Read the Full Video Transcript

Alicia Morgans: Hi, and welcome to UroToday and our online medical education program, Beyond Androgen Blockade to New Pathways and Novel Treatments in Metastatic Hormone-Sensitive Prostate Cancer and Metastatic Castration-Resistant Prostate Cancer. My name is Alicia Morgans and I have the honor to moderate today's discussion following a presentation by Dr. Oliver Sartor. His topic is tumor heterogeneity, selective pressure, why we need new targets. Thank you so much, Dr. Sartor.

Oliver Sartor: Thank you, Alicia. This is really an interesting topic and I enjoy the opportunity to be here and present today. I'm Oliver Sartor. I'm a disease group leader of GU cancers at Mayo Clinic and the director of Radiopharmaceutical Trials at Mayo.

Good to see you. What we're going to do is to begin looking at tumor heterogeneity with a slide that I think may be very familiar. This is from the Stand Up To Cancer, the very famous cell article back in 2015 that looked at individual patients and the heterogeneity and genomic findings after careful sequencing. This very famous slide is the one that shows that, sure, we've got a lot of AR, we've got a lot of TP53, we've got a lot of PTEN, but then there's all other mutations. This heterogeneity is really well documented now, and it presents both an opportunity and a challenge for personalized therapeutics.

Now there are a lot of different ways to look at heterogeneity. I'm just going to look at it quickly. First of all, you can look at circulating tumor DNA. There are about 100 different articles I could have chosen. I chose this one by Alex Wyatt, particularly nice, very recently published in Nature. What I'll say is that tumor heterogeneity is a fact and there is clonal evolution with therapies that are beginning to work and then to fail. At progression, you're going to have a different look than when you started off treating the patient, and it's one of the things that's very important. Whatever you start with is not what you finish with. There is this clonal evolution. There's more than just the typical copy number variations, mutations, infusions. There are now a variety of methylation markers that we're looking at, and those are going to be changing too. Lots that we can learn from circulating tumor DNA.

Now there are other ways to look at heterogeneity. This was presented by Howard Scher a while back. This is an interesting look at the circulating tumor cell phenotypes that are using facial recognition and mathematical grouping. I'll simply say, in this particular presentation, Howard talked about 15 different CTC phenotypes. That's something to keep in mind. There is a lot of heterogeneity out there. Or we can look at heterogeneity by PSMA and FDG PET. Or we could even use other PETs. We could use choline PET, we could use the fluciclovine PET. We could use AR-targeted PET. But the bottom line is, depending on how you look at it, you can come up with some different answers. Here with the FDG, you see a bunch of liver mets, and on the PSMA, you don't see nearly as much in the liver.

You understage the patient by looking at PSMA and this heterogeneity is very important. Now if you want to do heterogeneity at the immunohistochemistry level, you can. This is a presentation from Johann de Bono's group. I really liked it. What they did is to look carefully at a number of tumor biopsies. These were castrate-resistant patients. What you could see is some of these cells were strongly PSMA PET positive and other ones right next door didn't even express any PSMA. You see the two cell lines, PC3, which is negative, LNCaP, which is positive. But when you look at the tumor cells in vivo from the patients, these are very, very distinct. How do we document, how do we follow this evolving heterogeneity? We know that the tumors are evolving according to the selective pressures we apply. Selective pressures, that's the Darwinian term, but we call them cancer therapies.

Our cancer therapies are inducing changes in the tumors we treat. We could use serial tumor biopsies. That's impractical. You do it in a research setting, it's hard to do in the clinic. Circulating tumor DNA as a whole recapitulates the underlying tumor. But there are some deficiencies, particularly with copy number losses, CTC profiling, or we can think about research-oriented single-cell genomics, exosomal profiling, also multi-PET tracers research or we're going to think ahead and think about multiomics and molecular imaging and putting it all together. I think eventually that's where we're going to end up. Now when we look at this clonal evolution, and this is from Himisha Beltran, very nice article from hers back in 2020. You started with the primary tumor, these metastasize, then you treat and you have some of the cells developed treatment resistance, and then you're likely to have some polyclonal seeding.

