How Can We Leverage Our Understanding of the Tumor Microenvironment to Improve Systemic Therapy? - Kenneth Pienta

March 25, 2021

In a comprehensive discussion between Charles Ryan and Kenneth Pienta, the focus is on the tumor microenvironment in prostate cancer. Dr. Pienta describes the tumor microenvironment as an ecological system, emphasizing the interactions between cancer cells and various other cell types. He argues that current treatments, including immunotherapies, must be understood in the context of this microenvironment. Dr. Pienta also introduces the concept of eco-evolutionary principles, explaining how they can guide adaptive therapy and treatment regimens. He stresses the need to break down silos between different areas of research, such as immuno-oncology and tumor microenvironment studies, to develop more effective therapies. The conversation explores the complexities of studying this environment and suggests that the field is only at the "tip of the iceberg," with much more to learn and apply in clinical settings.


Kenneth J. Pienta, MD, The Donald S. Coffey Professor of Urology, Professor of Oncology, and Professor of Pharmacology and Molecular Sciences, The Johns Hopkins Hospital, Baltimore, MD

Charles J. Ryan, MD, The B.J. Kennedy Chair in Clinical Medical Oncology at the University of Minnesota and Director of the Division of Hematology, Oncology, and Transplantation, Minneapolis, Minnesota

Read the Full Video Transcript

Charles Ryan: Hello, and welcome. Today we are talking again about many of the provocative questions in prostate cancer biology, and there is no better person to start with than Ken Pienta, who is a professor of oncology and urology at Johns Hopkins. He is the Donald S. Coffey professor of oncology and urology at Johns Hopkins, which is highly relevant to today's conversation. Dr. Pienta and I are going to talk about the tumor microenvironment and how we should be thinking about this in our clinical world and in our deepening understanding of tumor biology.

Thank you for joining us. Always a pleasure to talk to you, Ken.

Kenneth Pienta: It's great to be here.

Charles Ryan: So let me start by just asking the question: What is the tumor microenvironment?

Kenneth Pienta: Well, the tumor microenvironment is really something that we should define ecologically, and when you think about an ecosystem it's all the abiotic and biotic pieces of the environment that are interacting with all the species in that ecosystem. So when we apply that to the tumor, what we are really talking about is all of the different parts of the body that the cancer cell touches, which includes the extracellular matrix and the blood vessels and all the pieces of the tumor or of the body that aren't cellular, but also all the different cell types that are interacting with those cancer cells at any given time. I wrote a paper way back in 2008 where I talked about how many cells a prostate cancer cell is interacting with at any given time in bone metastasis, and it's easily 27 and it's probably 40.

Charles Ryan: Is that 27 to 40 different-

Kenneth Pienta: Cell types.

Charles Ryan: Types of cells? Different types of cells.

Kenneth Pienta: Right. So if you think about a cancer metastasis in the bone, you're talking about that cancer cell interacting with osteoblasts, osteoclasts, osteocytes, endothelial cells, T cells, B cells, all the different STEM cells. What's really interesting, Chuck is that if you think about that and just pause for a second, okay. Now it's interacting with more than 40 different ones. So we have to think about the totality of that environment as we think about the tumor microenvironment, and that microenvironment lives within a patient and that patient in ecological terms is that cancer's biosphere. So just like we on earth live in our own little ecosystems of whatever cities we're living in, we live on the biosphere of earth, but for a cancer cell living in a tumor microenvironment, the patient is the biosphere. So what we are dealing with is really an entire system that is very interconnected.

One of my sort of personal missions right now is to remind everyone that when we talk about immunotherapy and immuno-oncology, that's traditionally been a T cell field, right? And yet, we're almost silo-ing immuno-oncology out from the tumor microenvironment, and that is not right. We have to think about how those T cells that are being affected by checkpoint inhibitors, etc., what other cells they are interacting with, and we forget that when we give chemotherapy or another targeted therapy, that we are not just affecting cancer cells, we're affecting all of the cells in that microenvironment, including T cells. So I think we need to remind ourselves that whatever we are doing in the IO field is totally wrapped up in the tumor microenvironment, and we should make a conscious effort to continue to break down those silos.

Charles Ryan: Do you think the treatments that we're using right now are in large part effective in part because they affect the tumor microenvironment, or are we just at sort of the tip of the iceberg where we have nice treatments that target cancer cells, but we have a whole lot to gain by targeting the microenvironment?

