Prof Thomas Powles – Barts Cancer Institute, London, UK
Prof Matthew Galsky – The Mount Sinai Hospital, New York, USA
Prof Bertrand Tombal – Cliniques Universitaires Saint-Luc, Brussels, Belgium
TP: Hello and welcome from Munich in 2018. It’s the ESMO meeting with over 25,000 delegates, I’ve got two of them here with me today, two old friends of mine - Bertrand Tombal from Belgium, a professor of urology with a special interest in prostate cancer, and we’ve known each other for a long time based around EORTC days originally, and my great friend Matt Galsky from New York. Matt and I have been working on bladder cancer for years, we went through the quiet periods of bladder cancer together and we’re now going through the noisier periods. I’m not sure outcomes have improved that much yet but we’re hoping that’s going to change hugely and some of the data we saw today suggests that might be the case and clearly new immune therapies and targeted therapies. But before we talk about bladder cancer, Bertrand, I want to come to you and maybe talk a little bit about what you’ve seen at the meeting, at ESMO this year, that you were maybe most impressed with.
BT: So there were a lot of, once again, practice changing abstracts coming from the UK mostly but when it comes to precision medicine we had a series of interesting studies. One of these was a study comparing an AR pathway inhibitor, abiraterone or enzalutamide, to cabazitaxel in these patients that are considered as aggressive phenotype or poor risk, selected because they have a short response to ADT, they have aggressive features. When we speak amongst each other there has been that contention that these patients would be better candidates to chemo than to abiraterone/enzalutamide. Every time everybody is speaking about sequencing and even at the APCCC, the big St Gallen conference, people would say, ‘No, these patients we would rather give chemotherapy.’ So Kim Chi compared abiraterone/enzalutamide to cabazitaxel and, I would say not surprisingly, they unfortunately all die very rapidly but there is no difference, at least no significant difference, between abiraterone/enzalutamide and cabazitaxel. So it’s a little bit counterintuitive.
TP: Can I push back a bit and say so what’s precision about that?
BT: It was about the fact that Kim Chi has tested a series of prognostic markers such as PTEN, RB loss, p53, DNA repair mutation, and it was thought that they would actually predict a phenotype that is most likely to respond to cabazitaxel than to AR pathway inhibitors. Actually, except for well-known already AR aberrations, when you come in first line treatment it doesn’t seem to make a difference. So basically I would say that it’s not going too much into using precision medicine tests at early stages of the disease.
TP: Why has precision medicine been so difficult in prostate cancer while breast cancer seems to nail it and haemato-oncology?
BT: I think for three reasons. The first one is that the tumour is unbelievably heterogeneous; the second is that we hadn’t had tests that were performing as well as in other cancer types. As you know, in prostate cancer because the disease is mostly in the bone it’s difficult to get a good biopsy. In this regard there was another poster showing promising results that you can get most of the information you need from ctDNA but when you look at that you see that we are still very, very, very far away from having a commercial test. The third one, probably the most important one, is that actually addressable mutations, it’s like winning the lottery. All together it takes 10-15% of patients all together. So I would say those are the three reasons why we are still struggling to be like… Even at EORTC I see all my friends doing lung cancer trials and then I say, ‘Why don’t we do the same?’ and I think it’s not because we are bad, it’s just because we’re dealing with a totally different disease.
MG: And perhaps to push back a little bit, the AR is probably one of the key examples of precision medicine in oncology in the history of oncology.
BT: But it’s a very basic one.
MG: And it’s a ubiquitous driver.
BT: It’s ubiquitous, yes.
MG: So that even though it’s very precise it’s present in the majority of patients and it’s driving disease...
BT: Yes, but then you could even argue that you don’t need to be a genius to know that if you give two or three lines of AR targeted therapy you’re going to basically deteriorate the AR that it’s going to make non-functioning. Actually if you take something like AR-V7 which is very popular, it was tested after two, three, four lines of treatment when now what we want, now that we move all these drugs very early on, what we want is earlier information. For instance, if you look at something like AR-V7 on prostatectomy specimens or on very early disease stage, that’s not necessarily helping a lot.
