Lawrence Fong: Thank you, Oliver. Really, a pleasure to be here and looking forward to our talk today.
Oliver Sartor: You published a manuscript that really caught my eye in Nature. So first of all, just to give our audience a background, immunotherapy, quite frankly in prostate cancer has been a bit of a bust, outside of the MSI high, mismatch repair division, high TMB subset, of which is probably only three to 5%. So, immunotherapy has been disappointing in prostate cancer because these things are cold tumors. But Larry, you're doing some work that might give some insight as to why the immunotherapies are not functioning as well as they should. And I'm going to have you start off by having you talk about tumor-associated macrophages, which I've paid no attention to, but you've been paying a lot of attention to. What are these TAMs doing and what subsets matter?
Lawrence Fong: Yeah, so as you know, Oliver, we've had multiple phase three clinical trials with immune checkpoint inhibitors, anti-PD-1 and anti-PD-L1 antibodies. And while those have been transformative in kidney cancer and bladder cancer, there's been very little activity in prostate cancer, to the point where all of these phase three trials have been negative. And as you alluded to, one of the things we think really suppress the activity of those agents is this immunosuppressive or cold tumor microenvironment. And people have known for a long time that these macrophages within tumors actually can be suppressive.
But one of the big challenges is actually figuring out which are the cell types that actually are mediating that suppression, and how can we actually target those. For a long time, people have thought about ways that we could suppress myeloid cells or these macrophages. And so, people have developed drugs to basically suppress all of their development and those are like the CSF-1 receptor or antibody antagonists, and those failed in clinical trials for immunotherapy. And I think that really reflects the complexity of these tumor-associated macrophages. Some of them could be good, but some of them could be bad. And if you actually try to eliminate all of them, you basically, on the whole, don't end up improving outcomes.And so, one of the things that we did was take a step back, because in terms of defining the macrophages within a tumor, we use very broad naming conventions. People may have heard of M1 versus M2, where people thought that M1 macrophages are good, M2 macrophages are bad, and we want to shift the balance between the two.
Well, the approach that we described in our recent Nature paper basically relies on a newer methodology that's called single cell analysis. And what that allows us to do is rather than trying to impose our classifications on what those macrophages are within the cancer, we actually now can assess every individual cell within a sample like a prostate tumor, and we can actually then see based upon its proteomic and transcriptional profile, what that cell actually does and how it functions. And that gives us a new way to actually now really inventory what are the macrophages that are within the tumors.
And what we did was look at biopsies from men with localized disease, with hormone sensitive metastatic disease, and then with castration resistant disease, and we identified a population of macrophages that not only are at high frequency within prostate cancer, ranging from 10 to 30% of the myeloid cells that are within the prostate tumor, we actually found that the same population actually increases as prostate cancer progresses. So that when you get to castration resistant disease, the frequency of these are actually much higher, and so that was the population that we ended up diving more deeply into in the paper.
Oliver Sartor: Well, it's interesting and you call them what? SPP1 high? And what is SPP1? I don't even know what that is.
Lawrence Fong: So, SPP1 is actually the gene that transcribes osteopontin.
Oliver Sartor: Okay.
Lawrence Fong: Yeah, and so osteopontin actually has a long history by itself of being an immunosuppressive protein, but it's expressed not just by tumor-associated macrophages, it can be expressed by tumor cells, it can be expressed by stroma, it can be expressed by a lot of different cell types. In this case, what we did was when we inventoried all those macrophages in the prostate tumors, we found that this population was quite unique in, that it actually expressed very high levels of these proteins. And I should point out that if you actually just looked globally at the macrophages and you didn't use these single cell techniques, it would be very hard to pick out this population because there are just so many different flavors of macrophages. But using that method, that allowed us to identify, oh, this is a very discrete population that's marked by this gene.
Oliver Sartor: Okay, so now you've got the SPP1 population, you got it isolated, it goes up in castrate resistance. It's maybe important, osteopontin, whatever. Well, what did these macrophages do in the tumor microenvironment that leads to immunosuppression? How does that work?
Lawrence Fong: Well, one of the things that we found was with this specific population, they actually express a gene program that's associated with hypoxia. And we know tumors are hypoxic environments, and we think that this is an adaptation of those macrophages to be able to persist and actually help reinforce that hypoxic environment.
And so when we think about hypoxia, we can think about some potential therapeutic targets that are there. Obviously, VEGF is one that's been known for a very long time. But one of the newer pathways that we know associates with hypoxia is the adenosine pathway. And what happens in tumors and in hypoxic conditions is you basically get an induction of genes that basically help break down ATP, which we all know is an energy source that's typically inside of a cell. Well, when you have a tumor, what happens is those tumor cells proliferate, they die, and they release ATP into the extracellular space. And ATP is typically immunostimulatory, but in a hypoxic environment, what happens is you get induction of these enzymes that break down ATP into free adenosine. And adenosine in the extracellular space is actually potently immunosuppressive and can hit different receptors, including the adenosine A2A receptor that we know suppresses a lot of different immune cells.
