AB: Good evening everybody, and welcome to tonight’s webinar looking at the latest developments in the treatment of bladder cancer. I’m Alison Birtle, and I’m one of your co-chairs for this evening. Tonight, I’m joined by my co-chair, ProfessorIgnacio Duran from the University Hospital Marqués de Valdecilla in Spain. Also, I’m delighted to welcome Dr Patrizia Giannatempo from the Istituto Nazionale dei Tumori Fondazione in Milan, in Italy. We were due to have a slightly larger panel this evening, but due to unfortunate circumstances, Robert Huddart unfortunately isn’t very well, so we will miss you, Robert. Get well soon. But I’m delighted that the three of us will be able to have a really focused discussion together, so welcome to both of you.
Thinking outside the box: Novel or unexpected data
AB: Our programme for this evening is quite packed, and so I’m going to move straight on to looking at thinking outside the box. This is a section that we put on where what surprised anybody this year? Were there any particular things that we weren’t expecting, or we were expecting, or perhaps might have been interpreted in a slightly different way at the meeting? So, Ignacio, I’m going to go straight over to you about this.
ID: Thank you, Alison. Hi, everyone. I’m very happy to be here today sharing this session with you and with Patrizia. So, as you mentioned, this thinking outside of the box is about what really made us a little bit moving in the chariot and not completely expected data. If I’m allowed, I’d like to comment on two press releases. The first one is the NIAGARA press release. We heard not so long ago that chemotherapy plus durvalumab in the preoperative setting, so chemotherapy, durvalumab, cystectomy, and then durvalumab, achieved positive results. We don’t know a lot. We know that the primary endpoint was event-free survival, and the study made it. We know the secondary endpoint was overall survival, and the study made it. We don’t know any more precise data. We don’t know anything about pathological complete response.
To me, that data was a bit surprising, because I’ve been carefully following the neoadjuvant context. There are so many studies that have been trying to improve the data from chemotherapy using different approaches: immunotherapy as single agent, combinations of immunotherapy, combinations of chemotherapy plus immunotherapy, and so far I haven’t seen any combo really outperforming the data that we have from chemotherapy. So I’m curious to see that full data, hopefully presented at ESMO in Barcelona, so looking forward to it. I’d like to hear then and learn which patients benefitted and what was the price that they had to pay for that benefit. That was one of the things that left me a little bit uncomfortable, let’s say.
AB: Okay. Are you going to keep the second thing that made you uncomfortable for later on in the discussion?
ID: I’m going to let Patrizia go, and then I’ll come back with my second one.
AB: Okay. Patrizia?
PG: Good afternoon to everyone. Thank you. We are going to speak about the news of this year. I think this year we have a lot of important presentations that could really change our practice in the near future. But let’s see what’s happened in the next hour to present our new data.
AB: There were two things that surprised me, one of which I’m going to talk about. One is just about the extent of surgery and extended lymph node dissection. I really didn’t see that coming. I’m going to talk about that in a moment. The second thing which I suspect, knowing you, you’re going to pick up on, which is the way that some of the data about adjuvant nivolumab was interpreted at that meeting. I was sat next to colleagues who were surprised, and said to me, ‘Well, who would have thought that the adjuvant nivolumab after cystectomy was going to come out with a survival advantage?’ And I sat there, and I thought, ‘But nobody has mentioned either statistically or clinically significant data here.’ They were very careful to say that, and very careful to say that this was interim data with longer follow-up. But yet I know people had interpreted that as being something different, that that was necessarily practice-changing. I guess that’ll come out in our debate a bit later, but I wondered if that was what you were going to pick up on.
ID: No, I mean that’s a very interesting point, and I had it in mind for the discussion, but we’ll come back to it later. I fully agree with you, and I’m sure this is going to feed the debate with Patrizia when she presents this data about neoadjuvant. My second point was another press release, and I was sad to hear that TROPiCS-04 was negative, a study with sacituzumab govitecan that was actually an approach that I was expecting to perform better, because there is a lot of information around ADCs, antibody drug conjugates. One of the concerns that we’ve been listening to is the concern about safety with enfortumab vedotin, which is a great drug, but there is that grey zone of unpredictable toxicity. I was very keen on sacituzumab, that looked more of a classic toxicity profile. So I was surprised to see that this study has been negative and did not actually succeed compared with chemotherapy, and there were some issues with toxicity. So, again, curious to hear the full dataset, because I think that’s going to help us designing other studies. I don’t think we should bury ADCs, or in particular sacituzumab. I think there are probably some methodological issues here that, you know, happy to discuss later with you guys.
