Prof Eleni Efstathiou - MD Anderson Cancer Center, Houston, USA

Prof Charles Ryan - University of California, San Francisco, USA

Dr Heather Payne - University College Hospital, London, UK

EE: Hello. We’re here in Orlando and we’ve been actually attending the GU ASCO meeting and we’re just finishing wrapping up the prostate cancer sessions. We’re actually on the second day of the meeting and we thought it would be a good idea to actually recap and share some of the thoughts that have come to mind after the presentation of several of the abstracts. I’m here with Heather Payne, Heather, I would call you a good friend, I hope, clinical oncologist. We actually share some patients. From London, she’s a Professor in the University College of London. And, of course, Chuck Ryan, he doesn’t need any introduction either. Chuck Ryan is from the University of California, he’s a Professor of Medicine as well. We’re all in the medical oncology field but Heather, on top of it, is also a radiation oncologist. So we’re really wanting her insight today given what was presented during this meeting. It was a diverse meeting, I think we got some information across the stages of the disease. There were several posters that didn’t really get the chance to be presented at the podium but were probably not worthy. So, Chuck, you actually during our discussion mentioned something we got ..?

CR: Sure, I mentioned a couple of posters that were important, we thought, for the general public. One had to do with a potential for an increased risk of high risk prostate cancer associated with a lack of PSA screening in the United States. I think it’s a little controversial and it’s a little early but it was nevertheless an analysis from the SEER and other databases presented by an author from the City of Hope that demonstrated that there may be an inflection, there may be an increased number of patients in the United States, who are being diagnosed with high and intermediate risk prostate cancer and that this inflection may have occurred after the time when the US Preventative Services Taskforce recommended against PSA screening. So far from settling the issue it just makes it important that we continue to monitor the incidence of localised prostate cancer and the risk factors associated with it.

EE: So actually this is an important point from you, coming from Europe, I’m just between the two continents and I will not comment on it. But would you like to actually take a stab at that? What is your feeling?

HP: I think obviously living in a society where we’ve never had prostate cancer screening it’s quite interesting because it’s like ripples in a pond, that you see a man who has prostate cancer and more and more men are coming forward in the UK because of charities and because of their experiences with members of families and friends to get tested. So I find that fascinating because it’s something that I know that many groups in the UK are working very hard to have a proper screening programme and obviously this is…

CR: Yes, so we’ll have to wait and see. It’s a little early to know what that means. Then the other factor that came across related to localised disease was the issue of active surveillance in an intermediate risk cohort and this is an area of controversy. I think there’s a relative consensus building that many patients with low risk disease do very well on active surveillance and I think it’s less controversial than it used to be. However, intermediate risk disease, PSA is greater than 10, Gleason 7 etc., there’s a very interesting presentation from Dr Loblaw from Toronto where they looked at their database over the course of fifteen or so years and found that there was a substantial proportion of men in the intermediate risk cohort who went on active surveillance who subsequently died of prostate cancer. So, again, it makes us think twice about this group and it calls for more research on molecular markers or other things that might tell us what are the differences in potential outcomes for patients with intermediate risk localised disease.
HP: They’re a huge group, aren’t they, intermediate risk?

CR: Yes.

EE: Actually, I’m sorry I interrupted you but you just really led us to the next topic – the presentation of the 4Kscore. That was something that created a lot of discussion following the meeting. It seems like a low-hanging fruit, an easily accessed signature combining the PSA readouts, the digital rectal exam and some other, phenotypic mainly, characteristics. So this is an old story that’s coming now, probably, to fruition but what would you anticipate to come off it?

CR: Well, as with every intervention or test or anything, it needs validation and a big study and a definitive endpoint. The question always in localised disease is what are we going to look at for the definition of clinical benefit? Is it reduction in the risk of death from the disease? Well that’s a long-term proposal that’s going to take many, many millions of dollars to test and many, many centres. Or is it decrease in the burden of treatment etc.? So it’s interesting, we continue to hash away at these very interesting issues related to localised disease.

EE: And your thoughts, do you think that it actually performs well? I wasn’t really happy with the performance of the test.

HP: It needs further investigation, doesn’t it, really -

EE: Probably enhancement with better molecular predictors in circulation or what have you. So actually that brings us to your domain – the intermediate risk disease. There was a little bit of a misunderstanding while the presentations were going on whether we were actually being presented with data that were referring to high risk disease or more so intermediate disease, localised of course, when it came to radiation therapy treatments. We saw results from three different studies, some more interesting than others. Some, as we discussed earlier, addressing more minor issues but of course of importance. So what would be your expert opinion in what was presented?

