Alicia Morgans: Hi, my name is Alicia Morgans, and I welcome you again to recording at Munich 2018 for ESMO. We're excited to have with us today Dr. Himisha Beltran who is a medical oncologist specializing in genitourinary malignancies at the Dana-Farber Cancer Institute. Thank you so much for being here with us today, Himisha.

Himisha Beltran: Thanks for having me, Alicia. Thanks.

Alicia Morgans: Of course. Your specialties, among other things, have really been an interest in the aggressive variants of prostate cancer, and you've done a lot of work in neuroendocrine differentiation, small cell differentiation, androgen indifferent populations of cells. Can you talk to us a little bit about what this really means, how you define these populations, so we can think about it in terms of the patients that we see on a day-to-day basis?

Himisha Beltran: As you mentioned, I'm a medical oncologist focused on taking care of patients with advanced disease, and a lot of my lab-based research and clinical investigation has focused on understanding the spectrum of the disease and trying to develop treatment strategies to sub-segment the disease to develop more precise therapies. And all the terms that you mentioned are often used interchangeably but are really trying, I think as a field, we're trying to define who are the subset of patients with advanced disease that may need to be managed differently because they have aggressive clinical features or molecular features or pathologic features, and how do these align and how could we use this information to really develop novel treatment strategies, or at least use this information to select patients for the current standard of care drugs as well.

There have been so many advances in the management of patients, which has been exciting. But now we have tools and we have choices, and so now we really, I think more than ever need to identify this subgroup, and by the subgroup, so you mentioned, aggressive variant, AR indifferent, neuroendocrine. We know most patients with advanced prostate cancer, their tumors are dependent on the androgen receptor, and this has been a major focus of drug and biomarker development.

A small fraction of patients, probably about 15 to 20% of advanced prostate cancer patients, evolve in the face of therapy to be less dependent on the androgen receptor. They may lose androgen receptor signaling pathways, AR itself. This is often associated with pathologic features like small cell neuroendocrine features, and when this occurs, clinically, patients tend to have more aggressive clinical behavior, often low PSA, PSA is androgen-regulated, visceral metastases. Sometimes we see brain metastases, and so this is really the subgroup that has been a major focus of my research as well as others.

Alicia Morgans: Absolutely. As you said, these patients experience very different clinical features, brain metastases, for example, so rare in most of our patients, particularly in those patients with traditional adenocarcinoma. But differentiating these heterogeneous groups as we think about really targeting certain treatments to certain populations of patients is so important. Your work has been focused not only on the clinical features, but I'd love to hear more about that as well. But you've also really looked at the molecular subtypes of these cancers, and I'd love to hear your thoughts on that. Where are we in terms of getting a molecular characterization of these patients into the clinical practice?

Himisha Beltran: I mean, I think that because this tends to be a resistance phenotype that emerges in late stages and there tends to be heterogeneity between patients with advanced disease, identifying the molecular features will really help us for biomarkers and selection of therapies. I've been interested in looking at the genomics and epigenetics of these tumors. And what it seems from analyses so far is that these, despite being so different pathologically and clinically, tend to evolve from a preexisting adenocarcinoma. They tend to acquire genomic alterations such as Rb and p53. These are common alterations in small cell lung cancer and other small cell cancers. There's been a number of beautiful preclinical studies published recently by various labs that have really highlighted the functional role of these two genes in cooperating to drive this lineage plasticity or this trans-differentiation towards this neuroendocrine phenotype.

So these are potential biomarkers and the advantage of those is that they're genomic and so potentially could be applied to ctDNA, and of course, would need more prospective clinical validation to use them, I think, usefully today for treatment selection. But there are also other pathway changes, epigenetic changes, and transcriptome changes that are really quite distinct that I think are providing new clues for new drug targets or new therapeutic avenues.

The West Coast Stand Up To Cancer PCF Dream Team has also been key in really understanding the frequency of small cell and after treatment as they identified also some pathway changes that I think are relevant. So I think as a field, and it's been quite gratifying that there's been a lot of collaboration in this space between scientists and clinicians to really try to understand this better. But I think those are really where I think a lot of the avenues are going towards more pathway changes, less genomics because they are evolving, epigenetic strategies and hopefully, and also learning from small-cell lung cancer. That's another area where I think we can learn a lot because there are some shared alterations there.

Alicia Morgans: Absolutely. Do you see a time when ctDNA might be useful in the clinic, or what are your thoughts on re-biopsy, for example, when we see patients with atypical spread, metastatic spread?

Himisha Beltran: If I see a patient that has an atypical presentation, low PSA, that would be a situation where I would consider re-biopsy. There are not really formal guidelines as to when to do this, but I think it's important today because that's the type of patient, if I saw a small cell, I might select chemotherapy versus an AR-targeted drug. And also there are other drugs like PSMA-targeted drugs and other implications I think for identifying the variants that are less dependent on the androgen receptor and lose PSMA or AR.

I think that non-invasive means of detecting this would be much better for our patients. Having repeat biopsies is not ideal for our patients, and I feel that ctDNA has a lot of promise for detecting the alterations we see in tumor biopsies. Part of ongoing work actually is looking at DNA methylation of ctDNA because perhaps that might be able to provide a little more insight into different resistance phenotypes. But I am quite excited about the use of liquid biopsy, CTC, ctDNA to diagnose not just neuroendocrine AR indifferent disease, but also the other molecular subgroups in advanced prostate cancer with important clinical implications.