Then what you have these clonal selections. Yes, you do have some of the lesions from the founding clone, but then you develop new things along the way. Here she's highlighting neuroendocrine prostate cancer, which is something we know that can develop also highlighting epigenetic alterations, something we know that can develop. This is one paradigm. However, there's more to the story. If we look at the AR and neuroendocrine treatment paradigms and we've established these paradigms, I'm now going to tell you there's more to the story. This is from a really nice paper in Science. I know it's a blot, so forgive me for showing a blot. But what we see is that some of these cells or xenografts or model systems are going to be AR-positive. Some are going to be synaptophysin positive, and some of them are going to be double negative.

Synaptophysin typically goes along with neuroendocrine, AR goes along with the typical adenocarcinoma we're all familiar with. But guess what? There's more to the story. What these investigators did, and I thought it was very clever, was to be able to look at the molecular classification in more detail going beyond this typical AR-driven and the neuroendocrine, and to an interesting degree, they found what they called a stem cell tumor, SCL. That was present either about 30% or about 20% of the population depending on exactly which population they examined. Then they found another subtype that was quite distinct genomically. These were WNT-driven tumors. WNT pathway are going to be important for a subset 6.8% versus 4.8%. But the bottom line is, yes, there's adeno and that may be half of them, and then there's some neuroendocrines and then there's more to the story. That's very important.

Now these categorizations to me, they're too neat, they're too tidy. In truth, they're continuums, they're evolving mixtures themselves. Conal evolution is real and subject both to the underlying disease because not everybody starts out the same. Then the distinct selective pressures called treatments that we administer that causes additional change, and that clonal evolution is something we're going to need to learn how to track. Now if we're thinking about AR and we're thinking about neuroendocrine, I'm going to tell you we need to think about new targets too because this is a complex disease and we don't have any cures yet.
Maybe we've got DLL3 that looks interesting for the small cells, they have lots of AR inhibition, but these tumors continue to evolve and grow. New targets are going to be key. Now what I did is to create a possible list of targets other than PSMA that we're going to think about because PSMA, we know about PSMA lutetium and all that.

What about B7H3, HK2, STEAP1, prostate stem-cell antigen. I mentioned DLL3. Then coming back, not only the STEAP1, but STEAP2, GPC3, interesting new target, HER2, FAP, GRPR, that's the bombesin receptor, TROP2, NTSR1, SSTR2 is a lot out there on the cell surface. We're just scratching the beginnings of all these new targets. But I think they're going to be imageable with PET or SPECT and we're going to be able to bring targeted therapies to them and we may even have some more internal targets like PARP and AR that we can do even better than what we've done before. With even cell surface targets, I think a lot about radioligands, but there's also antibody-drug conjugates, CAR-Ts, bi-specifics in addition to the radioligands. There's a whole bunch of ways that we can approach this. This is evolving rapidly.

Now I do like the theranostics because we can diagnose and we can look at imaging and then we can treat pretty much with the same molecule. You have a cell surface target, a ligand, a linker, and an isotope. You can see it, you can treat it. I really do love it. Because I think that this is a potential path forward for a lot of these new targets, but we have to prove it. It's not enough for me to say it. We have to prove it. One of the reasons I think that the isotopes are particularly attractive is that you have an action beyond the area of deposition. If you're binding to one cell and radiating the adjacent cells in the microenvironment, I think that's better than just killing one cell. They're two big radioligands that we can talk about, the alphas and the betas.

The alphas are the two protons, two neutrons. The betas are basically an electron. But if you come down and you look at the range in tissue, we're talking about 40 to 100 microns for the alpha, 50 to 5,000 microns for the beta. What I want you to think about is depositing on the cell surface of one cell and radiating not just that cell, but the surrounding cells and the surrounding microenvironment. I think that's going to be a key element for overcoming heterogeneity. One of the things that I like about radiation, it can kill them all. Particularly, with an alpha particle. You get an alpha particle in the right spot, it doesn't really matter what the underlying genomic alterations are, you're going to be able to kill that cell, which is one of the reasons that I have spent a lot of my recent career focusing on the radioligands.

With that, I'm going to put it back to Alicia and just say thank you for the opportunity to be here today.

Alicia Morgans: Well, thank you so much for that, Dr. Sartor. Certainly such wonderful data, and thank you for bringing forth some of the more recent data around disease heterogeneity. I wonder as you think about radioligand therapy and other novel approaches, I think one challenge that we've had in mCRPC in particular is that we don't always switch mechanism of action. Can you tell me how important it is to think through that? You've named a lot of really important data that suggests that we've got disease heterogeneity, but we tend to use AR signaling inhibitors in sequence sometimes. Where are we going wrong with that approach?