HP: I think firstly it was great to see radiotherapy up on the stage, to have that data presented.

EE: A lot of it.

HP: A lot of it. I think it brings several principles to light really. One is that we need better surrogate outcomes for radiotherapy and especially if you’re looking at low and intermediate risk disease. In fact, some of the patients in all of those studies would maybe now have active surveillance and so looking for small differences you’re going to have to follow them up a long time. To me, I think it was a shame that some of those studies weren’t looking at the real high risk patients because they are the group who most likely dose escalation is going to make the biggest difference because they die from their disease. They’re more likely to have extracapsular extension; they’re more likely to have microscopic metastases. But I think the take home message from all three papers were that you do need multi-modality therapy. I think it was a great relief for all of us who have given hormones with radiotherapy for intermediate and high risk disease for many years to see that that study did actually show that was the best group. I think we haven’t heard the end of dose escalation and I think for those high risk patients brachytherapy plus external beam is an attractive option. I agreed with Dr Martinez that high dose rate brachytherapy has got a big part to play as dose escalation, that’s what my standard of practice is – HDR boost, good doses of arcing therapy or IMRT to the pelvic nodes for high risk disease in combination with hormones. The big question is where surgery fits in as well in that setting.

EE: You actually brought up a great point, that urologists were quite happy to see the data because they actually think that they can, if needed be, intervene more after brachytherapy but of course they’re waiting for their data still. What are your thoughts?

CR: Yes, it’s control of the primary and even in the medical oncology circles we’re beginning to talk, and I know you are thinking about this at your centre as well, even in patients with oligometastatic disease treatment with radiotherapy or even radical prostatectomy, while it might have seemed crazy previously, there is now enough data emerging to say we need to rethink this and maybe this is a situation where we can, by controlling the primary, we’re going to affect outcomes positively across the board. Maybe there’s a subset of patients who even present with locally advanced, even small amounts of metastatic disease, who we can cure. Still very investigational but you hear this kind of conversation happening around data like this.

HP: And you’ve got to be pushing it forward, haven’t you, for these guys. The STAMPEDE study now, we’re irradiating randomised prostate radiotherapy versus no radiotherapy for people with multiple metastases to see if that’s going to have an impact on future bone mets.

EE: And having had a discussion with you earlier, you did bring up the fact that we have to wait for such a long time. You just told me that you’re getting proton therapy in London Hospital.

HP: We are.

EE: We’ve been waiting for ten years for these results; do you think that we’ll find some surrogate endpoints to use for these specific treatments? And this is outdated technology in some of the presentations used, as you’re aware.

HP: Absolutely. But I think the group who benefit most are the high risk group and they are the ones where you will see five year differences. They are the patients who are going to relapse a lot earlier because they have such a high risk of microscopic metastases and you stop their adjuvant hormones a lot of them are going to relapse very quickly. Looking at the Bolla data there was a difference of five years, wasn’t there, and all the big radiation and hormone data you saw earlier good outcomes from them. In this group PSA is actually much more important than in probably the castrate resistant patients because PSA recurrence does mean something after radiotherapy and hormones. So biochemical progression is something that we can look at in this group as a very valuable endpoint.

EE: That’s a great point. Before we move forward I wanted both your opinions on Anthony D'Amico’s presentation. During his presentation he went a little bit back and forth on the issue of androgen deprivation for high risk disease, presenting a little bit controversial data on whether it should be implemented more so or whether these cases would be more resistant eventually. We haven’t even brought in, as you said earlier, neuroendocrine features, even proliferation indexes. Where do you think the field is going to move? Because the unmet need, as you pointed out, is high risk locally advanced disease right now.

CR: I think what I took from Dr D’Amico’s talk is that the data on eighteen months of hormonal therapy is suggestive and it’s moving very, very close to looking like it’s definitive, is it fair to put it that way? He was saying it just needs a little bit more maturity or something, was his term to say, before we can consider this the definitive approach. This concept of whether longer hormonal therapy might create a more resistance phenotype down the road has always been there and we just don’t know. A couple of issues around that are one is we’re learning now a lot about what that resistant phenotype is, I just don’t know how we can apply it to this conversation about localised, locally advanced disease. Also, the trials are so complicated by all of the subsequent therapy that patients get, which is something we see in metastatic disease, and make it difficult to analyse some of the long-term outcomes of these studies at this point in time.

HP: I feel really strongly about adjuvant hormones with radiotherapy. Some of the early, the 8531, the Bolla study, 9202, they were before lots of further treatment so I think you’re seeing purer data in that. Also you had an overall survival benefit.

CR: In multiple studies, yes.

