Polyaneuploid Cancer Cells Drive Metastasis and Resist Treatments in Prostate Cancer - Sarah Amend

December 17, 2024

Sarah Amend discusses research on polyaneuploid cancer cells (PACCs) and their role in metastasis, published in Molecular Cancer Research. Dr. Amend explains how cancer cells under stress enter an adaptive endocycling state, increasing dramatically in size and genomic content without dividing. The discussion highlights their findings that cells in the PACC state are preferentially found in circulation and at metastatic sites, particularly in bone marrow, and can remain dormant before potentially developing into metastatic tumors. Through various experimental models and human patient samples, Dr. Amend's team demonstrates that these cells have enhanced survival capabilities and metastatic potential. The conversation explores the therapeutic challenges of targeting these non-proliferative cells and ongoing research efforts to develop effective treatments using an evolutionary double bind approach, combining standard therapies with novel strategies specifically targeting endocycling cancer cells.

Biographies:

Sarah Amend, PhD, Associate Professor of Urology and Oncology, Johns Hopkins Medicine, Baltimore, MD

Andrea K. Miyahira, PhD, Director of Global Research & Scientific Communications, The Prostate Cancer Foundation


Read the Full Video Transcript

Andrea Miyahira: Hi, everyone. I'm Andrea at the Prostate Cancer Foundation. Here with me is Dr. Sarah Amend of Johns Hopkins University. She will discuss her paper, "Cells in the Polyaneuploid Cancer Cell State Are Pro-Metastatic," published in Molecular Cancer Research. Dr. Amend, thanks so much for joining us today.

Sarah Amend: Thanks very much. It's exciting to be able to share this work. Today, I'm really excited to share with you this story that we recently published about cells in the polyaneuploid cancer cell state and their nature of being pro-metastatic. So by way of brief background, our group and several others around the globe have recently shown that cells that survive stress will enter an adaptive endocycling state.

So what you can see here on the screen is a prostate cancer cell line, PC3. They're growing, they're proliferative, and you can see they're really relatively small. And these cells were subjected to high stress, in this case, cisplatin. But we see the same phenotype with multiple classes of therapy and other tumor microenvironment forces, like hypoxia. And what you can appreciate really clearly from this phase image is that they increase in cell size. And we've previously published that this is an increase of more than 70-fold in their overall cell size. We found that this is universal to all types of solid tumors.

We have shown in previous publications that this is the result of what's known as an endocycle. So we are all familiar with the mitotic proliferative cell cycle where cells undergo growth phase, S phase to replicate the DNA, and then divide following mitosis. But an endocycle is where cells skip mitosis entirely and go through progressive G phases, growth phases, and DNA doubling through S phase. This is a highly conserved developmental program, evolutionarily conserved program where cells remain in interphase, and they undergo multiple rounds of genome duplication and increased cell size. And this is where we find the name polyaneuploid; aneuploid cancer cells are all aneuploid. So polyploid of an aneuploid genome, polyaneuploid cancer cell state.

We published several years ago now that the presence of cells in the PACC state in a diagnostic specimen—so in this case, this is a prostate cancer tumor microarray—that the presence of cells in this PACC state predicts the risk of recurrence, in particular biochemical recurrence and metastasis in prostate cancer patients. This led us to ask the question of whether these cells, cells in the PACC state, are indeed those cells that are driving the metastatic phenotype.

I'm just going to show you a few movies here. This is based off of published in vitro work, where we have parental cells here on the left and cells in the PACC state on the right. What I hope you can see from the movies is just how much more these cells are moving.

I'm not going to go into the data here today, but you can see from this paper—and this is the same lead author, Mikaela Mallin, who's now a postdoc at the NIH, who did her PhD work with us—that she showed that in vitro these cells have increased motility; increased directional movement under stress, so that would be important for invasion; increased chemotaxis, so they can sense and respond; they have this hyper-elastic phenotype that actually allows them to invade while protecting that large nucleus that we saw earlier; and they have increased vimentin that is necessary for the motility phenotype.

Now this is really beautiful work where she showed that they had this increased metastatic potential in vitro. But this, of course, led us to the question of what's happening in vivo? Do we think that these are really the cells that are driving the metastatic cascade? To test this, she employed a number of different animal models to test different steps of the metastatic cascade where we have our primary tumor site, of course, in the prostate. These cells then have to intravasate into the circulation. Here it would be a circulating tumor cell, or CTC, go through the heart and lung to enter the arterial blood supply and travel to a distant site where they would then extravasate. Likely undergo a period of dormancy, and then survive and eventually seed a metastatic lesion.

To address this, she first needed to come up with a way to actually evaluate these rare cells—already circulating tumor cells and disseminated tumor cells are already rare. But then how do we evaluate a rare subset of an already rare cell population? And to do this, Mikaela worked up a novel flow cytometry-based method, where we can, from a single mouse, quantify, count as well as phenotype, understand the ploidy and size of cells from the blood and the bone marrow from a single mouse.

