Real World Treatment Experience with Radium-223 - Dan George

May 3, 2019

Dan George concentrated on the patient experience in the treatment of metastatic castration-resistant prostate cancer (mCRPC). He described a typical treatment timeline where a new patient presents with elevated PSA is initially treated with androgen deprivation therapy (ADP) to reduce testosterone feeding tumor growth. While the patient may initially respond with a reduced PSA, it is most common that the diseases will progress. The next line of treatment is usually with second-generation hormone therapy such as enzalutamide. The patient may show an initial improvement, but after about 11 months, the disease again progresses with increasing PSA and reduced quality of life. A similar timeline is observed with abiraterone (often combined with prednisone) instead of enzalutamide. 

There have been studies of the sequential application of abiraterone/prednisone or enzalutamide followed by the other drug. The studies show no additional benefit of using both drugs sequentially.

In the patient treatment timeline, there is a "gap" between the initial treatment of enzalutamide/abiraterone and the application of another treatment like the chemotherapy agent docetaxel. This gap is a natural place for treatment with Xofigo® (Radium-223 dichloride). Prostate cancer overwhelmingly metastasizes to the bone, and since radium is chemically similar to calcium, it is preferentially absorbed in rapidly growing bone tissue. Xofigo is effective in reducing tumor growth and relieving bone pain. It is well-tolerated in most patients. Dr. George noted that Xofigo doesn't have much effect on PSA or the androgen pathway. He felt that it was essential to explain this to patients who are otherwise puzzled that their PSA continues to rise even though the therapy is working.


Daniel George, MD, Professor of Medicine and Surgery, Divisions of Medical Oncology and Urology in the Duke University School of Medicine and leads the Duke Prostate and Urologic Cancer Center.

Read the Full Video Transcript

Daniel George: I just wanted to thank the organizers for hosting this luncheon talk here. And I'm your entertainment for the next hour, I'm Dr. Daniel George, I'm a medical oncologist from Duke University, and specialize in GU cancers and prostate cancer in particular. And it's a pleasure for me to present to you, for the next 45 minutes or so, the real world, real life... whatever that means... treatment experience of Xofigo®, in patients with metastatic castration-resistant prostate cancer.

For those of you who are not familiar with Xofigo, this is radium-223, it was FDA approved, really widely approved in 2013 for the treatment of patients with advanced, symptomatic bone metastatic prostate cancer, and we've been using this now for five and a half years, and here to share some insights from our practice, and from the data that's surfaced.

These are my disclosures. I work with a lot of companies, in particular, Bayer Pharmaceuticals, who are the makers of Xofigo, and I'm a consultant and speaker with them. We are on schedule, this is the introduction period, I think we're just about hit it and ready to get started. We're going to leave a little bit of time for Q&A at the end, so if you have questions, hold onto them. If it's a burning question that just has to come out, I'm okay with that. Just put your hand up, stand up, wave, and I'll try to clarify whatever that is.

So we'll try to keep this informal, hopefully informative, and a little interactive at the end. So we're going to start with just a recap of Xofigo in metastatic castration-resistant disease, and really the broader treatment armamentarium for this disease, how we treat patients with this, what we term metastatic castration-resistant prostate cancer. And for those of you who aren't clinicians, aren't oncologists or urologists, don't treat this type of patient on a regular basis, let me just walk through for you a slide here that broadly summarizes what we might consider the natural history, or the progression, of this disease.

And it's really, to me it's the tale of two diseases. This is one course. These are the patients that present with localized disease, and in the US, we do a lot of prostate cancer screening. However, you feel about that, it is the nature of our practice in the US. So we diagnosed most of our patients with relatively low volume disease, and they'll undergo local therapies with surgery or radiation therapy. And about two-thirds of those patients are cured. And about a third of those patients will relapse at some point, and it's typically by a rise in PSA. And PSA, for those of you that aren't familiar with that, that's prostate-specific antigen. It's a blood marker that we use, it's kind of a biomarker of the disease burden.

We use it as a screen, but it's really developed primarily as a biomarker of the disease burden. So as that PSA goes up, the disease burden goes up, and eventually, we'll start these patients on androgen deprivation therapy. This is hormonal therapy, lowering testosterone from the testicles in particular, and that results in a drop in the disease for the vast majority of patients, sometimes for months, many times for years. And we'll typically do this in patients who don't have the metastatic disease yet, so that's that blue box there, it says M0 or non-metastatic.

And those patients will stay "non-metastatic", minimal disease burden if you will, for a period of time. And then we'll have, again, another one of these biochemical relapses, that's that curve going up. That's our definition of castration-resistant. So their testosterone's low, that's the castrate level. The PSA's rising, that's the resistance level, and it's still prostate cancer. And then those patients over time will develop potentially symptoms, they'll develop metastasis, they'll go on a number of different systemic therapies. Sometimes in the non-metastatic or biochemical only castrate resistant setting, sometimes with metastatic disease, sometimes for symptoms.

