S2 Episode 6: Sequencing Antibody-Drug Conjugates in Endometrial Cancer

Medscape

a day ago

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.
Ursula A. Matulonis, MD: Hello, I'm Dr Ursula Matulonis. Welcome to season two of the Medscape InDiscussion Endometrial Cancer podcast series. Today, we'll discuss antibody-drug conjugates (ADCs) and how they fit into the treatment of gynecologic (GYN) cancers, but also, importantly, we'll hear from our guest, Dr Leif Ellisen, about different aspects of ADCs that perhaps we haven't thought about in GYN cancer.
It really is an honor to have Dr Ellisen here today. He's a professor of medicine at Harvard Medical School and Program Director for Breast Medical Oncology. He's also the clinical director of breast and ovarian cancer genetics at the Massachusetts General Cancer Center. He's also the co-leader of the breast cancer program at the Dana-Farber Harvard Cancer Center. Welcome to the Medscape InDiscussion Endometrial Cancer podcast.
Leif W. Ellisen, MD, PhD: It's great to be with you, Dr Matulonis. I am looking forward to discussing this important and fascinating topic.
Matulonis: As you know, there has been an explosion of different ADCs in oncology in general. These drugs are now making a significant impact in GYN cancers with FDA approvals for mirvetuximab in ovarian cancer. The tumor agnostic accelerated approval for trastuzumab deruxtecan for HER2, 3+ advanced cancers, which really applies to all GYN cancers. Then, tisotumab for advanced cervical cancer.
There are also multiple phase 3 trials, as well as earlier phase 1 and 2 trials of ADCs in GYN cancers and in endometrial cancer specifically. These ADCs are targeting several different payloads, such as HER2, TROP2, and folate receptor alpha (FRα), CDH6. Some have the payload of topoisomerase one (TOP1), whereas some have anti-mitotic payloads. In phase 1 trials of TOP1 payloads, we are already seeing that previous exposure to previous TOP1 payloads are not allowed. We are eventually going to have to think about how to rationally sequence different ADCs and really understand the mechanisms of resistance and response to each of these drugs that we're going to be using.
You and your team at Mass General are leaders in ADCs. You recently published a paper in Clinical Cancer Research identifying the emergence of TOP1 mutation as a resistance mechanism to TOP1 payload ADCs. I really want to focus on resistance mechanisms and ask you what you think are the most common and impactful resistance mechanisms that cancers developed to ADCs and some of the challenges to using ADCs, not just once but repetitively.
Ellisen: Well, thank you again, Ursula. These are very important questions, and I think one thing that sets the stage for the discussion is that ADCs are quite complex molecules. When you think about an antibody binding on the cell surface, you think about delivering some kind of payload that's bioactive, and you think about how the antibody is linked to the payload, which determines how it's delivered. So, multiple and complex resistance mechanisms develop. One way to divide them up would be de novo resistance vs acquired resistance. Some of the work that we're doing in de novo resistance suggests that you can see a lot of intrinsic chemo resistance of the tumor cells, just like you would with standard chemotherapy as a mechanism. In some cases, low expression of the ADC target itself, although that doesn't seem to be a major contributor.
More interestingly, we're learning that in de novo resistance, many features of the tumor microenvironment that may affect both the biology of the tumor cells and the penetration of the ADCs into the tumor are important and may confer resistance.
When we look at the acquired resistance setting, we don't see a lot of loss of the target per se. However, we and others have demonstrated where loss or mutation of the target likely mediated acquired resistance. More commonly, we see some of the things that you might see with chemotherapy, such as activation of cell survival pathways in the tumor, upregulation of drug efflux transporters, and mechanisms of that kind.
More interestingly, though, and specific to the ADCs, is a specific mutation of the target of the payload. For example, as you mentioned, TOP1. When administering systemic TOP1 inhibitors, you don't really see the emergence of these mutations, but we are seeing them now in the setting of the ADCs. This may have to do with the fact that we're delivering much higher doses. It's more potent, and the tumor is pressured to have an on-target resistance mechanism.
