® titled “Advanced Endometrial Cancer.” This is 

®’s first endometrial cancer discussion. I am so excited to be here. My name is Brad Monk from the University of Arizona and Creighton University, both in Phoenix, Arizona. I’m also part of the US Oncology Network. My hospital is St Joseph’s Hospital and Medical Center in downtown Phoenix. I am joined today by 4 experts in the field from very prestigious academic medical centers. First is Dr Kathleen Moore. I’m going to call her Katie. Dr Moore is at the University of Oklahoma Stephenson Cancer Center. Katie, hello.

 Thank you. Dr Dave O’Malley from The Ohio State University Wexner Medical Center in Columbus, Ohio. Dave, welcome my brother.

 Excited to be here for this inaugural endometrial cancer discussion.

 Thank you. Dr Brian Slomovitz from Florida International University Herbert Wertheim College of Medicine and Broward Health in Fort Lauderdale, Florida. Brian, welcome.

Thank you very much for having me, Brad. It’s really an honor.

 You look very smart with all those plaques behind you, Brian.

 Thank you. Dr Krishnansu Tewari from the University of California, Irvine. I’m going to call you Krish. Thank you. Welcome, my friend.

 This is exciting. As I think all of you are aware, there have been several important advances in understanding endometrial cancer, both from a biologic standpoint but also from a novel therapeutic standpoint. We’re beginning to define frontline treatment, and now we have options in recurrent disease. We’re going to review these options, from a surgical, regional, and systemic opportunity. I’d like to get started on our first topic, and that is to consider the biology and the prevalence and incidence of endometrial cancer or uterine corpus cancer.

Krish, why don’t you go ahead and start about the increasing incidence of endometrial cancer and how that has affected your practice.

 Thank you, Brad. In 7 to 8 years we’ve seen about a 30% increase in the incidence of this disease in the United States. There are about 65,000 new cases anticipated for 2020. Maybe 8 years ago there was 45,000. So it’s an increased incidence. It accounts for probably 30% to 40% of my surgical practice, doing surgery for endometrial cancer. We’re not sure why the incidence is increasing. Certainly, for patients who are overweight, obese, and morbidly obese, not only has that been holding stable, but those conditions have been rising.

Diabetes is pretty stable, but what’s interesting is that ever since the Women’s Health Initiative was published in 2002, there’s been a decrease in the use of combined hormone replacement therapy. There’s a possible association between lack of progestins and the rise of endometrial cancer. Also, there are fewer hysterectomies being done for benign disease. Those 2 things need to be looked at in more detail. But predominantly, the obesity crisis in the United States is driving the higher incidence.

 It is the most common operation that I do. Virtually all these operations are done minimally invasive. Katie Moore, your partner, Joan Walker, did that pivotal trial called LAP2.

 Do you ever make an incision for an endometrial cancer? There’s been this evolution in cervical cancer that minimally invasive is not good in cervical cancer. But it’s still OK in endometrial, right?

 Right. Yes, occasionally make an incision, certainly if we can’t do it laparoscopically or more accurately robotically, we get the uterus out the old-fashioned way. But it’s pretty uncommon.

 It’s still not outlawed, but maybe someday open hysterectomy for endometrial cancer will be outlawed.

 It’s really interesting that you’re bringing that up, Brad. Dave can probably comment on this. Actually, all of us can as fellowship directors. When I do the numbers for the fellows now, they do not have a lot of open notes.

 That’s right. With sentinel lymph nodes now…

 Even more common lymphadenectomy is rare.

Overview of Endometrial Cancer

 I was asked the other day, “Brad, what’s the No. 1 word used to describe endometrial cancer?” One word. I said, “Heterogeneity.” It is a constellation of diseases. Brian, Dr Slomovitz, tell us a little about how we’re classifying these tumors at the molecular level.

 Sure. Thanks for that, Brad. Historically we’ve classified these primarily by traditional factors—histology and endometrioid being the most common—followed by serous and clear cell, which are some of the more aggressive histologic subtypes. We still classify them by stage and grade. But of all the cancers we’re treating, the molecular subtypes have really been well defined in endometrial cancer. Ever since Doug Levine published the first TCGA [The Cancer Genome Atlas] article on endometrial cancer, dividing endometrial into 4 distinct subtypes, we’ve really been able to define clear categories for disease and also come up with better treatment plans based on those categories.

Briefly, there’s MSI [microsatellite instability] high, and there’s the 

 mutations. The MSI high make up about 40% of endometrial cancers. 

 is in the single digits—not that common, with a good outcome. There’s the typical low copy number, we call them, endometrial cancers. Then finally, the most aggressive type, the type that are most likely to be ovarian cancers, the 

 subtype. But they’re really well characterized. Not only can we easily divide the cancers into those groups, but those groups really are prognostic and can determine outcome.

 It’s interesting. As gynecologic oncologists, we use the FIGO [International Federation of Gynecology and Obstetrics]staging system. We don’t use the TNM [tumor, nodes, metastases] system. Stage I is limited to the endometrium, stage II in the cervix, and stage III is in the nodes. But Katie, I’ve heard you talk about the molecular staging—maybe this is what Brian said—that there may be a molecular way to look at prognosis rather than disease spread. Tell us about that.

We still have work to do with that characterization. But certainly, alluding to what Dr Slomovitz just mentioned with the TCGA, there have been other interesting studies done in Europe that are consistent. When you look at endometrial cancers that have loss of 

, which we are very used to seeing in ovarian cancer. But in endometrial cancer, we’ve known about it for a while. It’s mainly those patients with uterine papillary serous or very undifferentiated grade 3 endometrioid cancers can have this 

 signature. They’re serous-like and they have a poor prognosis, even when you are controlling for stage.

