Targeting AR in Castration-Resistant Prostate Cancer

My goal today, we’re going to talk about targeting androgen receptor in castration-resistant prostate cancer. And these are my disclosures

Where I want to go with this talk is I’m not going to talk too much about drugs. Neal gave a wonderful presentation about all the newer agents that have been approved, specifically since 2010. I think most of you in the room have heard this, have read about all the new agents. But I really want to stress the fact that I think for—if you’re interested on managing prostate cancer, especially as we move from this volume to value-based medicine, as many of you know, we have been preaching for many, many years that we want the urologist to be involved in the total continuum of care. And specifically, especially as we become less and less invasive, you know, my father was the first urology resident at the Lahey Clinic. I’ve been around this all my life, and I love David’s video that he just showed. But I think as we are progressing to even though we still are a surgical subspecialty, as we become less and less procedure driven, and as we get more and more of these great agents, and the challenge of treating metastatic castration-resistant prostate cancer, I think it’s imperative that we start to understand the molecular mechanisms and the drivers behind what’s driving this disease. So we all know, it’s the Huggins and Hodges in 41 as the video just showed, basically they proved that prostate cancer is an endocrine-driven disease. And the mainstay of therapy, once patients progress with prostate cancer, has always been knocking out testosterone.

Tom Keane showed this slide earlier. We know that there’s basically three sources. Obviously, the majority of testosterone production that comes out of the testicle. You get about 3% or 4% that comes from, that has this paracrine production from the adrenal gland. And now we know that through mutational changes sometimes that is actually pressure treatment driven, like Dr. Pinthus talked about, that you get this autocrine production, where the tumors themselves are producing their own testosterone. And when I talk to my patients about what that means, the analogy that I use is that it would be like a car making its own gasoline. It would be great if you had that type of car, it’s not so good if you have that type of tumor.

Okay. So this actually is in a graphical form, not as cool at the video that you saw. It’s basically the history of how androgen deprivation, how we’ve been successful in lowering serum testosterone. And when you look what we’ve done in the past, especially for those of us that trained in the 90s, obviously, surgical castration, or excuse me, before 1990, surgical castration was always the mainstay. But again, a lot of us use diethylstilbestrol or DES, which was a great drug. The problem with DES at the 5 mg dosing, it did a wonderful job in controlling prostate cancer, but it was also very associated with thromboembolic events. And it also turned out be teratogenic. Then we’ve used anti-androgen. David, probably some really landmark papers on complete androgen blockade. And then as well as we have androgen synthesis inhibitors. But again, the mainstay, especially as patients advance, has always been knocking down the testosterone to less than 50, which is historically been the definition. Now, we can have a whole other discussion whether it should be less than 33, less than 20. I think most of believe the lower the better. But ADT still is the cornerstone, but at the same time, it is not without risk. It’s not without complications. And as you know, a lot of times it’s the urologist, we start ADT, especially in the setting of a biochemical recurrence for those patients who have been definitively treated. And again, it’s got some major problems. And I think all of us, we owe it to our patients to really discuss this with them. And again, this is becoming more and more—people are becoming more and more aware, and again, as we move into this post-macro world of alterative payment models, bundled payments, maximum-intensity projections (MIPs), one of the things that we’re going to need to be tracking is quality of life. So I think ultimately, this is just sort of a sidebar, I think we need to be, again, especially in the area of biochemical recurrence, we need to do a little bit better job, we need to be more judicious in our use of ADT. But again, we all know what the complications are, hot flashes, anemia, sexual dysfunction, neurocognitive dysfunction, which can be really bothersome, and I think we’re becoming more and more aware of that, osteoporosis, and I think Lenny Gomella is going to talk about that tomorrow, and obviously, this pseudo metabolic syndrome, which includes obesity, insulin resistance, dyslipidemia and hypertension.

We know that these patients, once we get them on ADT, they will progress. Not all of them, but they will progress. And when you look at the NCCN with this concept of what we call castration-resistant prostate cancer, we talked about earlier today why that nomenclature has changed. I’m going to give you some of the molecular reasons why. We’ve gone from this concept of hormone refractory, androgen insensitive to castration-resistant prostate cancer.