And she showed that this new strategy, using this flow-based panel, does allow us to have really high sensitivity. So we can count, we can detect and phenotype, even down in this 1 to 4 cell range of cells spiked into bone marrow, into blood, or just into media. So number of cells spiked and the numbers of cells recovered, and you can see a nice downward slope here. And we also show that our recovery rates are really quite high, sort of in the 75%-ish area of the number of cells we put in for a number of cells we get out. This gives us high confidence to actually be able to start to evaluate.

So in her first model, we wanted to understand how cells are leaving that primary tumor. And so we placed tumor on the back of a mouse using a subcutaneous model to understand metastatic dissemination. In this case, we have the parental cells, and then a PACC-injected population, like I showed you on that first slide. We allowed the tumors to grow, and at sacrifice, we collected circulating tumor cells and disseminated tumor cells from the bone and evaluated their number and ploidy.

Now, our hypothesis here was, of course, that tumors that were enriched for cells in the PACC state would have higher rates of circulating tumor cells and higher rates of disseminated tumor cells. The graph that I'm showing you here is the count of the number of CTCs, and then we have parental-injected, and then PACC-injected animals, about 15 to 60 animals per group. What you can see just by glancing at this, there's really not a difference in the number of cells, the CTCs in either group.

And indeed this also held with DTCs. There's no difference in the number of animals where we detected bone marrow DTCs either. Numbers are low here. You should know here that I'm just showing you one of many mouse experiments that we did. So have a look at the paper to see larger N.

To our surprise, though, when we started to actually phenotype what those circulating tumor cells and disseminated tumor cells looked like, what was their ploidy status, was that regardless of the starting tumor, whether they're enriched or parental, those smaller cells or the larger cells, that the majority of them are indeed cells in the PACC state. So in this case, the blue portion of the bars are cells that had greater than four ploidy. Those are cells in the PACC state, while red is a parental population. And you can see that the vast majority of circulating tumor cells that were detected were in the PACC state, as well as the disseminated tumor cells, DTCs, taken from the bone marrow, from multiple subcutaneous tumor models, as well as another model that I won't be showing you more about today, but the caudal artery model.

This is telling us that cells that are actually leaving the primary tumor and surviving circulation, surviving in that distant site, are in fact cells in the PACC state. Of course, this is a mouse model. What about the human situation? In this case, we looked at disseminated tumor cells from the bone marrow of patients with advanced prostate cancer. And to walk you through this panel, we have the Merge on the far left, we have DAPI to look at DNA, and then we have an Epi, epithelial channel, with a number of epithelial markers, cytokeratins, etc., and then vimentin, which we know is important for metastasis.

Across the top here from this one patient, this is a megakaryocyte. This is just to prove to you that even though this cell has high DAPI, a large DAPI signal saying it's polyploid—megakaryocytes are, of course, polyploid—they're negative in these other tumor channels. But this one patient, patient 6227, you can see very clearly has cells—have disseminated tumor cells from the bone marrow that are in the PACC state. What we're not showing you here is they also had cells that were not in the PACC state. So it's not just some artifact of this platform that we're using. And we showed this across a number of patients. So this isn't just an N of 1. But it's a vignette here that we are certainly seeing PACCs in a metastatic-relevant site in patients with prostate cancer.

The next question, though, is whether the cells are just disseminated and sitting there dormant but aren't actually growing up into a proliferative tumor—does it really matter? Or will those cells actually transition into a proliferative state? To test this, we used an intracardiac injection, in this case injecting cells directly into the bloodstream through the left ventricle. Again, using a similar model as we did before with parental cells versus those that are enriched for a PACC cell population. And then we set these animals on the shelf and just monitored their tumor burden over time.

Now, the first thing to pay attention to here, we have days across our x-axis and BLI, so tumor burden on our y. And in red, these are the control-injected animals. So we are able to correctly do this intracardiac model. Tumors do, in fact, grow out to a lethal tumor burden. So each of these animals were sacrificed at each of these separate endpoints.

In contrast, we have these PACC-injected animals in blue. And you can see that they kind of bounce around a little bit higher, a little bit lower for a really long time, up to about 135 days. And the reason why this is important is that these cells or these animals remain positive. That tells us that there's tumor there, but they don't necessarily outgrow, at least in the time frame that I'm showing you here. So, OK, that means that they can survive and lie dormant in the bone marrow or another metastatic site for a really long time.

The next question is whether they can actually grow and form clinically meaningful tumors. In this case, we have the mock-injected animals on the top of the panel, and then the PACC-injected animals down below. Remember, the parental animals are controls, had lethal tumor burden really quickly. And what you can appreciate here is that at the time point on the right of each one of these images, so 91 days, 98 days, about 100 days for these others, that you can see that while earlier in this time course, these were negative, so below a limit of detection. But we do, in fact, see BLI-positive tumors outgrow. And, in particular, you can see this one on the far right, where these tumors actually grew out to a lethal tumor burden before an experimental endpoint.