AB: Yes. I think it’s really important we see negative data as well as positive data, because I think only about 10% of studies with negative final data actually get presented. So I’m really keen on seeing that data as well because, certainly anecdotally, it did seem to be effective. There seemed to be some issues with toxicity, but I’m old enough to remember when we used to give irinotecan in colorectal cancer and we needed to adjust the drug when we first started using that. Although all of these ADCs are marketed as not chemotherapy or they’re chemotherapy just with a different mode of delivery, so they have side effects some of which are predictable and some of which aren’t. I think it’s a really good point. I’ll be fascinated to see that data, which I hope we’ll see the full set, maybe at ESMO.
So thank you both for those. As we go through today, please do send us any questions you might have, either via the chat box, or there is a URL that you can look at, and we’ll do our best to answer those either live or potentially after the broadcast if we don’t have time.
Non-Muscle Invasive Bladder Cancer
AB: So, I’m going to cover the section that Professor Huddart was going to do, which he was actually doing because one of our surgical colleagues also needed to bow out for tonight, which is why we have a very focused panel. I’m going to look at non-muscle-invasive bladder cancer, and then go on to a bit more about nodal management, both in one section, and save the main discussion for after we’ve done both of these two.
We know that patients really do not like treatment with BCG, so over the last few years, there’s been an explosion of interest in looking at different either drug installations, or different delivery of those drug installations directly into the bladder. And also studies looking at the checkpoint inhibitors, as we know, particularly with pembrolizumab, looking at patients both with BCG naïve and BCG unresponsive disease. It’s important to always be aware that these patients need to have had adequate exposure to BCG, so I would urge anybody who is uncertain about that to please have a look at either papers by Ashish Kamat or the EAU guidelines to look at what the definition of adequate BCG exposure is, so that we can actually separate the patients who are BCG exposed or BCG refractory. Just an important point to make, because that’s something that is in all of the trial protocols as well.
So, let’s look at what Andrea Necchi presented from the SunRISe-1 programme. SunRISe is this programme from Janssen looking in the main at some of their new delivery aspects, which is using the pretzel device which we have, the TAR-200 and the TAR-210. In this particular study it was looking at TAR-200, which was delivery of intravesical gemcitabine.
These were patients initially in four different cohorts that were going to particularly look at the TAR-200 alone, device cohort two. In the rest of the study, you’ll be familiar with this, some patients could have a checkpoint inhibitor as well, or the checkpoint inhibitor alone, and in some patients they had a non-muscle-invasive papillary only disease without CIS. But the data I’m presenting now is looking at patients with or without papillary disease on top of their carcinoma in situ, and they were unresponsive to BCG and didn’t want to have a cystectomy. The dosing of TAR-200 is looking at it as an indwelling device given every three weeks for the first 24 weeks, and then twelve-weekly up through to week 96, with the primary endpoint looking at disease-free survival. But obviously it’s very important for these patients: they will have had some BCG. They will already have a bladder that may have some late toxicity with the BCG. We often see this in terms of capacity, irritative symptoms. So it’s important that if the patients keep their bladder, and don’t want to have a cystectomy, that whatever we do to them actually continues to maintain quality of life.
So, if we look at the study, this was 54 patients altogether. A higher number of patients were current or ex-smokers, and two thirds of them had carcinoma in situ only, with the remaining third having CIS with high-grade papillary disease. The majority of patients actually didn’t want to have their bladder removed, so this is obviously a popular study for patients wishing to keep their bladder. You would infer from that that their bladder must have had baseline good function, because if they didn’t, they would have been excluded from the study, and also a patient might not want to keep a bladder if it was suboptimally functioning.
So these are the money slides, looking at the efficacy, and there was a very impressive overall complete response rate of 77% as we can see here. This is based on both cystoscopy and also urine cytology and a mandatory biopsy at weeks 24 and 48, with still a very impressive duration of response with a median that hadn’t been reached after a median follow-up of 48 weeks, as I’ll show in a moment. Out of the responses, 23 of those patients who were responders, 21 are ongoing with 11 of those patients having a duration of response of six months or more and six of them at twelve months or more. None of these patients have required a cystectomy, so pretty impressive data.
If we look at the duration of response here, with the dark bars showing the complete response, and these swimmer plots here, you can see that the median duration of response has not been reached in the majority of these patients right at the top. So again impressive data. But as I mentioned before, we need to make certain that we’re keeping quality of life good for these patients if they want to preserve their bladder. The majority of side effects were, as you might expect, urinary, because it’s a local treatment. There was an increase in terms of urgency and frequency, and dysuria in these patients, with a small number of patients having grade 3 or more.
We just need to pause and reflect, and see: for some patients, persistent grade 1 side effects would be an issue, so actually I’d like to see what the proportion of change from baseline would be. But actually, overall, probably a fairly good tolerability because the true measure of tolerability is the rate of treatment discontinuation. Only two patients stopped treatment due to adverse events.