A stage I, which is all confined to the uterus, endometrioid vs 1 of the loss of 

 tumors, that latter tumor is going to have a much different prognosis if left untreated. How best to treat it—we’re going to talk about it in this session a little later. But we know that they’re much higher risk for recurrence. They make up a much higher percentage of those patients who present with advanced and recurrent disease, so are high unmet need. On the flip side, Dr Slomovitz also mentioned POLE. That’s an interesting group and we don’t understand it really well yet.

We have a sense that even though they are commonly found in undifferentiated tumors, they still have maybe a better prognosis. There’s been some retrospective looks that they do better than they should histopathologically. But it really hasn’t impacted how we treat them. More to come on that.

Risk Classification of Endometrial Cancer

 Surgery is important. Anatomic, typical FIGO [International Federation of Gynecology and Obstetrics] staging is important. The molecular biology of the tumor is important. Dave, bring us back to practicality. I do a robotic hysterectomy as I did yesterday. What does the pathologist do with that tissue? What are the typical biomarkers that are the global standard in endometrial cancer?

 First of all, we need to assess the nodes. Most of us are doing sentinel lymph nodes at this point, and we think that adds to this dividing into stage III, which obviously is a higher risk. We have that. Let’s get the information we need surgically with the hysterectomy and at least sentinel lymph nodes, at the very least, mismatching pair proteins through IHC [immunohistochemistry] at the time of diagnosis. At this point, I would advocate—based on prognostic factors and potential treatment options, as well as identification of people with Lynch syndrome potentially or families with Lynch syndrome—that we do that as a baseline.

What is an evolution? Even though the NCCN [National Comprehensive Cancer Network] Guidelines changed last year, looking at this molecular classification based on the PORTEC trial that is still having a difficult time getting payers to reimburse and thus having my pathologist do these extra tests. What are some of the other things? Well, MSI [microsatellite instability]. Should we be doing that PCR [polymerase chain reaction] test if the MMR [mismatch repair] proteins are all intact? We’re not doing that yet, but that’s something I do on recurrence. How about the IHC for 

 that Katie and Brian have mentioned? That should probably be done routinely, but we need to make sure the payers are paying that. As we look at the 

 and the next generation sequencing, as well as additional mutational analysis, there are some other subgroups that maybe will identify their predictive in the future.

But at diagnosis, particularly in early stage disease, that’s not routinely being done, and we’re having trouble getting reimbursed. We really need to look and work with our pathologists as well as the payers to make sure these things are reimbursed so that we are able to perform those tests. I personally don’t think PD-L1 testing should be done. I see that frequently, especially in patients who are coming in for a referral, because we really haven’t seen that it is a marker in endometrial cancer. Don’t do a test if you don’t know what to do with the results.

 [human epidermal growth factor receptor 2] and ER [estrogen receptor], PR [progesterone receptor]?

 Great question. As we look at these, as 

 moves from the serous cancer where we have data, where it makes the clinical significance, clearly it should be done in all advanced-stage serous cancers. We’re not doing it at our institutions for the local or stage I or II disease, but it’s a consideration as we move forward. We’re not doing ER-PR routinely, but it’s really interesting information. I would argue that anybody with a grade 1 is going to be ER-PR positive. For that grade 2 and 3, you may bring a new set of treatments in that you otherwise wouldn’t use.

 I called you my brother when I introduced you. That’s way too complicated. If you tell your pathology what you just said, it isn’t going to work. Krish, I need you to boil it down here. Then I’ll get Brian. Boil it down—just a bite size. Go ahead. What biomarkers do you want done?

 I want TILs [tumor-infiltrating lymphocytes]. That’s what I want to know. Because that’s going to make a difference in whether immunotherapy is going to be effective for advanced disease.

 All right. Brian, help us. Come on, man.

 I appreciate that earlier you said we come from esteemed universities and that we’re all really smart, but I like to keep it simple. The advantage of molecular profile in these tumors is that 1 of the things we haven’t mentioned is that it makes it more reproducible for pathologists to come up with the same diagnosis.

: Prior to molecular profiling, it’s subjective. Here you get endometrial cancer, and No. 1, you do a 

 positive, you follow it up with a HER2/neu, regardless if it’s serous or endometrioid, because the pathologist could be wrong. If it’s 

 negative, you do ER-PR. It’s simple and straightforward. Anyone can do it.

 positive—in other words, mutated—then you do 

. If you’re not, you do ER-PR. You do immunohistochemistry on those—basically all patients. It’s basically the nonserous, but they might be wrong.

 We just made a very complicated subject matter very simple.

 All right. That’s rare for you guys. Wow. Hopefully we can keep the momentum going.

I don’t mean to insult my good friend, Dave O’Malley. But there’s a real ability here to be astute and smart, and you’re that, Dave. Thank you. I’m so glad to have you here.

Role of Biomarker Testing in Endometrial Cancer

 Krish, I’m going to come back to you. You said, “Let’s treat the patient.” What are some of the challenges you have in treating in an adjuvant setting, so after hysterectomy? Or do you have it all worked out in your mind? In other words, who should get adjuvant treatment? I’m not going to ask you to get into the studies. We’re going to that next. But just summarize, in a statement or 2, who needs adjuvant treatment for endometrial cancer.

 You’ve got a well surgically staged patient; you can use the GOG-99 criteria to determine if they adjuvant radiotherapy. If you’ve got a surgically staged patient with a positive node, they need chemotherapy.