TP: Just the AR-V7 story because you’ve mentioned it, perhaps just tell us where we are. It looked so promising four or five years ago, I remember getting off a plane from ASCO and it was the big thing. Now I was in a meeting earlier yesterday and it seems to have disappeared. What’s going on with AR-V7?
BT: I think that first of all the same observation I mentioned before, we don’t have a very robust test and there is a lot of inconsistency across platforms to look at AR-V7. Also, for instance, Johann de Bono has shown nice data where you have a correlation between the number of CTCs and the AR-V7 expression. So what people now believe is that actually it’s like a proxy for bad cancer.
TP: And it was thought to be a proxy for response to…
BT: It was to be huge. So basically it’s now not used that much because, as you may understand, the idea was that if you have received enzalutamide could you predict who needs abiraterone? But then we’ve got now two, three, four trials that show that on the whole only 4-5% of the patients are responding to the second line treatment. So having a marker that predicts who is going to respond to a second line hormonal treatment…
TP: Which was the aim of AR-V7.
BT: Which was the big game four or five years ago is not a question anymore. In the UK, for instance, you can’t get both.
MG: So the trial that you described, even though patients hadn’t been exposed to both next generation hormonal therapies yet, seems to be a perfect trial to test AR-V7. Was that tested, AR-V7?
BT: We still have preliminary results, it just shows some result, but we’ve seen since then that AR amplification in general, it’s more than AVR and it’s losing the ligand binding domain, amplification, many mutations. So it’s like one of many alterations which you know if you take a population which doesn’t respond to abiraterone and you look at AR-V7, yes it’s going to be enriched in AR-V7 but sometimes people are asking me, ‘Should I invest in my clinic into an AR-V7 test?’ I would say no.
TP: No at the moment. Bladder cancer, are we making more progress?
MG: I think we are making progress, I think the progress in bladder cancer has been after a period where we really had nothing that showed a lot of gains so there was a lot of enthusiasm about immune checkpoint blockade and certainly that has changed the landscape. But we’ve pulled back a little bit as we’ve really seen the proportion of patients that benefit from these treatments. That said, in addition to this huge field of exploring combinations to try and increase the likelihood of responding to immune checkpoint blockade, clearly there seems to be some value in moving these treatments earlier in the course of the disease. That was a big theme at ESMO this year, two key abstracts; one, the first abstract, looking at immune checkpoint blockade in patients with BCG unresponsive non-muscle invasive disease, KEYNOTE-057. An early look at this data but in patients with carcinoma in situ showing a 38% complete response rate at three months. It’s an early time point but the very fact that there is activity in this early disease state with this systemic therapy is certainly encouraging. This is a large patient population and a patient population who would otherwise be treated surgically with cystectomy. So this is a key unmet need and we’ll need to watch this data for a longer follow-up.
TP: There was some neoadjuvant data at ASCO too. We know the second line response rates and overall survival benefits are modest, why are we seeing this activity earlier in the disease? Is it greater than you would expect and if it was, why is that the case?
MG: It’s a little bit greater than I would expect. I think we have seen a hint of improved activity as the treatment is moving earlier in the course of the disease from platinum resistant to first line treatment to the neoadjuvant setting and now to the non-muscle invasive setting. The increase in response rates, it’s not doubling or tripling but it’s increasing a little bit as you move earlier in the course of the disease. I think there might be some biology there; we know in patients with metastatic disease, patients with lymph node only metastases seem to respond better to immune checkpoint blockade so that’s something about the biology. Maybe we’re seeing more of that type of biology the earlier we go. We know that happens with chemotherapy too in terms of the complete response rates seen with neoadjuvant chemotherapy versus in the advanced metastatic setting.
BT: And also, if I may, because as a urologist cystectomy is there because this is the only treatment we have. It’s not something you would like to have and we tend to forget that the bladder cancer population in most European countries is a population of elderly patients. It’s absolutely not rare that in the MDT we discuss cases of 82, 83, 85 years old guys. So far the view on the neoadjuvant has been to reduce the tumour load before you do the cystectomy and you gain something after that but I think that it’s basically opening a totally new dimension which is can we cure that disease by systemic treatment only? So can we endorse… and that’s where we urologists still have a lot of work to do trying to really define what is a complete response.
