Prof Nicholas James – University of Birmingham, Birmingham, UK
Prof Karim Fizazi – Institut Gustave Roussy, Paris, France
Prof Matthew Smith – Massachusetts General Hospital, Boston, USA
Prof Álvaro Pinto – La Paz University Hospital, Madrid, Spain
Dr Mark Beresford – RUH Bath, Bath, UK

KF: Hello everybody. I’m Karim Fizazi, medical oncologist from Gustave Roussy in France. I’m here in Munich for the ESMO meeting and actually it’s the last day of the congress. It’s been really great for GU malignancies and specifically for prostate cancer with lots of great news, new information. We’re going to try to summarise this data for you today at this ecancer discussion. So let me first ask all the faculties to introduce themselves so that I can just rest and hear from them. Please.

MS: I’m Matthew Smith, medical oncologist from Massachusetts General Hospital Cancer Center in Boston.

NJ: Nick James, I’m an oncologist from Birmingham in the UK.

MB: I’m Mark Beresford, an oncologist from Bath in the UK.

AP: And I am Álvaro Pinto, medical oncologist from the University Hospital La Paz in Madrid, Spain.

KF: Alright. So, I guess we’ve heard at this meeting lots of important news for prostate cancer patients and actually quite across the course of the disease. Let me perhaps start with a quite specific and actually debated space which is that of the CRPC M0 space. We’ve heard earlier this year the new data from the SPARTAN and the PROSPER trials at ASCO GU and now that we have more follow-up, more time to think about it what is basically your take home from this data? Can you please start?

MB: Yes, so it’s an interesting group of patients, as you say, because our impression was it was quite a sizeable number of patients who were established on androgen deprivation therapy and their PSAs were rising and we scanned and couldn’t find any metastases. These two trials, PROSPER and SPARTAN, that took two different agents, enzalutamide and apalutamide, in patients in that position. They had a PSA doubling time in both studies of less than ten months. In the PROSPER trial the PSA had to be above a level of 2 before patients were entered. There are other studies going on as well, there’s the ARAMIS trial with other agents which are similar. But we saw some new data presented looking at the risk groups within those trials at this meeting. Broadly speaking, both drugs improved the time to developing metastases from around a year to a year and a half as the baseline without the agents in both studies to around three years, 36 months and 40 months respectively for PROSPER and SPARTAN. So big improvements in time to developing metastases but obviously at the cost of having relatively expensive agents that do come with some side effects and some impact on quality of life for long periods of time.

It’s interesting the status of being non-metastatic. The harder you look, the more likely you are to find metastases and we now have new scanning techniques – PSMA scans, choline PET – which may be picking up metastases that we weren’t seeing in these trials with the standard imaging, perhaps, that was done as part of the inclusion criteria in the trial. That’s certainly been my experience in clinical practice that sometimes we think people might be appropriate for the studies, look harder and you find something that’s maybe equivocal or you’re not sure and you speak to the radiologist and it could be an early metastasis. So the group, perhaps, isn’t quite so big as we first thought. I think we need a little bit more time to pull out real survival benefits from these trials.

KF: OK, thank you. Nick, do you share this analysis or different thoughts?

NJ: Obviously that’s a very nice and thorough summary of the data. I have to say in my practice this group has pretty much disappeared because on the one hand you’ve got PET scanning picking people up who you wouldn’t have called metastatic before, we now know where the mets are because of the growing use of stereotactic radiotherapy that opens up salvage options that are not drug based. At the other end, I have to say, we’ve got more and more relaxed about not putting people on ADT just because their PSA is going up. There’s actually very old data, the Pound data, that shows that you can identify patients who are at very low risk of death and therefore are at guaranteed risk of harm if you put them on ADT and low risk of death from their cancer. So it might be that the population is disappearing from each end.

KF: OK. Matt?

MS: I’m not sure that the population is disappearing because these weren’t all patients with non-metastatic CRPC, they’re the upper half of risk, and there certainly are patients, if you look at the entire population, if you look back, there are patients you might wonder whether they ever should have been put on ADT, particularly the slow risers. The other thing I’d say is if you separate just the drug and all that and just say what’s the evidence, one of the really important things, and you pointed out this novel category of disease, which is novel imaging PSMA PET positive, conventional imaging negative, there are exactly no patients like that in the pivotal trials that led to approval of all the other agents for mCRPC. So it’s really quite an extrapolation to say that if you can find it on PSMA PET I now would treat them as though they had metastatic disease by conventional imaging. So in many ways one of the more important parts of PROSPER and SPARTAN is we now have the evidence. So these are those patients and more data will emerge about that from our individual clinical experiences that if you image folks like that, particularly the fast PSA risers, you’re going to find a lot of metastatic disease. We’re going to see more data coming out about that from large experiences to support that. So in a lot of ways PROSPER and SPARTAN are going to help us to optimally manage patients who fall into this novel category.