 I love it. See, we’re getting simple here, guys. If you have a deeply invasive or a grade 3 tumor or cervical extension, you give them pelvic radiation. In other words, it’s a chemotherapy world. I love it. Let’s talk now, Katie Moore, about what the standard chemotherapy is, published this year by our friend David Miller. Tell us about GOG-0209.

 Right. We’ve been waiting for this publication. We saw the abstract for it in 2011 or 2012, and we’ve been waiting for this paper for a long time. The reason it took so long is actually not laziness. It’s just that the patients did really well. It took a long time to reach the end points. This is how long it takes to do trials. This was a really important study. It’s very different from preceding studies that had set the standard of care in the past, which had been Adriamycin and cisplatin, after a series of studies done in stage IV recurrent disease. GOG-0209, led by Dr David Miller, had those patients, but also had adjuvant patients. You just referred to that, Dr Monk. For these patients who are fully staged and apparently stage I, then you find a node. Now we call stage IIIC1 or IIIC2. They were all included as well. It’s a very different prognostic group of patients included in GOG-0209. It took what was the community standard at that time—paclitaxel and carboplatin, which is identical to what we do in ovarian cancer—and compared it with what had been the preceding winning arm in endometrial cancer, which was a triplet: paclitaxel, Adriamycin, and cisplatin, which requires GCSF support to give. It was a noninferiority study. No one thought it was going to be better, but it is as good and certainly far more convenient and less toxic. GOG-0209 showed us that, indeed, it is noninferior. It’s pretty much identical if you look at the overall end point, which was overall survival. Why it took so long? If you look at overall survival, it’s completely noninferior and almost identical. If you look at the paper, the curves basically sit on one another with medians of 37 and 41 months.

Probably the most interesting part—and really why I wanted to see the paper—is that they did break out the adjuvant group from the measurable disease group. Because I’m interested to see what that looks like in a nonbiologic therapy era. We’re going to talk about the biologic therapies coming up. But if you use just chemotherapy, what do you get? These are benchmarking data. If you look at the measurable disease—identical, 20 to 22 months median—that’s better than what we’ve seen in the past.

 PFS for the nonmeasurable—basically, these patients are cured. They just never reached it. We had like 112 months vs nonreached. They had a lot of patients in this group cured with chemotherapy. It gives us standard toxicity comparisons that I don’t think we need to belabor here. We all know that paclitaxel-carboplatin is well tolerated and certainly better tolerated than the triplet. I believe we all moved to this many years ago. But if there was any question of what the superior regimen or backbone should be, it should be well established now that it’s paclitaxel-carboplatin.

 Very succinct. Thank you. Global standards. Dave, I’m going to ask you a tough question. I think I know the answer. I’m not sure. Are the endometroid tumors or the serous tumors more chemotherapy sensitive? Or you don’t know? Or they’re the same? Does it matter? Are the endometroid tumors less sensitive or more sensitive than the serous tumors?

We’re seeing across time that both of them respond to chemotherapy almost equally. As we look at some of the other challenges with the poor prognosis of serous cancers, they are usually explained by other risk factors. As we look at these trials, we see that both types respond to chemotherapy. Maybe it’s slightly more in serous, but it’s not enough to deny anybody chemotherapy based on their histology.

I like speaking after Dave because I get to look smarter. I agree. The only other thing is that the chemotherapy is the same for grade 3 endometrioid cancers. Grade 1 and 2 are more hormonally responsive. Fewer cells are in the cell cycle. The low-grade endometrioid cancer doesn’t respond as well to chemotherapy as high grade. But high-grade endometrioid, serous, and carcinosarcomas are in the same category.

The PFS for those patients with tumor is worth repeating. It’s measurable disease. What’s the PFS from GOG-0209?

You guys have done such a great job teaching the world that these cervical and endometrial cancers are chemotherapy sensitive. That’s what I was trying to get you to say, Dave. And that’s probably the right answer. With PFS of over 20 months, they both work.

GOG 0209 Study in Advanced Endometrial Cancer

: Krish, you had mentioned this high-risk, high-intermediate risk GOG-99 group, deeply invasive, maybe limited to the uterus, serous tumors, cervical involvement. You had participated and been an advocate for this study led by Scott McMeekin, MD, FACOG, FACS until his untimely death, GOG-249. Tell us what the intention was of GOG-249 trying to beat pelvic radiation as the standard in that high-risk limited to the uterus subgroup.

: It was taking a riff on what was done in cervical cancer and trying to see if chemotherapy could make the radiation work a little better and also eradicate distant metastases. There was a lot of thought that even serous cancers are going to respond better to this treatment approach because of the chemotherapy. But it really didn’t pan out. They had a good percentage of serous cancers in the study. But right now, if I have a high-risk, early stage endometrial cancer, I’m treating them with adjuvant radiotherapy.

: That would be the ASTRO [American Society for Radiation Oncology] guidelines. Katie Moore, so what it was is pelvic radiation, limited to the uterus, but still serious cancer, versus vaginal brachytherapy in 3 doses of chemotherapy, trying, as you said Krish, to get a systemic effect. Is the reason that that study did not show superiority of vaginal brachytherapy in chemotherapy because we only gave 3 cycles? What do you think?

: No, I don’t think so. I think we have a lot to learn from this study. I’m hopeful we’ll do some subgroup analyses from it. But I think the same problem plagued this study as has plagued other stage I endometrial cancer studies, is that we’re just not great at predicting that group of patients who are really high risk. The GOG-99 criteria, I respect them and the time, but I think they’re not the criteria that really identify the patients who need therapy. It’s probably all 3 risk factors. If you look at the breakdown, and I think we’ll see this from the GOG-249 exploratory. But if you look at those who have deep invasion, high grade and are of older age, all 3, that may be the group who benefits from chemotherapy. They’re all clinical factors. We haven’t incorporated the pathologic factors we’ve talked about, as well, to really identify that 15% who need more help. Then the rest probably don’t need it.