So the first area is this area is this area called M0, where by definition, the definition of castration-resistant prostate cancer, you’ve got to have a diagnosis of prostate cancer, you have to be on LHRH therapy, your testosterone levels need to be in the castration range, which right now by definition is less than 50, and you have to have a rise in your PSA. And again, as you know, most clinical trials define that as elevation 25% to 50% greater than the prior with between one to two weeks of it being drawn. If you then studied them radiographically we know that about a third of these people who you think are non-metastatic usually based upon their symptoms in fact are symptomatic; excuse me, are metastatic. So it really is incumbent upon the urologist for us to be early identifying these patients. But we know that there is this subset of patients who really do not have radiographic spread based upon bone scan and CT. Now, I think many of us believe especially with some of the new advanced imaging that are coming out, and I think Dan is going to touch on that soon, is that this area, this niche of patients will probably become smaller and smaller. But sufficive to say right now if you truly are in this M0 CRPC category your treatment options are clinical trial or observation. None of these cool drugs that we have are currently approved for the M0 patient. You have to have metastatic disease. And the older therapies, including the first generation anti-androgens don’t have a survival benefit.

But then as you get into the M1 disease space, which again is this CRPC, but they do have radiographic mets, then the whole world opens to us in terms of all of these new therapeutics.

These are kind of the key clinical trials that have been put forth since 2004. These are the positive trials. But again, I think when you sort of look at this, all these different agents all are different mechanistically. And again, this is sort of a summary slide that many of you have seen looking at the survival benefit of all these particular agents. However, and this is where the meat of the talk is going to come, many of these agents, what still remains central to the progression of these patients is this androgen receptor. And I will say that if you would have told me ten years ago, Conceptcion, if you want to manage advanced prostate cancer you need to be fully aware of the androgen receptor and where it’s located, and those types of things, I would have told you you’re out of your mind. I don’t think you can begin to understand or begin to treat and know about these drugs until you really have a good working knowledge and definition of the AR.

So, sufficive to say is that it’s found on chromosome 11.   And the androgen receptor actually has four major domains. It has the N terminal, it has the DNA binding area, it has the ligand binding domain, and then there’s actually a hinge region. So this is all found on chromosome 11, and these are the various exosomes that are associated. And how the androgen receptor works is on—what I want to point out to you is on the left side of this slide. This is what I sort of call the canonical pathway. And so we’re going to go through this pretty quickly.

So what tends to happen in normally functioning prostate cells in men is that testosterone really circulates. Most of it is bound by sex hormone binding globulin, but most of it, but the functional activity of it is actually free testosterone. So what happens is that testosterone disassociates from sex hormone binding globulin. It diffuses through the membrane of the prostate cell. Once inside the cytoplasm it then encounters the androgen receptor. Now the androgen receptor when it’s not locked into the ligand it can be found with what we call a chaperone molecule, which is usually heat shock protein. Testosterone, once inside the cytoplasm gets converted intracellularly by the enzyme 5alphareductase to dihydrotestosterone. Dihydrotestosterone is a much more powerful androgen. When it comes into association with the androgen receptor you get disassociation of the heat shock protein, you get binding of the ligand to the ligand-binding domain of the receptor. You then get this dimerization of the ligand receptor complex. What then happens is that you get nuclear translocation. Once inside the nucleus the N terminus then binds to DNA, which then starts the whole process of transcription and translation. That’s how it works. And what do we get for that? Well, we get PSA production, we get cell growth, we get cell survival. Remember, and it’s a very important point, and we forget this sometimes, especially as we’re treating advanced prostate cancer patients, PSA is a function of androgen binding, it is an androgen driven event. But as we get into these pathways of castration-resistance, except for—and as Dr. Pinthus talked about, except for the true neuroendocrine dedifferentiations in the small cells, which often lack AR expression, a lot of the escape mechanisms still revolve around the androgen receptor. So that’s why we have moved into this change in nomenclature of these patients being hormone resistant, androgen insensitive. I think Dave had a very nice suggestion a couple of years ago, that these should be called endocrine resistant.

So when you look at this, and this is a series of slides that actually I borrowed from Phil Kantoff, is that these are all the various targets that we could be potentially looking at on how to control this disease. Well obviously, the first one is to do what we always do, which is look at circulating testosterone. So we can do that with LHRH analogues, castrations, surgical castration. And again, what that’s going to do is drive down circulating testosterone.   You can knock out the paracrine or intracrine production, which is basically at the adrenal gland. And again, we know that ketoconazole used to do that, and now we also know that abiraterone acetate has its effect there as well.

You can actually degrade the androgen receptor as well as blocking some of these chaperone molecules. You can, obviously, block the androgen receptor, and obviously, the earlier ones were these first generation anti-androgens, including flutamide, bicalutamide, nilutamide, but then also we now have the second generation, which includes enzalutamide, also what you’re going to get is ARN509, also now known as apalutamide.

You can prevent AR nuclear translocation. And how do we do it. If you heard the lunchtime talk we know that enzalutamide can block nuclear translocation, but one of the great questions that Dan Petrylak has said is that, true or false, taxane chemotherapy has a role in the androgen receptor, and the answer is true, because we think that because of its microtubule inhibition it can also block nuclear translocation. You can interfere with AR enablers. And again, these are some of these escape pathways. You can block coactivators, as well as you can block these downstream effectors, which again, get into these whole areas of molecular drivers.

