Read the Full Video Transcript

Andrea Miyahira: Hi, I'm Andrea Miyahira here at the Prostate Cancer Foundation. With me today is Doctor Xiaolin Zhu of UCSF. Doctor Zhu will discuss his recent paper, Genomic and Transcriptomic Features of Androgen Receptor Signaling Inhibitor Resistance in Metastatic Castration Resistant Prostate Cancer that was published in the Journal of Clinical Investigation. Doctor Zhu, thanks for joining us.

Xiaolin Zhu: Thank you, Andrea. It's my pleasure. And I would like to share our recent paper. It's titled Genomic and Transcriptomic Features of Androgen Receptor Signaling Inhibitor Resistance in Metastatic Castration Resistant Prostate Cancer. My name is Xiaolin Zhu. I'm a medical oncologist specializing in GU medical oncology, and I do research in prostate cancer. Our study carefully looked at paired metastatic biopsies obtained from patients with metastatic castration resistant prostate cancer or mCRPC.

Our pretreatment biopsy was obtained before the patient was ever treated with an ARSI, like abiraterone or enzalutamide. And our post-treatment biopsy was obtained at disease progression on that ARSI. It's a long paper, so I would like to highlight three novel or we think most interesting findings from this paper. And they are listed below. So I will go through them one by one.

The first interesting finding is we found a putative enhancer downstream of the androgen receptor gene is amplified after ARSI treatment, and the field is already known that the upstream AR enhancer, as identified by David Quigley and others, is the most recurrent DNA-level change in mCRPC. So we follow this idea using our paired data by looking at their whole genome sequence. And we carefully looked at the AR gene and its flanking regions, actually, by taking advantage of multi-omic data we have generated not only for this project but over multiple years with the Phoenix phones lab and the West Coast Dream team.

So we overlaid the data of whole genome sequencing. As you can see, we specifically looked for tandem duplications and copy number alterations in this locus. And then we also overlaid data from the Dana-Farber group. And we generated over the past few years from cell lines and PDX models, as well as patient samples of ChIP-seq data of key transcription factors, including AR, FOXA1, HOXB13, and enhancer marker H3K27 acetylation.

So basically, when we overlay the data together and focus on our paired data by looking at the specific locus where there is a copy number gain, we found that these four regions—we labeled them as C1, C2, C3, and C4—in addition to the well-known AR promoter and the upstream AR enhancer are also amplified and show all these characteristics of active regulatory function. So especially when we also looked at the HiChIP data, we found that C4 is the most interesting enhancer, and it's distal of the AR gene.

And in the HiChIP data we found that C4 actually contacts the AR promoter, which is not shown here. But the more detail is in our paper. So the second most interesting finding we found is from a paired analysis of RNA-seq data. We actually identified a transcriptional downregulation of a gene called SSTR1 in ARSI-resistant prostate cancer. As you can see in the volcano plot on the left side, SSTR1 is the most significantly downregulated gene—the most significantly altered gene—when we compare the paired RNA-seq data.

And if we look at 31 pairs for RNA-seq, we found that there is a statistically very significant downregulation of this gene. What is SSTR1? So SSTR1 encodes a gene called somatostatin receptor type 1. So actually, it's a GI-coupled and G-protein coupled receptor. And when it's activated by its ligand, which is somatostatin, it will transduce an anti-proliferative signal in the target cells that express SSTR1. So we further showed that SSTR1 is expressed in prostate cancer cells rather than the immune cells or stromal cells in the tumor microenvironment.

So next, we have our third interesting finding. So this is a little bit unexpected. So we found that SSTR1 actually is upregulated in preclinical models and in patient samples in the setting of bipolar androgen treatment. So basically, on the left side is a paper published by the Fred Hutch group a few years ago. They showed that in these two enzalutamide-resistant PDX models, LuCaP 35CR and the resistance model, as well as the LuCaP 96CR. And in these two models, when the mice received treatments with supraphysiological testosterone—meaning very high dose of testosterone—the tumors will regress in these two models.

And interestingly, we found that SSTR1 expression is upregulated in both models. On the right side is from a recent paper from the Johns Hopkins group, which reported that BAT plus nivolumab treatment in patients with mCRPC—so basically, what we found—and this study actually also collected paired biopsies, pre-BAT treatment, and on-BAT treatment. I think this is after three cycles of BAT and nivolumab or four cycles.

And what we found is that in using the paired RNA-seq data, we found that for patients who actually had a PSA 50 response, their SSTR1 expression in the post-BAT metastatic biopsy is increased, whereas in patients who did not have a PSA 50 response, the SSTR1 expression is unchanged. So we find this fascinating, and this might be a link between SSTR expression and response to BAT.