: Yes. Krish, you wrote a very nice subset paper of GOG-122, chemotherapy versus radiation in node-positive disease. You said, “Well, what if there’s just 1 node? Maybe radiation is good enough.” But your findings suggested even where there’s 1 node, that 3C1 tumor, that chemotherapy is better, right?

: That’s right. It’s exploratory. But you’d think that a single positive node, radiation should be just as effective. But we found that if it was a single positive node, chemotherapy is still better. We also found, even if there was just isolated cervical involvement, chemotherapy is better.

: Again, to Brian and Dave’s point, chemotherapy works. It works regardless of the cell type as Dave was saying.

GOG-249 Study in Early-Stage Endometrial Cancer

: Brian, I think this study GOG-258, which was first authored by Daniela Matei, MD, published in 

, of both—we’ve been talking radiation versus chemotherapy—but this is both versus chemotherapy. I think the reason it was published in 

 is because it did not show that chemotherapy and radiation was better than chemotherapy. It was published to encourage people to stop giving chemotherapy and radiation. But I don’t know. Brian, tell me what you think about GOG-258 and what’s the real interpretation.

: To Katie’s point, GOG-258’s really a group of high-risk individuals. In GOG-249, there was no difference because if you’re not going to recur, it doesn’t matter what you give. But GOG-258 was truly high-risk individuals. They gave the best chemotherapy that we thought at the time. The study made an assumption going in that Taxol/carbo [paclitaxel and carboplatin] was going to be the best chemotherapy, because it was prior to the publication of GOG-209. They compared it to a regimen of a heavy dose of cisplatin, 50 mg with radiation, followed by 4 doses of carboplatin and paclitaxel.

What they found is the addition of the radiation for the entire group did not have a superior outcome to those patients who just got chemotherapy alone. Now there’s a subgroup analysis in those patients that had a target to radiate. Meaning, if they didn’t give whole pelvic, if it was more target directed, those patients may benefit from some radiotherapy. But in general, the typical advanced-stage patient, node positive, node was resected, endometrial cancer, the study is definitive evidence that says chemotherapy alone is enough. We don’t need to radiate those patients.

: Thanks, Brian, that’s very succinct. I keep coming back to you, Krish. In your practice, when do you use chemotherapy and radiation in endometrial cancer?

: All right. Well, that’s pretty simple.

: I told you, if they’ve got advanced disease then chemotherapy. If they’ve got high-risk, early stage disease, it’s adjuvant radiotherapy.

: I love the practicality and quite frankly, I agree with you. Dave, there was this paper called PORTEC-3, which I don’t understand. I’m not even going to start. I’m turning it over to one of my smartest friends. Tell me what this trial is. It’s gotten a lot of attention. Put this PORTEC-3 into perspective.

: Well, PORTEC-3 was a European trial. It’s actually 2 publications. The first publication looked at the clinical outcomes in a very—you used the word heterogenous group earlier in this conversation. This was a very mixed bag. So stage I, grade 3 with deep invasion, or lymph vascularization, or both. As well as stage II, stage III or any patients with stage I to III serous or clear cell endometrial cancer. So, a very mixed group of patients. Basically, what they compared was whole pelvic radiation versus chemotherapy plus whole pelvic radiation. We were having a conversation….

: Wait, Dr Tewari said if I got a positive node, the control arm is chemotherapy.

: Well, you have a very strong argument with regard to data that we’ve already reviewed, that the control arm should have been chemotherapy. In some of these patients, the control arm should have been nothing or vaginal cuff brachytherapy to support the previous PORTEC trials. So when we look at this, we call into question, but when you bring in stage III, it’s a diverse group of patients, to say the least. But interestingly, they showed improvements with the chemotherapy. As we’ve already talked about, that should have been no surprise. To the radiation therapist on the clinical trial, I think it was.

So 5-year overall survival benefits by about 7%. The hazard ratio was about 0.7. Failure-free survival, the same thing, about a 6% improvement. When we look at those groups of patients, the addition of chemotherapy to radiation showed improvement in this entire population. We can talk more about the population, the molecular changes, but I’ll stop there.

: I think I’m going to keep moving. To your point, and thank you for being so good natured, it shows that chemotherapy works when used in the adjuvant setting, which is what Dr Tewari said. He said it when he subanalyzed GOG-122 and he practices it, as all of you do, and Dr Slomovitz says that the addition of radiation doesn’t do much to it. I think we’ve set the standard.

GOG-258 and PORTEC-3 Studies in Endometrial Cancer

Let’s discuss what you guys are excited about. What I’m excited about is immunotherapy. Immunotherapy in ovarian cancer has been very disappointing.

But in cervix and endometrial, it’s very exciting. There are several trials, and I hope the next 

® can be cervical cancer based. But we’re going to talk extensively here about checkpoint inhibitors in endometrial cancer. Brian, maybe you could talk about the role of PD-L1 monotherapy in MSI-H [microsatellite instability-high] endometrial cancer, how it has changed your practice, and how it has evolved.

 It’s changing the world of endometrial cancer. In our first section, we were high on chemotherapy. My goal in my practice is to even eliminate chemotherapy for these patients. I discussed earlier that 39% to 40% of patients with endometrial cancer will be MSI-H. With immunotherapy, particularly checkpoint inhibitors, we know that there’s a very high response rate to tumors that are MSI-H, which is tumors that have a high tumor mutational burden [TMB].