So again, what’s key here, what’s what common here, again, it’s still the presence and the working of the androgen receptor. If you want to read a great article, this was published in the Journal of Clinical Oncology in May of 2013. This came out of the University of Michigan. This was Arul Chinnaiyan’s group. Arul basically is obviously one of these most talented guys when it looks at looking at molecular drivers for prostate cancer. And this is a very nice review article. And what he was able to do, he basically summarized looking at clinically relevant genomic alterations. And I’m not going to read all of these, but this is what basically we’re starting to talk about TMPRSS2-ERG, transcription factors, obviously, alteration in the androgen receptor, PTEN loss. We talked Aurora kinase and those types of things. And then these are some of the targets that you can go after. So again, this is a very nice article that I would encourage all of you to read to sort of get a baseline in terms of trying to understand these molecular drivers.

And again, this is just a little bit of a different way looking at it. And again, we’re looking at obviously the androgen receptor. But then you can also look at basically the effect of tyrosine kinase receptors in all these alternative pathways, including P13K, again PTEN loss, and again, in green here are potentially all the potential drugs that we can use to alter these pathways.

So again, what I’m trying to get you to understand here is that we’re understanding more and more, and it’s becoming more complex. Neal talked about today that what happens is that these cells, they’re going rogue and they figure out a way to evade apoptosis. And so that’s really the problem that we really have now, is that we have lots of cool agents, all of them different mechanistically, but there’s nothing to help us to guide these therapies. And again, we know that these cells with treatment pressure selection will actually undergo mutational changes. And one of the things that a lot of people are focusing on right now are these androgen receptor splice variants.

This depicts just a few of them. Again, there’s probably 40 or 50 of them out there, but a lot of attention has been towards this what we call this ARV7 splice variant. So again, this is where the understanding of the makeup of what the androgen receptor does becomes very, very important. If you don’t understand the makeup of the androgen receptor you really, it’s hard to kind of get head around why ARV7 is so important.

So basically, AR splice variants are associated with poor prognosis and treatment resistance. The translation of splice variants results in protein with altered activity, visa vie the androgen receptor. So again, we know that in exons, so again, this is the full length, this is the wild-type AR like we talked about. This area right here is the N terminus, it has nothing to do with ligand binding. This is sort of the DNA binding domain, as well as the hinge region. This is all involved in transcription. Now, the C terminus, this ligand-binding domain is coded for by exons 5 through 8. And so what happens, one of the mutations that happens is that you get this drop out exons 4 through 8 which results in the loss of the ligand binding domain. So you would think that’s a good thing, right? You can’t bind dihydrotestosterone, you can’t bind testosterone, nothing should happen. Well the problem is what have figured out, what they have figured out is that the receptor stays active. It stays active. And what’s the data to suggest that?

So, many of you have seen this. Neal presented some of this today. This was basically the work done by Emanuel Antonarakis at Johns Hopkins. This was written up in the New England Journal of Medicine. And so what this was, this was a prospective study of M1 CRPC patients who were eligible for either treatment with abiraterone acetate or enzalutamide, and through the use of circulating tumor cells, looking at the transcript they were able to identify who was ARV7 positive or negative. And so what I want you to do is go down to this bottom line first. So the patient treatment standard, what was a prevalence? And this is an interesting number to remember. What’s the prevalence of ARV7? So before any patients got either enzalutamide or abiraterone we think the prevalence of ARV7 to have this mutation is about 12%. So obviously, that’s just due to a patient being on androgen deprivation therapy.

And so what they did, is that they measured baseline ARV7 using circulating tumor cells. And in the treatment group that was going to get abiraterone they found 6 of 31 patients, about 20% were ARV7 positive, and in the enzalutamide group, again, 31 patients, it was a little higher at 39%. So basically in the ARV7 positive category those were the numbers. And then what they did is that they started them on therapy and then looked at these various responses. And so remember, if you’re ARV7 positive manipulating the androgen access, whether you put them on an androgen receptor blocker or you put them on a androgen synthesis inhaler, it’s not going to make any difference because it can’t bind the ligand. So low and behold, when they looked at the response rate looking at drops in their PSA of greater than 50%, what they found in the abiraterone treatment group is that if you were ARV7 negative, so you still had the wild-type ARV7 receptor, low and behold about 70% had a PSA drop of greater than 50%. The number was zero if you were ARV7 positive. Overall survival, excuse me, radiographic progression-free survival 2.3 months in the ARV7 positive group, greater than 6 months in the negative group, overall survival better, obviously, in the ARV7 negative group. And the same pattern held true with the enzalutamide treatment group. Again, if you were ARV7 negative the chances of you having a drop in PSA, which again would make sense, was over 50%. You had a longer RPFS, and the overall survival had not been reached, but again, when you look at the ARV7 positive group the numbers were markedly less. And then when you look at the patient treatment status, again, there was about a 12% prevalence at baseline. Post enzalutamide, that went to about 25%, 50% for abiraterone, greater than 60% if you had both.