Kenneth Pienta: So I think that we are at the tip of the iceberg. I think there is a continued growing appreciation of how the microenvironment affects tumor growth and prostate cancer has actually, of all the cancers, solid tumors have actually been a leader in this field, but mainly because we started talking about osteoclast-inhibiting therapy and those kinds of therapies that affect bone metastases and how that relates to how do you complement your chemotherapies and your other therapies with those types of therapies, radium-223 and all these other therapies that are affecting the microenvironment, certainly of prostate cancer bone metastasis, very dramatically. And again, when we talk about IO we're talking about tumor microenvironment therapies. Those are not... right? We're trying to turn on a cell in the microenvironment. So I think if you look back to the 2000s when anti-angiogenesis therapy was sort of starting on the rise and we were all atwitter about avastin and drugs like that, and I think there was a tendency, "Oh yeah, anti-VEGF therapies are targeting the tumor microenvironment", and that has sort of tailed off of late. So now all we do is talk about IO, IO, IO.

But we have to continue to have the dialogue and make sure we are thinking about how all of these cells are working together to either support or inhibit cancer. For example, macrophages. Macrophages generally come in two flavors, right? M1s and Th1s and M2s, Th2-type things. M1s will mediate cell death and kill a cancer cell. M2s as we and many others have shown support the tumor microenvironment. For example, the main source of VEGF in the tumor microenvironment is M2 macrophages. The main source of matrix metalloproteinases to break down that environment to help the cancer escape are M2 macrophages. So we're doing trials where we're inhibiting M2 macrophages in combination with other therapies. We're going to see more and more of those things. I'm on a roll here, so I won't let you ask a question. I'm just going to keep going.

The other thing about the microenvironment, I think there is a real merging now in cancer biology between what's been classically considered the genomic folks, the cancer evolution people. As we look at the lineages of how mutations occur across time and space within a tumor, well, how much of that mutation rate is done because of a sort of intrinsic genetic instability versus how much of it is driven by the tumor ecosystem.

So evolution for us on this planet is all ecologically driven, right? You don't evolve because you are inherently genetically unstable. You evolve because you have to crush harder seeds or you have to run faster. So this idea of tumor evolution, cancer cell evolution, how much of that is driven by the fact that you are hypoxic where your pH is a little low all the time or you don't have enough nutrients?

We've been applying optimal foraging theory, as you know, to cancer biology. Well, why are we applying optimal foraging theory? Well, if you're a squirrel and you're sitting in the forest and you have enough acorns and you have a mate and you have a warm nest and there is no predator, you never leave that 20-foot patch, right? So why does a cancer cell leave the primary and metastasize? If it's got enough oxygen, enough nutrients, and there's no T cell predator around, why would it ever metastasize? Is metastasis truly a random event or is it driven by the cancer swamp? So I think there's a growing sort of understanding that we need to understand what the environment is not only doing to support cancer, but also to help it mutate more.

Charles Ryan: It's almost like a volitional attribution to cancer. Cancer wants to find a better environment. It's probably more than the cancer cells are randomly spewed out of the prostate and those that are able to find a better environment are those that are able to survive, perhaps. But what is it? Go ahead.

Kenneth Pienta: So you just used a word I can't let pass, which is volition.

Charles Ryan: Right.

Kenneth Pienta: Right? So you just-

Charles Ryan: Philosophy major.

Kenneth Pienta: Well, you anthropomorphized cancer, which is something we can't do and we don't need to do, right? Cancer doesn't think, but nor does bacteria, right?

Charles Ryan: Yeah. Well, that's what I said is, it's actually probably that the cells are randomly spewed out and those that are able to find an environment where they are able to survive are those that are able to thrive and proliferate.

Kenneth Pienta: Yeah, but I don't think it's random. I think cells leave because the environment they're in turns... For example, we know if you're hypoxic and turn on HIF1 alpha, you turn on epithelial mesenchymal transition programs that allow you to be more motile. And when you do that you can be more motile, but you're going to look for gradients, right? You're going to look for more oxygen. You're going to look for more nutrients. So I don't think it's a random spilling. I think it's a look for a cell saying, "I'm not getting enough of what I need here", and that creates a negative gradient so you look for other places to get it, and then they find the bloodstream or the lymphatics, and then they get to distant sites.

But I actually think you... The reason why I'm bristling a little bit about the randomness is that I think if you, again, looking at circulating tumor cells, I think it's pretty clear that 99.9% of circulating tumor cells never successfully metastasize. I think those are the cells that maybe you get a little bit of necrosis and are just sort of dropping into the circulation. Those random events are going to die. It's the cell that's been ecologically pushed to be an active emigrant that is going to make a difference here.

Charles Ryan: So I think a lot of people are listening to this and like me, they are enthralled by this idea, and then they're wondering as I am how do you study this? We are so used to the laboratory model of growing cells in a dish. We have animal models which don't have the same microenvironments, and there's only so much we can do with human tissue in real-time. What is your sort of vector for testing and model for hypothesis testing in this?