NJ: I guess my concern, Matt, is that obviously the data are completely consistent with STAMPEDE, abiraterone, and LATITUDE as well, which is that early use of effective therapies is better than late use in terms of disease related things. But obviously the data you presented from the ERA 223 trial showed quite high, for example, fracture rates with abiraterone in quite a late setting and that was one of the features of SPARTAN and PROSPER as well, fractures and falls and things like that. So if you have very long durations of these very active androgen depleting/blocking agents is for sure we’re having impact on prostate cancer related effects but we may well be having impact we don’t want on other things.

MS: Yes, so that part of the equation of individualising management is age old, we’re going to be making decisions considering benefits and harms. I think you’re right to point out one of the themes that emerged from this meeting was really that factures in particular are important. It’s not just in SPARTAN and PROSPER but if you look back at the mCRPC studies of abiraterone and enzalutamide there’s excess fractures there too. So one of the consistent things too is that there is really under-utilisation of osteoporosis prevention of fracture prevention strategies so many of the patients entering these trials were probably not well managed, for example, in terms of fracture prevention. Only about 10% of the patients in SPARTAN came in on a bone sparing agent, despite the fact that they were in their mid-70s and had been on long-term ADT.

KF: Actually it’s fun, well not necessarily fun but it’s interesting to see that we’re learning things in parallel to our trials. We’re aiming to something with the ERA 223 combo or the early use of x agent, CRPC M0 trial, and we’re learning that actually bone targeting agents should probably be used much more earlier and more frequently than what we typically do. That’s really interesting so we’re still learning. Let me move to a discussion about metastatic disease. That’s really a field where, in the last five years, we saw a tremendous amount of good news for the patients from docetaxel trials, abiraterone trials. Yesterday at this ESMO Congress, STAMPEDE reported two abstracts at the Presidential Session which is quite exceptional, early congratulations, a fantastic achievement. So can you just briefly summarise what was presented?

NJ: Thank you. The first one was the abiraterone data and, as of course you’re extremely well aware, the LATITUDE trial was high risk patients only and that changed the licence and the guidelines for high risk patients but didn’t include low risk patients. Now STAMPEDE included the whole spectrum of metastatic patients but we didn’t categorise the disease by that classification. So the first abstract that we presented was our data re-analysed according to the LATITUDE criteria. I guess it wasn’t a huge surprise to see that the effect was broadly very similar in the two groups. So although it’s not within the licence strictly, the licence will be interpreted, we think, quite liberally once this drug is off patent, which is it in some places and will be pretty much everywhere. So it’s important because it extends the use into both ends of the spectrum. There has been controversy about whether the docetaxel benefits are equally distributed across high and low volume so we did think it was important to show whether it was or was not the case with abiraterone.

KF: Before we move to the second abstract, let’s maybe comment on that one. Others, what do you think about this new data?

MB: I think it’s fascinating, really. We’re moving to a point now where abiraterone seems to be the best drug for up-front metastatic disease. My concern, of course, a little bit, as we were talking before, is that there’s a whole cohort of patients for whom that might be suitable and we’re going to be looking to have these patients on treatment for a long time. So it’s a clear impact on capacity, on resource and issues like that, but these are data that we can’t really ignore given the huge benefits that are seen for the patients.

KF: OK.

AP: I agree, you have the evidence, you have the data. When you didn’t have them it was more speculating about why not low risk patients were not going to benefit. But there you have it and that links with the next part of the STAMPEDE, the second one, maybe you even could think about giving everything even in low risk because that’s the ones that get the benefit – ADT, radiation and abiraterone. Of course this won’t be from 0-100%, we will still have to individualise, to decide which patients may be the ones that benefit most from that strategy but now you have data, I think it’s the most important thing.

KF: What about the potential approval or reimbursement and excess in other countries or do you think for Spain this indication?

