and from Tamur, of all the interesting stuff going on in the peritenoneum. The uninvolved, apparently uninvolved peritenoneum of women with endometriosis has much higher levels of estrogen receptors actually ending up in the peritenoneum as well. So you can start to understand how these vascular changes might be driven by the presence of estrogen receptors.
Now, sort of on the other side, what about the effects of progesterone in endometriosis? So progesterone, again, they're kind of a classic ovarian steroid hormone. The hormone diffuses into cells. It binds to nuclear receptors, quite similar to estrogen receptors that come in sort of two different flavors. In this case, PR progesterone receptor A and progesterone receptor B. The B form is a little bit of a longer, larger form of the progesterone receptor, and it has a predominantly anti-inflammatory effect on cells. So when it binds to cells, or when the progesterone binds to the receptor, and the progesterone receptor binds to genes involved in gene transcription, you'll see those anti-inflammatory actions.
Now, we have taken advantage of that to some extent by using progesterone as a way of treating endometriosis associated pain. This is not a randomized study, but a prospective study done by Paolo Vercellini and his group in Milano quite a few years ago now, but showing that in women who are undergoing a second laparoscopy or who have had a prior laparoscopy and have had recurrent pain, they were either treated with surgical therapy, removing endometriosis lesions, very much like what you've seen and saw god pain relief over sort of a 12 month period versus those that were treated chronically with progestins and had a fairly similar response. Now, again, this wasn't a randomized study so patients self-selected for which, whether they'd get surgical treatment or medical treatment afterwards, but again, suggesting that these progestins can be effective for pain relief after surgery.
Really, this kind of paradigm was changed a little bit when Serdar and his group made a really important observation, now some years ago, that progesterone receptors are not expressed in a normal fashion in endometriosis, and particularly, this anti-inflammatory progesterone receptor B, even though it's expressed at nice, healthy levels, this is the messenger RNA for PRB in normal endometrium, in endometriosis tissue, this is actually very, very ow in expression. It really kind of means that when progesterone is around, and other synthetic progestins are around, in many cases they can't activate an anti-inflammatory pathway to help relieve pain, and this might be one of the reasons that so many women break through medical therapies as I think Harry has suggested.
So what about the inflammatory immune cell system? We felt that this, and you've heard a lot about that really already this morning, the importance of inflammation in immune cells. We believe that these are really kind of critical cells to a lot of aspects of endometriosis growth. There are lots of different cell types, mass cells, dendritic cells, neutrophils that all probably have an important role to play, but just for time's sake, I'm gonna focus on macrophages, a very important, innate, sort of immune cell, and something that we've really discovered is quite important in endometriosis, and might be an appropriate target for new drug therapies that could be used for the treatment of endometriosis pain.
So this is a little bit of a view that we've had about endometriosis evolving, maybe, over the past 20 years or so from work in my group and many others. Again, we're starting here with retrograde menstruation, not necessarily the only way the endometriosis comes about, I would agree, but at some stage, we know that we get these endometriosis implants growing, and we believe that a really critical function of these implants is to really recruit an inflammatory response these macrophages, when they actually infiltrate into the lesion, can stimulate through the synthesis of a variety of different cytokines, and we'll talk about these in a little bit. The growth of nerves and blood vessels into those lesions, the fibrosis that you see, that you've seen in lots of these examples already this morning, that's really critical.
Recruitment of other immune cells, and ultimately effects on the endometrium that might really be important in terms of infertility. We're not talking very much about infertility today, and I think it's really quite clear that medical therapies don't have much of a role to play in infertility, current medical therapies, but I think important to kind of keep this in mind as part of the overall syndrome. We think that these kind of angry, activated macrophages that are purple here, make these kind of purple cytokines, and I'm gonna use interleukin one beta as a prototypical one, as a way of kind of influencing these inflammatory responses.
