two talks, one that I gave at ASTRO and the other at SGO, and I will get started. So I want to discuss a radiotherapy named endometrial cancer in the era of molecular profiling. So to begin, this very interesting trial showed no benefit of radiotherapy in stage 3 endometrial cancer. So this was GOG258. It was chemotherapy only, six cycles of two-agent chemotherapy versus chemoradiotherapy, concurrent cysts and four cycles of carbotaxol. And as you can see, there was no difference in overall survival. So as a radiation oncologist, we're a little defensive and I want to present evidence for why radiotherapy may be effective in endometrial cancer. So I'm going to provide three avenues for evidence. Number one, neoadjuvant radiotherapy. Two, GOG238, a randomized trial. And three, the data in medically inoperable cancer. So for neoadjuvant radiotherapy, this is an uncommon, but a somewhat historic way to treat endometrial cancer. So for patients with big tumors who may not be amenable to, to a hysterectomy, one approach is to give preoperative radiotherapy. And that's what these colleagues did from, from university of Pittsburgh in the U S these were stage two and three patients, relatively small number, a retrospective series, 36, all had cervix involvement. 50% had parametrial involvement. They gave standard doses of radiotherapy and they also did high dose rate brachytherapy. Over half had concurrent cisplatin, 24% had a pathologic CR, 36% had micro microscopic residual disease. And the three-year local control rate in this group of relatively advanced tumors was 96%, which I would say is successful. The next piece of evidence is GOG238. So this was for a randomized trial of pelvic recurrence, plus or minus concurrent cisplatin for recurrent endometrial cancer, standard doses of radiotherapy. And as you can see here, there was no difference statistically between these two curves. The radiation curve is in blue, slightly above the, the arm of radiation and cisplatin. This was just published in JCO. These data were released prior to publication because the, the NRG oncology investigators felt that it was worthwhile to let people know of these results. Albeit, you have to know that most of these were very early stage. Most of these were recurrences at the top of the vagina. And most of these were grade one and two over 80%. Toxicity of, did favor the radiotherapy as opposed to the radiotherapy and, and cisplatin. And the authors concluded that those with low grade and vaginal apex recurrences may be best treated with radiation therapy alone. The last piece of evidence is medically inoperable endometrial cancer. This is a paper by Schwartz et al., a summary published in brachytherapy. They looked at 11 studies with stage one and two disease, disease, disease, a free survival averaged 81%. So I would conclude that radiation therapy is a curative modality in the upfront and sometimes in the recurrent setting. We've learned a lot in recent years. These are the PORTEC3 data published in JCO in 2020, four years ago. As you can see, poly mutated patients did very well. Patients with P53 mutations did markedly worse and patients who were MMR deficient and patients who had no specific molecular pathology or NSMP were in the middle. If we look closer at relapse free survival and, and overall survival panels, A and B are the P53 curves. We see a major benefit for the addition of, of chemotherapy in those patients. In panel C and D there is, all patients did very well regardless of treatment. In panels E and F, MMR deficient, there was essentially no difference with the addition of chemo and within the NSMP group in G and H, again, very little difference in those two curves with the addition of chemo. In the PORTEC3 results, little benefit in early stage patients, but, but a, a fairly marked improvement in stage three patients, as you can see. And overall survival for stage three patients at 10% improvement with the addition of chemotherapy. If we look at toxicity in the, for the global health status, both groups did quite well for physical functioning. Radiotherapy alone did slightly superior compared to chemo radiotherapy. And you can see that on this curve too, patients who underwent chemo radiotherapy compared to radiotherapy had somewhat worse toxicity over time. So the 253, excuse me, 258 results, which I showed you briefly earlier was a similar population. And those patients had chemotherapy alone versus chemo radiotherapy. And there was no difference in those two curves. There was a difference, however, in the, in the, in local regional recurrence, about a 10% improvement in pelvic and periortic recurrence and improved vaginal recurrence in patients who received radiotherapy, but no benefit as mentioned for overall survival. GOG-249 was a, was a trial in, in early, earlier stage disease, stage one and two. That trial was radiotherapy alone, pelvic external beam, standard doses versus three cycles of carbotaxel and vaginal cuff brachytherapy. And there was improved pelvic nodal control in patients who received the external beam radiation. There was more toxicity before