Matulonis: That is very interesting, and as we step back as clinicians to try to navigate both the de novo and the acquired mechanisms of resistance, what are some of the strategies that you're pursuing — both from a laboratory standpoint, but also from a practical, clinical standpoint? How to overcome these ADC resistance mechanisms?
Ellisen: My bias since we started working on ADCs a number of years ago was that, because of the complexity of these molecules, a lot of the work that we needed to do was going to be analysis of patient samples themselves to really understand among all the possible ways that resistance could occur, which are the ones that really occur importantly in vivo.
For example, we've launched a number of preoperative trials in breast cancer, where this is a common method for treatment now, where we can profile the tumor before treatment and after treatment. We can identify, through things like single-cell analysis, these tumor phenotypes in tumor states that are chemo-resistant. Then, we can do things like spatial analysis to identify what we call cellular neighborhoods or ecosystems within the tumor that govern how the tumor cells themselves behave. Are they hypoxic and chemo-resistant? Can the drug make its way into the tumor, governed by things like angiogenesis, which may preclude effective delivery of the ADC?
And then, when we switch to thinking about how we can study mechanisms of acquired resistance. One very useful tool has been the increasing use of circulating tumor DNA analysis, that is, blood-based analysis, because we can do repetitive sampling in the setting of progression and really get an overview of the genetic evolution that gives rise to resistance. That was how we identified these TOP1 mutations.
But it's also the case that we need to look at tissue; that can be challenging in the metastatic setting. And so, what we've done here is develop a number of patient-derived resistant tumor models that we can then take back to the lab, propagate them, understand resistance mechanisms, and then do things like really complex and comprehensive screenings to ask in an unbiased way, 'How might that resistance state be overcome?' I really think, fundamentally, it's going to take this kind of coordination between the laboratory and the translational work that is going to give us the best insights that will be most relevant to patients.
Matulonis: Before we move on to biomarkers, given that you've identified — these TOP1 mutations and changes — I want to ask you the question: Do you think it's possible to overcome these different resistance mechanisms?
Ellisen: TOP1 is a particularly fascinating one because, as we know, TOP1 itself is so essential to the fundamental processes in the cell transcription replication. What happens with a TOP1 mutation is the enzymatic activity of TOP1 is not lost, but it's substantially modified. That modification of the activity requires the cell to adapt in very complex ways, and we're just beginning to understand it. However, the adaptation that results from, for example, a TOP1 mutation is going to create additional vulnerabilities in that tumor cell. We're doing a lot of work, and others are as well, to try to understand those evoked vulnerabilities that occur after such resistance mutations that might lead to ways to overcome that particular state.
Matulonis: That's really fascinating. I am going to move on to biomarkers and what we're seeing in GYN. We're not as far along as in some other cancers. We have a lot of work to do, but we really need to understand the heterogeneity of the biomarker expression in different patients and different sites of recurrence. And this certainly happens in endometrial cancer, where patients will sometimes have peritoneal disease resulting in ascites. They may have lung metastases, or they may have nodal metastases. How do you think about the biomarker changing over time within a patient and their different metastatic sites?
Ellisen: This is the big challenge of advanced and metastatic cancer as a whole. Just to give you an idea of what we're facing, a number of years ago, we published a paper describing an individual breast cancer patient who had received a sacituzumab govitecan. They had a really fantastic response to the drug but progressed at multiple sites. We were able to do an individual sampling of those sites. In one subset of metastatic sites, there was a mutation in TROP2, which was the target of the ADC, which caused loss of expression and resistance to those particular sites. In another subset of metastases, there was a mutation of TOP1 resistance to the payload.
We hope that this kind of phenomenon, you know, will not be occurring in all of the patients, and I do believe that to the extent that we can make ADC therapy more effective and more potent, we have a better ability to combat this heterogeneity that's intrinsic to the cancer. That might be through a better selection of ADCs. That might be through combinations that we might talk about momentarily, to really try to expand the way that the body responds and expand the ability to target the tumor, in particular through combinations such as immunotherapy, where you can then activate the immune system to combat the heterogeneity of the metastatic tumor. But there's no doubt that it's a daunting challenge for the whole field.