TP: So Bertrand, on this issue what you’re saying that’s important is you’re saying we’re going to try and replace cystectomy?
BT: At least, I would say, postpone cystectomy because we learned something very, very interesting and especially with Ronald de Wit’s pembrolizumab study…
TP: That was the non-muscle invasive study?
BT: That most of the patients when they progress they do so through non-invasive cancer, meaning that you still maintain the window of opportunity for radical surgery. That was very well presented by Shahrokh Shariat, who is a urologist as well, saying that that’s reassuring that when they progress it’s rather a minority will develop muscle invasive disease. So now we can slowly start with the concept of saying you give neoadjuvant treatment, you have a lot of options for patients that are fit, unfit for chemotherapy, because that’s once again when they are 85 they have already a lot of comorbidities, and that instead of jumping on them you create a kind of window of opportunity when you can apply active surveillance. Because if tomorrow you can reduce the rate of cystectomy by 45%, which would be in line with the complete response rate, for these patients that would have a major impact.
MG: I have to admit, I was a little bit nervous about giving systemic treatment that has a relatively good safety profile but some bad things can happen in patients with non-muscle invasive disease but this point about delaying cystectomy whilst also perhaps providing some systemic coverage, because we know that giving these patients intravesical therapy has been done in the past, not only has that not been associated with robust response rates but you always worry a little bit about the disease becoming systemic.
BT: That’s always what we say to the patient. The patient who has six cycles of BCG, has not complete response, we say the problem in bladder cancer is we are always too early or too late. If we give you another six cycles it may work but it maybe also that if it doesn’t work you’re going to get lymph nodes by that time. So immune therapy by BCG has its limits, so clearly this is opening a totally new… And this is a message for urologists who are looking at us today, I hope we’re going to be clever enough to see what are the potentials to basically better select patients who don’t need surgery.
TP: Sorry, I’m going to ask Matt that question. Talk to me about patient selection in bladder cancer right now – are we going to use biomarkers or are we not? If we are, what are they going to be?
MG: There are three trials right now that I know of, and probably more, seeking to define a patient population with muscle invasive bladder cancer that might be treated with systemic treatment alone in a curative fashion. Two of those trials are using DNA damage response gene alterations to select patients to achieve a clinical complete response for surveillance rather than cystectomy. One of the issues with those trials, although incredibly innovative and paradigm shifting trials, one of the challenges is that the list of genes in both trials are different and that really begs this question how robust do those biomarkers need to be validated before they enter a trial like this? The third trial is really asking an old question in a new way and it’s asking what is a complete clinical response, how do you define that, what tests do you need and if you really define it robustly can you select patients who don’t need to have a cystectomy?
BT: That’s why we have work to be done on imaging and post… You know many of the urologists, they do the cystectomy not having re-resected the patient. Maybe we should do that and then do the cystectomy. We need to have trials to validate the definition of complete remission because as long as we don’t have the specific trial any further development will be made very complicated.
MG: And this is a biomarker, complete clinical response is a biomarker, and it should be defined and validated as robustly as any other biomarker.
TP: I’ve just got thirty seconds to go so one key message from both of you about what you’ve seen today, or the last few days at the meeting, in precision medicine in prostate cancer, the take home message from your perspective, Bertrand.
BT: I would say that for prostate cancer last year I was still thinking it would be a dream, actually it’s not a dream, not because of the drug, the drug works in prostate cancer, but because we’re getting much better at developing tests to find the few patients that are likely to respond. However, the bad side of it is that for the prostate it’s still a little bit like winning the lottery…
TP: Because there’s so few patients?
BT: There’s so few patients but these patients can live for ever.
TP: And Matt, bladder cancer, where are we?
MG: I would say that earlier use of immune checkpoint blockade is going to be the near-term future but so far we have single arm phase II studies and we really need randomised studies to establish the benefits of this strategy.
TP: I’m going to thank you both and I’m going to tie this up. Thank you for listening and see you soon I hope.
