Hello, my name is Jubilee Brown. I'm really excited to be here with you today. I am a professor of gynecologic oncology and division director here at the Levine Cancer Institute at Atrium Health in Charlotte, North Carolina. Today, I'm so excited to be talking with you about the genesis of modern uterine cancer treatment. I'm excited about this topic because it is such, it's so relevant to everything that we're doing now, and so much has changed just recently. So some of this will be older data, some of this will be newer data, and I think it's all relevant to what we do in practice. So these are my disclosures, none of which are relevant today. And I guess one of the things, I'm so happy to really work worldwide with such an amazing organization that we have here, so many opportunities for fellowship and sharing of ideas. So today, we will talk about the transformation of surgery from laparotomy to minimally invasive surgery in uterine cancer. We'll detail the metamorphosis of medical therapy for uterine cancer, and then we'll kind of marry those concepts and see how the field is just so different now than it used to be. And I think that, you know, the real key here is seeing how medical and surgical oncology really mesh very importantly together here. So back in the day when, you know, when I was a fellow, everything that we did for uterine cancer looked like this. It was all open surgery. And so we were very excited when we saw dissections, when we were able to obtain great dissections like this in an open fashion, where we saw the aorta overlying the inferior vena cava, where we were able to dissect out the ureters so well and take all the nodes off, right? Everything that we did was open. It was all about exposure and radical dissections. And that was way back in the day when I was very young. Now, and at that time, staging was very simple. In 1988, the staging that we had really involved just anatomic locations. Where was the tumor involved? Was it just the endometrium? Did it invade the myometrium and to what degree? Did it involve the cervix? Did it involve the adnexal structures? Were the nodes positive or were there distant metastases? That's the only thing that staging incorporated. And after surgery, we really didn't have many options. We had cisplatin, doxorubicin, then came along carboplatin and paclitaxel, and that was exciting. And we had radiation therapy. And really, that's it. There was nothing else. Outcomes were very divided. I would say it was stormy seas. Either patients did very well, so they were early stage and had so-called type one tumors. You can see here that those patients basically did extremely well, almost no matter what. They were caught early. And these were the patients who had estrogen-dependent, more obesity-related types of endometrial cancer. Patients who had late stage or type two disease did not do well. And you can see the five-year survival rate really depended on stage. Patients who were localized had about a 96% five-year relative survival rate. But by the time you got to distant disease, that five-year survival rate was down to 20%. That's really all we knew. And we, of course, always want to strive to improve outcomes. So how did we do that? Well, in 2009, Joan Walker published LAP2. And this is actually how I started doing laparoscopic surgery. I actually went to a pig lab and just went pig to pig to pig and learned how to do nodal dissections. And that's what allowed me to start doing more intensive laparoscopy. In her study here in 2009, she compared laparoscopy with laparotomy for comprehensive surgical staging of uterine cancer. And this was a gynecologic oncology group study. What she initially published was that it was feasible. It was safe. There were fewer complications than in open surgery. And of course, there was a shorter hospital stay. Interestingly, they provided this graph, which shows that as your body mass index goes up, the risk of conversion to laparotomy increases. And what you can see is that sort of the inflection is right around a BMI of 35, which now today we really don't consider that obese. That's almost normal in our community. And so times certainly have changed. Now, of course, it's pretty routine for us to do large BMIs without conversion well over that 35 BMI. Subsequently, she published the outcomes, the cancer outcomes in 2012. And what we see is that it's hard to see. There are actually two lines, laparotomy and laparoscopy. As you can see, those lines really overlap. There was no visible difference there between laparoscopy and laparotomy in approach to cervical cancer in terms of overall survival. Now, interestingly, when you actually dive into it, the hazard ratio was 1.14. And this actually fell short of non-inferiority bounds. That's actually sort of similar to what we saw with the LAC trial. Here, however, we accepted that these lines are essentially overlapping. And even though the number of events made this fall short of non-inferiority bounds, we can see that these techniques are essentially equivalent enough to say that this is the standard of care. There is a clear benefit to minimally invasive surgery, and this has really become the standard of care. So this is how our patients routinely look now. I personally do three-port hysterectomies with nodes. These are all five-millimeter ports. So you don't even have to close the fascia. There's very little discomfort. And really, what a huge change this is from a vertical midline incision and a five-day hospital stay. So the next step in improvement had to do with lymphatic mapping. And so lymphatic mapping is