Tumors that, in laymen’s terms, are recognized by noncell by the body. Something that shouldn’t be there. The first tumors that benefited from single-agent checkpoint inhibition, melanoma, non–small cell lung cancers. We learn that there’s a large subgroup of endometrial cancers, those with MSI-H, that respond quite nicely to single-agent therapy with checkpoint inhibitors. 57% response rate, pembrolizumab. Dave O’Malley, 1 of our speakers, reported that 57% response. Which is crazy, because what you got labeled for was about 38%. They found as they got more and more data, it worked better. When I give it to my patients who are MSI-H, their tumors have a high likelihood of melting away.

 It’s a class effect. Nivolumab, another story through the NCI-MATCH trial. In the MATCH trial, a large NCI [National Cancer Institute]–sponsored trial that is tumor agnostic, looking at not a specific disease but a mutation. Those patients that were noncolorectal—MSI-H, when given nivolumab, another checkpoint, high response rate. It’s changed the way, particularly in the MSI-H patients, and we’re going to spend a lot of time trying to make cold tumors hot. But in the hot patients, MSI-H, single-agent checkpoint therapy has changed the world.

Katie, is there any way that someone has not heard of this? You hear Brian talk about it, and you’re like, wow. But this is a message that’s 3½ years old. This was approved on May 23, 2017. Everyone knows it. I know you don’t know what everyone knows, but I’d be shocked if a treating oncologist has never heard of an MSI-H tumor.

 Yes. This message—the medical oncology community was using immune checkpoint inhibitors long before we were. Colorectal is probably the nearest relative to our endometrial cohort. They’re well versed in this and saw the data as they came out. Really, as the standardization of testing for mismatched repair, as we alluded to earlier, on all patients has come into play. It’s unusual to find a patient, unless they were diagnosed long ago and now, they’ve recurred distantly and we’re testing them. That happens, but this message has permeated the community and people are doing a good job. If anything, when I talk to medical oncology, they’re being more creative with their use of immune checkpoint inhibitors for gynecologic cancers than we are.

A couple of things. IHC [Immunohistochemistry] for mismatched repair deficiency is not perfect. Maybe only 90%. I think you said, maybe Brian, that the 

 mutations, which are hypermutated, are not detected by IHC. Now, as of June 16 of this year, we will have a TMBindication based on the FoundationOne CDx. Again, IHC is not enough. At least sometimes. Misses 

 for sure. Now you have TMB. Krish, when we get a recurrent patient who might be eligible for pembrolizumab, do we need to send it to FoundationOne? I’m just using that because that’s the companion diagnostic to look for TMB and 

. Or do you say, “Oh, it’s mismatched repair, enzyme IAC[intra-arterial chemotherapy]? It’s not there, so I’m going to move on.”

 I don’t send it. If I can do it with IHC and get them going, it’s easy for me. The bottom line is that with the combo approved, every patient with advanced endometrial cancer can get pembrolizumab.

Yes. No question. When they’re mismatched repair deficient or MSI-H on IHC, that’s easy. But I may not want to give them the combination. Sometimes I’ll send it for next-generation sequencing [NGS] just to make sure I’m not missing a single-agent pembrolizumab opportunity. What do you think about that, Brian?

That’s a crucial point. Everyone’s excited about the combo, and we’ll talk about that. But we’ll also mention that there are some toxicities associated with the combo. To get a better response rate and fewer toxicities, we’re not doing our patients justice if we’re not looking for MSI-H.

 Yes. I would argue for advanced recurrent disease, where unfortunately you know the likelihood is that if they’re stage IV recurrent, so I guess we can discuss this. I don’t see a reason not to do next-generation sequencing. Because you get the 

, you get MSI-H, you get tumor mutational burden—I’ve actually just found 1 who’s not MSI-H, so she’s going to get pembrolizumab. But you also get everything else that we’re going to talk about, including ERBB2 amplification and mutations. Which are also targetable. I don’t see a reason not to send next-gen. I don’t send it on stage I, but everything else is getting NGS.

I’m going to put more pressure on you. Also in June, we got to 2020, first-line pembrolizumab for MSI-H CRC [colorectal cancer], first line vs chemotherapy. A randomized phase 3 trial. If you have a metastatic case that’s MSI-H, you’re going to give her a carboplatin-paclitaxel or checkpoint or all 3. I know you wanted to put her on a clinical trial, but what’s the right treatment for that newly diagnosed, MSI-H, widely metastatic patient?

 We don’t know, to be honest. The trial needs to be done. Some of our colleagues have proposed it several times. It keeps getting turned down, which is frustrating. It’s not standard of care, and you have to counsel your patients about that. However, patients are smart and they know the data.

 What’s the harm, Brad, about just starting with pembrolizumab? Because if they progress through it, you can give them chemotherapy. That’s the message.

Yes, I never disagree with Katie. But we don’t need a trial. We got single-agent approval in the second line on, what, 17 patients?

A several-hundred-patient trial. We have to report on just 17 patients to get approval.

 I am fine using I/O [immuno-oncology] in the front line in a well-counseled patient. We’ve learned that lesson the hard way with immunotherapy. MSI colorectal and MSI endometrial, are they more similar? Because if it is similar, we can just move to I/O. I don’t know that we can say that definitively, but I would not fall on the sword against I/O in the front line.

 To your point, Brian, the accelerated approval for MSI-H was mostly CRC. Right? The CRC drug and the endometrial cases with it. Here, vs chemotherapy, to your point, they also sort of dragged endometrial cancer with it. To what Dr Tewari said, what’s the harm? You know it’s going to work. The argument, Krish, to your point, is triplet. Carboplatin-paclitaxel plus pembrolizumab vs pembrolizumab alone. What do you think of that?