the, in the group that received chemotherapy. So for stage one and two across these trials, there was excellent pelvic control with radiotherapy. There was a benefit from radiotherapy seen also in patients that were staged with a lymphadenectomy and pelvic radiotherapy remains standard for stage one and two non-serious cancers. We do use vaginal cuff brachytherapy in many subsets I might add. For stage three disease, combined chemo radiotherapy provided a 12% improvement in failure-free survival and a 10% improvement in overall survival. So we need shared decision-making to come up with the right treatment. And for patients with serious cancers, we do advocate chemotherapy and frequently radiotherapy too in those patients. So there are several remaining questions. One is the sequencing. Another is potentially less extensive surgical staging for serious cancers when we know they're going to get chemotherapy and a differential approach by molecular subgroups. And that's what I'd like to talk about a little further here. CORTEC-4 will be presented hopefully later this year. Dr. Coittsberg has shared that with me. This was a trial where patients were allotted to various therapies depending upon their, their risk. So favorable patients, those, that is those patients that were pol-E mutated, as you can see in the blue boxes, those patients were observed. Patients with an intermediate prognosis in the green boxes were to receive vaginal cuff brachytherapy and patients with higher risk features, IEP53 mutations were to receive external beam radiotherapy. Those results are, are forthcoming. So what's the best treatment for patients with mismatch repair deficient tumors? That is currently, I think, unknown. Here's one report. This was a retrospective multi-center trial for stage one and two patients, grade three disease with mismatch repair deficient and mismatch repair proficient patients. Those that were mismatch repair deficient had a benefit to the radiotherapy, whereas there was no benefit for the patients that were mismatch repair proficient. Those are, that's just one trial. This was an excellent report, again, published in JCO. This is a summary of the PORTEC1 and PORTEC2 data looking at molecular classification and the response to radiotherapy. As you can see in panel A, those that were pol-E mutated all did very well. In panel B, omitting radiotherapy, excuse me, the benefit may be rather limited for radiotherapy in the mismatch repair deficient tumors. In panel C, those that were P53 mutated had a pretty marked benefit for pelvic external beam. And those that were NSMP seem to benefit from vaginal cuff brachytherapy. So we have almost an embarrassment of riches now. There's been really remarkable trials in endometrial cancer with immunotherapy agents. This is the GY018 trial published by Eskander et al in the New England Journal in the past year. Large number of patients. You can see in panel A in the deficient MMR cohort, there was a major benefit for the addition of pembrolizumab to chemotherapy in patients with stage three and four endometrial cancer. And in panel B, even in the proficient MMMR cohort, there was a statistically significant benefit. Similarly, Mansoor Mirza published the RUBY trial, which again showed very similar results with dastarlimab compared to pembrolizumab in the previous report. And again, major advantage for those two drugs. And again, in the overall population, you can see that here. This has been investigated more recently in early stage disease and stage one and two, the data is not out yet, but this was performed by the NRG oncology group, the GY020 group. I believe the trial is closed, but it was a standard of care, essentially vaginal cuff brachytherapy randomized to pembrolizumab for early stage disease. Those trials are awaiting yet. Another targeted form of therapy in endometrial cancer is lenvatinib and pembrolizumab. And those data are shown here showing again, a pretty marked improvement with those drugs. Another a form of targeted therapy for patients with endometrial cancer is that in uterine serous carcinomas, about 30% of those cases will overexpress HER2. Dr. Fader published several years ago that HER2 targeted therapy may show a pretty significant benefit for those patients that do have overexpressing serous carcinomas for HER2 new. These data have led the GCIG, the gynecologic cancer intergroup to embark on a variety of trials shown here with dose escalation in the P53 abnormal subset with the addition of a laparib in the MMRD deficient, looking at immunotherapy in the NSMP group, looking at hormonal therapy to see if a benefit can be achieved. And then the polybutated cases, we'd like to see if de-escalation of therapy could potentially be beneficial. So the conclusions are that molecular classification predicts response to radiotherapy. This was shown in PORTEC1, 2, and 3. There are new drugs for molecular defined subsets. These are here and more are coming fast. It's a fairly remarkable space. Even in wealthy countries such as mine, sometimes these are very difficult to come by. Colleagues I've heard in