EE: Good morning, I’m Eleni Efstathiou from the University of Texas MD Anderson Cancer Center. I’m a medical oncologist and I do focus in prostate cancer, clinically and research-wise. I’m joined here in Barcelona just wrapping up ESMO by three colleagues, two from Europe and one also from the United States east coast. So I’ll let them present themselves.

NC: My name is Noel Clarke. I’m a urologist and Professor of Urological Oncology at The Christie Hospital in Manchester. I have a specific research focus in prostate cancer and I’m part of the STAMPEDE and RADICALS trials teams.

EC: Good morning, I’m Elena Castro. I’m a medical oncologist at the Spanish National Cancer Research Centre. My work and research is focussed in prostate cancer and I have a special interest in DNA repair.

MS: My name is Matthew Smith. I’m a medical oncologist and a Professor of Medicine at Harvard Medical School and the Director of Genitourinary Malignancies at Massachusetts General Hospital Cancer Center in Boston. My clinical and research practice focusses on prostate cancer.

EE: Thank you very much. So we are here to discuss our take-home messages from this ESMO meeting. The weather was great in Barcelona but somehow everyone stayed in the rooms and listened to the data so it must have been good. No, truly. I have to point out one thing – we managed to get the gender balance today but we’re not very balanced in specialty so I’ll start with the urologist. So you did not present because you were not here but it was your data from STAMPEDE. Some very interesting data that, correct me if I’m wrong, is a post-hoc analysis but it’s a post-hoc analysis that we really wanted. We wanted to see whether indeed there is a difference from the STAMPEDE standpoint of outcome when treating a man with de novo metastatic high volume disease versus low volume disease with docetaxel, in other words, if the data from CHAARTED are replicated in your dataset. So tell us a little bit about that.

NC: Well we were coming from a background whereby the original data from STAMPEDE looking at docetaxel, published in 2015, didn’t really match what we’d seen in GETUG-15 and CHAARTED, that’s to say that there was any volume effect. We saw complete across the board effect. On this analysis what we had was an extended follow-up of the docetaxel data, so just over 78 months median. We’d sub-stratified the patients based on volume using CHAARTED criteria. What we found was really in keeping with our original findings from 2015 was that there was no volume effect. We saw activity right across the spectrum of M1 disease. The hazard ratio for the high volume cases, using CHAARTED definitions, was 0.81 and, in fact, the hazard ratio was rather better for the low volume patients – 0.76. What’s important to note also is that in previous work with abiraterone we’d had exactly the same finding which was heterogeneity of effect. When we looked in STAMPEDE at the number of LATITUDE-type risk stratification and CHAARTED-type risk stratification patients who weren’t receiving therapy with docetaxel or with abiraterone we found that that’s about 45% of the patient population. So it’s quite a substantial number who were not eligible for treatment using CHAARTED criteria who now should be considered for treatment.

EE: To that point, you were there in that room, Matthew, when Karim Fizazi went on to discuss that he had a concern. He viewed the data a little bit differently and he had a little bit of a concern in using docetaxel in the de novo metastatic hormone naïve setting. Even though we have from last year a very good publication from Matt Sydes who is your, let’s say, wizard statistician on STAMPEDE. We could see clearly there that there was no overall survival benefit difference between using docetaxel and abiraterone for multiple reasons. What are your thoughts? Are we over with docetaxel? What’s going on there?

MS: I don’t believe so. In the absence of head-to-head comparative data we can’t be sure which is the most effective initial treatment for metastatic castration sensitive prostate cancer. The best data arguably comes from STAMPEDE where you had contemporaneously randomised patients and probably have the best ability to compare between arms of STAMPEDE to get a sense for whether there’s a difference. As you summarise, there was no notable difference. But Karim Fizazi, Professor Fizazi, did a beautiful job of being the discussant and trying to make the most of available data and doing the impossible which is to compare between trials and to make a case that AR targeted therapy may be preferable to docetaxel as initial treatment in castration sensitive metastatic prostate cancer.