We'll talk about metastatic patterns in the future. You see the dotted line on the top, that's the hormonal therapy not fully controlled. These patients are responsive, still, to newer or novel anti-hormonal therapies, and we'll show you some of those, but they're not cured. So this is a population of patients that we will be treating to prolong survival, but no longer to complete, indefinitely cure patients.

I mentioned this is a tale of two diseases, because there's another narrative that patients can follow, and that is, those patients that present with metastatic prostate cancer from the get-go. So there's a group of patients, and we might draw this curve out here to begin with, and it might come down somewhat but not all the way to zero, maybe it does go all the way to zero, but it comes back up again. It tends to have a little bit more of a rapid course, but for those patients who present with metastatic disease at the start, they may only be 5% of our overall patient population with prostate cancer, but they make up about a third of the patients that ultimately die from metastatic castration-resistant disease.

So they are much more enriched to develop this kind of progression over time, and usually over a shorter period of time. And we'll show you some of that data as well. But that's a group of patients that never undergo this localized treatments, or at least not initially, and they don't go through this period of M0, non-metastatic, biochemical relapse at all. They typically will have a higher burden of disease, and the higher the burden, the worse the prognosis.

So that's just the backdrop, in just five minutes, of prostate 101, that you need to know now as we talk about the armamentarium for castration-resistant prostate cancer. The other piece that you need to know is about prognosis, and we know a lot about prognosis in cancer. And some of that comes from different biomarkers, or different symptoms, or settings. Some of that can come from the patterns of metastasis, and prostate cancer is a bit unusual in that it has this high proclivity for bone. For reasons known and some unknown, prostate cancer really does have a high likelihood of spreading to the bone. And we'll see almost 90% of our patients with bone metastasis at some point, that develop this kind of metastatic castration-resistant disease.

There's 10% or so that have lymph node exclusive disease, and they have a much better prognosis. The group of patients that have bone disease, that present with high volume bone disease, historically have done very poorly. And historically, these patients have a median survival of maybe only two years, and very few of them living past five years. That's changed recently with the development of chemotherapy, and novel anti-hormonal therapies added to the mix in the upfront setting for these patients. But still, the prognosis is not great for these patients.

And then probably the less common but more worrisome group are the patients that have higher visceral content. Liver metastasis, or large lung metastasis, adrenal metastasis, other soft tissue sites of disease. These patients really do have a more aggressive phenotype, a more mixed phenotype, and tend to be less fully responsive to therapies. Sometimes they have different forms of prostate cancer during by neuroendocrine cells, or other phenotypes. So a lot of different biology going on, but the bulk of the patients we're going to talk about today are these patients, the bone dominant prostate cancer patients.

There's a lot of ways to mix the treatments and sequence the treatments in prostate cancer. We have five different classes of agents, not agents, but classes of agents, from chemotherapies to hormonal therapies, immunotherapies, bone-targeted radiopharmaceuticals, et cetera. A lot of different types of treatments for these patients. Sometimes multiple agents within each class is available, and so you can imagine that there is not one pattern that fits all patients. In addition, there are different physicians and physician groups that are comfortable with some treatments more than others. And the data doesn't really tell us. Most of the data is compared against one treatment or another, or even against placebo, and so in many cases, we're left really struggling with understanding what the right head-to-head comparisons are.

In some cases, aggressive cases, we've overlapped, or what we call layer therapies. And that is a pattern that may actually become more common as we have more and more therapies to offer these patients. But this sort of unclear guidance with head-to-head trials, I think really does create a lot of diversity in the practice patterns.

Xofigo is really unique in class. Not only is it unique because it's a radiopharmaceutical, but because it's an alpha therapy. Yes, that's right, you heard right, an alpha therapy. I don't have to tell you guys what that means. We're very excited because this is really one of the first settings where alpha therapy has shown in an unselected patient population survival benefit. I say unselected, but as you will see, it was a bone dominant patient population that we targeted. But we didn't select for a certain marker, we didn't select for a certain genetic alteration, this was really a late setting study, and showed a survival benefit.

Its mechanism, as you all know, is pretty unique because alpha particles have such a short range of radiation. And in this case, with radium-223, we're talking about probably 100 microns or less. But within that tumor microenvironment where radium, as a calcium mimetic, gets deposited, wherever there is bone being laid down. And again, one of the unique features of prostate cancer is this remarkable sclerosis that occurs in these bone dominant lesions. They tend to be highly sclerotic, and in that setting is where we'll see calcium deposition, and consequently radium deposition. And in that bone microenvironment we'll find not only prostate cancer cells, but osteoblasts, other stromal cells, and potentially other key subtypes of prostate cancer that alpha particles will kill indiscriminately.