A trial in progress is TAR-200 plus cetrelimab in patients who are actually BCG naïve. I’m sure we’ll be able to report this study at future meetings, but just to give you a flavour of this, this isn’t a study which compares patients who are BCG naïve with high-risk disease but they can have CIS with papillary disease, and are randomised to either the TAR-200 device. Again, this is intravesical delivery of gemcitabine through this novel applicator, compared with a combination with cetrelimab or with standard BCG. I’ll await that study with great interest.
Finally, if we look at this particular study, this is again giving intravesical treatment, and this is the novel recombinant adenovirus vector-based gene therapy for patients, again, who are BCG unresponsive. This was a poster presentation at EAU this year from Stephen Boorjian. Pretty impressive data. If we look at this, and this was again with patients with or without papillary disease on top of their CIS. This is the efficacy analysis of this. If we look at patients at 3, 12, 24, and 36 months you can see that the responses that we’re achieving is about half the patients at three months, and this is in the group with CIS with or without TA or T1 disease, going down to about 13% at 36 months.
But the durability of those patients who had a complete response, we can see that it is durable for some of these patients. By 36 months, of those patients who had achieved a complete response, one in four of them still remained free of a high-grade recurrence with this treatment. If you look at those patients without CIS and TA/T1 disease, the results are better. So again, of those patients who had a complete response, 31% remained disease free at 36 months. We can see that there’s quite a big difference here between those patients with carcinoma in situ and with papillary disease, and you can see on the bottom curve, figure 5 here, that there is a difference between these two groups that you start seeing by the time of that three-month cystoscopy.
So we can select patients based on this and it’s certainly a useful drug for us to think about. And this has now got licensing, particularly in the US. Three patients discontinued treatment out of the total population of 151 patients, and that was due to, in the main, because of bladder symptoms.
Nodal Management
AB: I’m going to move on to the nodal management next, and then we’ll come to those three abstracts a little bit later on. When we look at nodal management, I’m thinking about those patients who present with locally advanced disease in the main, but I’m going to start off with a subgroup analysis looking at how we should manage the nodes of patients who are going to have a radical cystectomy, which is a bit different. This was Seth Lerner’s paper that was presented at AUA. Now we know for many years now, based on the Mansoura data, the standard of care has been to look at a template dissection with an extended lymphadenectomy for patients with radical cystectomy. This has been mandated in many guidelines over the years. This has come in based on single-centre data, really, from the Egyptian group, but this was the first study looking at the benefits of this extended lymphadenectomy compared to a standard one.
This was the SWOG 111 study. These are patients with clinical T2 to T4a, N0 to N2 tumours, stratified to see whether they had neoadjuvant treatment and whether they had locally-advanced disease or not, and performance status. They would have had a standard bilateral pelvic lymphadenectomy up to the obturator, external and internal iliac. The experimental arm was this extended lymphadenectomy up to the bifurcation, which would have also included the pre-sacral and pre-sciatic nodes.
You did have an increased nodal yield, as you would expect, and a higher pathological node positive stage because you are going to be taking more nodes out. But, interestingly, there was no disease-free survival and no overall survival benefit with that extended lymph node dissection. What there were were a few more VTEs, but there were no other real high-grade adverse events. Interestingly, among the second endpoints, only having a conduit was associated with a worse overall survival. I was really interested in this, because for many years my surgeons have been telling me that there was a survival advantage in doing this and now this really turned it on its head.
I’m going to continue just looking at the node-positive group now, and look at these two particular subgroup analyses: one with avelumab and one with gemcitabine/ cisplatin/nivolumab. This is the avelumab study, this is a subgroup analysis from JAVELIN Bladder 100, with patients with a low tumour burden, some of whom would have had node-positive only disease, within the pelvis or beyond the pelvis. In the discussion, I’m really interested to see how my colleagues would manage these in real life, so I’m just going to present the data now. Remember that JAVELIN 100 was this study of patients who had locally advanced or metastatic urothelial carcinoma who had had standard chemotherapy, a minimum of four cycles, and then went on to either maintenance avelumab on a two-weekly schedule or best supportive care. With a primary endpoint of overall survival, we know that this significantly improved overall survival by an average of nine months for all comers.
So, what did the study subgroup analysis do? This looked at patients overall with non-visceral metastases, which is the curve on the left of your screen. In the middle, we have lymph node only disease and then on the right-hand side we have those with either pelvic or retroperitoneal lymph node only disease. You can see that for all of these they do better with maintenance avelumab, but actually it seems to be that the survival is actually better in these patients, as you would expect, with lower-burden disease.