the injection of an identifying substance that travels to the so-called sentinel lymph node and allows removal of that specific node or nodes, which is predictive of metastatic disease. The idea behind lymphatic mapping is that we can achieve improved outcomes due to improved detection of metastatic disease and due to reduced surgical morbidity. In this technique, rather than taking out all of the lymph nodes, we instead inject dye around the tumor. The dye goes through the lymphatic channels to the regional nodes. And therefore, we can see the so-called sentinel node. That one node is going to tell us the status of all the nodes. If it's negative, all the nodes are negative. If it's positive, that indicates some positive nodal disease. In this technique, only the tumor and the sentinel node are removed. And the idea is that this reduces surgical morbidity. Most patients with early disease actually have negative nodes. And we know that there's no survival benefit from a complete lymphadenectomy. And so this allows us to really select the node-positive patients for regional or systemic therapy without having to perform a full lymphadenectomy. This then prevents over-treatment of unstaged patients with risk factors. It also allows us to have a more accurate diagnosis of metastases because we can do ultra-staging of the sentinel node and actually improve the predictive ability. And the whole idea behind this is that it's better for the patient. So even though it wasn't much talked about, it was subsequently studied, we can really decrease the risk of lymphedema, lymphocysts, and vascular and neurologic injury by just doing sentinel node biopsy rather than a complete lymphadenectomy. In fact, the odds ratio is 0.5 for any or severe complications compared with full lymphadenectomy. And the hazard ratio is 2.75 for lymphedema with a full lymphadenectomy compared with sentinel node biopsy. One study showed 26 versus 49%. Another took the risk of lymphedema from about 40% down to less than 4%. And you can see in the picture, lymphedema is a significant issue for patient satisfaction and quality of life. In addition, and perhaps one of the things that I think is most important, is that we're not going to overlook a positive node. Mapping allows us to find that positive node. There's some options for lymphatic mapping. This has been a journey. We looked at technetium-99, given either preoperatively and or intraoperatively. We looked at blue dye, whether it was lymphozerin or isosulfan blue. We looked at indecine in green or ICG10. And these different techniques can be used either independently or in combination with each other. This all started with melanoma. We use it, of course, in breast cancer and in cervical cancer. And injection of technetium-99 then allows us to perform lymphocentigraphy so that we can see, in fact, where the nodes that drain the area of interest are located. Again, it doesn't mean that those nodes are positive. It just means that those are the nodes that we want to look at. This is an early, gosh, this is very early, example of a Geiger counter, basically, a neoprobe that allows us to look. You can see the blue dye here and here through the peritoneum. And then we have this laparoscopic Geiger counter, basically, that allows us to look for the nodes that contain the technetium-99. Here we can see lymphozerin go up. You can really see these nice blue channels. That blue dye, though, it's very limited. It goes away very quickly. Here we see lymphatic mapping of, on the right, there's a nice blue right external iliac lymph node that lights up there. You can see that track, that afferent lymphatic that goes and then fills that node, which you might miss otherwise. Basically, technology enables everything. When we had better technology, mapping was better. Now, of course, we use near-infrared imaging with ICG-10. Basically, we inject into cyan and green or ICG-10 in the cervix and lower uterine segment. We look with a camera that has a near-infrared range, which is 700 to 900 nanometers. Rossi, in 2012, actually did the feasibility study with a robot in patients with cervix cancer. And I think that, you know, really the take-home for me is that this avoids some of the difficulties present with conventional mapping. In contrast to blue dye, the ICG-10 stays present much longer, so there's no need to rush. There was always this, you know, pressure to rush and get there before the blue dye was excreted. In addition, many patients had an allergic reaction to the blue dye. And that's incredibly rare. I've never seen anybody have a reaction in a cyan and green. You can also inject it at the time of the procedure, which is very different from technetium that had to be injected beforehand. And, of course, technetium-99 is radioactive. There's no need for radiation safety precautions because ICG-10 is not radioactive. I think this is just a great picture. And so here you see this is what was published by Dr. Rossi in 2012 just under white light. And after ICG-10, boom, there's no way that you're going to miss those sentinel notes. A number of authors have since published on either the combination of ICG with blue dye or technetium or just ICG alone. And the bottom line is it really works in both low- and high-grade tumors. There have, in fact, been three prospective trials, the FIRES trial, the CENTOR trial, and the SHREK trial, looking at both high- and low-grade cancers. Most of these patients had periodic lymph node dissections. In total, this encompassed 798 patients. And across the board, the sensitivity was excellent, 96 to 98 percent, a very