HP: At ten years, yes. So an overall survival benefit of ten years has to be the best thing to do. OK, some men are going to relapse but if you’re actually curing a group of people then I think you have to go. If eighteen months is now as good as indefinite versus three years versus two years, that’s fantastic because obviously you don’t want to give them all the morbidities of hormone therapy if there’s no need. But you do have to have adjuvant treatment; you do need multi-modality therapy.

EE: Moving on to the negative studies of the meeting, and there were several, that were presented and in the same disease group, localised, locally advanced disease. The Scandinavian group presented the final analysis of their bicalutamide 150mg versus placebo. Of course it was a negative trial and we also discussed it earlier as probably anticipated. What’s, in general, your view moving forward with the issue of bicalutamide and what else is there to expect out of bicalutamide?

CR: I don’t offer bicalutamide as monotherapy for somebody with locally advanced disease, it’s never been something that I’ve even thought of and that was based on the data from fifteen years ago or twelve years ago which showed an adverse outcome in that group of patients. So this is a confirmation of that. I don’t go much further than that conversation. It also brings up the issue of controlling the primary versus not controlling the primary.

HP: I use a lot of bicalutamide, actually, as monotherapy, probably because I was involved in the EPC studies. I think if you looked at those studies at the beginning obviously when they were designed things had moved on when you were looking at ten, fifteen year data from EPC. But if you looked at the beginning the only group where there was going to be a survival difference was adjuvant to radiotherapy, as we were just saying. Certainly locally advanced adjuvant bicalutamide there was a survival advantage for that subgroup. The other groups, there was never going to be a survival advantage. In fact, I think for the localised group they didn’t need any treatment at all so if you give somebody who doesn’t need treatment at all a treatment there is going to be more toxicity than there will be benefit. But I think there are advantages, certainly adjuvant to radiotherapy or with locally advanced disease itself, is that men don’t lose bone density. Certainly for some of the older men that can be a big issue. They keep more sexual interest and some maintain erections so I think there is a place for bicalutamide.

CR: I believe there’s a place for it, for sure. What I meant, just to clarify what I said, I don’t give it in place of local therapy or as the primary therapy for anybody.

HP: No, no. Yes.

CR: But we use it, for example, in certain patients who wish to avoid castration based approaches in recurrent disease, rising PSA etc. I believe it to be a fairly safe treatment at low cost etc.

HP: We often give it concomitant with prostate bed radiotherapy because that young group of patients and there was the 9601, wasn’t it, the study looking at two years versus no bicalutamide. So we tend to follow that.

EE: So it’s quite interesting, though, if we were to project how such data would look in light of enzalutamide or ARN-509 and I think it just raises more of an interest towards that point. Moving on now to the metastatic, purely metastatic, disease, another negative trial that was updated. It was actually a nice spin that they gave to it. Dr Gravis presented the analysis of GETUG, the follow-up, it’s a more mature study now and also did a post-hoc analysis applying the Chris Sweeney, let’s call it, classification, it’s the Glass classification. It changed a little bit the percentages of high volume disease in that trial, it came up to 50%. No differences even in the post-hoc analysis. I think Eric Small did a terrific job reviewing it and I wanted to get your thoughts on that and how he put it across the Sweeney data.

CR: The fact that it certainly throws into question the ECOG data because there was not a significant difference seen in the high volume cohort of patients. So it’s certainly not the confirmation we would have maybe hoped for but it was reaffirming that actually that study, that group of patients in France, had a much longer survival than the ECOG study. So that’s what Dr Small threw into the question here which is maybe these patients weren’t as matched as we would think.

EE: And your thoughts, we’re waiting on something. We’re waiting.

HP: We’re waiting. Obviously we’re waiting for the STAMPEDE data.

EE: Oh boy, yes.

HP: I think it was hinted that all the data might be put together, wasn’t it, and looked at in…

EE: Hopefully.

CR: It would be great to see those two studies, the data combined if it’s possible.

EE: We were discussing earlier with other colleagues from countries in Europe that we can’t even really use docetaxel earlier in hormone sensitive metastatic disease based on the guidelines and the specifications in some of those countries. So we desperately need first to see the published data and we also need to see the STAMPEDE data.

CR: It’s really important to point that out with the ECOG data. It looks very impressive but it hasn’t been published yet.

EE: Yes, of course. Probably very soon.

HP: Certainly my practice is that if somebody’s PSA isn’t going down quickly on their first line ADT then we’d add docetaxel in very early within three or six months.

EE: We do, don’t we? It’s in our algorithm essentially.

HP: Yes, and that was different to the study, wasn’t it, where it was more a delayed edition.