Long follow-up… sorry, my slides aren’t moving at the moment, they’re enjoying seeing you so much. Progression-free survival, also very different, isn’t it? What you can see is that actually in terms of those patients on best supportive care compared with maintenance avelumab, the curves diverge very quickly – by the time of the first follow-up scan for many of these patients. So a hint that maybe chemotherapy alone was not really holding things. You can see the patients that do the best are those with lymph node-only disease within the pelvis.
What the authors felt was that these are an important group of patients, and that this is safe. They didn’t find any new signals in terms of side effects. This leads us on to the second study, the CheckMate 901 subgroup analysis, which was the patients who had gemcitabine/cisplatin versus gemcitabine/cisplatin/nivolumab and looking again at the lymph node only group. Just a reminder of this, this is a 1:1 randomisation between standard chemotherapy with or without the addition of nivolumab or placebo. We know that this study was positive. There’s been discussions about whether this is standard of care or not, or whether we would still look at doing chemotherapy switch.
How many patients within this study had lymph node-only disease? We could see that out of the population in total there was around 18% in both groups, but in terms of complete response, half the patients who had lymph node-only disease had a complete response with chemotherapy and nivolumab, but also about half the patients had a complete response with gemcitabine/cisplatin.
How durable was this, and how good was the response? We can see that if you look at patient randomised to nivolumab and gemcitabine/cisplatin, they had a better best assessed response: 57.6% compared to 43.1%, with a higher rate of complete responses in all groups, but in the lymph node only patients, twice as many people had a complete response overall and they did better as a cohort than the main study.
If we look at the best overall response again, you can see that in patients either with pelvic lymph node-only disease or retroperitoneal lymph node-only disease, or distant lymph nodes, they’re doing better with this combination of treatment than with standard chemotherapy. Data, I would say, fairly similar to the maintenance avelumab data as well, so these patients do better if they have lower-volume disease.
The duration of response was twice that of the gemcitabine/cisplatin alone group in those that had gemcitabine/cisplatin/nivolumab. We can see this in terms of overall survival data, significantly better, and progression-free survival significantly better. There were no new safety concerns with this, and what the authors, Matt Galsky and colleagues, said was that this provides additional support, and this maybe should be a area that we focus on with these patients with low-burden disease.
Panel Discussion
AB: I know what I think about this data, and I’d be really intrigued to see what my two co-panellists think, whether this should be what we do for patients who have… I’m going to give you a patient with T4, N2 urothelial cancer. What would your standard management be? Would you adopt either of these two approaches?
ID: Thank you, Alison, and thank you for sticking to the time and opening the discussion. So, if you let me, I’m going to start asking you about the non-muscle invasive and I’m going to say this so people can start thinking what would be their management. Then we can discuss more profoundly about the lymph node positive disease because I think the story really deserves that time.
So, Alison, if I can start with the non-muscle invasive, you presented these interesting studies and if you look at the whole picture we’ve got so many approaches now in the non-muscle invasive, both in the BCG naïve where we’re waiting for further results of all these studies ongoing like POTOMAC, ALBAN and KEYNOTE, CREST and all those. But you focused on SunRISe-1, that is quite interesting, and then the nadofaragene study. So when you look at the whole combo, I was trying to organise my mind and I thought, okay, we’ve got now for BCG unresponsive approaches with immunotherapy systemically, the first data was pembrolizumab a while ago, 40-something percent responses. Then we had some data from atezolizumab with about 27 complete responses. Now we’ve moved into IO locally, that will be nadofaragene, it’s actually a way of doing some kind of stimulation of interferon alpha with a vector. We have the BCG plus N-803, the interleukin-15 agonist, and then the TAR-200 which is cytotoxic treatment locally. This one you presented data of 76% complete responses.
So how do you look at the whole picture? I know that populations are different and the data could not be completely compared transversely. But I was quite surprised about these numbers. I don’t know if the numbers are going to decrease along time - the initial excitement of pembrolizumab got diluted over time. What do you think about TAR-200?
AB: I think it’s very exciting. I think that the particular aspect in terms of toxicity, because grade 1 toxicity is annoying but the fact that there were only two discontinuations because of toxicity is what is the true measure of tolerability. That’s far lower than patient discontinuation rates with the other different agents in this space. I work a lot, as you know, with some of the patient advocacy groups and they really hate BCG but they are terrified to tell you about their side effects because they think that if they stop it they are going to die the following day. That’s really what they think. So many patients have this horrible BCG affected, shrunken, poor capacity, urgent bladder and so anything that makes that worse is not worth doing. So the fact that actually patients weren’t stopping treatment and they didn’t object to quite the frequency of the installation, because obviously it’s every three weeks to start off with, and patients don’t like cystoscopies and this device is put in by a flexible cystoscope done under local. So I was intrigued by that and I think, in terms of how do you choose between them, it is going to come down to all of the efficacy rates probably are going to balance out, to be honest. As you say, we’ve seen with longer term follow-up with pembrolizumab it’s exciting but not as exciting as it was when we looked at the early results. So is this going to stay consistent? I don’t know, but that’s what we need to find out with longer follow-up.