These are waterfall plots. And again, Neal had mentioned this today, you did have some outliers, but again, in blue are all these patients who are ARV7 negative. And for the most part you had significant drops in your PSA. So that was the enzalutamide treated group, and this is the abiraterone treated group.

Again, the point here is that if you’re V7 positive, and that’s where it helps to understand the makeup of the androgen receptor, you’re not going to bind ligand, it’s not going to have an effect, the receptor stays active, and as Dan had mentioned earlier, you could make a case potentially of putting those patients, if you knew that upfront, you would skip abiraterone, you would skip enzalutamide, and potentially you could make a case for going right to a taxane.

And again, this was survival benefit. And again, this is for the enzalutamide group, the abiraterone treated group, and again, this held true.

So that’s just one mutational defect. That’s just one resistance mechanism that these cells develop. The problem is, as we know, prostate cancer, it’s a heterogeneous disease, its heterogeneity. This concept of it starts here and that every tumor cell enjoys the same characteristic, and then as it progresses it maintains, it’s not—that’s just not the case. That’s why we know looking at these clones, there’s areas of Gleason 3 plus 3, but within the same tumor itself you’re going to have Gleason 4 plus 4, you’re going to have 3 plus 4, you’re going to have a 4 plus 3. At a molecular level once these patients start to progress the challenge is is try to figure out which mechanism, which driver is causing these to progress.

Another mutation is a gain of function mutation in dihydrotestosterone synthesis in patients with CRPC. So remember what we talked about is that dihydrotestosterone is formed from the conversion from testosterone to DHT by the enzyme 5alphareductase. And so what they found is this mutation in 3BHSD1 which actually will produce dihydrotestosterone from variant precursors, not testosterone. So in other words, are you taking other ligands that’s not testosterone and converting it to dihydrotestosterone. So again, it’s another mutation, it’s another way to drive the androgen receptor without altering testosterone production.

You can also activate the glucocorticoid receptor, which confers resistance due to anti-androgens. And again, this was a study, these were patients on enzalutamide. And what they found, and this was actually out of bone marrow biopsies, these patients got enzalutamide, they continued to get it. And then what happens is that they actually tested, and what they found is that you had activation of the glucocorticoid receptor. So even though you were blocking the AR it was escaping, the cells were escaping by activation of a glucocorticoid receptor.

So the bottom line is that this is a challenge. And again, we have to be very aware of all these mutations. This was really great, another great article published in Cell back in 2015. This is basically the Standup to Cancer Group. Funded heavily by the Prostate Cancer Foundation. Robinson and Charles Shores and these guys, and again, this is a wonderful article that I think everybody should read.

So I’m just going to wrap up. And so these are the challenges. All these different drugs, how do we find, how do we know who to treat, how do we know which drugs. David has convened another working group, this Radar Two, trying to come up with some semblance, some order on how we need to think about this. And this becomes very daunting. So here’s the prostate cell, we have to deal with the bone microenvironment. We can manipulate the immune system, which I’ll talk about later. You can also have angiogenesis. So all these things come into play. And again, they’re probably all in play. And we have to figure out a better way, liquid biopsy, CTC, cell-free DNA, tumor DNA, micro-RNA, all the things that we talk about in terms of monitoring, what’s the best way to sort of dictate how we manage these patients?

So I was telling I think Gerry Andriole, so this actually a really good book, I would suggest you read it. This is Mukherjee’s book. This was his followup. He’s a medical oncologist at Columbia. He wrote Emperor of all Maladies and 2011, which won the Pulitzer Prize. And this book actually is called the Gene. I think it came out the middle of last year. And he really talks about this whole concept of genomics, transcriptomics, proteomics, and epigenetics. And really what drove him is that—what drove him to write this book was his family history of schizophrenia. So he was trying to figure out, is there some sort of genetic profiling. So again, I would strongly encourage you to read this.

So in conclusion, the AR, the androgen receptor continues to be a majority driver in the growth and survival of prostate cancer cells, even in the CRPC patient. As urologists we really need to understand the nuances of resistance mechanisms and genomic alterations. And despite all the recent advances, there remains multiple challenges and opportunities for researchers to better understand the disease and possible development of novel targeted agents. Thanks so much.