And I think that's one of, really, the amazing aspects of this treatment mechanism, is the ability to kill cells agnostic to their phenotype. Whether they're neuroendocrine, or whether adenocarcinoma, whether they're osteoblasts, or clasts, stromal cells, mesenchymal cells, alpha particles will kill anything in its path, but for this very short, microenvironment setting.

It's remarkable to me, then, that we see an overall survival benefit with this treatment, but this is exactly what we see, and if you start at the bottom of this slide you'll see our schema for this, this was a 900 patient trial, which at the time, for prostate cancer, that's a big trial. Patients had to have two or more bone metastasis, no visceral disease known. And lymphadenopathy was allowed because again, that was less of a prognostic driver in this field. So bone or bone and lymph node disease, stratified by prior docetaxel chemotherapy. That's one of our mainstay chemotherapies at the time.

And patients were also stratified by whether or not they used a bone health agent, a bisphosphonate, and also by their baseline alkaline phosphatase, a bone turnover marker, which also has independent prognostic significance in this disease. So stratifying this population around prognosis, recognizing bone dominant patients, randomizing them to Xofigo, and this was given on a once every four week intervals. Just a small, one minute injection through a peripheral IV six times, or a placebo saline, with best standard of care support.

So it allowed for other therapies to be layered on top of that as indicated, but not chemotherapy, specifically. What you can see here, I can just go back for a second, on this curve, this is the treatment period. This is the six doses, five month treatment period, and you see these curves separate. And this Kaplan–Meier, the one on top has a longer survival, and you can see throughout this entire population, from the first six month split of these curves all the way through to 24 months or more, we're seeing a separation of these curves. At the median, this is represented about 3.6 months improvement in overall survival, but as a hazard ratio, how we like to this of this, how much overall benefit did we see? We see a 30% reduction in the risk of death associated with this treatment across the population.

We see that benefit persisting all the way out two years or more. It's really a remarkable sustained benefit, in this setting of a therapy that is so really specifically targeted to the bone microenvironment. Going along with that is toxicity. And it's important to recognize with any efficacy, what are we taking on for added side effects? It's interesting that overall, we actually saw greater rates of what we call grade three or grade four. These are the serious adverse events, complications that we'll stop therapy or admit people to the hospital for, were actually higher in the placebo group.

That's really driven by disease progression, and I'll just remark... sorry to keep going backward. The control arm here, median survival was less than a year. These were patients that were really at their last year of life on average, and so not surprisingly we saw a lot of those complications occurring on the study. But what we didn't see with radium-223 was a lot of other complications related to treatment. So you see overall the toxicity rates here for again, those higher grade, grade three, grade four rates, of 6% or less. Thrombocytopenia, not surprisingly a bone marrow toxicity was 6%, everything else was really 1% or 2% or less, in terms of serious adverse events.

And even the less serious all grade things were relatively rare. The events we saw with GI toxicities were similar, maybe a little bit higher than placebo, but similar. Again, this is a patient population that's symptomatic. And then overall, our cytopenias were relatively low even looking at lower grades.

If you look at the quality of life we have a lot of different quality of life tools that we look at, and depending on the tool we can see different levels of sensitivity. But overall for our general tools, our EQ-5D and our FACT-P for prostates and more general health symptom scores, we saw an overall improvement in more patients on radium, on Xofigo, than on patients who got a placebo. So this is looking at who actually improved on the study. I've shown you who got toxicities and who died, this is actually who actually felt better over the course of time, and a higher percentage of patients on the radium Xofigo arm actually felt better.

So again, I mentioned to you the two different phenotypes of prostate cancer patients presenting. On the top, we have those patients diagnosed with progression from metastatic hormone-sensitive disease, those patients presenting with stage four disease. And our typical treatments for those patients are these novel hormonal therapies, hormonal therapy alone, combined with those novel hormonal therapies, or chemotherapy. And when those patients progress into the castration-resistant setting, they have a number of treatment options, and Xofigo is one of those. For those patients that start out with non-metastatic castration-resistant disease, they can be treated with novel hormonal therapies, or ADT alone, typically not with chemotherapy.

So they'll come into that castration-resistant metastatic disease setting where we would consider chemotherapy as an option for some of these patients. But as you can see with some of the novel hormonal therapies and chemotherapies already being used in this setting, Xofigo is a reasonable choice to be used a little bit earlier in that disease course. For these patients, we might opt for some of these other approaches before moving on to Xofigo. So you can see how sequences might vary, not just based on the physician and the practice, but also based upon how this patient was previously treated.