The other longer follow-up aspect is are those people keeping a well-functioning bladder? Are we going to see late toxicity? Because, as a clinical oncologist, I am fully versed in acute and late toxicity with radiation changes. So are we going to perhaps be accumulating more late toxicity with any of these direct installations and with the device installations than we perhaps are with some of the agents that we’re giving systemically? Because only time is going to tell with those, so that’s what we need. We need the longer term toxicity data and the patients that have cystectomies not because of recurrence but because of toxicity. Because there will, I suspect, be some of those in the same way with any other bladder preservation technique in muscle-invasive disease which, as a Brit, we’ve been doing for years in muscle-invasive disease.
ID: Alison, you mentioned there a word that is very critical and it’s functional. I remember once discussing with surgical colleagues, with our urologists, about how in other tumours we decrease the aggressivity of the surgical approach as we’ve done in breast cancer. One of the comments was, ‘Don’t forget that bladder is a functional organ and many times we have to perform surgeries because patients cannot actually live with the bladder in place because of limitations when those bladders are very symptomatic.’ So I think it’s something to keep in mind, as you very well mentioned.
My experience with these devices has been good overall. It’s true that some patients complain about some burning feeling and they come to the clinic sometimes asking if maybe can you remove it before week 3. We did it in a couple of the patients. It’s been different, we participated in the SunRISe programme across non-muscle invasive and muscle invasive. In the non-muscle invasive, for whatever reason, the tolerance has been better. I don’t know if there is any explanation or it’s just random. But the data is quite striking and it’s probably a good approach. The next generation, as you briefly mentioned, the TAR-210 targeting specific tumour types with FGFR mutations, it is quite encouraging. Patrizia, I don’t know if you were exposed to any of these studies in Milan or when you have experience with your patients with TAR-200?
PG: Yes, we have experience with both trials, with the pembrolizumab trial, nivolumab trail and the SunRISe trial as well. The problem is sometimes you have to remove the bladder even in patients with a complete response because they still have functional problems. Because we have to remember that these patients receive a lot of TURBT, BCG before, so the bladder is not like a naïve bladder. So sometimes to wake up 70,000 times during the night could impact quality of life for our patients, not because of the device, not because of pembrolizumab, just because they have been treated too much. So all these treatments will need to anticipate because otherwise the bladder could be still there but maybe the patient could have some problem in a functional way.
ID: So let’s go back to Alison’s question – what do we do with a patient that presents with metastatic lymph node disease? Because I got the impression that sometimes the way the data is presented tries to… I don’t know how to explain it, induces a way of thinking. I think we need to know our history and we need to know that some patients have done very well, regardless of the new approaches. So what would be your approach, Patrizia, you’ve got this patient in the clinic with N2 disease. How do you guys discuss it in the tumour board and what would be the approach?
PG: Yes, I think it depends on the patient. Because for these patients we have different opportunities, as Alison said – avelumab, radiotherapy – so three modalities, strategy with lymph node irradiation or surgery with chemotherapy before. So normally we go ahead with avelumab for these patients because I think if the patient can have their bladder there it’s better than removing it surgically. We have a lot of trials that were published. Even our surgeons that do surgery in lymph node disease, patients like something that has a less worse outcome compared to radiotherapy plus chemotherapy or avelumab as maintenance. So normally we go ahead with avelumab as maintenance.
ID: So chemotherapy plus avelumab, that would be your approach? Yes, you mentioned it very well – these cases are the cases we discuss and initially it’s okay, we’re going to give these guys chemotherapy normally as a systemic disease. Then sometimes when you achieve a great response then it comes back to, okay, should I consolidate? Then it’s when we get into the discussion. As you very well mentioned, the data is not definitive and it’s true that the higher the lymph nodes probably the higher the chance of relapse, even if you feel pretty confident because the response was so satisfactory. So probably most of the patients would follow this path of chemotherapy plus or minus maintenance. I think that would be our approach too but I have to admit we have the individual cases where sometimes we want to consolidate either with surgery or RADS. I don’t know, Alison, what do you do?