Kenneth Pienta: Well, I'll be the first to admit, it's hard. So one of the things that we've done is continue to create more and more sophisticated extracellular environments to use in the lab. For example, many folks as you know have organs on a chip, and those are helpful. One of the devices we used for culture actually puts a gradient on the cells so that you can mimic what it's like to be, here's a blood vessel, here's a cell, here's a cell, so you can see how nutrients and oxygen and therapy all work along a gradient to mimic that sort of thing. I don't think in this setting... so more sophisticated in vitro models. I also don't think mouse models are bad here. I mean, they're not human, but our genomes are very close.

Charles Ryan: Yeah. You take a little bit from the CTC data, you take a little bit from genomics, you take a little bit from the mice, and I think you are synthesizing that into your comprehensive theory that you're talking about, basically.

Kenneth Pienta: Yeah, yeah.

Charles Ryan: So one of the questions that come to my mind is I see a lot of studies going on now where they are combining an anti-VEGF with immunotherapy, and I think those are interesting studies. I always wonder, is the concept not yet ready to go into the clinic? By the time we spend two years doing that clinical trial and scratching our heads with the results and saying, "Well, we maybe are seeing a signal here, maybe not." I don't know, but I mean we could spend the next 10 years talking about ecology and developing better models for it, but at some point, we do need to be in the clinic. What is your advice for those of us who designed clinical trials and how we could incorporate this thinking into our designs and our outcome measurements?

Kenneth Pienta: Yeah. So several things there. First of all, I think that we have to recognize that, again, all of these other inhibitors whether they be metabolism inhibitors, VEGF inhibitors, are affecting T cells as well as cancer cells and every other cell in the body. It's so fascinating that we are combining those now with sort of a new lens for that. But I think some eco-evolutionary principles have already made their way into clinical trials. I think as part of your series you need to get Robert Gatenby on because he is really the father of now adaptive therapy, which he started doing in prostate cancer, and The Moffitt has shown very well that you can use eco-evolutionary mathematical modeling to better dose patients with second-generation anti-androgens for much better results in patients and they are doing now much larger studies, and now those studies have spread to other diseases. So understanding how to do adaptive therapy is based in eco-evolutionary principles.

Then from our standpoint, again, thinking about therapies, for example, that what we call an evolutionary double bind. So in that setting, if you use one therapy, what you have to recognize is, what are you driving those surviving cells to look like. When you do that, then if you recognize what that phenotype is of those cells that survive, then you can direct the therapy against that. So for example just to put this in more specific terms, we all know that cytotoxic therapies kill proliferating cells. So what you are left with is why don't all those multi-cytotoxic regimens work against solid tumors, the drug, things that work for lymphoma things like that. Well, it's because all of them are based on anti-proliferative cell-cycling drugs. Well, therefore you select cells that are hibernating or selecting for cells that have stalled their cell cycle to avoid that type of therapy. So, therefore, there are therapies that maybe you could direct against non-dividing cells, but we've never designed a study like that, right?

Charles Ryan: Yes. There's a couple of responses to that. One is, there are many studies looking at alterations in the dosing schedule of cytotoxics to target not the tumor cells, but the T regulatory cells and other components, and that's a whole field that is making some headway outside of the context of prostate cancer in conjunction with immunotherapy and may merit further consideration. Then the other concept is something that's happening at your own institution, which is the idea of trying to bring cells into the cycle so that you can hit them, the use of testosterone, things like that. So I think maybe we are just at the beginning of that process of trying to manipulate. Maybe those are examples where we're manipulating the microenvironment, trying to scare the pheasants out of the bush so we can shoot them, so to speak.

Kenneth Pienta: That's a great analogy. Yeah.

Charles Ryan: That's a Minnesota analogy. Yeah.

Kenneth Pienta: Yeah. I like it. I like it a lot.

Charles Ryan: Anyway, it's always a pleasure to talk to you. I could listen to this for a long time. But your final thoughts on sort of the direction you want to see this field go both for the clinical researchers as well as the basic researchers.

Kenneth Pienta: Yeah. I think we've covered it, right? In the short term, I think we need to bring IO back into the microenvironment fields to make sure we merge those at every level of science through the clinic because again I think we are silo-ing too much as a field in the short term and that is something we can help. Then in the longer term, I do think we are going to see the rise of eco-evolutionary-driven therapies over the next few years as these ideas get more and more accepted.

Charles Ryan: That's really fascinating. I wish you the best, and it's always just great to listen to your thoughts on this topic. So thank you for joining us.

Kenneth Pienta: Thank you.

Charles Ryan: One of the biggest provocative questions we have in the field.

Kenneth Pienta: Thank you.