Matulonis: I just wonder how we use ADCs in a patient's treatment and where that patient is at the start of their cancer journey vs later on. Obviously, resistance mechanisms are likely to increase as the tumor changes over time. One would think that the ADC that you use first has the best chance of response. Should we be thinking strategically about how to position the ADC sequencing?
Ellisen: That's exactly right. There's no doubt that we have to be smarter about that. For example, coming back to these TOP1 mutations, it seemed to be quite clear from the relatively limited but quite compelling clinical data so far that, if you have one of these mutations, you're very unlikely to respond to a second ADC that harbors a TOP1 inhibitor payload. The more we can understand resistance, the more we can test for resistance at the time of progression on one ADC and the more likely we are to be able to make a better choice in the second ADC.
ADCs are being tested in so many different settings, and there are going to be numerous ADC options in the future. We'll have many ADCs to choose from and making the smart choice hopefully will be informed by knowledge about specific mechanisms that arise in individual patients.
Matulonis: That totally makes sense. What do you see as the differences and differentiators regarding the other parts of the ADC? You mentioned this before, but specifically the linker and then the type of the antibody.
Ellisen: These are really important considerations and speak to the complexity of these ADCs. Linking the antibody to the payload is important, because it determines how much of the payload is released outside the tumor cell. If you take two examples on the extremes, sacituzumab govitecan has a pH labile linker, so it's quite labile and has a short half-life. You have a lot of payload, SN-38, a TOP1 inhibitor released into the circulation, and therefore, sacituzumab has a lot of chemotherapy-like toxicities.
On the other end of the spectrum, you have an ADC, like trastuzumab emtansine, which has quite a stable linker that requires intracellular processing and lysosomal processing to release the payload. And so there, you can see how a very labile vs very stable linker controls the side effect profile to a large degree of these two drugs. It turns out that features of the antibody are also quite important. One of the particular ones that's coming into focus is the issue of once the antibody binds, whether it stays on the cell surface or whether it's rapidly internalized. With sacituzumab govitecan you have rapid internalization and no ability for the antibodies to stay on the surface. We know that antibodies that coat the surface of tumor cells attract immune molecules that can induce things like antibody-dependent, cellular cytotoxicity, and phagocytosis that can engulf and kill tumor cells.
So you have this balance between rapid internalization, which can deliver more payload, and this staying on the cell surface, which can recruit the immune system. There's no perfect formula or a secret sauce for what everybody believes is the best way to design your antibody. These are things that are being tested systematically to see whether there really is a best practice or best design along those lines.
Matulonis: It's really interesting, and there is a lot to learn about all these different ADCs and how unique they are. You just never know until you start testing them in patients and see that they're safe and effective.
This is really exemplified in breast cancer, but there are some ADCs that really don't require minimum levels of biomarkers, as we've seen with HER2 and trastuzumab deruxtecan, TROP2, as you've mentioned, CDH6 and in certain circumstances, FRα as examples. Is the efficacy of these ADCs really about the presence of just tiny, low levels despite there being 0 or 1+? Or is there some other mechanism of action that these ADCs are imparting against the cancer cell? These are questions really around where the field is headed, because clearly, not requiring tissue, not requiring a biomarker, is much easier for clinicians and patients than requiring perhaps new biopsies, sending testing out for biomarker assessment, et cetera. What are your thoughts about that?
Ellisen: This is a really fascinating aspect of ADCs. Currently, there's a debate in the field of ADCs about how much of the ADC activity is really related to this canonical ability to bind to the target and deliver into the tumor cell and how much of ADC activity relates to a slow and persistent payload release in the microenvironment, kind of what we might call a metronomic release of payload, in the tumor microenvironment that is going to be less target dependent or particularly less dependent on very high levels of target. I would say that, undoubtedly, most ADCs exhibit some aspect of both of these. You see this in clinical data where, at very high levels, there is a correlation with target expression and duration of response. But at lower levels, there's not a great correlation. So, both of those exist depending on the design of the ADC and the target itself. Which is better, high specificity and target dependence vs not screening? In the short term, it's certainly convenient that we have ADCs that don't require prescreening. But in the long term, particularly as we have more ADCs becoming available, it would be quite valuable to identify patients who are highly likely to benefit from a given ADC based on target expression. In other words, as more ADCs become available, it would be valuable to have the means to select the best ADC among many for a given patient, either by the target expression or some other biomarker, the personalized, patient-selected cancer therapy that we all like to give, ultimately with a lot of ADCs available. I think that's what we'd like to do.