AB: I’m going to put the fly in the ointment here because I think if someone has had a spectacular response with chemotherapy then I’m going to offer them chemoradiation. I’m going to give them mitomycin C 5FU, fractions 1-5, 16-20, I’m going to radically irradiate them. I don’t tend to irradiate the nodes unless they’re easily encompassable in a radiation field. But we know that some of these patients do really well and I’d have a discussion with them at the start. I would say, ‘Look, we can either give you avelumab or we can look at going for broke.’ I’d have the surgeons involved in that discussion as well; some patients might really, really want to have a cystectomy. But I would explain that actually we saw in the BC2001 study, which admittedly the patients didn’t have N2 disease but they could have up to T4a disease and N1 disease, we didn’t see a high rate of pelvic relapses in that study, despite the fact that we only treat the bladder, we don’t treat the nodes. That’s because of the systemic treatment that you’re giving. So I’d be very comfortable doing that and that’s what we’ve done many times before. But now, since we’ve got avelumab, I would have a discussion with the patient and say that’s an option as well. When they ask me what I would do I say, ‘If it was me, I would have radical treatment at this point.’
ID: Alison, last question because we’re running a little bit behind schedule. For some reason we’ve been listening, okay, patients with lymph node disease they should be treated with a combined approach with chemotherapy plus nivolumab. I don’t know how much of that would you agree or do you think we’re interpreting the results of this study?
AB: I think that, yes, are we going to potentially miss some patients that don’t respond to chemotherapy and don’t go on to maintenance avelumab? There may be a few but actually if we look at the results of those patients that we give gemcitabine/cisplatin to and then put onto maintenance avelumab it looks very similar to the gemcitabine/ cisplatin/nivolumab data. I think we’ve got very comfortable with giving that as a uro-oncology community. I don’t feel that there’s enough of a difference in benefit, certainly for us to change that. That’s our position at the moment, really.
ID: That would be my approach. I don’t know if Patrizia has a different one but I don’t actually think the data is enough to change my approach. If you look at the differences, even in responses, it’s just the additive or the addition of the immunotherapy effect. So I wonder whether the results that we’ve seen in CheckMate-901 is basically due to the maintenance compound, as you mentioned. So, to me, the standard is still chemotherapy plus or minus avelumab maintenance if there is a good response until we get access to the new combinations that will come in later. Patrizia, anything that you want to add about this before we give you the podium?
PG: Yes, I absolutely agree. The take-home message, to me, from CheckMate-901 is start maintenance as soon as possible. That’s the main message for me.
ID: Great, okay. So we’re going to have to move on. There are many questions there that we can go through then but we’ll have, I’m sure, another occasion. So I’m going to move on and we’re going to invite Patrizia Giannatempo, a medical oncologist from Milan, as Alison said at the presentation of this session. She’s going to talk about the perioperative data that has been recently presented. So, Patrizia, the floor is yours, thank you.
Perioperative therapy in UC
PG: Thank you very much Ignacio and good evening to everyone, to all our colleagues that join us online. I am happy to present updated results focussed on the perioperative setting.
So, the first abstract that I would like to present you is the avelumab trial with the use of avelumab in the neoadjuvant setting. We called this trial the AURA trial. The AURA trial is a multicentric non-comparative randomised phase II trial where patients were randomised to receive avelumab plus chemotherapy or avelumab alone, according to cisplatin eligibility. We had two cohorts, the first cohort is with patients randomised to receive gemcitabine plus cisplatin plus avelumab or dose-dense MVAC plus avelumab. In the second cohort the patients are cisplatin ineligible, this means they are randomised to receive paclitaxel plus gemcitabine plus avelumab or avelumab alone.
The primary endpoint was complete pathological complete response and, as you can see here, the rate of pathological complete response is slightly better in the dose-dense MVAC plus avelumab arm compared to gemcitabine plus cisplatin plus avelumab. When we look at the results in cisplatin ineligible patients we have that avelumab alone is doing better compared to the chemotherapy combination.
Looking at cisplatin eligible patients, the event free survival, as you can see here, there is an advantage in dose-dense MVAC. They got the preliminary 36-month event free survival at around 79% in the dose-dense MVAC compared to 62%. Overall survival in the same population is higher in the dose-dense MVAC plus avelumab arm.
Moving on to cisplatin ineligible patients, surprisingly there is no difference according to event free survival and overall survival when they add chemotherapy to avelumab. We have the same event free survival and overall survival in patients that received avelumab alone or paclitaxel, gemcitabine plus avelumab.
Something very nice and something that maybe we already know that patients that achieve a pathological complete response in the cisplatin ineligible cohort, they have better, longer overall survival compared to patients that did not achieve a pathological complete response.