Couple of things we know about these sequencing come from some of the responses to therapies after clinical trial. So this was a trial at the top where they looked at chemotherapy response after one of those novel anti-hormonal therapies, abiraterone, and you can see that there's a significant percentage of patients that show a PSA decline. I mentioned PSA is a marker of disease progression, of growth and progression. It can also be a marker of disease response. It's not necessarily the only marker of disease response, we'll look at, sometimes, radiographic responses. Sometimes we'll look at the time to disease progression as a marker of clinical benefit. But another way of looking at clinical benefit is at a decline in PSA, particularly when we're looking at agents that target the hormonal pathways, like these novel hormonal agents, and even chemotherapy.

And you can see that with chemotherapy, a sizable amount of patients showing some decline in PSA. What's frustrating for us is that in practice, a lot of people will sequence one hormonal therapy after another, and our data rarely shows a clinical benefit to that. Most of those patients are having PSA progression through that, and if that were the only measure of disease progression we could live with that if there are other signs of clinical benefit, but there aren't. There are no studies that show a benefit, a survival benefit, to one novel hormonal therapy after another. And yet this kind of practice still is very persistent in many of our clinical settings.

If you look again at another measure of that PSA progression, time to PSA progression, it was modestly good here in this population of patients that got docetaxel following abiraterone. It took six months or more for the majority of those patients to progress, 7.6 months on average. 2.6 months for those patients getting abiraterone after enzalutamide, so again, a really, really short time. Not just no sign of PSA decline, but actual PSA progression, and many of these have also been shown to be markers for shorter overall survival in this population. So concerning to us that this practice pattern still exists out there, despite the lack of data to support it.

So where does Xofigo fit into this population? I think that's one of the things that a lot of physicians have struggled with because it's got a little bit of a different mechanism of action. Because it's not something that they necessarily prescribe themselves, it may be a radiation oncologist, or a nuclear medicine physician. But if you look back at our ALSYMPCA study, docetaxel was given in about 60% of those patients previously, but 40% of the patients were chemotherapy-naïve, and what they found was in those patients, placebo versus Xofigo, those that went on to get chemotherapy actually had a longer survival if they had prior Xofigo.

So it does look like we are seeing some benefits to this sequence of Xofigo following chemotherapy. 22% of the ALSYMPCA patients went on to receive chemotherapy post-Xofigo, and overall, although we don't have a lot of data from that study, chemotherapy following Xofigo appears to be well tolerated in other, smaller clinical settings. So overall, we're seeing a relatively long time from the time on study to the start of chemotherapy.

Here's what's happening now. So when we look at practices out there, patients are being primarily treated with these novel anti-hormonal therapies, things like enzalutamide or abiraterone. We're seeing PSA declines, and then progression. We're seeing functional declines in some of these patients with radiographic progressions, and then we're seeing chemotherapy given late in this course of disease. I'll revisit this later. But there's a window, a gap here, that could be as long as a year or more, where again, we're sort of looking for treatments to help slow the disease progression, prolong survival for this population, while they're having perhaps a more insidious decline in their functional status and other signs of disease progression.

That to us is sort of the window for radiopharmaceuticals like Xofigo to target this disease, particularly bone-targeted, bone dominant patients. So if you look at our treatment armamentarium overall, Xofigo's got a clear indication for these patients with multiple symptomatic bone disease, no visceral disease. Its shown a safe, reliable safety signal for these patients. There's overall survival benefit, there's quality of life benefit. This is a therapy that we've now used in our practice in a number of settings, but particularly in these patients that have the early symptoms associated with this disease showing benefit. And we'll get now into what we initially started with, the real world use of Xofigo.

So what's happened over the last five and a half years since that approval? Well, a couple of things are concerning with prostate cancer today. This is looking at now some real-world data, this is from the FLATIRON Health Registry. This is a large database collected retrospectively from electronic health records across the United States in different practices and health systems. And they've looked at a number of prostate cancer patients that meet the definition of metastatic castration resistant prostate cancer. And what's concerning is that only about 50% of patients are making it to second line, and if you look at those patients, again, a smaller percentage are making it out to the third line or so.

Patients who have died without receiving subsequent lines of therapy, 23% here, following just one line of therapy, another 20% following two lines of therapy. So a lot of these patients are going on to receive what we consider life-prolonging therapies. Again, a novel hormonal therapy following a novel hormonal therapy isn't life-prolonging. You get one overall survival benefit from that strategy. But Xofigo would be, but if people aren't using it in these settings, then we're not seeing those necessarily additive benefits associated with these mechanisms.

But it's still a practice, I think, that we're still understanding how best to do that, and to do it in a safe way. It's interesting if you look at... oh, sorry... enzalutamide, and what's used after that? Again, most commonly we see abiraterone, but again, some combinations. Either Xofigo in combination, or docetaxel. And then those patients receiving docetaxel in that first line setting, again, more of that symptomatic high-volume patients, again, enzalutamide, abiraterone are common in chemotherapy. With cabazitaxel, a second chemotherapy is common. You don't see radium use as much here as we do in the other settings. So interesting to see where radium is being used in these different sequences.