AP: Yes, in Spain, for example, it’s still not formally reimbursed and it depends. In each institution it’s taking its way to lead with this issue. In my case, for example, you may ask for a special petition for individual patients and patients have been approved. So it’s really following the LATITUDE schedule. I don’t know with the widening of the group of patients if things will be as easy as have been until now.

KF: Matt, what about the US? What do you think about access for those particular indications of low risk disease?

MS: I don’t think it will be a barrier, it’s already FDA approved and payers really aren’t too concerned about the extent of disease.

KF: OK, so very important. I think there was also a second abstract from STAMPEDE at the Presidential.

NJ: There was. Actually, just to stick with the first one, just for a bit longer, we showed data at ESMO last year which we haven’t published yet showing really quite marked synergy between radiotherapy and abiraterone in, for example, N0, N-plus disease. The interesting thing about that was that the test for heterogeneity of effect was significant, suggesting that there was actually a bigger effect in that group than the metastatic patients when you gave radiotherapy as well. The other thing about that was when you gave two years of abiraterone, I’ve got patients in my clinic now who have been on abiraterone for seven years and I’m am concerned that we maybe don’t need to give them for that long. Obviously we don’t know but I think that’s a question for the future.

So the second abstract, yes, we were obviously delighted. We’ve incorporated radiotherapy for newly diagnosed metastatic disease in the trial around about the same time we set the abiraterone arm up, slightly afterwards. Because of all the stuff around docetaxel and volume we’ve reclassified, or we got the scans in and reclassified them, and secondly the HORRAD trial had suggested there might have been a benefit difference between high and low volume disease. We prospectively reclassified the scans and did all of this before we did the analysis and what we saw was a very clear effect on failure free survival across the whole population but that was bigger in the low volume patients, a lot bigger, and no survival effects in high volume patients but for men with up to four metastases the hazard ratio is 0.68 which is somewhere in between the abiraterone and the docetaxel benefit in the same population. So highly significant effects from actually a very simple treatment. Around 20% of these patients had docetaxel as part of their standard of care and the hazard ratio was the same with or without docetaxel so it does suggest that it would be reasonable, and this is the conclusion we have come to for our next randomisation, that for this group the standard of care or a standard care would be ADT, docetaxel and radiotherapy to the prostate.

KF: OK. Is that practice changing outside the UK or inside the UK?

MB: I think that is because coming back to your previous point about the approvals and the regulations it’s easier to get radiotherapy through than it is a new drug. We don’t have the same restrictions so all we need to do is rewrite our radiotherapy protocols locally and we can do it. So it’s something that we could start doing effectively next week. Is it the right thing for everybody? The same conversations we were having before – there may be a group of people who don’t need it. Radiotherapy does have side effects. Interestingly in the trial the incidence of UTIs and bladder symptoms wasn’t greatly bigger in the radiotherapy group and, in fact, there’s a suggestion that it might prevent further trouble down the line in terms of urinary tract symptoms. But it does have some short-term effect and it has an impact, of course, on the patient travelling up to the hospital for treatment, particularly if they’re elderly. The fact that there is a very pragmatic allowance to use different radiotherapy schedules is one of the strengths of the trial in that we could say actually in this group of patients we’ll just give the 36 in 6 dose, it’s minimal impact, most people can manage that, we’re getting some significant benefit from that at no great cost. So I think it’s a nice practical one.

KF: Right. So will it change practices in Spain and in the US?

AP: I think so. At least tumour bursts will be much more interesting in our case. There have always been this kind of very selected patients that you would like to, or you would even decide to, give radiation therapy to the primary but, as before with the first STAMPEDE abstract that we discussed, now you have the data and very solid data. Something that was really highlighted in the presentation, I think it’s important, is it’s not a subgroup analysis like something bad. It was a pre-specified subgroup analysis, very solid, very robust data so now you have data to do what you have always wanted to do in some selected patients.

KF: OK. Matt, do you buy it?

MS: Much the same, yes, I do buy it. We’ve done so in selected patients without good evidence so now we’ll do it more patients with greater confidence that we’re providing net benefit to our patients.

KF: OK, so we’re making progress. Right. So let’s continue moving on with the natural history of the disease. Metastatic CRPC – harder to treat patients, I guess we’ve heard some good news, some bad news. ERA 223, a combination of abiraterone and radium, can you comment on the results?