So, in conclusion, I think that this trial, even though the numbers are very small, we are talking about a very small number of patients included in this phase II trial, it’s something that could be interesting to notice that the dose-dense MVAC plus avelumab had better outcomes compared to cisplatin and gemcitabine and even the toxicity correlated with when we add immunotherapy, avelumab in this case, to the dose-dense MVAC. Similar results in terms of toxicity. In the cisplatin ineligible cohort there is no effect of the addition of chemotherapy to immunotherapy in this specific population. Once again, the use of pathological complete response is something that is very useful because it is correlated with better survival outcomes. There is going to be a phase III trial where they try to validate a biomarker and try to select patients for these treatments.
So let’s move to the next presentation. The next presentation was presented by Professor Tom Powles at EAU this year. It was a very nice presentation about the use in this subgroup population of IMvigor011 and it focussed on ctDNA negative patients. Just to remind you, the overview of the trial, they included in this trial a high-risk population that received neoadjuvant chemotherapy or if they did not receive neoadjuvant chemotherapy they are not fit for adjuvant chemotherapy. At the time of cystectomy they had pT2 to pT4 or pT4a with lymph node positive disease.
The patients were randomised where they collected plasma for one year and they categorised patients according to ctDNA status. In this presentation Professor Powles focussed only in the ctDNA negative patients, those are patients that are still negative for all year during the control of the entire year of the trial. So they ended up with 171 patients, ctDNA negative patients. If we look at the characteristics we see that they are quite similar to what we see normally in our daily practice, daily clinic – patients with lymph node disease, patients that are treated with chemotherapy before or patients that they removed a lower number of lymph nodes. So normal patients that we see normally in clinic.
For the disease free survival in the ctDNA negative population only around 10% of patients had experienced some relapse during the follow-up. The median follow-up of the trial was 16.3 months. Let’s move to see the PD-L1 status – there is no difference according to PD-L1 in patients with ctDNA negative during the follow-up period. In terms of overall survival, almost the entire population, the ctDNA negative population, is alive during the whole entire follow-up period. That’s very important in practice.
When we try to find some clinical characteristics to compare patients that experienced some recurrence during the follow-up or the patients without recurrence during the follow-up period, there is no difference in terms of clinical characteristics. What does this mean? It means that the ctDNA negative is something that is superior compared to clinical data like lymph node positivity, type of surgery that they received, neoadjuvant chemotherapy that was received or not. So the ctDNA impacts overall survival irrespective of clinical data, clinical characteristics of the patients.
So, in conclusion maybe ctDNA testing could be a very important test to use in the risk stratification tools to identify patients that we could propose adjuvant treatment. Of course, they are doing a phase III trial to validate this data. Maybe there is a part of the ctDNA negative patients that maybe could be spared, they could avoid adjuvant treatment for this specific population.
Let’s move to the other abstract, the third abstract. Once again it’s focussed on adjuvant treatment. This is the CheckMate 274, this is the first overall survival data that Professor Galsky presented at EAU this year. Just to remind you, once again it’s a high-risk population that received cystectomy and this means they could have received chemotherapy, neoadjuvant chemotherapy, or they were unfit for adjuvant chemotherapy. Once again, pT2 to pT4a or lymph node positive patients. So exactly the same population. They were randomised to receive nivolumab versus placebo.
We already know the disease free survival data, it’s already published. The nivolumab arm had an advantage, a statistical advantage, in terms of disease free survival. But in EAU 2024 they presented the preliminary overall survival results. This is the interim overall survival data, it’s in favour of nivolumab. The nivolumab arm median overall survival was 69.5 compared to placebo where the median overall survival was 50.1. When we look at PD-L1 status there is a higher advantage in PD-L1 positive patients treated with nivolumab compared to patients treated with placebo.
The subgroup analysis is similar to what we already know. The huge advantage is in patients with bladder cancer, patients with PD-L1 positive status and with pT3 disease compared to pT0 or PT2.
In conclusion, this is the first data about overall survival but we should stress this is only the interim analysis of overall survival. Of course we need time to see the follow up, to see the final overall survival data. But to date, nivolumab as adjuvant treatment seems superior in terms of disease free survival and maybe overall survival as well.
Let’s see the fourth abstract, this is once again an adjuvant treatment. This is the AMBASSADOR trial. It was presented by Professor Andrea Apolo at the ASCO GU of this year. The same population as the previous two trials but they randomised patients to receive pembrolizumab or observation. The primary endpoints were disease free survival and overall survival.
Here mainly the difference is the observation arm. The problem of this trial is that they used observation as the control arm and in the meantime they go ahead with the trial – FDA approved nivolumab as adjuvant treatment. So during the AMBASSADOR trial there were a higher number of patients that withdrew consent and they went ahead with a different treatment, maybe nivolumab immunotherapy in other ways.