EE: But it was interesting what you brought up. Eric Small pointed out that only 30% of that cohort on Chris Sweeney’s dataset actually received docetaxel.

CR: For CRPC.

EE: For CRPC, and that’s important and how much of it. A very interesting analysis and I really asked him for his slides immediately.

CR: I’m hoping I can get them too.

EE: So moving on, cabozantinib. The big negative data that we were all hoping to be different, big spread on how it changed the characteristics on imaging. Your thoughts, would you like to tell us first?

HP: It was very disappointing, wasn’t it? I think it was the drug for the future for the last few years that people have been waiting for the data from the two studies. Again, a fantastic presentation. There were just those little hopeful moments that there may be subgroups of patients who are going to benefit from the drug and also, I guess, the whole dosing issue – that maybe if the dose had been a bit lower people would have carried on with it because there were significant toxicities.

CR: Most of us who have used cabozantinib, myself included, have seen responses, have had a belief or a feeling that it was benefitting patients. There are always two questions that come up in studies like this: one is was the patient population too advanced in light of the availability of enzalutamide and abiraterone and cabazitaxel. And number two is was the dose correct. Some of the initial studies were at 100mg and many patients are dosed as low as 40mg. So we don’t know, we’ll probably never know. I don’t think there is a plan to have another study, a phase III study in this patient population, so we’ll have to see. It is a disappointment. What I always liked about cabozantinib as the story had evolved was that it wasn’t a hormonal therapy so it had the potential to either interact with the hormonal axis or it had the potential to be available for those patients who immediately fail the enzalutamides and the abiraterones of the world.

EE: So a tumour microenvironment agent that doesn’t find its place even though the biology is in line. Don’t you see a pattern? We’re going in the past couple of years through negative trials, where’s the failure? We’re failing after all this wonderful drug development.

CR: Well, there are a number of reasons why trials could fail and we could talk about this for the rest of the day if you want.

EE: Just a couple of things.

CR: Yes, I would say that, number one, the need for overall survival to be the primary endpoint, I think that that was appropriate five years ago but it may not be appropriate any more. And so we can’t restrict what patients get after progression on a therapy in one of these trials and so we need to have a surrogate endpoint. The study that I published, and others, demonstrated that we can develop a primary endpoint of radiographic progression free survival so that’s something that we would hope to see incorporated into some of these studies. Some of this has to do with regulatory structures in various parts of the world that demand survival differences and that’s a problem and a challenge that we face in the field.

EE: Actually I would chime in saying that we need to be a little bit more selective in the populations -

CR: Absolutely, yes.

EE: And we’re not even close, probably. Just a couple of trials coming up that will try to do that. And that brings us to actually the data that you presented during this meeting and it follows, of course, the recent presentation at ESMO of the updated analysis. That gave us a lot of lessons to be learned and, in fact, the wonderful thing about androgen signalling inhibitors is now we have four trials in line which is the best you can get.

CR: Yes, exactly.

EE: So tell us a little bit about [?? 21:16].

CR: I would say the headlines are basically that now we have the COU-AA-302 study has gone through demonstrating a radiographic progression free survival benefit, an overall survival benefit which didn’t emerge at the second interim analysis that was presented and published two and a half years ago. But the overall survival advantage was there when the data was fully mature. There were a variety of reasons for this but the data we presented today looked a little more closely at the issue of crossover. So in the placebo arm of the COU-AA-302 study 44% of patients received abiraterone at some point in their lifetime. It could have been post-docetaxel, it could have been on protocol right after they came off placebo, that was only 9% of patients, but the point is these trials in which we think it’s randomised and there are two different arms there’s a lot of crossover that happens in many trials and I think this trial was a great example of that. So the data we presented was a statistical exercise, really, that looked at what would be the expected survival of the placebo patients had they not experienced abiraterone exposure. With the analysis overall the hazard ratio is 0.81, that’s the final data, statistically significant but when you add in the iterative parameter estimate, which is the statistical methodology, the hazard ratio drops to 0.74. So that’s a post hoc analysis, I fully grant you, but it does help us to think around this issue of how much of an impact crossover has on survival in these kinds of studies.

EE: I lost you when you put in the stats but would you now say that earlier is better?

CR: Well we’ve known this for a while, that the patients with earlier phase disease have better survival and that’s emerged from a number of different studies. So yes, when we’ve looked at, for example, baseline PSA by quartile and those kinds of things we see longer durations of benefit and preserved benefits over placebo.

