Read the Full Video Transcript

Andrea Miyahira: Hi, I'm Andrea Miyahira at the Prostate Cancer Foundation. Please welcome Dr. Matthew Dallos of Memorial Sloan Kettering Cancer Center, who will present the paper-- androgen deprivation therapy drives a distinct immune phenotype in localized prostate cancer, recently published in Clinical Cancer Research. Dr. Dallos, thanks for joining.

Matthew Dallos: Great to be with you, Andrea. All right. So it's great to be with you today to present some recent work from our group, looking at the ways that androgen deprivation remodels the immune system. And my name is Matthew Dallos, and I'm an assistant attending at Memorial Sloan Kettering.

So as a bit of background, we know that prostate cancers are these immunologically cold—prototypically immunologically cold—tumors. But as they develop, we know that immune tolerance is a key step in both their initiation and progression.

And prostate tumors actually escape the immune system through a number of different mechanisms, including decreased antigen presentation, increase in T cell exhaustion, and then infiltration of a number of different suppressive immune cell subsets. And this is all really well-known.

But we know that prostate tumors are not all the same. And they change immunologically over time and across the disease spectrum. There's a great variability in the immune infiltrate by both site of disease and also within different treatment contexts.

And so the question that we asked was, how does androgen deprivation, this backbone of prostate cancer therapy, remodel the immune TME? And really it's driven by this goal, collective goal, to figure out what is the optimal timing to intervene immunologically, and how do we do so?

And so we had access to a really unique cohort that was put together across two clinical trials, one that was done at Johns Hopkins and one that was done at MSK, where patients were treated with ADT using degarelix given as a single dose either 4, 7, or 14 days prior to radical prostatectomy, which gave us an opportunity to look deeply at the immunological changes within these tumors.

And so we performed targeted DNA sequencing using the impact test. We performed whole transcriptome sequencing, as well as multiplex immunofluorescence, and then compared these time points to a cohort of untreated matched controls.

And so the first thing that we observed was really to look broadly in an unbiased manner at gene set enrichment analysis using GO enrichment. And it was really striking to us that actually the top pathways that were modulated by androgen deprivation therapy were primarily immune, rather than some of the traditional ways that we think about how androgen deprivation therapy works, which is shown here on the left.

And then when we looked at selective pathways of immune activation, comparing untreated here at the top in red versus our ADT treated samples, we saw upregulation of a whole host of different immune pathways, including immune checkpoints, components of antigen presentation machinery, as well as markers of T cells and macrophages.

And when we looked at an enrichment score at baseline or in our untreated cohort, as well as at 4, 7, and 14 days, we saw an increase in immune activation following ADT. And so to further evaluate this at the protein level, we performed immunofluorescence. And we're able to validate that actually after androgen deprivation therapy, you see an influx in CD8 T cells. These CD8 T cells produce granzyme B, so they appear to be activated.

But we also see an increase in tumor-associated macrophages, primarily the M1 or pro-inflammatory type as indicated by CD68 positive macrophages. And then finally, we observed that actually the antigen presentation machinery increases in tumor cells following ADT.

And then lastly, one of the top down-regulated genes from our unbiased analysis was CD47. CD47 is the “don’t eat me” signal to macrophages. And we validated this at the protein level. And what you see here on the left is CD40 expression on tumor cells. And then on the right following ADT, a downregulation of this “don’t eat me” signal, perhaps suggesting that ADT can actually enhance macrophage killing of tumor cells.

So in summary, we think that ADT actually within days can turn prostate tumors from cold to hot. These immune changes after ADT are quite complex. We see changes in T cells, macrophages, and many others that I didn’t have a chance to present. And then ADT seems to change the tumor cells themselves to enhance antigen presentation. And that’s both by class I and class II presentation.

And so based on this work, we have a number of ongoing studies combining immunotherapy with ADT, particularly in the neoadjuvant setting where we think it may be most effective. And so with that, I’d like to acknowledge the folks who did this work, as well as my institution, MSK, and the Prostate Cancer Foundation for supporting this project.

Andrea Miyahira: Well, thank you so much, Dr. Dallos, for presenting this really interesting study. So did you evaluate expression of immunotherapy targets like PD-1, PD-L1, CTLA-4, and how do these findings intersect with prior multiple clinical trial failures of checkpoint inhibitors and other immunotherapy?

Matthew Dallos: I think it’s a great question. And we did look at expression of a number of different immune checkpoints. And we actually saw upregulation even within days of PD-1 and CTLA-4. And I think it does raise the question about how we and when we use immune checkpoint inhibitors. We know that they failed pretty much across the board in advanced disease and some early disease settings.

But I think we don’t fully understand whether or not there’s a role for combination therapy in the very earliest setting, really in this neoadjuvant setting where our goal is to cure these patients. And there may be a role for sort of in the very earliest—when we use ADT very early and combine that with immune checkpoint blockade.

But I don’t think immune checkpoint blockade alone is going to be the answer to overcoming this tolerance that really develops over the course of tumor initiation and progression.

Andrea Miyahira: Thanks. And based on these data, did you observe any new immunotherapy approaches that may be more promising or other possible I/O targets that showed up?

Matthew Dallos: Yeah, so I think one of the things that was so striking to us was that we’re seeing so much increase in antigen presentation machinery. We know that this is one of the key escape mechanisms and resistance mechanisms to a lot of these traditional immune checkpoint blockade or immune checkpoint inhibitors.

And so I think the idea of leveraging strategies that can further enhance immune cell priming and further engage antigen presentation would be an attractive approach. And there’s a number of ways that we’re doing that and ongoing trials, and I know a number of others are as well. So I think that’s an exciting area for the field to go from an immunologic perspective.

Andrea Miyahira: Thanks. And what do you think the biggest lessons are from this study, particularly in light of prostate cancer being largely considered to be immune cold?

Matthew Dallos: Yeah, I think it really teaches us that there’s a lot of heterogeneity, and we don’t fully understand this blanket idea that all prostate cancers are immunologically cold is probably not true. And disease setting matters, prior treatment matters, and the individual patients.

Even at baseline, we observed that a number of the untreated tumors actually showed a more active immune phenotype. And so there’s probably quite a bit of heterogeneity. That’s how to pick up those patients and develop a biomarker that we could potentially use to more carefully select patients for immunotherapy trials. I think it’s going to be an important future line of work.

Andrea Miyahira: Thanks. And what are your next steps? And do you have any translational plans?

Matthew Dallos: Yeah, so sort of as I mentioned, we have a few different neoadjuvant approaches that we’re taking based on this, I think in combination with hormonal therapy—so ADT plus different immunological strategies in the neoadjuvant setting. And I think the other question that our study didn’t really address is how long some of these immunological changes last.

So we observed—we looked at very early events at 4, 7, and 14 days. And at all of those, we saw an increase in the immune infiltrate. So what happens at three months and six months of maximum hormonal blockade in this setting I think are other areas that are certainly worth exploring, and we’re starting to do so and looking back at some other tissue that we’ve collected along the way.

But I think we’re learning a lot more about how the immune system is actually interacting with prostate cancers, and hopefully we’ll get to the point where we can exploit it more effectively.

Andrea Miyahira: Thank you for sharing this really interesting study with us.

Matthew Dallos: My pleasure. Well, thanks for the opportunity and for having me.