EE: That was a head turner. Interestingly enough in low volume we’re looking at almost seven years, 6½ years now, median follow-up. Over 40% are alive which is outstanding. Maybe we’re changing the disease history, maybe other things. Elena, you’re probably the true representative now of Europe since they’re ready to get out, European Union I meant to say. So what are your thoughts? Is docetaxel here to stay? Did this give us the message that we should consider it for low volume? What do you think about that low volume/high volume segregation, is it applicable anymore?

EC: I don’t think it’s applicable anymore. We are also very aware that imaging techniques that we are using, we don’t really know exactly which patients are really low volume because in some cases we may only be seeing the tip of the iceberg and classifying them as low volume when they really are not so low volume. So in these studies I think we are seeing a benefit for treating all patients, even those that we are artificially classifying as low volume. So I think that it’s a benefit with docetaxel for all patients. The other thing is whether we choose one over another based on side effects, patient preferences, etc. but I don’t think we have evidence to say that docetaxel doesn’t benefit those patients.

EE: So we have an agreement that…

MS: We should add that in STAMPEDE almost all the patients were de novo metastatic disease. That may actually be a more important biologic construct than volume in that rationale.

EC: I agree.

MS: I’m glad. So we should think about a population that may not benefit for intensification of initial treatment and those may be patients with lower burden recurrent metastatic disease, particularly if there’s a long interval from primary therapy. We don’t know that for certain yet but best available data argues that there’s the least, if any benefit, in that group.

EE: And hindsight is 20:20 in the sense of when volume was used, or risk, it was just a poor surrogate for biology. We still, to this day, have not come to biologic determinants even though you STAMPEDE folks have promised some things. I don’t know when you’re going to deliver.

NC: We’re learning as we go and the more data we have the better understanding we have. Matthew is absolutely right and I agree completely that if we look at GETUG-15 and CHAARTED roughly about 40% of the patients were progressors rather than de novo which is a different biology, I think, to that in STAMPEDE. Another paper which was presented at the meeting from STAMPEDE looking at volume thresholds and burden really shed more light on our poor understanding of really what’s happening in the primary and what’s happening in the metastatic sites. We looked at the burden issue in arm H, which is the radiotherapy and metastatic disease, where we’d published last October 2018 that radiotherapy to the primary site improved outcome but only in patients who had four or fewer metastases on standard bone scan. We have done another analysis of this. Adnan Ali, who is one of the research fellows working with STAMPEDE, looked at this in detail looking at nodal volume, something which isn’t considered in CHAARTED, looking at visceral disease and looking at bone metastasis number. We’ve also looked at volume, actually. What that showed is that patients with lymph node only disease, they benefit from primary radiotherapy; patients with visceral disease had no benefit at all and the threshold for bone metastases on bone scan really falls off after three and at five it’s completely lost. That suggests that there is a threshold effect and that this is biologically related. We don’t quite understand that but certainly we’re getting better at understanding patients who will respond to certain types of treatment and who won’t. Now, to the point of your question which is when will we have a biomarker analysis, well, we do have two big groups within STAMPEDE looking at this. Gert Attard is looking at the molecular aspects, we are matching that with the imaging aspects and we hope to have a bit more information that will help guide treatment in the future.

EE: Hey, you’ve done already so much. I’m not criticising in anyway, I wish we could have delivered at similar or even much lower level. So, just really in the interests of time, Matthew, you presented an interesting update from the SPARTAN trial. Putting in context, for me it’s not just the SPARTAN trial and, of course, I want to hear your view on the data that you presented, but it’s also the whole field of non-metastatic castrate resistant prostate cancer. I’ve heard you before put it in perspective and what exactly that disease setting encompasses. Because earlier Elena brought up the whole idea of novel imaging being introduced, I’d like also your view beyond the data that you presented because it came up as a question so many times – so what about PET PSMA imaging?