So how does IL-1 work? The ideas here and now is that we've got macrophages that are infiltrated into the endometriotic lesion, the growing endometriotic lesion that can synthesize cytokines such as interleukin one beta. One of the things that interleukin one beta does, is that it stimulates the production of protein subunits that are called nuclear factor kappa be subunits, that's not too important, they're represented here as these blue and kind of purple balls. Those are normally sequestered in the cytoplasm of cells, and I'll show you some examples of this in a second. Due to the attachment of this inhibitory trimer. When cells are treated, endometriosis cells are treated with interleukin one, the 65, sort of the, I guess that's the blue one here, and the 50 KD, I don't have my glasses on, so I can't really quite see what these are, but both subunits of these are increased each time we treat these various, and these are from four different endometriosis patients with interleukin one beta. So interleukin one beta stimulates the production of these, what are transcription factors, that then, actually, the dimer, and I'll show you kind of how that happens in just a second, goes into the nucleus, binds to DNA in the nucleus of the cells and activates pro-inflammatory responses within the cells.
This is kind of the process by which this comes about. Again, this is this trimeric thing. You've got this dimer of a P65 and a P50 dimers of NF kappa B associated with an inhibitory subunit called I kappa B. That I kappa B is actually phosphorylated and degraded when interleukin one binds to it's receptor. You can see this little red protein actually gets degraded and goes away, which allows this green dimer to diffuse into the nucleus and activate the genes. This is kind of showing that I kappa B is present. This inhibitory subunit is present. Within five minutes, it's almost 80 percent degraded. After 15 minutes, it's really completely degraded, and it doesn't come back again for an hour. So interleukin one causes the degradation of this little red subunit, allowing this green subunits to move from the cytoplasm of the cell, which you can see here before the cells are treated with interleukin one to 20 minutes after interleukin one beta, this diffuse stain actually goes into the nucleus as these nuclear transcription factors actually become translocated and activate genes that are associated with inflammation.
In endometriosis cells, relative to normal endometrial cells, there's much more of a sensitivity or a responsiveness to interleukin one beta. Under the same conditions in cells derived from normal endometrium versus endometriosis, if we treat those cells with interleukin one beta in the black bars, vascular endothelial growth factor, potent angiogenic factor goes up quite dramatically in the endometriosis cells, not so in the normal cells. Interleukin six, another pro-inflammatory cytokine goes up a little bit in normal cells, but quite dramatically in endometriosis cells. Part of this kind of hyper responsiveness is actually due to the presence of higher levels of messenger RNA. For interleukin one beta receptor, we measured, back in these older studies the message being up about two and a half fold. In a later study by Aliya Kumin and his colleagues in Quebec, this has also been shown at the protein level. Interleukin one receptor protein has actually increased about two and a half fold in endometriosis cases. So these cells are actually primed to be more responsive to cytokines.
A little bit of a model as to how we're thinking about this, endometriotic lesions develop either kind of in cyto or can develop, potentially, from a retrograde menstruation kind of perspective. I'm a little bit agnostic about either of those, but lesions are growing sort of within the peritoneal cavity. They can recruit an inflammatory response by the secretion of proteins. Those immune cells, macrophages for example, can actually secrete inflammatory proteins such as interleukin one beta that actually stimulate these endometriotic cells to do the same thing over and over again. So you've got a feed forward kind of process occurring.
We also know that receptors for interleukin one beta are actually higher on endometriotic cells, making them sort of more sensitive. They ultimately activate enough kappa B proteins go to the nucleus to sort of turn on a variety of genes that are associated with growth prostaglandin production, pain, and the growth of blood vessels and nerves into these lesions.
Now, why might this be important. This slide is a little bit harder to read, but one of the things that's really quite interesting is there are a lot of drugs that have already been developed actually, that have quite potent NF kappa B inhibitory activity. These really might be repurposed in settings of endometriosis and could be kind of interesting for future drug development.
I've listed or underlined a number of these. Bortezomib, is a protozoan inhibitor. It's being used, actually in clinical trials of multiple myeloma currently. Digitoxin is a cardiac glycoside, obviously used for the treatment of heart failure. This is a drug that actually has some NF kappa B inhibitory activity. Sunitinib is a tyrosine kinase inhibitor used in the treatment of renal cell carcinomas. Even this kind of third generation antifungal agent, Tioconazole, which is for a yeast vaginitis, has some activity in terms of inhibiting NF kappa B, which is what this slide kind of shows the pharmacology of that. The idea is that there might be quite a few drugs out there that could be models for new drug development for anti-inflammatory kind of pain treatment.