 It would be nice to get the trial back on because that whole thing got torpedoed because of COVID-19 [coronavirus disease 2019]. So I don’t know.

Right. We have a trial funded through NRG Oncology called GY018, which asks this very question. Not just in MSI-H patients but in all comers. But there is a specific MSI-H end point. Thank you, Krish. I know you’ve been a passionate advocate for clinical trial enrollment throughout this pandemic.

Immune Checkpoint Inhibitors in Advanced Endometrial Cancer

 Dave, I’m saving the best for you. I have news for the world that pembrolizumab isn’t the checkpoint inhibitor. Right? Shocking. I know sometimes you think that way. But we have 6 FDA-approved checkpoint inhibitors. You work on probably No. 7 and No. 8. You work on balstilimab and dostarlimab. Tell us a little about dostarlimab in endometrial cancer. Because that’s likely to be the next FDA-approved checkpoint inhibitor.

 I would love for that to happen. I love competition. Competition is good for our patients. It’s good for the cost of health care. I hope that that occurs. This is a nice presentation, but these data on endometrial cancer are basically 2 trials. One on endometrial-proficient patients and 1 on deficient patients. These were 100 in the deficient arm, 140 in the proficient arm. What we showed was a 45% response rate in the deficient group and 13% response rate in the proficient group. Very consistent with what we’ve seen, particularly in the deficient. But we also have seen a group of patients here, about 13% response rate with single agent. We’ve alluded to the lenvatinib-pembrolizumab data, which had about a 37% response rate. It doesn’t compete with that combination. But we do see some activity.

Emerging ICI Therapy for Advanced Endometrial Cancer

Let’s transition and talk about the best treatment for second-line non–MSI [microsatellite instability]-high recurrent endometrial cancer. Katie Moore, you mentioned that pembrolizumab got accelerated approval for all MSI-high tumors, that that message is out there, that we’re using it, we’ve seen fantastic responses, sometimes over 50% in some cohorts. But most patients are not MSI-high. This has created a high unmet medical need, a need that we’ve been working on intently. Then ultimately, there was a study that led to accelerated approval again, but this time not in the MSI-high subgroup. Dave, tell us about this opportunity.

: When we look at this group of patients, we have lenvatinib and pembrolizumab, which has, as I said, about a 37% response rate [RR]. The problem is the toxicity, which we’ve all alluded to. You’re talking about 70%, maybe even more, dose reductions, and about 90% if you include the interruptions. When we look at these patients, we get some amazing responses. In the waterfall plot of this original paper, basically everybody responded. That’s unheard of in second-, third-line endometrial cancer. We see this group of patients receive these opportunities. But we must get better at managing the toxicities.

Katie Moore, I’ve heard people say that if you get frontline carboplatin and paclitaxel in the adjuvant setting, and if it’s been, I don’t know, 6 or 12 months, because of some of these adverse events of the pembrolizumab, lenvatinib combination, that maybe you should go back to chemotherapy first. You talked about the expectation for GOG-209, that this was chemotherapy sensitive. What’s the expectation of retreatment let’s say with carboplatin and paclitaxel, so that we can make an assessment? Should we do that in the second line and relegate pembrolizumab, lenvatinib to third line, or do doxorubicin, something completely different?

: Yes. That is the question. The answer is, we don’t know. We all guess. There is no defined platinum-free interval for endometrial cancer. Although people talk about it, it’s all made up data. There’s no 6- or 12-month interval that predicts re-response to platinum. It’s never been prospectively studied. The ongoing studies that we’re going to talk about at the end of this, do allow patients who had adjuvant paclitaxel, carboplatin, and they have variable periods, usually it’s a year,when they can come on these frontline metastatic studies. We will have those data coming. We’ll see if anybody responds. Is it like ovarian cancer? I bet not.

: I don’t know if I should say I wish it were, but we kind of wish it were, where we could repeat use platinum. I don’t think we can. We’re going to learn it. That’s coming. But it does present this clinical dilemma where you have these patients who received adjuvant therapy on the GOG-258 regimen or whatever, and 2 years later, they recur. Do you go right to this lenvatinib, pembrolizumab combination, which has the highest response rates we’ve ever seen in second-line endometrial cancer? Or should you start with, and this is with MSS [microsatellite stable] of course, should you start with chemotherapy again and see if they re-respond?

Krish, you’ve been on all targeted therapies, all of you have,you’ve been an advocate of this anti-VEGF checkpoint combination in cervical cancer and also here. Tell us about the biology. Why would an anti-VEGF make IO [immuno-oncology] work in a cold tumor where it otherwise is not working?

: Yes. It’s interesting. This is what’s exciting about it. Because we know that VEGF is immunosuppressive, especially in the tumor microenvironment. It’s this combination that allows for the tumor immune microenvironment to be reprogrammed, if you will. That’s where the synergy comes in. But we always say cross-trial comparisons are never good. But I like to cross tumors and stuff. If you look at PAOLA, you have patients who are HRD [homologous recombination deficiency] positive but they’re not necessarily 

mutated. But they need something else other than just the olaparib to work in the maintenance setting, so you add BEV [bevacizumab]. In this case, you have a microsatellite stable patient and they could benefit from pembrolizumab, but they need something else. That’s where the combination with anti-VEGF comes in.

I do that all the time. I try to look at other tumor types and get smarter and learn from my colleagues in ACC [adenoid cystic carcinoma], so atezolizumab, bevacizumab.

But it didn’t work in ovarian cancer. Why, Katie Moore, why? It didn’t work in IMagyn050, presented at ESMO [the European Society for Medical Oncology annual meeting].