MS: Sure. So the study was designed to look at this combination of concurrent abiraterone and radium-223 with the primary endpoint of SSE free survival. The study was unblinded early after more deaths and fractures were observed in the radium group and then the study proceeded to its final pre-specified analysis. The final results of the primary analysis are there is no significant difference in SSE free survival and in the preliminary OS analysis no difference in overall survival. But there is harm – there were more fractures in patients who received radium-223 and then by a variety of analysis it looks like most of those fractures are osteoporotic. So certainly based on these findings we would not recommend the combination of abiraterone with concurrent radium as first line treatment for mCRPC. Regulatory information has changed in various countries based on these results. The other interesting observation of the trial was that only about 40% of the patients came in to the study on bone health agents which begs the question of why so low but that’s what has been seen across trials. Patients on bone health agents had fewer fractures in both groups although it didn’t spare the patients entirely from the radium, the adverse effects of radium on bone.

KF: So, others, will you change the way you’re treating patients with radium? Will you use it in later stages, post abiraterone or enzalutamide? Will you systematically use denosumab or zoledronic acid together with radium?

MB: I think throughout the UK radium has very much been a third line treatment once we’ve exhausted other options. People have tended to do it sequentially rather than in combination so I don’t think it will have a big impact on our practice. But the point about the bone health is really important. Ever since the early days of STAMPEDE when zoledronic acid didn’t seem to show much benefit…

KF: In the hormone sensitive setting?

MB: Exactly, yes. We haven’t really focussed very much on bone agents in prostate cancer, whether it’s denosumab or bisphosphonates. There has been a smattering in the trials but it has been very variable as to the percentages of people that have been on these agents. Yet it’s a big problem for the…

KF: Sure. Nick, same thing?

NJ: Yes. I think in STAMPEDE obviously we were focussing on cancer related bone events and there was very clearly not benefit in STAMPEDE from doing that. But what Matt has highlighted is that this was not cancer related bone effects that were increased, it was the…

KF: Mostly osteoporotic factors.

NJ: It was the damage to the normal bone that was the problem. It’s answered a very important question because at the St Gallen consensus meeting we had a lot of debate about if you’re on abiraterone and you’re responding do you need to give anything else and the answer is very clearly yes, you absolutely do, now from your data.

MS: Well said, yes, I agree. The other part of the study, the endpoint for the trial was intended to be a cancer-related event which is this SSE but when you dig deep into that you find out that some of these events which are thought to be pathological fractures really are osteoporotic fractures. I’m as sure as I can be that this is not unique to this trial, it’s probably the case in a lot of other studies as well.

KF: I guess we’ve heard also good news at this congress in men with metastatic CRPC, PSMA targeting and PSMA assessment, PARP inhibitors, the TRITON2 trial for example. Any comment about all this data?

AP: I think that mainly for PARP inhibitors there was a promise that this could be the first targeted therapy for prostate cancer because the incidence of the previous studies about 11-15%, somewhere in the middle, of germline DNA repair defects involving mainly BRCA1 and 2 but also ATM and other related genes. So the data of TRITON2, as we were commenting before, revealed that if you have a good biomarker the drug will work but if you don’t the drug won’t work at all. So maybe we have been misled by the hype of ovarian, breast cancer etc. seeing those high responses and very durable responses. The first data by TRITON2, although appealing – 40% of responses – also showed you that maybe all the DNA repair defects are not the same so we have people been putting together BRCA1/2, ATM, PALB2, CHEK2, whatever and, for example, there were no responses in the ATM cohort of patients. So we may have to rethink which patients are included in these trials and which patients could be potential candidates for the drug. Of course it’s a step ahead, we are trying to better select our patients based on biomarkers but maybe the biomarker is not still perfect. It gives you a hint of what you should do but BRCA2 may be the most important one, BRCA1 is very uncommon in prostate cancer and ATM and other related genes may be less active and may be less important.

KF: Right. Any comment on PARP inhibition or PSMA?

MB: I suppose the only thing is in general with these biological aspects in tumours is how much we’re changing the biology of prostate cancer by more and more upfront treatment. We may see that the resistant clones are coming through earlier on, perhaps, now that we’re using all our best agents up front. So it could be that the pathway of the disease is changing and these agents might be more appropriate after failure with lots of up front therapy.

AP: In fact the rate of somatic mutations in BRCA and related genes grow. Each line you give the higher the percentage of mutations you find in the tumour so it makes sense.