low false negative rate of 2 to 4 percent. You know, most would agree that anything 4 percent or less is an acceptable false negative rate. And the negative predictive value taken together is 99 percent or higher. So I think we can very definitely say that, yes, these techniques are accurate for both low- and high-grade cancers. The technique. So once we inject, then we open the retroperitoneum just under regular white light. We switch to near-infrared light and obtain the sentinel nodes. Usually this is about 10 to 30 minutes after the injection, and that's pretty normal operative time. If there's only one side that takes up the dye or if we don't see either side take up, give it a little more time, like five minutes, before you disrupt all the lymphatic channels. And now, you know, the technology's improved even since this time. There's the 1588 or 1688 scopes, and so you can use a single camera with excellent resolution. Even I use one that's a five-millimeter scope, and the resolution is excellent. Very importantly, if one side does not map, you basically use the traditional criteria to determine if a full lymphadenectomy is required on that side. So in other words, basically, if you're in there, you see that your right side maps, you can take your sentinel nodes, but your left side doesn't map, well, should you just do a complete lymphadenectomy? No. If possible, you should send your uterus off for a frozen section. Look to see the traditional criteria. So look to see the traditional criteria. So is the tumor less than two centimeters? Is it grade one? And is it less than 50 percent myometrial invasion? If it meets all those criteria, you do not need to do lymphadenectomy. If it doesn't, then you do a complete lymphadenectomy on that side that didn't map only. So here's our laparoscopic platform. And so this is a few years ago, but you'll get the idea. It's not subtle to see these sentinel nodes. And so here we are dissecting on that left side. We see our sentinel node that we can just elevate off of the external iliac artery. We can just spread the pedicles, making sure that we see all the relevant anatomy. And honestly, you can see that this is not that challenging to do. We just lift the node up. You detach it. And it's nice to seal these lymphatic channels. You can see all those nice lymphatic channels. You don't want to get a lymphocyst. And by sealing those lymphatic channels, it decreases lymphocyst formation. Here's another example of how at first glance you might not see a sentinel node. So it's important to truly do your full dissection. So open up your planes. And if at first you say, oh, I see that little lymphatic channel down there that's green. But you don't want to just stop there. You really do need to look further and make sure that you dissect out all the different areas, again, in a very safe manner, spreading, letting your laparoscopy, your CO2 help you, and lo and behold with further dissection now we see that sentinel node in that sort of obturator space there. So now that we get that, and I'm doing this with three ports, you can do more ports, but you can see that there is other nodal tissue there, but that sentinel node turns green, so we can take out just that node, and that's all you have to do. This can also be done with a robotic platform. If you have a robotic platform, you can see here, again not subtle, where these nodes are located. We can open our retroperitoneum, and here of course we are on our right side. We're just opening that retroperitoneum the same way that we typically would. We open up these spaces, and then we switch, and we can see where those nodes are. We take out the sentinel nodes again by just lifting, spreading, and you can do this either under near infrared or white light. And sometimes it's helpful to switch back and forth just so you make sure that you see all the relevant anatomy, and that you don't miss any of your nodes. Once it's detached, we place it in the back and bring it out through either the port or through the vagina. Ultrasectioning actually does find more cases, so this increases detection of positive nodes by about four percent, and you can see why. Before we did mapping, we just took one longitudinal section of each node, and you can see if you had just a cluster of cells, you would miss it. Now we do ultrasectioning, as you can see. When we find these little tiny foci of positive cells, this is really mostly in low-risk patients. There so far has been shown to be no benefit for treatment and studies to date. There is a low risk of positive nodes and non-sentimental nodes, and at present, we don't reoperate when isolated tumor cells are identified. Okay, so surgical development. Wow, it changed everything, but mortality remains the same, and in fact, if we look at what happened with uterine cancers over time, it actually increased. Look at that. That's going the wrong way, and in fact, it's worse in women of color. We actually went from light microscopy to the cancer genome atlas. What does this mean? Well, our understanding changed. Previously, I mentioned type 1, type 2 tumors. All we had was light microscopy. Well, now with the cancer genome atlas, what happened was that we were able to map and fully describe the genomic signature of multiple patients with uterine cancers, and what we found was that tumors segregated into four separate cancer phenotypes, polyultramutated, MSI, hypermutated, copy number low, which is typically like the grade 1 types that we see, and copy number high, which is serous type, and what we saw. Now, we have multiple trials that look at this, but