: Yes. That’s why we don’t want this disease to be more like ovarian cancer….

Any comments on that, Katie? This is not an ovary discussion, but it’s weird, isn’t it?

: I thought it was going to be positive. Ovary is just not these other tumors as much as want it to be. The synergy that was seen in HCC [hepatocellular carcinoma] and renal cell and increased T-cell trafficking and all the things that should have worked, and maybe they did happen, we don’t know. Maybe they did happen in the ovarian cancer tumors, but it didn’t matter. It probably comes down to lack of understanding of biomarkers. We have a lot to learn, but it didn’t work in ovary, unfortunately. But here it does work in endometrial cancer.

: Yes. It’s cervical cancer. It’s chemotherapy with or without bevacizumab, plus or minus pembrolizumab. It’s for the first line of recurrent metastatic cervix cancer. I’m hoping it’s going to be a positive trial. Like I said, we should know in 2021.

Yes. IMagyn050, we were hoping; it didn’t work. But we haven’t given up. KEYNOTE-826 is your regimen, Krish, GOG-240, chemotherapy, bevacizumab, first-line metastatic cervical, adding pembrolizumab to it.

Combination Therapy for Recurrent Endometrial Cancer

Brian, I want you to weigh in on what I asked Katie. How do you decide in a pretreated patient? The label for pembrolizumab, lenvatinib is you’ve got to have the molecular signature, you have to have had prior systemic therapy, and you can’t be cured with surgery or radiation. Who is it not right for? Or is it basically right for most patients in the second line?

 It’s going to be right for most patients in the second line. We always have to talk about the future, until it gets approval in the first line. Because we’re doing the trial, which is accruing nicely. It’s a Merck-sponsored trial that we’re doing. It’s outside of the GOG [Gynecologic Oncology Group] but it’s accruing nicely. It’s looking at pembrolizumab, lenvatinib in the first-line setting.

Versus chemotherapy, versus carboplatin, paclitaxel.

: Versus chemotherapy in the first-line setting. Now remember, the other thing about what you just said is the label for pembrolizumab, lenvatinib as written is for a prior systemic therapy.

: They could have gotten 1 dose of cisplatin and 40 mg of…with 1 week of radiotherapy, and they’re indicated to get it. They could have gotten 1 week of Megace and be eligible for pembrolizumab, lenvatinib. It’s a game changer. I know we’re going to talk about the ongoing trials. I think that this one has the potential to have the most impact. Because as I mentioned earlier, it may knock chemotherapy off the pedestal for the first-line treatment of endometrial cancer. It’s exciting.

The toxicities that Dave was talking about, they’re there. In the trenches, we’re learning how to better handle those toxicities. We’re starting at lower doses of the lenvatinib. We’re learning how to deal with some of the diarrhea and hypertension, which happens. It’s an extremely exciting time. Every second you were talking about cervix and ovarian cancer, I’m thinking, let’s get back to endometrial.

 Let’s try not to use the word toxicity, because that word is toxic. These are adverse reactions. All systemic therapies have some adverse reactions associated with them. I just want to review with you, this is a combination. You have the immune-related adverse reactions related to the pembrolizumab, and then those related to lenvatinib. Lenvatinib really is 2-fold. It’s an oral anticancer therapy, so it has some fatigue and some GI [gastrointestinal] disturbances. But it also has the anti-VEGF effect, which includes hypertension, proteinuria, wound healing, and so on. You have to balance the IO [immune-oncology] aspect, the oral aspect, and the anti-VEGF aspect, and we’re good at it. 

All of you said you were excited about it. All of you inferred that it’s probably better than chemotherapy. Then you said, Brian, well, we’re comparing it to chemotherapy, in the existing label indication KEYNOTE-775, and comparing it in the front line in LEAP-001. So Dave, 88% of patients have to interrupt or dose reduce their lenvatinib. You start at 20 mg, and you go 20 to 14 to 10 to 8 mg, what’s the right starting dose? Is it 20 mg? Or is it 14 mg?

: I don’t start at 20 mg on anybody. But it’s more a practical reason. I start at 18 mg. There’s a package insert, or a tab, of 18 mg. It gives you a 10 mg and two 4 mg. Thus, if you know that 70% are going to be dose reduced, and as you said, almost 90% would have dose interruption, it allows you to get down to 14 mg much easier than waiting for a new script and a new co-pay and all that. I start at 18 mg. If they need dose reduction, they have the 4 mg there. It works perfectly. There are people who are starting at 14 mg. There may be some induction. That’s why I’m a little apprehensive to go all the way down to 14 mg on all my patients, but it’s also very reasonable. I like 18 mg.

But the study was 20 mg. Where are the data on 18 mg?

: There are no data on 18 mg. There are no data on 14 mg….

There are no data starting at 14 mg, but there are lots of data that if you start at 20 mg you get to 14 mg in the induction effect. I’m not trying to conflict against you, I’m just saying, I try to follow the recipe. Go ahead. Any of the 3 of you, Krish, Brian, or Katie, who want to chime in on this.

: I start at 20 mg, and we’ve had to drop down to 14 mg and down to 10 mg on a few patients. But I’ve got quite a few patients who are tolerating it pretty well. The worst I’ve seen is maybe some grade 1 or 2 hypothyroidism.