Oliver Sartor: Okay. Now, I'm sorry, but I lost you, Larry, because you went straight from the tumor suppressive macrophages to adenosine, but how are those two linked? Do they make the enzymes to break it down? What's the link between the macrophages and the adenosine?
Lawrence Fong: Absolutely. So, these macrophages that we identified that express these high levels of SPP1, as it turns out, they also express high levels of these proteins that break down ATP into adenosine. And so, CD73, CD39, those are proteins that break down ATP to free adenosine. And so, that was how we identified that there could be a link there, where not only were there these special myeloid cells that were there that have these hypoxic programs, they actually had the proteins that could break ATP into free adenosine.
Oliver Sartor: Got it. So, the TAMs are there in the hypoxic environment, they express a variety of ways to break down the ATP to create the free adenosine. Now, what does that adenosine do? You mentioned a receptor, what was it, the A2A receptor? What does that do?
Lawrence Fong: So, what that does is that receptor is actually broadly expressed on a lot of different immune cells. And so, what it can do is on a T-cell, it can suppress a T-cell so that it no longer works, and those T-cells can become exhausted. It also drives the CD4 helper T-cells to become regulatory T-cells, which also suppress immune responses. It can actually hit other macrophages within the tumor microenvironment and make those more suppressive. And so basically, if you can think about all the cell types that you think about taking out to allow a tumor to survive, adenosine can actually do that within the tumor microenvironment.
Oliver Sartor: Very interesting. So, now you've got a way to block the adenosine receptor that is implicated in this immune suppressive microenvironment. So, now I got the TAM, I got the activation, I got the adenosine, and I got the receptor, and now you're going to block the receptor. Tell me what happens next.
Lawrence Fong: Yeah, so as it turns out, there are a few companies that actually have drugs that can block this receptor. And so what we did was we started out in the lab and actually looked at whether or not you get these macrophages out of a tumor, you get T-cells out of a tumor, you put them together, and we confirmed that these macrophages actually can suppress the T-cells. Right? So then what we did was experiments in the lab where we add in this receptor antagonist and we could show that we now block the immune suppression that's mediated by these TAMs.
So then what we did was study it in vivo, and we always start out in mouse experiments, and we then showed that if you use these adenosine receptor antagonists, you can actually improve the efficacy of checkpoint inhibition in mouse models. But most importantly, what we also did was look in patients with castration-resistant disease. And so we treated patients with atezolizumab, an anti-PDL1 antibody, in combination with an adenosine antagonist called ciforadenant, and what we found is that when we use the combination, we could now see clinical responses with this combination in patients with castration-resistant disease.
Now... yeah?
Oliver Sartor: Larry, I now know why you're the Bezos Distinguished Professor, because I think in 1,000 years I could never put this one together, but you did. This is really nice.
Lawrence Fong: Well, thank you for that, Oliver. I think the important part is that when we did that trial, we saw that there were responses, but it wasn't a complete home run in that we got responses in about 20 to 30% of the patients. But what we found was when we biopsied those patients, the patients who responded actually had high levels of these macrophages, whereas patients who didn't respond did not have high levels of these macrophages.
And so, what we believe is going on now is in prostate cancer, there are going to be those patients who actually have a lot of these macrophages, and we need to target that mechanism a specific way. And then in other patients, there may be other operative mechanisms that we may need to target in another way. And so, I think this gives us a better sense of how to really develop immunotherapies for prostate cancer.
Oliver Sartor: All right, so, Larry, I'm thinking now what I want is an imaging agent to be able to find tumors that are enriched in this TAM microenvironment so I can have a predictive biomarker for the use of the adenosine antagonist. And so, now I can put that one together.
Lawrence Fong: Yeah, absolutely. I'd really love to work with you on that, Oliver. This will be terrific.
Oliver Sartor: Well, that'll be fun. Larry, so let's do a little wrap up and I'll wrap it up and you correct me. So, what you've identified beautifully in human beings is the existence of a tumor-associated macrophage population that is enriched in CRPC, that is immunosuppressive, particularly in the hypoxic microenvironment. By generating free adenosine and by blocking that adenosine receptor, you find that the activity of a PD1 antagonist is actually elevated. Hence, you've created a whole new model for creating immunoresponsive prostate cancer. Larry, fabulous, fabulous work. What do you have to add to that little summary?
Lawrence Fong: I think that really is a great summary. I think the importance is that while a lot of folks had sort of written prostate cancer off as a cold tumor that immunotherapy would not work in, I think what's really clear now is we now are figuring out ways that the immune system can reject prostate cancer, and this helps in our understanding of how we can really achieve that goal.
Oliver Sartor: Larry, it's been incredibly informative. Congratulations on beautiful work, and thank you for being on UroToday and helping to explain this to our listeners. I think this is really high-impact science, and appreciate all your contributions.
Lawrence Fong: All right, well, thank you for your interest, Oliver, and looking forward to the future developments in imaging agents to figure this all out. Thank you.
Oliver Sartor: Thank you, Larry.