There are a lot of natural compounds as well. Lycopene you might be familiar with. Resveratrol, you've heard quite a bit about that. Our group has been quite interested in curcumin, the sort of the spice from turmeric, which is used a lot in south Asian cooking. He's an example of kind of using some of these curcumin as a way of kind of treating, in vitro at least, inhibiting inflammatory responses in endometriosis cells.
These experiments, which are currently being completed in our group show that interleukin one beta as well as other cytokines and growth factors, such as vascular endothelial growth factor, shows that in normal endometrial cells, there are relatively lower levels of these cytokines under control conditions in the open bars, relative to cells derived from endometriosis samples, but that when you treat these with curcumin in vitro, there's a dose responsive inhibition down to the level of where we can kind of decrease to the level of normal cell expression, normal endometrial cell expression, how much of these inflammatory cytokines are made. This is at least sort of suggestive that there might be natural types of therapies that could be beneficial as well in the future.
Now, the pharmaceutical industry has been quite interested in trying to antagonize interleukin one effects. So these might be useful for new drug development in endometriosis. Again, I sort of describe this to you a little bit, interleukin one beta will bind to it's receptor, send a signal through the cell membrane to sort of activate these downstream pathways, which would include NF kappa B that I've talked to you about. Pharmaceutical approaches to sort of inhibit this process include monoclonal antibodies, a sort of Herceptin like concept of using monoclonal antibodies to bind to the interleukin one to prevent it's ability to sort of activate the cell. Oxciobal receptors such as Humira can be used to bind interleukin one beta. There's actually endogenous, naturally occurring antagonists of IL one that can be made recombinant protein can be made to sort of block that effect as well. So there may be, in the future, some development of drugs that could be effective in those areas.
In the last couple of minutes, I jut wanted to talk about some experiments that we've done, really based on a lot of very good work in breast cancer to look at the potential use of estrogen receptor modulators, drugs that actually bind estrogen receptors in a mouse model of endometriosis. From Doctor Nudeson this afternoon, you're going to be hearing more about mouse models of endometriosis, and maybe also form Serdar as well. These syngeneic mice, so these are inbred mice, they can actually tolerate transplants from other members of their family. We can transplant uterine fragments into the peritoneal cavity. We've used ovariectomized mice in this particular case, and then those mice can be treated with subcutaneous release of different hormones. Estradiol, which is our positive control which stimulates the endometriotic tissue to grow in these mouse endometriosis lesions, as well as two selective estrogen receptor modulators. OBHS, here, which is a drug, both of these were actually developed by John Katzenellenbogen. I'll sort of talk a little bit more about that, but this binds to estrogen receptor alpha preferentially and this chloroindazole to estrogen receptor beta.
These are the kinds of drugs that we're studying in this mouse model. Here's a picture of Benita Katzenellenbogen, really a world class breast cancer investigator and a collaborator of ours in this endometriosis project. You can see these, again, I said these are ovariectomized mice that have these lesions that have been surgically placed now in the peritoneal cavity. When they're treated with vehicle because there's really no estrogen around to stimulate them, there's no growth of the lesion. When the animals are treated with estradiol, over a two week period of time you can see nice linear growth of these four implants that we've actually sewn into the peritoneal cavity. If we do estradiol in the presence of either of these two selective receptor modulators, we inhibit that process.
Vascular endothelial growth factor messenger RNA in these lesions is stimulated by estradiol, but it blocked when estradiol is administered in the presence of these two estrogen receptor blockers. So this shows, at least, that there are some biological activity in the system. If we, these are some experiments now done in intact mice. These mice have their own ovaries. They're making their own estrogen. Over a two week period of time, you can see that there's nerves. This is the PGP 9.5 is an indicator of nerves growing into these lesions. Substance P is a no susceptive signal, a pain signal in these nerves. So nerves grow into the endometriotic lesions, but you can prevent or block nerve growth by giving either of these two estrogen receptor modulators.