: Yes. I honestly do think there are regional differences in patients. But we have a lot of patients who present late for care and with very advanced disease and a lot of comorbidities. I don’t start at 20 mg and neither do my partners for recurrent disease. They’re so sick, and if they get sick on 20 mg, they’re not going to dose reduce, they’re just going to come off and go to hospice. For us, we start at 14 mg, although you might have just persuaded me, Dave. I like what you just said about starting at 18 mg. Maybe I can start at 14 mg and then escalate up. But I start at 14 mg, and I get 2 months in, and then if they’re doing fine, I escalate them to 18 mg. I’m going to be honest, in our sick population, I can think of 3 patients who were told they needed to go on hospice who are now going to have more time because of this drug.

I was real worried about the toxicity, and I was not a super believer, but it works. I like 14 mg because I haven’t harmed anybody, and I can escalate them. That’s what we do.

: What about the induction? What about the induction piece?

: Well, the induction piece is interesting. Does it help to get that higher level? I’m not sure of the answer to that. I don’t know we’ll ever have the answer to that. But one thing, Katie, that you just said, they’re going to get to this next holiday. There’s an 11% complete response [CR] rate. I have some sustained partial and near complete response using this combination. Are we talking that we may be able to cure recurrent endometrial cancer?

In the label, 94 patients, 38.3% response rate, 69% of them remained progression free for 6 months. Krish, to your point, your induction, you taught me that this anti-VEGF effect impacts ascites. If I have a recurrent endometrial cancer that for sure is not MSI [microsatellite instability]-high because that doesn’t exist, I want her to get some serious anti-VEGF activity to help her with her symptomatic ascites. It’s individualization, there’s no one-size-fits-all, but that’s an example where I want to get some anti-VEGF effect because what’s hurting her is her cancer. If I don’t get control of that ascites, she’s going to get a malignant bowel obstruction.

: I totally agree with you, Brad. The patient I was just talking about presented with 8-liter paracenteses for 3 consecutive weeks. Then she came to me for a clinical trials consult. She’s dying. Now we’ve got her on study, and we’re at like 2 liters a week now, from 8 to 2 liters. It’s not perfect, she’s not a CR, but she’s walking and functional and eating again. It’s that VEGF, it’s amazing.

It is. To your point, if she throws up her lenvatinib, you haven’t accomplished anything.

: I don’t switch to 18 mg, I keep it at 14 mg. Brad, to your point, we don’t know the activity, come on, we just got over niraparib. We almost lost niraparib because we started at 300 mg and there were too many adverse events. If they’re doing great, have them go up to 18 mg. It works.

Krish says, what about Doxil, or liposomal doxorubicin? Liposomal doxorubicin, for all intents and purposes is inactive as a single agent in ovarian cancer. With response rates of around 10%, or I don’t know what it is.

Both. I was talking ovarian cancer. But it’s like 0% in endometrial cancer.

: Yes. Zero in endometrial. Look at the AURELIA trial, it’s 10%, 13% with bevacizumab.

Right. My passion, though, for dose intensification of pembrolizumab, lenvatinib is because I get 40% response rate for it. I get a durable duration. To Dr Moore’s point though, if it’s intolerable, it’s going to be 0%. You’re all right.

: My patients are not that sick when we put them on the combination. But I’m doing this talk from my St. Joseph Hospital office. I just remembered what my nurse navigator always says, when this combination works, it really works. Disease isn’t gone, I can scan them and it’s still there, but it’s not hurting them, it’s stable. They’re doing great on the combo.

Any closing comments on the activity and the tolerability of pembrolizumab, lenvatinib in non–MSI-high endometrial cancer?

: The message needs to get out. This is something that we need to push out. The question about, do you go to chemotherapy first or do you use this? I don’t know. I feel like maybe we need to use this first. It’s probably more active, is my gut feeling. But that’s something we need to sort out and get into the community a lot more, and tailored instructions as to how to manage the toxicity. Although, again, medical oncologists know how to use lenvatinib. They’ve been using it for a long time. They understand the hypertension’s fast, they know how to manage the fatigue, and they know how to manage the diarrhea. And they understand pembrolizumab. They’re probably ahead of us on that. It’s just remembering to use it.

Dave, you were going to talk about the importance of the multidisciplinary team, go ahead.

: Yes, absolutely. I have an amazing nurse practitioner who works with me, and they take these phone calls and they have strict SOPs [standard operating procedures] in place, with regard to hypertension, with regard to diarrhea. If you’re working together, it can really help control these symptoms.

 I agree with that completely. Pembrolizumab, lenvatinib doesn’t start when you take the medicines, it starts when you decide to take the medicines. From that point, you’ve got to get the blood pressure under control, to Dave’s point, getting your nurse extenders involved, and preparing for the diarrhea. If you’re prepared, I find that it’s a tolerable regimen, and obviously, we talked about the efficacy.

Advanced Endometrial Cancer | Peer Exchange

EP 1: Overview of Endometrial Cancer

EP 2: Risk Classification of Endometrial Cancer

EP 3: Role of Biomarker Testing in Endometrial Cancer

EP 4: GOG 0209 Study in Advanced Endometrial Cancer

EP 5: GOG-249 Study in Early-Stage Endometrial Cancer

EP 6: GOG-258 and PORTEC-3 Studies in Endometrial Cancer

EP 7: Immune Checkpoint Inhibitors in Advanced Endometrial Cancer

EP 8: Emerging ICI Therapy for Advanced Endometrial Cancer

EP 9: Combination Therapy for Recurrent Endometrial Cancer

EP 10: Recurrent Endometrial Cancer: Safety of Combination Therapy

EP 11: Sequencing for Advanced Endometrial Cancer

EP 12: Emerging Agents: Treatment of Endometrial Cancer

EP 13: Endometrial Cancer: Chemo-Free Regimens on the Horizon

EP 14: Novel Targets for Advanced Endometrial Cancer

EP 15: Future of Advanced Endometrial Cancer