Alicia Morgans: Absolutely. I think in clinical practice there's such variability in our ability to actually re-biopsy and get viable tissue, particularly in bone metastasis, that it can be really challenging. And there's such a time delay to biopsy scheduling and then time to evaluation of the tissue that if we could get a ctDNA approach in place, I think we'd all be really grateful for that, especially our patients, of course.

So what do you do when you see patients that you define either clinically, because of atypical spread, low PSA, a lot of visceral disease perhaps, or molecularly where there's the triad of P10, Rb, p53? Do you treat these patients differently from the get-go? Do you treat them differently over time, or how do you approach thinking through that patient population?

Himisha Beltran: I mean, I think if I had a patient with small cell that has aggressive clinical features, I would treat with a more aggressive combination chemotherapy approach. I would continue ADT or add ADT if they haven't had it because there is heterogeneity that might not be captured on a biopsy, and we know that tumors often do have mixed features before they become small cell. I think what's more challenging is this middle group where they have aggressive features but they don't have small cell, or they have neuroendocrine features, but they're not that aggressive clinically. That's where I believe it's much more challenging.

I do continue ADT and use this information to potentially guide chemo versus AR-directed therapy or even the chemotherapy regimen, if I might see a mixed tumor, to do taxane plus platinum potentially. I mean there's, it's a kind of a fuzzy area, and I think that it really depends on the clinical context for most of the time in that middle ground. That's where I hope molecular biomarkers will help us understand this better because of that variability on single-site biopsy that you might get or even in clinical features.

We know not all patients with visceral mets have small cell, and most of them don't, and many of them will respond to AR-targeted drugs. And so I think that there isn't necessarily one clinical feature that would lead me that way, but it's almost a constellation, at this point, in that middle ground. And I think this needs to be better defined, for sure.

Alicia Morgans: Absolutely. I really appreciate you saying that you continue ADT. I remember a conversation a few years ago. I had a patient with true small cell, which is different than just high-grade neuroendocrine. But he had true small cell and you had a clinical trial open, and I was talking to you about trying to engage him with that trial. I said that "We're trying to get him on ADT," and you said, "Please do, because there's such heterogeneity of tumors, your biopsy may not have hit the area that is AR driven, and so to have that person on ADT is still important." I definitely locked that tidbit away, and I appreciate you saying that today.

But where do these trials stand right now? Where are you in terms of your research or the research of others that's ongoing, because this is definitely an area of clinical need that we want to highlight?

Himisha Beltran: Yeah. I recently led a clinical trial of an Aurora kinase inhibitor for this subgroup of aggressive variant neuroendocrine disease. The trial was designed many years ago before we really had good pathologic or molecular biomarkers. The rationale for that trial was that Aurora kinase A, which is a cell cycle kinase, can also indirectly inhibit N-Myc signaling, and N-Myc is overactive in these neuroendocrine prostate cancers, similar to neuroblastoma and a subset of small cell lung cancers.

That trial was quite gratifying in that it really was a collaboration. There were so many people that came together to enroll and a very difficult population to enroll in. Median PSA was one on that study, most of them heavily pretreated. And we did identify exceptional responders in this study and some patients with long-term durable benefits. However, I don't think that drug is moving forward, mainly because it doesn't necessarily effectively inhibit N-Myc, necessarily in patients we feel, and also, in the end, we found the disease group to be really quite heterogeneous for inclusion.

So I would support the next trials in this space really including more of a molecular definition, and that's where I think the challenge is, is because it's hard to find the biomarker for patient selection. I am involved in Rova-T trials. Rova-T is an antibody-drug conjugate for small cell lung cancer that targets DL3. There's a Phase I basket with a prostate arm. The advantage of something like that is that it requires the target to be present, so DL3 has to be present. So having trials that are enriched for patients most likely to respond will be useful in this space. DL3 is interesting in itself, it's a notch inhibitory ligand, and so there's some biology there that needs to be sort of elucidated as well.

And then immunotherapy is in the clinic, not necessarily specifically for small cell neuroendocrine prostate, although I think there have been some discussions about that. It's still yet to be seen whether these are more like small cell lung or prostate in this realm and if combination therapy may be useful. And then, of course, there's platinum chemotherapy trials, which is actually more upstaged from some of these things that I'm talking about, the anaplastic aggressive variant trials led by MD Anderson. That's where I think we are mainly. There are other interesting targets, more in a preclinical, early stage of development, epigenetic strategies, and mainly focused on, and again, immune strategies, but these are more, I think, earlier stages of the disease.

Alicia Morgans: Fantastic. Thank you for sharing all of this. I think that it's a little bit of a mystery or a black box to some people, even myself when we think about these patients in our clinical practice. And it's wonderful to hear that we continue ADT, that we continue to strive to molecularly characterize them, that people like you and collaborators are really working on therapeutic advances in this population of patients, which is pretty distinctly different than the rest of the patients that we see. And it's important as we continue to treat men with prostate cancer that we take what looks like just a big bucket of people and really break them into the heterogeneous groups that they are.

So do you have any final thoughts or messages for the viewers, closing thoughts on this?

Himisha Beltran: Yeah. I mean, I think that it's been such a privilege to be part of the prostate cancer research community to see so many advances in the treatment of our patients. I think understanding like you said, these disease subgroups and understanding that not all patients are the same is going to be very important, and really incorporating some of these research findings into the clinic is really where I think the trials are going. I think we'll see where things go in the years to come, but it's been a very, I think an exciting time in general.

Alicia Morgans: Absolutely, and we need to encourage our patients to get on these trials-

Himisha Beltran: Yes.

Alicia Morgans: ... because we can't answer the questions without the patients. And we're looking for your next breakthrough, Himisha. So thank you so much for your time.

Himisha Beltran: Thanks very much.