If you look here in our real world experiences, that ALSYMPCA study that I showed you was done at a time before we had these novel hormonal therapies like abiraterone, enzalutamide. If we look again at this FLATIRON data, it's encouraging to us to see that following abiraterone or enzalutamide, patients treated with Xofigo are having a prolonged survival associated with that approach, regardless of whether or not they've got prior abi or enza. So it doesn't look like that we've missed a population of patients here who are not benefiting from Xofigo, and it makes sense. That ALSYMPCA study was done at a time where older hormonal agents were used, but mechanistically they still targeted the androgen receptor pathway.

And Xofigo showed a survival benefit there, it's not surprising to us that it shows a survival benefit as well following abi or enza. This is not a repeated phase three trial, but this just looking at retrospective real world data to suggest that that benefit is holding up in this setting. What's probably most concerning to us is how much Xofigo patients are getting, and for those patients that are getting four or more cycles of Xofigo, these are the patients that are able, not just to tolerate repeated doses, but are clinically stable enough with their disease to go on to month, after month, after month, after month Xofigo treatment, they have the survival benefits of 20 months or so.

If you use this agent in a patient population that is so rapidly deteriorating that they don't have time to get multiple injections, more than four, four or more, that survival is much shorter. So somewhat self-fulfilling to say this, but it's also, I think, important to recognize that for these patients, we're probably too late, and the window to treat with an agent like this needs to be in a population that's likely to benefit, likely to be stable enough to benefit from this agent.

I mentioned layering of therapies before. So at ESMO, one of our European Medical Oncology meetings this past fall, there was a presentation on the ERA-223 data. This is up here. A study that was done in patients with metastatic castration-resistant disease, starting on abiraterone, one of those novel hormonal therapies, and either being randomized to just placebo or combination of abiraterone plus Xofigo. And in that population of patients, we didn't see any difference in the survival benefit or time to progression, but we did see a difference in fractures in these patients. Many of these, their bone fractures, many of them pathologic fractures, some of them may not be, but that fracture rate was higher in the combination arm, which was really alarming. And it's led to a warning in the FDA label associated with concurrent use.

So we went back to that FLATIRON database and looked to see, was there any real world data that would substantiate the timing of Xofigo? Patients who were maybe started first on abiraterone, and later given Xofigo, versus those that started the two drugs concomitantly. So we looked at it, we called one layering, if you started it more than a month one after the other, or if you started them truly concurrently, we called that concurrent. And we actually did, we saw greater rates of fracture, we saw greater rates of complications and shorter overall survival associated with the concurrent versus the layered.

So it does seem that timing is really important. The other thing that came out of that was that the patients receiving bone strengthening agents, things like bisphosphonates, or denosumab, a rank ligand inhibitor, but that we actually saw a lower rate of these symptomatic skeletal events, particularly fractures in the need for radiation in that population. So we've learned from that study, and we have an ongoing study in Europe called the PEACE 3 study, which is now looking at a very similar patient population starting on a novel hormonal therapy, in this case enzalutamide, and then adding radium. But there's a little bit of a delay, so it's more of layering than a concomitant, so we've learned that is important. And these patients are required to have a bone health agent to help prevent fractures, and so we've learned from that.

And hopefully this study now will give us the opportunity to see, is there actually a benefit? Can we actually identify a patient population that gets a survival benefit associated with this up-front use of radium? So I think it's a really important study, it's important for us to understand is it safe to layer these therapies. If we think it's not safe to start them at the same time, how much of separation do we need, and can we identify these patients at risk? This will be a very informative study for us, but it was eye opening experience, and it was great to be able to learn from that ERA-223 study, and be able to change our practices accordingly.

Just a couple other things we've learned in the real world setting and one of them is that PSA doesn't seem to be affected by radium, by Xofigo. This is the ROTOR Registry, it's a Danish registry, 300 patients that were followed with on use of the standard of care radium, and we see a PSA decline in about 3% of patients. The vast majority of patients showing some increases in PSA. And I just showed you earlier that one hormonal therapy after another didn't really decline the PSA, and this is similar to that. What's different is we see this marker here, alkaline phosphatase, a bone turnover marker, significantly decreased in the majority of patients receiving radium. So it is having a treatment effect that we can measure, it just happens to be by a different marker than something driven through the androgen receptor.

So I think in summary here, we'd say that right now in our practices in the United States, most of our patients are not going on from one life prolonging agent to another, and that would be from a hormonal agent to chemotherapy or a hormonal agent to a Xofigo. Instead, we're seeing sequencing of hormonal agents, and then maybe no treatment at all in these patients. Xofigo is used both as monotherapy, but it's also been used in practice in combinations, and I think we have to recognize the safety of that with some future studies.