Oliver Sartor: Alicia, I think that we've learned to enjoy treating patients through AR-targeted pathway because they've been able to make a big difference. Whether or not you're looking at abiraterone, enzalutamide, darolutamide, when you start treating with these drugs, you make a real difference in the survival of patients, particularly when you bring it early on. I think there's a tendency to go after the AR axis again and again. Is that wise? Is that smart? I'm not sure that it is. I think as we move forward, we do need to think about alternative mechanism of action. Yes, let's hit the AR, but maybe you want to hit the PSMA and then you want to maybe hit the microtubules. Now I think all the above might be better than one of the above.

Now it's one thing to talk about it, it's another thing to show it. We do have data to be able to show, say for rucaparib in patients with broad communications as compared to AR signaling that a PARP inhibitor is better. I think we have data to show that something like PSMA with Lutetium is better than an AR signaling agent. But we've got to be able to move not only these therapies forward but a whole variety of new targets as well. I love AR, I love PSMA, but let's keep working so that we can add maybe the B7H3s, maybe we can add the TROP2s, maybe we can add the GPC3s. There's a lot of work to do, and everybody who does clinical oncology appreciates the needs of our patients. These patients are not cured with AR, they're not cured with PSMA when they have advanced disease. We need to do more for them.

Alicia Morgans: Absolutely. I think one of the ways that we have been able to do that is sometimes through combination approaches. What are your thoughts around some of the combinations that are currently being tested and explored and sometimes even demonstrating benefit when single-agent approaches are not so successful?

Oliver Sartor: Great point. We have some very legitimate combinations now with the use of AR inhibition in parts for selected genomic defects. I was actually impressed with the enzalutamide talazoparib trial that looked at a variety of HRR mutations and showed what I thought was some pretty substantial benefit. We need to be thinking about combinations where the data indicate that they provide benefit, but we can't endorse combinations just because we think it's a good idea. In the end, we have to show the data and show it convincingly because combinations can bring more in the way of adverse events and more expense. We've got to do it smart, not just in everybody.

Alicia Morgans: Well, I guess along those lines, you've certainly seen firsthand how sometimes when we think we're doing good and we're helping patients, we're not always doing it. So clinical trials are really required to demonstrate not only that efficacy that we think is going to happen with a synergistic approach or a combination approach but also the safety. I wonder as you think about some of these agents, particularly oral approaches, things like PARPs and other targeted agents that may be oral, are these just another pill, or are these agents that can come with toxicities, not that are prohibitive to use, but that we do need to understand and ensure we're supporting our patients through?

Oliver Sartor: Absolutely, Alicia. I think all of our therapies have toxicities, and some may have more than we originally anticipate. I'm going to talk about the AR-targeted agents. AR inhibition over a prolonged period of time basically leads to an acceleration in human frailty. People lose their muscles, they lose their mobility, they lose their get-up-and-go. This is important. Under the ideal circumstances, we need to be very sensitive to what our drugs do to patients in the adverse reactions they experience, not just the benefits that we report at the conferences.

Alicia Morgans: Absolutely. But also so important I think that as we continue to expand our options, we also I think expand our ability to support patients and we'll do our best to continue to keep people safe. Because ultimately, we want the best outcomes in terms of disease control and quality of life. That's how our patients really win. As we wrap up, I would love to hear your summary of this and how you think about attacking disease heterogeneity in prostate cancer.

Oliver Sartor: Alicia, first of all, we need to be able to monitor better the disease that we treat. I think at times we continue ineffective therapies longer than they should be continued because we're not on top of the evolution that is inevitable when we use these newer therapies. I think as we move forward, we need to monitor better and then we can intervene better. But it comes hand in hand. We can't do one without the other. We can't intervene better without monitoring better. If we monitor better and don't intervene successfully, then what have we done? It takes a combination and these biomarkers are really critical. It's going to take a lot of us working together, and not just you and me, but big teams to be able to work and create the progress that these patients need.

Alicia Morgans: Well, thank you for that. Thank you for sharing all of these thoughts about novel targets, new ways we can really try to attack the heterogeneity of prostate cancer whether we see a stem cell phenotype, a WNT phenotype, or our traditional AR signaling-driven phenotype or even our neuroendocrine phenotype, there is going to be more to come and you have really helped set the stage so that we have a sense of where we're going. Thank you so much for your time and for your expertise.

Oliver Sartor: Thank you, Alicia. Pleasure to be here today.