So taken altogether, it's been a bit of a whirlwind tour here. But what we showed was that cells in the PACC state are preferentially intravasating into the circulation as circulating tumor cells and extravasating at a metastatic site. Most of the data I showed you today was from the bone marrow. And that after a period of dormancy, a number of those cells, at least in some animals, can revert back to a proliferative phenotype and actually seed a metastatic lesion.

Now in our group, we're really interested in understanding cancer metastasis and why it is incurable. And we here at the Cancer Ecology Center apply evolutionary ecology paradigms to try and understand this. And from an eco-evo perspective, the phenotype of metastasis and the phenotype of resistance likely represent a shared adaptive strategy—this endocycling cancer cell state. We're doing more work on this now, and it's really the focus of our ongoing work to understand this endocycling cancer cell state with the goal of eventually applying an evolutionary double bind.

So like I showed you before, can we stress cells to enter this PACC state where here they are going to be pro-metastatic and also resistant to other forms of classic chemotherapy? Can we then come on board and specifically direct an anti-cancer therapy to these endocycling cancer cells with our goal of cure? And thanks to Prostate Cancer Foundation, we were recently awarded a Challenge Award to very specifically test this hypothesis and identify and apply these directed endocycling cell state therapies.

So with that, we just have to thank the team, including Ken Pienta, the Co-director of the Cancer Ecology Center, and really applaud Mikaela Mallin, who's the great grad student who led this work. And thanks, of course, to our generous funding sources, including the PCF and the DoD.

Andrea Miyahira: Thank you so much, Dr. Amend, for sharing all of this. So have you been able to evaluate human CTCs to look for PACCs in patients and how they relate to different disease stages and treatment responses?

Sarah Amend: Yeah, so this is actually other work that is under review that perhaps UroToday may be interested in a little bit down the line. But in collaboration with Jim Hicks at USC, also a PCF investigator, we have started to look at circulating tumor cells as well as disseminated tumor cells from the bone marrow of patients with prostate cancer. And in a very small sample set thus far, we found that the presence of PACCs in the bone marrow of patients is associated with worse prognosis. So we are working on being able to, number one, evaluate them in the circulation and in the bone marrow, and then we're starting to understand some of the biology.

Andrea Miyahira: OK, thank you. And why do you think CTCs or DTCs are often in the PACC state? Is there some kind of advantage?

Sarah Amend: Yeah, so we know from in vitro studies anyway that in this PACC state, the endocycling cancer cell state is adaptive. That means that cells can adopt it and then survive preferentially. And so while in this state, we know that cells survive in the primary tumor better, but then also that they can survive the stressors of the circulation better. So they can invade this—even biophysically, this larger cell size can be protective. They have hyper-elastic properties, so that means that they can both be stretched and compressed and retain the necessary integrity—to retain cell integrity as well as nuclear integrity.

We also have some hypotheses that there may be a selective advantage for having a highly whole-genome-doubled, high genomic content cell at the secondary site. So once cells can undergo that ploidy reduction and [inaudible] metastasis, you could imagine that coming from a high ploidy state to a low ploidy state may also be advantageous.

Andrea Miyahira: OK, thank you. And are PACCs good therapeutic targets to prevent metastasis and treatment resistance? Or how can targeting be achieved?

Sarah Amend: So one thing that's challenging about this cell state is that it's non-proliferative. It's sort of a novel form of cell dormancy, where cells are continuing to go through the cell cycle, but they're not dividing. And what's not often appreciated is that most of our standard-of-care therapies, cells are actually dying at that G2-M checkpoint. So it's when they enter cell division or when they try to undergo cytokinesis.

So our challenge here is how do we target a non-proliferative cell? We have a lot of strategies ongoing. We're thinking about how to do this by actually blocking the endocycling process itself. So how can we force cells into mitosis prematurely? We're also thinking about the stress response. We know that these cells are dealing with reactive oxygen species and are uniquely susceptible to ferroptosis, for example. And so we're thinking about novel strategies to try to kill these cells directly but using an evolutionary double bind approach where chemotherapy works, ADC works. We want to continue the standard-of-care therapies, but then add on a subsequent therapy to eliminate these endocycling cancers.

Andrea Miyahira: OK, thank you. And what are the next steps in your studies?

Sarah Amend: So right now we are really focusing on understanding how these cells are surviving in the endocycling cell state and really defining this novel form of dormancy where it's an active cell cycle but in the absence of cell division. And we think that by understanding this, we'll be able to target these cells more easily.

Andrea Miyahira: OK, well, thank you so much, Dr. Amend, for coming on and sharing this with us today.

Sarah Amend: Thanks so much for having me.