what we see is that those four phenotypes, they actually predict the clinical behavior, so progression-free survival is incredible for patients who are poly. No matter anything else, these patients are going to do great, literally no matter what you do. On the other hand, patients with serous tumors or copy number high tumors do much, much worse. The other two groups kind of fall in the center, but they are the characteristics predict what we may be able to treat these patients with, and we'll talk more about that. Uterine cancers, we know, have a very significant tumor mutational burden, and in fact, these patients who are DMMR, they have deficient mismatch repair. We also call them MSI high, okay? When we look at that, patients with endometrial cancer actually have a very significant tumor burden, and so this really explains how cancers form, so our patients who are deficient in mismatch repair have a defect in the cancer, in the machinery that prevents cancer formation, so basically, if we look at normal cells, mismatch repair machinery recognizes the MSH2, MSH6 complex, which we see right here, and repairs mismatch basically by recruiting this MLH1 and PMS2. In MSI high cells, basically deficient mismatch repair, it disallows the DNA mismatch repair within the microsatellite, and so mutations accumulate, and there are defects in DNA that happen, and that is how cancers in these patients happen, so we talk about either proficient mismatch repair or deficient mismatch repair. Patients who have deficient mismatch repair are MSI high, and that is targetable. It's targetable with immunotherapy. It turns out that when we talk about MSI or microsatellite instability high patients, patients who are deficient in mismatch repair, we're going to call those MSI high or DMMR. That represents about 4% of cancers in adults. MSI is a result of inactivation of the mismatch repair system, and so that, again, is the phenotype. What we see is that when this repair situation doesn't work, we have a very high frequency of frame shift alterations in the microsatellite DNA, and so we see this sometimes in Lynch syndrome, and these are the genes that we test for, or I should say proteins that we test for now, because germline alterations in these genes, MLH1, MSH2, MSH6, and PMS2, we see those in Lynch syndrome, and when we see these alterations, we know that Lynch syndrome can be hereditary, and we're going to test the patients to see if they have that. We also see hypermethylation of the MLH1 gene. This is a sporadic mutation, and this basically, yes, confers DMMR, but it's not through a hereditary mechanism. Either way, this is prognostic and predictive, and it correlates with PD-1 and PD-L1, and we see a graphic representation of that here. Here's our tumor cell. It expresses an antigen, which the T cell receptor on the T cell recognizes. The thing is that the breaks have to do with the PD-1, PD-L1 complex, and a PD-1 inhibitor or a PD-L1 inhibitor can recognize these antigens. What does that mean? Well, this means that we can exploit this technology and test all of our patients who have endometrial cancers at the time of diagnosis to see where they fall, and this is one example of a mechanism or an algorithm where we test patients, and we're working our own out in our institution, as you may be as well. We test patients for poly first. If they're poly positive, we know they're going to do well. We don't need to know anything else. If they're poly negative, which most patients are, then we test them for mismatched repair, and that's just immunohistochemistry. If they're MSI high, so again, remember if any of these genes are abnormal, MLH1, MSH2, MSH6, PMS2, then they're considered MSI high, and then they need to be referred for Lynch testing. If not, if the expression is retained, then we check P53. If the P53 is abnormal, then they're copy number high or serous-like, and we know to treat them like a serous carcinoma. If they're copy number low, they kind of fall into that type one or wild type pattern. All of this is sort of converging. This is all sort of hot off the press, you know, and so we're really anticipating a change in recommendations, in how we test, in how we stage, and so all of this will likely factor into who we treat. We can see that the FIGO is likely going to update their staging for uterine cancer. This really matters. So, for example, what we will see, for example, is that patients are going to have a lot of changes in terms of their stage based on their, not just anatomic location, but also the way that they stain. So basically, life just got a lot more complicated. Life does that, right? You saw me with a baby back there in the very beginning when I was young, and now here's my whole family. Life is complicated. The advent of immunotherapy has really just changed the world. It is truly a sea change, and this all started really just a few years ago. When we look at all the trials that have been done for single agent immunotherapy, what we see starts with pembrolizumab with a response rate of 71% in the initial report from Lee. This holds true when we look at Amanda Fader's study. She looked at pembrolizumab in patients who had DMMR tumors and recurrent endometrial cancer, and here she found a response rate of 56% and a disease control rate of almost 90%. The Garnett trial, which we were fortunate enough to be on, basically looked at dastarlimab, previously called TSR-042, and in MSI high recurrence and advanced endometrial cancers, found a response rate of 52%. This is literally, this is unheard of. We never had findings like this. In addition, this immunotherapy produced deep and long-lasting responses, even in a recurrent setting. Look here in keynote 158. This is pembrolizumab. These are advanced patients, and what we see in the updated data from 2022 is these very deep responses. You know how to read these. Basically, this is where your tumor started and the percent change. If the tumor decreased, if they responded, this is what we see, and you can see these are deep responses. This swimmer's plot basically looks at how long the control lasted, and you can see the further out the line goes from here, the better. These are long, long responses. The Garnett trial showed the same thing. Basically, for DMMR and PMMR patients, there was a confirmed response in 30 patients. Again, you see in graphic formation these deep responses. The overall response rate was 42.3%. 