Real world settings, real world survival, seem to be comparable to what we would expect now, even in the setting of abiraterone, enzalutamide. So it looks like that survival benefit is still there, associated with this. And again, it looks like we're seeing benefit associated with, not just with overall survival but quality of life, and overall good safety profile.

So what is it like for patients? When you really get down to it, how are our patients looking at this? So I want to go back to that slide I showed earlier, this is just data from two different phase three studies. International studies, US, Canada, Europe, and others. This is the PREVAIL study that led to the FDA approval of enzalutamide, and this is the Cougar-302 study. It's interesting to me, and mea culpa on this one, because we were part of both of these studies, and part of these practice patterns. But these patients on average, this is just looking at the patients treated with enzalutamide, on average their PSAs began to rise around 11 months, and so too do their quality of life score start to decrease.

So patients reporting on their quality of life scores a decrease, and their overall well being, and their PSA is rising. We kept them on study, as per protocol until they had radiographic progression, typically about six months later, and they didn't get chemotherapy. Radium wasn't available at this time, for another year plus. On average chemotherapy was given out here at 28 months, average overall survival, 32 months. Chemotherapy was given late. Now maybe the chemotherapy patients lived longer than the average patients, so don't necessarily look at this as four months, but it's just showing you this gap, here, between the time of disease progression on enzalutamide, this is not placebo patients. These are the enza patients.

And what else were they receiving if they weren't getting chemotherapy? Probably abiraterone and other drugs like it that we haven't really shown beneficial in that setting. And then the same thing in the Cougar-302 study. Here, actually, we also have the added marker of ECOG performance test. This is the physician's grade, our independent assessment of their functional status decline on average, happening about the same time the PSA is going up, as the patients are reporting their functional status decreasing. And what are we doing about it? Well, we're treating them until radiographic progression, and then no cytotoxic therapy for another year or so.

So this is really the missed opportunity, the gap, and to me this is where radium falls now, Xofigo falls in our treatment patterns for these patients. We showed the quality of life improvement, but to me what's probably more relevant is the decline in quality of life, because we know that on the placebo arm, their median survival was 11 months. These are the patients that we want to really impact overall, and what we can see is their median time to skeletal event is about 10 months. We can see almost a 50% improvement or more in that quality, in that time to that symptomatic skeletal event, that bone fracture or need for radiation. But more importantly, we can also show you for the general population, a slower decline in their quality of life associated with radium.

I'm not saying this is sort of putting this disease back into a complete remission, but these patients were living longer, and they're living better during that treatment period, and well beyond. So it really suggests that there is, again, sort of clinical benefit in both settings.

How do I talk to patients about radium? This is probably one of the really difficult discussions with patients. Not because the therapy is hard to explain, or because it's difficult for them to tolerate, but it's hard sometimes for them to grasp the fact that this is not going to lower their PSA. And if you look at our median time on the ALSYMPCA study to PSA progression, it was exactly the same placebo and Xofigo. And yet we see this overall hazard ratio improvement in survival. And I explain that this medicine is not going to target that same androgen pathway that's driving the PSA, but at the end of the day it's really important for patients to understand what does that mean in terms of them?

So one of the things I try to do is put it in practical terms. And what I mean by that is, if I've got a patient that's been on a hormonal agent, like abiraterone or enzalutamide, and say their PSA's been rising, now they've got radiographic progression, and in the last two months, their PSA has gone from 50 to 100, in two months. We're stopping that hormonal agent, we're putting them on Xofigo, and we're going to treat them for six months. I tell that patient when we finish treating it, their PSA's going to be 800. And they look at me, kind of like you all are looking at me like I just lost a little bit of credibility, I get it, but the reality is that this is really important, is that setting expectations.

They kind of shake their heads and leave, they don't say too much about it, but they'll come back and see me at the end of that treatment, and guess what, they'll be all smiles. And I'll say, what are you smiling about? They say, "Well I feel great, and guess what? My PSA's only 500. You were wrong." And it's a great way to be wrong, it's important for them to recognize that PSA isn't the end all, and in fact, yeah, we do see this leveling off or slowing of PSA. I can't really say that that's a clinical benefit in and of itself, but it helps the patients cope and live with those expectations.

So setting expectations around what might happen in the future with this marker is important. It's important to know that regardless of that rise in PSA, there's a survival benefit associated with this. I'll leave you with a picture, this was a patient of mine that was hanging on and waiting for Xofigo. It was June of 2013, it had just gotten approved a month before. You can see under June 2013, his bone scan, it's literally covered throughout his pelvis here and in spine, ribs and even his appendicular spine, his arms, and his legs.

