Read the Full Video Transcript

Andrea Miyahira: Hi, everyone. I’m Andrea Miyahira at the Prostate Cancer Foundation. I’d like to welcome Dr. Bianca Cali from the Institute of Oncology Research in Switzerland to discuss her paper “Coagulation Factor X Promotes Resistance to Androgen Deprivation Therapy in Prostate Cancer.” And this was just published in Cancer Cell. Dr. Cali, thanks for joining us.

Bianca Cali: Hi, Andrea. Thanks for inviting me. Just to start, I just have a little bit of introduction. We all know that a high neutrophil-to-lymphocyte ratio is a valid prognostic marker in metastatic castration-resistant prostate cancer patients. And this is associated with shorter survival.

And it’s important because this high neutrophil-to-lymphocyte ratio also reflects the infiltration of these inflammatory neutrophils in the tumor microenvironment. These neutrophils are pathologically activated. They have immunosuppressive capacity, and that’s why they’re also called PMN-MDSCs. And the lab has demonstrated earlier that PMN-MDSCs can promote the onset of castration resistance in prostate cancer via paracrine signaling that is partly due to interleukin 23. And these myeloid cells reach the tumor, and the infiltration can be prevented by the administration of CXCR2 inhibitors.

And these CXCR2 inhibitors have been recently shown to be effective in reversing resistance to androgen deprivation therapy also in patients. And despite that, I just want to remind all of you that patients affected with different types of tumors have been shown over the years, actually in the last century, to have blood hypercoagulability. So what we did was try to understand if there was any association between these polymorphonuclear cells and coagulation factors, because we tried to dig into these polymorphonuclear MDSCs in the different mouse models of castration-resistant prostate cancer. And we found that in the phase of castration resistance, these cells do upregulate F10, which is a coagulation factor that you expect to be released and produced by the liver. But actually, we found this extrahepatic source in these tumors. And what was really interesting for us is that this coagulation factor expression significantly increases in different models, but it’s also associated with the prolonged treatment with enzalutamide. And we also found it in PMN-MDSCs from tumors from mice that were treated and were not responding to CXCR2 antagonists, as well as in mice that were treated with enzalutamide plus interleukin 23 antibodies. So we tried to understand this better and to characterize this subset of PMN-MDSCs. So we reclustered them, and we found that those ones that were expressing high levels of F10, which we can characterize here in this blue region of this UMAP, are also those ones that express very low levels of CXCR2.

That suggests that these cells are insensitive to CXCR2 inhibitors. They express F10, and they can also be tracked because we tried to characterize these cells on their surface by a marker, which is CD84. So, but this coagulation factor, we know, does not trigger thromboembolic events in our mice. We excluded the involvement of other coagulation factors in the clotting. So we tried to assess whether there was a direct effect of this F10 on prostate tumor cells. So we started in vitro.

And we exposed the cells that were kept in androgen deprivation to this recombinant molecule, and we saw that exposure to F10 was able to induce androgen-independent cell proliferation. And we tried to understand the mechanism. And we found that among the receptors expressed in the tumors, F2r1, encoding for PAR2, was a critical receptor upregulated in mouse samples that was also significantly expressed by tumor cells in human prostate cancer single-cell RNA sequencing data.

So we found that F10 could work through PAR2 in these epithelial cells. And what we realized is that activation of this signaling was able to promote phosphorylation, and we blocked that phosphorylation when we blocked the receptor by using different antagonists. And then we also moved in vivo, and we verified in different mouse models—CRPC, xenografts—that exposure of tumors in castrated mice to exogenously administered MDSCs was sufficient to promote tumor growth despite androgen deprivation.