Australia have difficulty obtaining many of these drugs. It may be some time before we all have access to these agents, but I do believe it's fundamentally changing how we think about endometrial cancer. In early stage disease, omitting radiotherapy I think is safe in poly-mutated cases. It's not easy to get poly-mutated patients or to know of those cases. To do that, you need to sequence the poly gene. Benefit for radiotherapy may be limited in MMR deficient cases. External beam radiotherapy may be preferred in the adjuvant setting for P53 abnormal cases and vaginal brachytherapy may be preferred for NSMP cases per the PORTEC1 and 2 data. Concurrent radiotherapy and immunotherapy is safe. That was shown in both of the RUBY and GYA018 trial. And incorporation of radiotherapy in these cases will be potentially a challenge. So I'd like to move on to a different subject. This is a little bit controversial and this is the new staging system adopted by FIGO. I presented this at the Society of Gynecologic Oncologists in the United States about a month ago. I was fairly apprehensive because it's not very popular in the United States. I was a co-author on the publication and so I participated in a debate and it was an interesting experience. The charge to our committee was to update the staging system and to include molecular classification. In the most recent staging manual by the AJCC, the American Joint Commission on Cancer, they describe several principles of staging. And one is to compare like and unlike groups, particularly in regard to different therapeutic procedures. And the other is to build the important bridge from a population base to a more personalized approach to patient classification in the era of precision molecular oncology. As I present this, I want you to understand that as a co-author on the publication, I am biased and I am in support of the FIGO 2023 classification. The major critiques of this system are the following. That is people complain that cancer staging should remain anatomic. It is too complex. Validation is needed. Molecular-based prognostication should not be part of staging. LVSI should not be part. Histology should not be part of it. Is it equitable and is it cost-effective? So here are the two previous staging systems. In 2009, we had nine sub-stages. We currently have 19. And that is clearly more complex, but I would argue it's not too complex. Shown here is the current staging system by FIGO, the 2023 staging. And due to the limitation of time, I'm not gonna go through all of those sub-stages. I do firmly adhere to what we call the KISS philosophy. That is keep it simple. I do think we need to keep things simple. And I do think this system is sufficiently simple. I don't think it's overly complex. So to address the first point, should staging remain anatomic? Well, the train has left the station and many other disease sites, we use non-anatomic factors such as head and neck, breast, melanoma, skin, et cetera. We frequently use non-anatomic features. For example, in head and neck cancer, if you're P16 positive, that is if you're HPV positive in your T3 or T4 and N1, you are stage two. If you're P16 negative and you're the same, you're stage 4A, markedly different. In head and neck cancer, there are 10 sub-sites, all with separate T and N staging. It's moderately complex. Breast cancer is not simple. There's 14 T stages, 13 N stages, three M stages. To get the prognostic stage system, you need to know the T stage, the N stage, the M stage, the grade, the HER2 status, ER status, PR status, and oncotype. And then you put those together over these various four pages and you come up to the appropriate prognostic staging group. Nearly 20 years ago, we published in this paper the following statement, that future work is needed to continue to delineate clinical and biologic factors, which can guide treatment decisions and account for disparities in outcome between varied substance of patients. In the United States, we have these large databases that have EOD codes or extensive disease codes, but they in the past have include anatomic features and we can do lots better. And in fact, we have done lots better. If, for example, as I've shown you earlier, the PORTEC data encompassing the TCGA classification, totally revolutionized the way we think about endometrial cancer. And in my own practice, and I would think many others, this informs our practice every day. I don't have the luxury of knowing if my patient's wholly mutated, but we can get a P53 stain, one immunohistochemistry test to tell if the patient has a P53 mutation. I would submit that we would be remiss not to use this important information. So does this system need to be validated? Shown here on these two slides are 10 different reports that indicate support for the new system. So I would also say that in the history of FIGO staging, this has probably been more analyzed than any other. And the data so far is in support. This is one report. This was from one of the residents in my own group at the University of Utah. This was a SEER, excuse me, an NCDB database study on over 