He had been through chemotherapy, novel hormonal therapy, he was dropping some of his blood counts with pain and fatigue, and we put him on Xofigo. He actually got a pain flare, but we were able to manage through that, and then over the course, he was able to kind of stabilize. We did have to transfuse him, red blood cells, but his platelet count was stable, his white count was stable, and he finished therapy in February of 2014, and he lived another 15 months. We didn't treat him with anything else afterwards. His alk phos declined like a champ, we were super glad to see that go down. We didn't really see much with the PSA, it kind of went up, it kind of went down, it kind of leveled off. It always stayed around that... this scale was a little bit higher, it said seven, but it was actually about 700. He stayed in that 500 to 700 range for PSA.

So it's an example for me of somebody who had a modest marker effect, a pretty dramatic change in his scan effects, and if you looked at his bone scan, you would just see white throughout all these areas. What we had done is sort of sclerosed a lot of that marrow, and it was no longer taking up technetium. And it was like that for a long time for this patient, so a pretty dramatic treatment effect. I don't want this to leave you with, this happens in everybody. We see a lot of different patterns, and we really don't know how to interpret the bone scans in these patients subsequently, but I do know that in these patients that we stabilize this for a long time, that was clearly a life-prolonging treatment effect.

Overall, a couple of other questions that come up from our patients, they're worried about how they're going to feel on Xofigo treatment, and we again, go over that side effect profile, but overall, our patients' worry and preconceived notions of what those are going to be like is actually worse that what most of their experiences are. How do we monitor these patients? I get repeat scans after three doses. I'm looking for soft tissue progression, particularly visceral progression, but I'm also re-gauging that bone scan to see if there's a mixed response, what that will look like after six.

How often do I see these patients? My nurse practitioner will see the patients every time they come in, I may see them at the end of the six doses. I may see them halfway through to look at those scans. Depends on how they're doing. We recognize that a lot of these patients are symptomatic, they're going to need symptom management, so that's the need for the visits, and again, that's why some of our advanced practitioners do a great job with this. They want to know about lifestyle. A lot of these patients, it's really about quality of life as much as quantity of life, and I think that's where some of those quality of life measures are really helpful, to help them understand that they may actually improve during this time, but if they don't, we think that the decline will be slower with this treatment.

So I think to end on this experience overall, we stress the overall survival benefit. Therapies that are new in class, different from what they've had before, particularly in these bone dominant patients, can extend survival. Quality of life is super important for these patients, and we talk a lot about that, and then helping them disassociate the biomarkers is the other key message. So with that, I'll stop there, and take a few questions, thank you very much. Is there a question in the back?

Question 1: Thank you for the wonderful talk. Do you know if in the clinic anybody has tried to inject mesenchymal stem cells following the therapy? And if it offered any benefits to the patients.

Daniel George: So if I could repeat your question, the question was, has anybody tried offering mesenchymal stem cells to patients with prostate cancer, following Xofigo treatment, yeah. So stem cell transplant has been tried in prostate cancer in the past. There was a wave in oncology around breast cancer in particular, where stem cell transplant was really the craze, in order to give more chemotherapy. And that actually turned out in subsequent randomized trials not to be beneficial. So the field has moved away from that approach as a way of resuscitating marrow in these patients.

I haven't heard specifically about that being revisited, my concern would be with a lot of sclerosis that we're causing that it may not necessarily take. The patient case I showed you to me in some ways it's a remarkable case, and it really validated for me the clinical benefit of this approach. But it was also a case of concern, because when we wait on this therapy to that point of disease... And again, he had no visceral disease, it was all bone, but it was so encompassing of his marrow that it was really to me kind of late. If I had my druthers I would have used the therapy six months or a year earlier in him, it just wasn't available. But it's a great question in ways that, how can we help support complications associated with this therapy, and to me the best way to do that is to use it earlier, before we're at risk for that real marrow suppression.

Pretty good lunch, I can tell, everyone's a little quiet.

Question 2: Does a rise in PSA predict overall survival?

Daniel George: I heard a voice. Where's that? I'm sorry. I'm sorry, I couldn't see you there. Does a rise in PSA predict...

Question 2 cont’d: Overall survival benefit.

Daniel George: Overall survival benefit. So what was...

Question 3: And the second question is, if consecutive hormone therapy is not effective, why do clinicians continue to give it?

Daniel George: So that's a really good question too. Let me answer the first one, because the first one's a little bit complicated. In some settings, a rise in PSA is a pretty good marker for disease progression. And as I showed you in those two studies, with enzalutamide and with abiraterone, the radiographic progression followed the PSA progression. The PSA progression happened first, and the radiographic progression happened second, and that was fairly reliable in those cases. But those are agents that specifically target the androgen receptor, and PSA is driven through that androgen receptor activity. So it stands to reason that it would work like that.