So to summarize our paper, we showed that by systematically comparing the whole genomes and whole transcriptomes of paired metastatic castration resistant prostate cancer biopsies, we further validated that the AR locus is the primary substrate of convergent evolution under ARSI-induced pressure. And we also identified that transcriptional downregulation of SSTR1 is a common feature in ARSI-resistant prostate cancer. So the take-home message here is we are able to identify recurrent DNA-level and RNA-level features of ARSI-resistant prostate cancer by looking at paired patient samples.

And some of these features we think are quite exciting, and they are recurrent. For example, the amplified enhancer downstream of the AR gene and the downregulation of SSTR1 expression in ARSI-resistant prostate cancer. We are planning to study both of these features in the wet lab, and we think these findings are relevant to patient care and worth further investigation. Thank you.

Andrea Miyahira: So thank you for sharing this, Doctor Zhou. So were there any other surprising or novel findings involving AR alterations in mCRPC?

Xiaolin Zhu: I think the most novel finding is the amplified putative enhancer downstream of the AR gene. So AR is the most important, I would say, genetic substrate of evolution in prostate cancer because there are recurrent AR gene or upstream AR enhancer amplifications. And there are also recurrent AR hotspot mutations, especially of the ligand-binding domain. So collectively, they affect a lot of patients who progress on hormonal therapy.

So the finding that distal enhancer, which is downstream of the AR gene, is also amplified, I think further validates that AR is so important as the genetic substrate, but also it points to the complex regulatory landscape of AR expression itself. We think not only the gene is important, but also the flanking regions, including putative regulatory elements, are important. I think it further validates AR as such a critical target in prostate cancer in mCRPC.

Andrea Miyahira: Thank you. And what might be the function of SSTR1 in prostate cancer and ARSI resistance?

Xiaolin Zhu: Yeah, it's a great question. As I alluded before, SSTR1 is—I would say that might be an unexpected finding, but it was identified through a transcriptome-wide analysis. So it's kind of a hypothesis-agnostic analysis. So SSTR1 encodes Somatostatin Receptor Type 1. Somatostatin is a natural hormone that's in the human body. And importantly, when it activates its receptor, there are actually five subtypes—SSTR1 to SSTR5. So they share a lot of structural similarities and functional, I think, importance.

But SSTR1, I think, is what we found in prostate cancer specimens. And once it's activated by its ligand somatostatin—or there can be artificial, I would say, chemical agonists—it transduces an anti-proliferative signal. So actually, in our paper in the cell lines, in 22Rv1 cells—those are castration-resistant prostate cancer cells—we showed that if we overexpress SSTR1 in those cells, actually, it slows down the growth and actually kills some of the 22Rv1 cells. And if we knock it down, it actually accelerates cell growth. So I think this is in the supplement of the paper, and it shows there is some causality of SSTR1 being a functional gene in mCRPC. We are studying its antiproliferative and anti-tumor and potentially anti-ARSI resistance effect in cell line models and PDX models of mCRPC.

Andrea Miyahira: OK. Thank you. And do you have translational plans for SSTR1, agonists, or as a biomarker?

Xiaolin Zhu: Yeah. So for SSTR1 agonist, there is an FDA-approved drug called pasireotide. It is not an SSTR1-specific agonist. It's a pan SSTR agonist. So it activates SSTR1 to 5, but it is the only FDA-approved SSTR1 agonist, although it has effects on other SSTR subtypes. So there are SSTR2 agonists—for example, octreotide and lanreotide. These are commonly actually used to treat certain types of cancer. And also there are clinical trials actually studying them in prostate cancer many years ago with, I would say, borderline evidence of efficacy.

So we are trying pasireotide, although it's not a specific agonist. We think it's because it's the only one that's FDA approved. It's actually allowed to be used to treat acromegaly and Cushing's disease. These are the tumors of the pituitary, the endocrine gland right beneath the brain. So we are testing it in preclinical models of mCRPC to see if there is any efficacy of inhibiting cell growth. As for the biomarker perspective—so we are looking at actually SSTR1 mRNA and protein expression in patient samples.

Not only for patients with mCRPC, but also actually for patients with hormone-sensitive disease because we try to assess the expression of SSTR1 across the different disease states of prostate cancer because it's downregulated in mCRPC. So we expect, actually, its abundance is higher in earlier states of disease. For example, in primary tumor and metastatic hormone-sensitive disease. So actually, if you look at TCGA, SSTR1 mRNA expression is very high in primary prostate adenocarcinoma compared to other cancer types. So we are working on that as well.

Andrea Miyahira: OK. Thanks. And have you evaluated SSTR1 in NEPC?

Xiaolin Zhu: No. So I would say that's a great question. So we are actually planning also to use NEPC-PDX models. For example, LTL 331. It's a well-established NEPC model. Yeah. So we are working on this model to see how SSTR1 is regulated and may impact tumor biology and tumor growth.

Andrea Miyahira: OK. Well, thanks so much for sharing this with us today.

Xiaolin Zhu: Thank you, Andrea.