KF: OK. PSMA or PSMA targeting, any comment?

NJ: It’s a very interesting concept. You can image the disease with a PSMA based radionuclide and if you just hook a warhead up instead of an imaging isotope obviously you’re targeting radiation. So it’s a very appealing concept and there was one of the posters was data from this. The response rate looks high but the durability of response is rather like the PARP inhibitors – it doesn’t look quite so impressive and, of course, these are heavily pre-treated patients. But actually there was a very nice poster from Michael Morris looking at characterising patients with positive PSMA-PET scans, the PSMA expression on the circulating tumour cells. A lot of the CTCs did not have PSMA expression so there is, of course, a mixed population in the cells. So I guess it depends how mixed because if they’re mixed enough you’ll get bystander effects, if they’re not mixed enough you won’t. It’s in phase III trials so it looks exciting but we’ll see.

KF: Absolutely. Now, let’s go back to localised disease. One big question is whether we should use a more sustained systemic treatment for these men. We’ve been using a combination of ADT, three years, two years, a year and a half, whatever, to give enough radiation for many of these men who are not treated locally with radiation. At the moment it’s fair to say that we’re wondering whether we should integrate to this kind of combination the new agents including abiraterone, docetaxel, enzalutamide, maybe others. So we could report at this meeting all our own data from France, from the GETUG-12 trial, with a twelve year median follow-up now. That was comparing radiation ADT versus the same plus four cycles of docetaxel-estramustine. We had already reported the RFS data and relapse free survival was significantly improved, we already knew it, with about a 30% reduction in risk. For the first time we were able to report clinical relapse free survival counting only local relapses, metastases or death. What we saw was a statistically significant 25% reduction in the risk. I guess the STAMPEDE data will be also analysed soon and hopefully in the US the RTOG data will be also re-analysed with a longer follow-up. So, based on this data, I would be very curious to know what you think – should we already integrate docetaxel-based chemotherapy for these men, or at least some of these men, with high risk localised disease or is that too early, we need more? Matt, what do you think?

MS: Well, we’re talking about benefits and harms and given the long natural history of this disease, even high risk disease, it’s really a situation that demands careful consideration of the patient in front of you. So I think of the individual patient with high risk disease – what is his cancer specific mortality? In other words, what are his competing risks? So I look very differently at a 55 year old guy with T3, Gleason 9 and then a 75 year old guy.

KF: So what about your first scenario, would you?

MS: Would I ever, that kind of question?

KF: Or future?

MS: Yes, we certainly have that conversation and the younger they get and the risk ramps up, the probability of them actually dying of their cancer increases so that is a patient who I would have and have considered such treatment and certainly will consider more of it in the future, based on your evidence presented at this meeting. There’s going to be a lot of other evidence to follow. It’s behind, what they’re going to lack is the long-term follow-up of GETUG-12 but there are certainly other studies looking at AR inhibitors, etc., looking in a similar patient population.

KF: Right. Nick.

NJ: So we presented data at GU ASCO, the beginning of the year, which we published a couple of weeks ago looking at the quality of life as measured by EQ-5D in these patients. Because they relapse later and because there’s an effect on failure free survival it means if you’ve given radiotherapy and a fixed duration of hormone therapy you’ve got a considerable period of extension to your time off treatment which turns out to be very good for your quality of life. The other thing that we showed that has previously been reported and is very consistent in all the docetaxel and abiraterone trials is you reduce SSEs by about 40%.

KF: So does this mean that you’re already treating patients in your [?? 27:16] practice with high risk localised disease with docetaxel-based chemo?

NJ: Yes. So I was going to say, men going into STAMPEDE with M0, around a third are having docetaxel as part of their stated standard of care so it’s clearly widely done in the UK.

MB: Yes, I’ve only done it within the trial so a number of patients, the same.

KF: What about in Spain?

AP: I don’t think it’s standard, of course, but we’re almost there. Adjuvant chemo in a lot of types of tumours, high risk tumours, has reached a sustainable benefit, not a huge benefit but it’s a remarkable benefit in extending survival. So in prostate we’re almost there.

KF: Cool. So we have good news. So with that I would like to thank the four of you, this was really fun and a very interesting discussion. I would like to thank ecancer for having us tonight wrapping up the data from ESMO 2018. Thank you very much, bye-bye.