Matulonis: As the ADC is nearing the cancer cell, what happens, either that internalization or perhaps the attraction of other molecules to, or other cells to that interaction, really would depend upon perhaps the target. So, maybe B7-H4 may have a more immune response vs also where the cancer is. Is it located nodally, or is it somewhere else where maybe there are not as many immune cells, or the immune cells just come regardless of where the cancer cell is? Do you have any thoughts about that?
Ellisen: It's going to be important. I agree with you, and I think it's going to be particularly important as we think about some of these more novel combinations. For example, payloads that may activate the immune system combinations with immune checkpoint inhibitors. Site-specific tumor immune microenvironment and context are going to be even more important drivers of the responses that we see.
Matulonis: Another one of my questions to you is what types of combinations of either ADC/ADC, or ADCs plus something else are you most excited about?
Ellisen: It's a very exciting time because I think we have so many opportunities to combine ADCs with, as you said, with other ADCs potentially, but with other molecules and therapies and even established therapies that are mechanism-based and potentially synergistic and even potentially combinations which might have been tried in the past but didn't work because of prohibitive toxicity. One example is a clinical trial that we did with sacituzumab govitecan together with a PARP inhibitor. It was known that TOP1 inhibitors and PARP inhibitors were synergistic systemically. However, the problem was that there wasn't a great therapeutic window, because it was highly toxic to hematopoietic and other cells. But we reasoned that the specific delivery to the tumor cells might give more specificity and widen the therapeutic window. It turned out that was only partially true, and in fact, we had to do the sequencing of the true drugs, the sacituzumab and the PARP inhibitor, given sequentially, not at the same time. But still, the idea that this kind of combination, which, if you deliver it just systemic drugs, totally prohibitive, could be feasible. We were able to complete the clinical trial successfully with an ADC-based delivery. As I mentioned earlier, I do think that in the near term, there were already many clinical trials going on combining ADCs with immune checkpoint blockade.
I think this is particularly exciting because we know that one of the major features that can stimulate the immune microenvironment is immunogenic cell death. We do know that ADCs are really effective in many cases at inducing this kind of cell death and potent tumor responses, which is a great stimulation to the immune system. Combining these with checkpoint inhibitors in the right context is a very exciting possibility, at least in context. And we're gonna be having more data emerging about that in the coming months and years.
Matulonis: That's really exciting. I want to move to ADC payloads and focus on two questions. Obviously, the ADCs that we're using mostly clinically are either comprised of TOP1 inhibitors, or anti-mitotic agents. What do you think are some of the future payloads that are unique and different from these two? Much is discussed about drug antibody ratios (DARs). Everyone thinks higher is going to be better, but sometimes higher is toxic. What are your thoughts about the sweet spots of the DAR? Or does that not matter as much?
Ellisen: Starting with payloads, I think we shouldn't overlook the ability to just try new chemotherapy payloads. One of the reasons I believe that in breast cancer the approved ADCs have been effective — these are all bearing TOP1 inhibitors, by and large — is because we've never used TOP1 inhibitors in breast cancer. Typically, no breast cancer has ever seen that drug before. Then we come in with an ADC and a totally new chemotherapy that can be successful — so leveraging on that paradigm and linking the ADCs to other chemotherapies, which we know can be successful in particular contexts, is certainly one thing that's being tried and that's reasonable. Others that I think are very exciting are things like immunomodulatory agents and immune agonists: for example, delivering those in a specific way. We know that, systemically, immune system agonists have been tried, but in some cases, they have prohibitive toxicity. When delivered selectively, they might be quite successful. So, conceptually, it is very interesting. And then other classes of target inhibitors that we know have efficacy in the right setting. But again, it might have been shown to be relatively toxic systemically, such as various DNA repair inhibitors, DNA damage checkpoint inhibitors, all these things, given our ability to deliver them in a more specific way with ADCs, I think now can be considered and reconsidered. That's a very exciting potential, I'd say.