MS: Yes, it’s a great question. So first the data that was the updated survival data from SPARTAN. There are now three drugs approved in the United States for the treatment of non-metastatic castration resistant prostate cancer based on positive phase III trials. They all showed a marked benefit in metastasis free survival and in each case the overall survival data was immature at the time of the primary efficacy analysis. So in SPARTAN specifically we only had about a quarter of the number of events required for the final overall survival analysis. So this was a pre-specified interim analysis for overall survival. It had about 285 mortality events out of the 1,200 patients in the trial and showed a 25% reduction in risk of death although it just missed statistical significance. The fact that the strong trend in favour of overall survival is holding up despite crossover of the patients from placebo to apalutamide and a high rate of subsequent life-prolonging therapies in the placebo group. So it’s holding up that this is not just a benefit in metastasis free survival but later endpoints including perhaps overall survival and, as previously demonstrated, time to symptomatic progression. We often get this question of this disease status disappearing because of newer imaging like PSMA PET. It’s a bit of a nomenclature issue and some folks say,’Why do you call this non-metastatic CRPC?’ My simple answer is because mCRPC was already taken and it actually has regulatory implications. So, in my view, these three studies, SPARTAN, PROSPER and ARAMIS, provide us the best evidence to intervene in patients who have detectable metastases by PSMA PET only because no such patient was enrolled in the previous mCRPC trials. We now have evidence published from PSMA PET series that the vast majority of patients who would have enrolled in these three trials in fact have detectable disease by PSMA PET, much of that being distant metastases.

EE: So, Elena, of course Matthew being the PI may have a little bit of a… let me not say bias, more of an endearing position towards using in this setting these agents. But there are still the naysayers that are saying, ‘Well, even though the primary endpoint and most of the secondary endpoints are beyond met and all three trials are aligned, I want to see overall survival data that are compelling.’ And it was just missed by a hair, I think, just a few events didn’t reach that interim result of positivity. What are your thoughts? Are you happy with just the primary and vastly most of the secondary endpoints being met to treat earlier these men?

EC: Absolutely. I think we are discussing a population, we are in a setting that trials will take a very long time to have results on overall survival, but from all trials it’s quite clear these patients benefit from the early use of androgen receptor signalling inhibitors. So I think we should maybe stop discussing only about overall survival and just accept that these are of real benefit for these patients. If we are not increasing their overall survival at least we are delaying the time to castration resistance, we are delaying the symptoms and I think we are improving their quality of life.

EE: He’s going to show us data like in 302 we had to wait for several years. We’re going to see the data, it’s going to be positive.

MS: There will be a final analysis, so thank you for your comments. The one additional thought, when Professor Fizazi discussed these other trials in metastatic castration sensitive prostate cancer he made this particular point of noting that some of the trials exceeded a greater than 50% subsequent life-prolonging therapy rate and many of the trials were less than that. But in this disease state, particularly in SPARTAN, nearly all the patients received subsequent life-prolonging therapy, 19% of placebo patients crossed over to apalutamide and 69% of patients in the placebo group received subsequent life-prolonging therapy, mostly abiraterone acetate plus prednisone. So there’s a strong trend in favour of overall survival we’ve seen despite a lot of subsequent therapy.

EE: I’m going to cut the non-metastatic CRPC discussion short in the interests of time because I want to move on to the big data that was presented during the Presidential Symposium. I don’t know if I should call you the Queen of DDR…

EC: No.

EE: Because there’s Rosie [...] there too. So, do I call you a princess since she started so many years before you?

EC: You know I trained with her?

EE: I absolutely know that.

EC: So I’m very happy being anything after Rose.

EE: So I followed you work and then we learned so much about how men who specifically carry mutations such as the BRCA2 mainly have a poorer prognosis. We saw some fascinating data suggesting that using cytotoxic chemotherapy such as taxanes may not work at all as compared to the general population. I think you’re the right person to put into perspective the data we saw from the profiling study, the olaparib activity in this disease setting. So what are your thoughts?