135,000 patients. And the likelihood ratio favored the new staging system. For 1A3 patients, that is those patients with low risk synchronous endometrial and ovarian tumors, 10-year overall survival was 73% compared to 52% in the old staging system, a marked difference. For patients that are 3B2 that had pelvic peritoneal involvement, survival was 49% compared to 19%, a 30% difference. For patients that were 3A2, cirrhosal involvement compared to 3A1 adnexal involvement, there was a 10% difference. So we're seeing, I would say large differences in survival. So what were the main changes? Well, there was an expansion of stage two in the new system and a reduction in stage one. There was an expansion in stage two because incorporation of LVSI in stage 2B and stage 2C was adverse histology. So those patients with cirrhosal or clear cell were incorporated in stage 2C and that expanded the stage two group. For patients that were stage one, there was differences in overall survival that ranged over 20 percentage points at 10 years. I would say that is important. For patients that were upstaged in the new system, there was a marked difference in overall survival at 10 years. In later stage disease, Matsuo and colleagues demonstrated that macro metastases to the lymph nodes did worse than micromets. And in stage four, there was more discrimination which was clinically relevant for surgical and adjuvant decision-making. So remember that molecular-based prognostication is already done in many cancers. Molecular testing in FIGO2023 is optional. Additionally, imaging and pathology according to the FIGO system in cervix cancer is optional. So these data are looking at molecular classification for early stage disease, stage one and two. This, when molecular testing was performed, these were statistically significant. It was not statistically significant when molecular testing was not done. The new FIGO staging system holds immediate clinical relevance. So molecular classification has strong prognostic value in high-grade stage patients with no adjuvant treatment. In this important study, when molecular classification was used, stage was no longer significant on multivariate analysis and LVSI remains significant. So what about lymph vascular space invasion? There's many such reports in the literature as this one. This is from Washington University in St. Louis. And the single greatest prognostic factor was lymph vascular space invasion. And in fact, in this cohort study out of Sweden in nearly a thousand patients, LVSI presented the strongest association with survival in staged patients. And lymph vascular space invasion remains significant for survival in staged patients with no lymph node metastases. So should histology be part of FIGO staging? As we all know, it's a major component of prognosis and defines treatment relevant subgroups. Treatments are vastly different between non-aggressive and aggressive histologies. In this study from Duke University from nearly 25 years ago, for those patients with serious and clear cell carcinomas, five-year survival in early stage disease was 56% compared to 92% for endometrioid grade one tumors. So is this system equitable? We do know that black patients are more likely than white patients to have P53 abnormal cancers. Unfavorable TCGA molecular subtypes are more common in black patients. I would suggest that more emphasis should be placed on accurate diagnosis and less on treatment. Adjuvant therapies are expensive. FIGO 2023 has already influenced some payers to support the optional molecular testing. So is this system cost-effective? There's two reports in the literature from the University of Pittsburgh indicating that in early stage disease and stage three disease, that molecular classification is in fact cost-effective according to the authors. So other critiques are that new staging systems break history with previous ones. We've had two FIGO endometrial staging system changes in 36 years. That is not a rapid rate of change. So there were perhaps some mistakes that may be over saying it. Some things that didn't happen optimally. The staging changes should be prospectively analyzed. We should solicit input from multiple societies around the world. We should have mandated written standard operating procedures or rules for our FIGO committee. Here are suggested treatment stages by stage. I won't go through each of this in the interest of time, but no staging system is perfect. I dare say FIGO 2023 is better for our patients. It provides improved discrimination. It identifies prognostic and treatment relevant subgroups. More patients will receive correct care. Some or many gynecologic patients are not treated by experts. They are treated by generalists. We hope this will encourage more precise pathology. It will permit more tailored therapy for patients. It is a bit more complicated for us and lots better for our patients. With that, I'd like to thank you and would be happy to try to address any questions if you have any.

radiotherapy

endometrial cancer

molecular profiling

FIGO staging system

patient outcomes

tailored therapy