Matulonis: It's really quite impressive to think about this field moving forward.
Ellisen: I will make a comment on the DAR because I think — that's a drug-to-antibody ratio — I think that's another fascinating one. The field has not settled on an answer. Because when you think about it, if you increase the drug to antibody ratio, what it does is mean that you're actually infusing fewer molecules of antibody, right? Because if you have more drug per antibody, you're infusing fewer molecules of antibody for a given toxicity that can be tolerated. And what that means is that there's a consideration called the saturation front where, if you have a lot of targets in a certain part of the tumor, if you don't have enough antibody molecules there, they get saturated. And so you never get the tumor penetration to actually kill all of the tumor, and that leads to a generation of resistance. On the other hand, with lower DAR, and more antibody, you can overcome that saturation front. You can get into the tumor, but maybe not as potent per ADC molecule. So, there is a sweet spot there. Whether it's gonna depend on the individual payload and target or the individual tumor type, I think, remains to be seen. But, there's gonna be a lot of investigation of that area, and it's going to be a really fascinating one to watch.
Matulonis: Tell us about some of the ADC research that you are working on right now that you're really excited about.
Ellisen: The real excitement in ADCs comes from their complexity. It's the opportunity to really merge what we're seeing in patients about how response and resistance are governed with what we can learn in a systematic way in the laboratory. When we think about de novo resistance, neoadjuvant studies understanding evolution in microenvironmental determinants that we typically don't think of when we're thinking of drug resistance. We think of the tumor cell but thinking about how the individual cells are arranged into a neighborhood within the cell. Maybe it is not at the top of our list for determining whether we're gonna give a drug or not, but I think it may be in the setting of ADCs.
Secondly, this idea is that we can really think about mechanistic and rational combinations with ADCs that might not have been possible before for toxicities or other reasons. Then, we can allow the selective delivery that we can get with ADCs to make this possible and really get potent and synergistic tumor cell killing in the way that we always hope to do.
Matulonis: What an exciting field. Congratulations to you, Leif, for all you and your team have done. You're pioneers in this field. Once again, GYN cancers are learning from breast cancer. And you're leading the way on this. Thank you so much for being here today. Today, we've talked to Dr Ellisen about ADC resistance mechanisms, ADC sequencing, and the different components of the ADC, including what's exciting in the field and novel payloads. Thank you so much for tuning in. Take a moment to download the Medscape app to listen and subscribe to this podcast series on endometrial cancer. This is Dr Ursula Matulonis for the Medscape InDiscussion Endometrial Cancer podcast.
Listen to additional seasons of this podcast.
Endometrial Carcinoma
The Clinical Landscape of Antibody-Drug Conjugates in Endometrial Cancer
FDA Grants Accelerated Approval to Fam-Trastuzumab Deruxtecan-nxki for Unresectable or Metastatic HER2-Positive Solid Tumors
Mechanisms of Resistance to Antibody-Drug Conjugates
TOP1 Mutations and Cross-Resistance to Antibody-Drug Conjugates in Patients With Metastatic Breast Cancer
Single-Cell Analysis Technologies for Cancer Research: From Tumor-Specific Single Cell Discovery to Cancer Therapy
Spatial Transcriptomics for Tumor Heterogeneity Analysis
Exploring Circulating Tumor DNA (CtDNA) and Its Role in Early Detection of Cancer: A Systematic Review
Current and Emerging Prognostic Biomarkers in Endometrial Cancer
Parallel Genomic Alterations of Antigen and Payload Targets Mediate Polyclonal Acquired Clinical Resistance to Sacituzumab Govitecan in Triple-Negative Breast Cancer
Antibody-Drug Conjugate Sacituzumab Govitecan Enables a Sequential TOP1/PARP Inhibitor Therapy Strategy in Patients With Breast Cancer
Antibody-Drug Conjugates as Targeted Therapy for Treating Gynecologic Cancers: Update 2025