Inflammatory responses in these lesions, this is looking at chemokine, these are increased in the presence of estradiol. These are intact animals, I'm sorry. In animals that have their normal ovarian function, there is activation of these inflammatory chemokines, which can be inhibited by the two estrogen receptor modulators. Even the production of tumor necrosis factor alpha, another pro-inflammatory cytokine, which is transient clinically increased in vehicle treated animals, driven by their own ovarian estrogen can be inhibited. If we stain from macrophages or immune cells in these endometriotic lesions you can the yellow, green color here indicates macrophages infiltrating the lesions in the vehicle treated animals, but it's prevented that macrophages influx is prevented in the presence of these selective estrogen receptor modulators, and that's indicated here in this slide.
Now, we can take endometriosis cells in vitro, human endometriosis cells in vitro, and ask some of the same kinds of questions. This just shows that apoptosis, that programs cell death, can be actually induced by these selective estrogen receptor modulators. Estradiol promotes cell growth. It doesn't induce apoptosis, but in the presence of these receptor modulators, there's a apoptotic effect. Cytokines like interleukin six, which can be stimulated with estradiol and tumor necrosis factor alpha for example, can be inhibited pretty dramatically with either of these two selective estrogen receptor modulators.
Now, I do have to say it's a little sobering that what about sort of clinical experience in this area. This is a study from Pam Stratton and the group at the NIH from a decade ago, where they used Raloxifene. You might be familiar with Raloxifene. This is a selective estrogen receptor modulator that's used clinically for the prevention of breast cancer. In this particular setting, they took women who'd had endometriosis, who had returned with recurrent endometriosis symptoms of pain. They underwent laparoscopy, and then were randomized either to Raloxifene treatment or to treatment with placebo. There are about 50 patients in each of these two groups, and they're followed. They were treated for six months with the medication, and then followed for a couple of years. At one year, there was really no difference between the two groups, but past a year, as they got out to the two year period and beyond, actually the women that were treated with Raloxifene, back at this poin in time, had a more rapid recurrence of their symptoms than did the placebo group, and the data safety monitoring board made them stop the study.
This was a little bit of a sobering reminder that even though we had good principles based on mouse studies, sometimes kind of a parallel ... Now, this is a different compound, obviously, than the ones that we've been studying in the mice recently, but you can see that it's really important to do these clinical trials to be sure that we can confirm in clinical studies what we think is going on, either in the laboratory or in a laboratory animal.
So to summarize, endometriosis cells are estrogen sensitive and progesterone resistant. I've described that a little bit to you in terms of the differential expression of these receptor proteins that are present. Selective estrogen receptor modulators in a mouse model look very, very promising in terms of lesion growth and some of the inflammatory changes and nerve and blood vessel effects that we see, but a clinical Raloxifene trial, a single clinical trial using a different drug did fail to show efficacy, and actually might have shown some harm. So again, a sort of a reminder that we've gotta sort of pay attention to these things.
Immune cells and the cytokines they produce are really important, we believe, relative to the endometriosis symptoms, pain and likely infertility as well. Many of their effects are mediated through this NF kappa B signaling pathway. We know that a number of existing medications, including some natural herbal compounds can inhibit NF kappa B and might be repurposed or developed or extended in endometriosis therapies. Then, other inhibitors of interleukin one beta, which is sort of an example of one of the other cytokines that might be important, could be promising in the future.
I think what's really kind of an interesting concept is that some drugs like some of these experimental serms that we've been studying in the mouse model, they have dual effects. They might both effect the estrogen receptor pathway, as well as the inflammatory pathway. If we could kind of identify ones that were particular efficacious in that setting, it would really kind of make a lot of sense that these could be developed in time as new sort of drug therapies to supplement surgical treatment, or in cases where surgical treatment fails.
So I'll end at this point. I want to thank the support from the NICHD that I've had over a number of years for doing these studies and lots of collaborators from all over the world, but for these particular studies that I've described to you, particularly Benita and John Katzenellenbogen. John was the one who was responsible for synthesizing those new serms, and [Uachaza 00:30:37], who was the post-doc in their laboratory that did a lot of the mouse studies. So thank you very much for your attention.