EE: So I’m specifically speaking with regards to androgen signalling inhibitors because I’m just going to try to move into the splice variant presentation with that. My personal view is earlier is indeed better but we could perhaps be biased. I think your data supports it 100%, it’s what [??] said, it actually has its own control in there to compare. But let’s go into Emmanuel’s data set. Emmanuel Antonarakis and Jun Luo just presented us some very interesting and provoking data, I would say provocative, with regard to the impact of taxane treatment on patients whose cancers harbour or not AR-V7 splice variant. So your thoughts and, I guess for the audience, a summary of what you saw?

CR: The V7 is essentially a mutation of the androgen receptor that renders it able to signal without the presence of androgen. At ASCO and in The New England Journal of Medicine he showed that those patients who have circulating tumour cells who have that mutation are unlikely, almost zero percent, responded to enzalutamide or abiraterone. That makes complete biological sense, it’s a nice story. There are some questions about the assay and whether or not there are some methodological issues that still remain questioned but what he did now, which was the next question that came up, was, OK, we know that those drugs don’t appear to work but what about the other stuff like chemotherapy? So he showed that cytotoxic chemotherapy with docetaxel and even some cabazitaxel patients appeared to benefit at a rate that is pretty much what you would expect from historical controls with the V7 negative. There was a slight difference in the V7 positive versus the V7 negative patients in terms of the response to docetaxel but they were both within a range of what we might expect. So for the clinicians out there, should the V7 test become something you can use on a regular basis it should be something that could potentially select away from the use of the androgen axis drugs and select towards chemotherapy earlier rather than later.

EE: Your thoughts, Heather, on how that data was presented and what your take home message would be, given that we do not have access to AR-V7 assessment maybe for a few years?

HP: I think it’s probably one of the big hopes for the future, isn’t it really, coming out of this meeting. There has been a lot of data which has consolidated or confirmed things that we thought but the big question is always going to be sequencing. To have a piece of information which may help you to sequence in a scientific and rational way is a big hope for the future. That’s where we would hope that personalised medicine is going to come in and that we’re going to be looking at more and more stories like this.

CR: I would add, Eleni, that you and I know this, we all know this, that credentialing a drug through a clinical trial is a very difficult, labour-intensive, time-consuming process. What fewer people appreciate is that credentialing a biomarker is equally laborious. We don’t want to get too enthusiastic about this V7 story yet; this was an academic lab running this, great centre, great lab, great investigators, all of that but the question of whether it’s applicable, its universality and its reproducibility in how we handle the specimens and things like that is another important question. I think that’s one of the challenges that that group has over the next years, and we as a field have, which is to validate that data.

EE: Actually you’ve brought up a great point because we all remember the time it took and the presentations we’ve seen over the CTC Veridex platform and how [??] has really persevered through the years to get it there. My only, and actually Emmanuel brought it up at the end of his presentation, my only slight concern was that these were only 37 patients that were looked at. I would be very reluctant to come to any conclusion from the finer graphs that were showing, putting all these patients against the abi enza. This was a prospective study of the biomarker but not a prospective study of the agent’s performance. So I think we’ll need to look very closely because the p values are strong but we need to look at the details. So, final, final comment, I really liked the story on the [??] exported in the association of statin use with a delay of time to PSA progression. Phil Kantoff presented and it was very elegantly done. Your thoughts? Is there something we should be looking for?

CR: Yes, these are very common questions we get from patients about available agents, statins, anti-diabetes drugs and those kinds of things and could they benefit patients. There’s a series of analyses that seem to all be pointing in the same direction that statins may have a benefit and I’ll throw metformin in there as something that’s frequently discussed. So all of this, for me, begins to point to the fact that for many patients, not all, but for many patients advanced prostate cancer is a metabolic disorder and it’s a metabolic disorder where there are steroids being made, there are other intratumoral metabolisms, events that are happening. It is possible that some of these available agents could actually be integrated into the standard of care. They suffer the problems of everything else too, which is validation and appropriate trial designs and things like that. So I’m a little bit conservative and not recommending all these different therapies for patients. I don’t put patients on statins just because of what I see at ASCO, I put them on statins if they have hyperlipidemia. But anyway that’s open to question and may change over time.

EE: Heather, I think that STAMPEDE is trying to address these issues?

HP: Absolutely. I think STAMPEDE just rolls on and on, really.

CR: That’s why they call it STAMPEDE.

HP: Absolutely, stampeding through prostate cancer. It’s fascinating, isn’t it? It’s quite unusual for me to see a patient who isn’t on a statin actually, that’s the only thing, that the majority of patients in my clinic are already on a statin. But they’re keen to carry on now.

EE: So we started off on a good year on prostate cancer therapy development, I think, with GU ASCO. More to come in a month, the rest of the urology meeting and then on to ASCO. Thank you very much for joining us.