12.7% of these patients had complete responses, and 30% had partial responses. Also, again, here we go. Swimmer's plot. These patients, if they respond, they're going to respond for a very long time. Look at the numbers down here. Can you see it? 72, 76, 84. This is out to 102 months of treatment. That's miraculous. Disease, median duration of response was actually not reached at a median follow-up, 11.2 months. These patients did so well. Said another way, the likelihood of maintaining a response was 96% at six months and 77% at a year. That's remarkable. We also saw an overall survival benefit even in patients who had proficient MMR. Again, unheard of. Study 309 looked at Linvatinib and Pembrolizumab versus chemotherapy. Remember, chemotherapy used to be the only thing we had. Now, we have this long-term data with follow-up extended by over 16 months. Look, the curves continue to separate. The overall survival favored Linvatinib plus Pembrolizumab, even though there was some crossover in the chemotherapy arm. That's pretty remarkable. After excluding patients who crossed over, the PMMR overall survival hazard ratio was 0.64. That's just crazy. It's really good news for all of our patients. Incredible at the SGO this past March, we heard about two studies simultaneously released and published in the New England Journal of Medicine. We had the NRG trial GYO18. This tumor looked at Pembrolizumab plus paclitaxelone carboplatin followed by Pembrolizumab versus carboplatin paclitaxelone placebo. Wow, look at this. The hazard ratio in these patients with advanced endometrial cancer was 0.3. That's incredibly significant. If patients respond, again, they're responding for a really long time and only rarely do they subsequently recur. It's hard to use the word cure in these patients, but it's tempting to do that based on when you see curves like this. We saw exactly the same thing in the RUBY trial, which looked at distarlomab. It's sort of like the GARNET trial, but up front here was RUBY with distarlomab and paclitaxelone carboplatin followed by distarlomab maintenance compared with just the chemotherapy portion. Again, what we saw here, this is crazy. Even in the MS-stable population, the proficient MMR hazard ratio of 0.28 and look at that tail. It doesn't ever drop down. Patients continue to do well many, many years out. This is 34 months here at this point. We'll continue to see the data as the data mature. Interestingly, in the RUBY trial, they included carcinocercomas or MMT patients. Why does this matter? Well, this is a patient of mine. A 69-year-old woman, stage 1a, grade 1, endometriated adenocarcinoma. She was DMMR. She had a robot-assisted laparoscopic hysterectomy, BSO, and staging in 2015. She should never have recurred. She had 2 of 16 millimeters myometrial invasion, so not much at all. She had negative LVSI. She had 20 negative nodes. It was right before we started mapping. Less than a year later, she recurred at the vaginal apex. She got radiation therapy and brachytherapy. She had a complete response initially. Then she again recurred about another year later. You can see this FDG AVID area here. She got six cycles of paclitaxel carboplatin, had a persistent mass. I tried to resect it, but it was very infiltrative. It was an incomplete resection. She had significant vaginal bleeding and pain. She literally couldn't sit down. Here's this mass. She got pembrolizumab. She's had 52 cycles. She had a complete response, and she's been without evidence of disease since that time. I mean, this is remarkable. When patients do well, they do well for a very, very long time. When you put all this together, this is what it's been. We can truly stratify and individualize treatment. When we are considering the upfront setting, we now do minimally invasive surgery with lymphatic mapping and sentinel nodes. We have adjuvant therapy that really works. I mean, I think we can say that it actually cures a lot of patients. In advanced stage and recurrent disease, we now have chemoimmunotherapy to shrink the disease down. We saw those incredible curves. When we put all of this together, it really allows us to have less aggressive surgery. There are fewer what we would call big whacks, large laparotomies, less morbidity, better outcomes. I think it's essential that as we take care of our patients with uterine cancer, we have to know the entire disease process to make appropriate clinical decisions about surgery and medical oncology. Happy to say, smooth sailing ahead. It's good times to take care of women with uterine cancer. Thank you very much. It's been a pleasure chatting with you today.

uterine cancer treatment

minimally invasive surgery

medical therapies

laparoscopic surgery

lymphatic mapping

genomic testing

immunotherapy