With docetaxel chemotherapy, it's not as clear that the PSA progression correlates with the overall survivor. So that's one where it's a little bit murky. There is some suggestion, but it's not as strong. And then with agents like radium, or agents like our immunotherapies, like Sipuleucel-T, there's really been no correlation between the PSA, and the overall survival benefit. So to answer your question on PSA, it depends, kind of like what I say to my kids. Maybe. It just depends on the context of the therapy that you're doing it in.

Now, the second question is really, I think, a great question. And again, why do physicians use therapies one after the other when there's no data to show a survival benefit, and when other agents are available? And it'd be easy for me to say, "Shame on all those physicians, they are so wrong," but I've done it too, so I can't really say that in good faith. So why would I do it? Why do I do that? And the reason is, again, it's a little bit complicated.

Sometimes it depends on tumor volume. If these patients have really, really small tumor burden, and they're truly asymptomatic, I may think that the use of cytotoxic chemotherapy might be too early, so using a hormonal agent is attractive. Again, we don't have data there, but sometimes our progression on hormonal therapy may be incomplete, meaning the PSA is rising slowly, or there's maybe one or two spots of radiographic progression, but it's not that convincing. And maybe they've responded to the therapy for a really long period of time.

So not all progressions are the same, not all tumor burdens are the same, not all patients are the same. Some of these patients might not be able to tolerate some of these agents, some of them may have lymph node only disease. So there really isn't one size fits all. The problem is, is that I think physicians like one size fits all, so in the community, if you're an oncologist, and you treat all different types of cancers, you're kind of looking to our guidelines to say, "What should I do next?" And a practice pattern might emerge where you're just going to do that each time. And that's where we run into trouble.

So I think physicians who treat prostate cancer exclusively, or predominantly, get this nuance of when it might be appropriate and when it isn't, but for those in settings where they're not dealing with it all the time, it's more tempting to have one pattern that you follow. And so that's the slippery slope. And again, I'm not judging because I've done it, but I think it really, to each of those questions, it really depends on the setting. I'm sorry that's kind of a wishy-washy answer. Yeah.

Question 4: Really a nice presentation, thank you. How confident are you to suggest that Xofigo can prolong survival in the post-NADs era? Because on your slide where you showed for both NADs the progression without Xofigo it was about 34, 35 months overall survival, and you showed the time point when you would maybe suggest to use Xofigo, which was at about the 16 months time point.

Daniel George: Yes. Yes.

Question 4 cont’d: So 16 plus 20, in the best case, if they can receive all the doses, it's about... no? Wouldn't be that much different.

Daniel George: No, it's a fair question, and I think ideally, we would redo that ALSYMPCA study to answer that question, but ethically it would be hard to give people placebo in that setting now, number one. Number two, you could think about it in terms of maybe just doing what we had talked about, crossing over from one hormonal agent to another, if we don't think there's a survival benefit, but there's still an attraction. And then maybe having Xofigo in the other arm. The real issue is, this comes up with all of our agents. Docetaxel chemotherapy was FDA approved prior to the development of these novel agents. Almost any solid cancer you think about, from lung cancer to colon cancer, as we develop new agents in the frontline setting, all of our subsequent therapies come into question.

So this is not an unusual problem to have, and it's kind of impossible to keep repeating all of our data. It really comes down to the mechanism of action, and rationale. Is there a reason to think that radium wouldn't be effective, mechanistically, in patients who had had, say, abiraterone instead of, say, ketoconazole, an older androgen metabolism targeted strategy. And there really isn't, mechanistically, a reason to think that, but that all said, there is a change in phenotype when you use these novel hormonal agents, that may be different than what was studied.

So it's really important for us to be open minded regarding this, and as we develop more data with Xofigo in these settings, and we will. And there are newer phase three studies going on, it's important for us to be able to reevaluate that exact question, and say, "Is this still consistent with a benefit, a clinical benefit, a survival benefit, a drop in alk phos benefit," as we'd seen. Are we seeing the same patterns of response that we'd seen in the ALSYMPCA study? So I think those are the things that give me confidence when we still see the AHOs dropping, and we still see radiographic stability. Whether it be on an individual case basis, like my patient, or whether it be on a large population study. 

But it's a very valid point. The data we show is really not fully answering it, and to answer your question, how confident am I? I'm pretty confident, because of the mechanism, because of what I've seen, but I can't tell you that I'm as confident as I am with ALSYMPCA, that this drug in the pre-novel hormonal agents improves survival. So confident enough to say, "Yeah, I want my patients to get it," but not confident enough to say we shouldn't keep reevaluating these questions.

Okay, well, since I'm the moderator as well, I'm going to close this session. Thank you guys very much for your attendance and everything, and I hope you have a great rest of the conference.