And blocking F10 coming from these cells by using rivaroxaban, which is an F10 inhibitor, was sufficient to prevent this pro-tumorigenic function of MDSCs in vivo. And then we verified this in multiple other models, and we also confirmed that these F10 inhibitors can ameliorate, can improve, the efficacy of enzalutamide in multiple mouse models because we saw that the tumors were less aggressive, and also phosphorylation of their kinases was reduced. Then all these preclinical data needed some clinical validation. So we tried to verify if this was also conserved in patients. So we measured F10 plasma levels in patients affected by mCRPC. And we found a strong association—actually, there was a correlation between plasma levels of F10 and neutrophil-to-lymphocyte ratio. And we found actually that F10 was prognostic towards disease in a multivariate analysis. And we found that this F10 in the plasma was also reflected in cells expressing F10 belonging to the polymorphonuclear subset in the tumor microenvironment. Indeed, the counts of these cells were significantly higher in tumors as compared with benign hyperplasia tissues. And what was really interesting is that a signature of these PMN-MDSCs expressing high levels of F10 and also the surface marker CD84 was associated—high signature was associated—with shorter survival in prostate cancer patients. And we also looked at the receptor. And by analyzing tissue microarrays from patients affected by prostate cancer, we found that the castration-resistant tumors showed even higher PAR2 expression levels as compared with the hormone-sensitive ones and benign hyperplasia. And the tumors that were negative for PAR2 were also those ones that showed longer disease-free survival.

So how we can summarize this is that we found a novel crosstalk between PMN-MDSCs and prostate tumors, and we found actually an extrahepatic source of F10 in the tumor microenvironment. And this F10 is critical in promoting tumor cell proliferation despite androgen deprivation. And we can track these highly protumoral cells by the CD84 marker. And all these players involved in this axis, including CD84, F10, and PAR2, predict shorter survival in cancer patients. So what I want to stress here is that we finally found that direct role for coagulation factor X, at least in prostate cancer progression. So it’s not something involved in clotting in vascular events, but just within the tumor. And there is a subset of cells that are highly tumor-promoting that might be insensitive to CXCR2 inhibitors because actually, they express very low levels of this receptor. And both high plasma levels and intratumor levels of F10 can predict poor survival in castration-resistant prostate cancer patients. And we have to keep in mind that, at least in mouse models, administration of F10 inhibitors was sufficient to improve the efficacy of ADT but also of enzalutamide.

So with that, I really need to thank all the collaborators and, of course, Professor Andrea Alimonti, who hosted me in his lab for this amazing project, I would say.

Andrea Miyahira: Well, congratulations on this study, and thanks for sharing it with us. So based on that, your lab and others have found multiple mechanisms by which MDSCs can promote prostate cancer, what do you think are the most important mechanisms of MDSCs in prostate cancer, and how do we target them, considering it seems like they are different subsets?

Bianca Cali: So actually, I think that they are a pool of different clusters that somehow, by releasing different factors, trigger different mechanisms in the cells. We already demonstrated in the lab that—actually, the lab has demonstrated that, for instance, interleukin 23 triggered the STAT3 pathway.

In our case, we found that there is no involvement of androgen receptor—I mean, a strong involvement of androgen receptor, but just this activation of the AIRE phosphorylation cascade. So we are really digging into it now to understand what is really the mechanism, because we just excluded some of them, but we are not really aware of what is going on. I mean, work is ongoing on that.

Andrea Miyahira: OK. Do the MDSC-derived F10 cells have any effects on anti-tumor immune responses?

Bianca Cali: Yeah. Actually, we know. We did multiple experiments. We also performed some bioinformatic analysis of mice that were deficient for F10 in their myeloid compartment, and we confirmed that these cells have immunosuppressive capacity, as well as the mice that have proficient F10. What we can say is that these cells keep their immunosuppressive ability, but they are not more immunosuppressive than others. So it’s really a direct effect on the tumors that we highlighted, let’s say, with this work.

Andrea Miyahira: So does F10 and the CD84-positive MDSCs drive resistance to specific treatments, such as AR-targeted therapy, or are the contributions more in a treatment-agnostic way to disease progression and treatment resistance?

Bianca Cali: So actually, we found in our mice that these F10^hi CD84 cells increased in numbers in resistance also to enzalutamide and to CXCR2 inhibitors. I cannot exclude that they are increased in other treatments, but it’s something that I didn’t prove experimentally. So I cannot reply.

Andrea Miyahira: OK. Thanks. And what are your next steps for these studies?

Bianca Cali: So actually, like I said, we are trying to understand better the mechanism that involves ERK kinases that triggers this androgen-independent cell proliferation. And then we are also trying to—we just designed a trial to start to assess whether, in patients, the administration of F10 inhibitors can, let’s say, improve the efficacy of current therapies, the current androgen deprivation therapies.

Andrea Miyahira: Well, thank you again for coming on and sharing this study with us today.

Bianca Cali: Thanks to you.