EC: What we have seen at ESMO this year is finally precision medicine coming for prostate cancer. So in the PROfound study what we have seen is that the treatment with olaparib for patients with some DNA repair defects, particularly for those with BRCA2 mutations, result in an improved progression free survival, significant progression free survival. There may be benefit also for other groups of patients but this is not as clear as for patients with BRCA2 mutations. We have to remind that these patients have previously received at least two lines of therapy, most of them, so from my point of view it’s a quite good result. We have also seen the results of another two studies, preliminary results of the GALAHAD and the TRITON study and the conclusion should be that PARP inhibition after a number of therapy lines achieves durable and significant responses for patients with these DNA repair defects, particularly those with BRCA mutations either somatic, germline, monoallelic or biallelic. It is true that probably the number of patients that will benefit from these therapies is small, it’s not great, but we have to think to what is happening in other tumour types such as lung cancer where we also have therapies and we screen patients for alterations that are only present in 3% or 7% of the patients. So even when a small number of patients may benefit from these therapies we have to be proactive in looking for, trying to identify who these patients are.

EE: Matthew, I’m not going to come to you since you’re the GALAHAD PI but I’m going to go first to the independent reviewer here. Noel, do you agree? Are we coming to the point where we’re moving? Because, for me, I agree with Elena that this is actually the contribution of this trial. It’s not just about the drug, it’s about moving into a more targeted therapy, precision therapy, development grounds. Every start is rough and bumpy but what are your thoughts? Is this a good start?

NC: I think it’s a very good start. What it shows us is signal and a clear message. It also identifies the sort of problems that we have in using targeted therapies like this and that is identifying the patients. Firstly, how do we pick them out from the population and then how do we test? Because we know that tissue-based testing has its problems, there’s a significant failure rate, there’s a significant negativity rate. We also know from some of the data which we published that in other settings with the use of other drugs, abiraterone, you can potentially give this to a wider population. But, as you said, it’s a bumpy ride at the start but we have a positive trial and we have to accept and congratulate the team that pulled this off.

EE: Your work on olaparib and abiraterone was great but it opened the floodgates again for an all comer approach. I’m sorry to say that.

NC: Well, we’ll see.

EE: So we’ll see that. But I’d give the last word to Matthew who will tell us more about his view specifically with regard to the advent of all these assays that will be more pinpointed in telling us which patients should be treated with what.

MS: Sure, so I’ll try. The field had a lot of enthusiasm about this based on a series of observations about relatively high rates of germline DNA repair defects in patients with metastatic disease, relatively reasonable rates, 20% or so, of patients in tumour biopsies with refractory disease. And then the compelling observations from TOPARP-A providing preliminary evidence for olaparib in patients with DNA repair defects. The PROfound study is extremely important because it really confirms those series of observations. The field has converged, not just from olaparib trials but studies of other drugs as well that really most of the activity, nearly all the activity, appears to be in the patients with BRCA mutations. It will remain a challenge to determine the optimal way of identifying those patients. In our practice we routinely refer all patients with metastatic disease for germline testing so you can find those patients, some of those patients, up front. You then need to do some form of tumour genetic testing. It is worth noting that we’re not only looking for BRCA and other DNA repair defects, there are less common observations of mismatch repair and potentially actionable mutations, at least for referral to clinical trials that you might find in other pathways. So while there’s going to be a lot of focus on the companion diagnostics with each of these drugs, the real question for the patient is it’s going to be a minority of your patients in front of you who are going to be eligible for treatment with a PARP inhibitor but you also want to find other pathways. So the field has a lot of work to do in defining the best path to identify appropriate patients for treatment.

EE: I think that’s an excellent, excellent observation and comment. I want to put also in perspective the fact that others before us, such as breast cancer, ovarian cancer and lung cancer, have worked more with targeted therapies. There have been some disappointing reports that even though we in more scholarly settings have access and use these assays, in more community based practices the education side of things is lagging or the time allocated to these efforts is limited. So we need to spend a lot of time on educating the community based practices and actually treat the bulk of our patients moving forward. There was this report, I don’t know if you saw it, that only 50% of patients with actionable mutations in other solid tumour settings are actually being treated accordingly. We do not want to have the same thing happen to prostate cancer. We can learn from the failures of others, right, and look better at least. Thank you all very much for joining us and we hope we have brought you some of our own insight. Whether you agree with it or not you let us know. Have a great day.