So we can see the disease free survival, so it’s a positive trial in terms of disease free survival advantage. In particular pembrolizumab, the median disease free survival with pembrolizumab was 29 months, so slightly superior to CheckMate 274. The median progression free survival in the observation arm was 14, hazard ratio 0.69. So a positive trail in the primary endpoint, in one of the two primary endpoints.
We had a similar subgroup analysis to the CheckMate 274. There is a major advantage for patients with bladder cancer compared to patients with other types of tumour. There is no difference in terms of PD-L1 status or in terms of disease free survival advantage, no difference once again according to if the patients received neoadjuvant chemotherapy or not.
Once again something strange, in this case with the use of pembrolizumab instead of nivolumab there is no difference in terms of PD-L1 status. So PD-L1 positive patients performed slightly better to the observational arm but when we look at PD-L1 negative patients they still have a benefit even in this subgroup population.
They showed interim overall survival data that was the second primary endpoint. As you can see here, it was quite disappointing because there is no difference in terms of overall survival. But we have to remember that there are a lot of patients that will have immunotherapy during their life after the randomisation because in that time there is a high number of patients that withdraw consent or receive immunotherapy not in this outcome.
Once again the PD-L1 status, even though with the overall survival result there was no difference in terms of PD-L1 status and advantage in overall survival.
So, in conclusion, I think that could be an option in the adjuvant setting. We had an advantage in disease free survival. Of course, we have no reliable results in overall survival so that’s still something we have to think about when we propose adjuvant treatment to this kind of population. The PD-L1 status, in this case, it is not helpful to set the picture for the adjuvant setting, even though it’s exactly the same population as CheckMate 274. Of course, pembrolizumab is well tolerated, there are no new safety signals. So let’s wait until the next trial, the next publication, the next presentation about the use of ctDNA or the other additional correlatives of this trial.
Let’s move to the last paper. This was a very nice paper that Professor Alison published this year in JCO. It’s the results of a phase III randomised trial focussed on adjuvant therapy in upper tract tumours with lymph node disease or muscle invasive cancer. This is, I think, the most important trial in the adjuvant setting in upper tract tumours. As you can see here, when we have patients with upper tract in the adjuvant setting we have an advantage of using chemotherapy in terms of disease free survival, disease specific survival and overall survival. Even if we propose to these patients cisplatin or carboplatin we still have an advantage in terms of control of disease and overall survival.
So thank you for your attention. I think we can open the panel discussion now.
Panel Discussion
ID: Thank you Patrizia, grazie mille. We don’t have a lot of time for questions. I’ll remind people in the audience that they can send their questions electronically; some questions already came in we’ll be happy to answer. Just one question to you, Patrizia, you’ve presented AMBASSADOR, you’ve presented CheckMate 274, we already spoke that the data in overall survival is still immature and it didn’t cross the prespecified boundary. Also we have another study, 010, so three studies in the adjuvant setting with immunotherapy and we saw that they are not completely aligned: 010 completely negative, AMBASSADOR positive for disease free survival, negative for survival, CheckMate 274 positive for disease free survival with a PD-L1 signal and we’re still waiting for OS. Is there any explanation? Are we comparing pears with oranges? How can we explain these kinds of inconsistent outcomes, if there is an explanation?
PG: Yes, thank you for this important question, it’s a very tough question. There is no really clear explanation about that. We know from the metastatic setting that the immunotherapies are not the same for urothelial carcinoma – we had a negative trial with atezolizumab but we have a positive trial with pembrolizumab in the metastatic setting. So there is something that is not completely clear in urothelial carcinoma and we should better understand the mechanism of action of each of these drugs in urothelial carcinoma.
ID: Allison, thank you Patrizia, I cannot remain quiet. Patrizia presented the POUT study – you’re the first author and the mother of this study. We didn’t see the overall survival impact in the overall analysis but you presented some data on restricted mean survival time. Can you expand a little bit on that? Can you share with the audience what does it mean?
AB: In POUT the primary endpoint was three-year disease free survival because that was identified as a really important endpoint for patients in an adjuvant study and the majority of patients do relapse, if they’re going to do, within that timeframe because UTUC is a more lethal disease than its sister, bladder cancer. Very different in terms of its mechanisms, as well, and its biology. So a secondary endpoint was OS but we met the primary endpoint of three-year disease free survival early, we met it at two years. The independent data and monitoring committee met and their recommendation was that we should shut the study at that point. We did have a really long discussion in the trial’s management group about whether we should continue to recruit patients to the study. But when you see a study were there’s a 21% difference in disease free survival which is then maintained at 17% at five years, the same in metastasis free survival, it was very difficult to be able to say ethically we should continue recruiting just so we could get an overall survival endpoint for this study. So the study shut at 261 patients of the planned 346 we were going to originally look at. So it’s underpowered for overall survival which is why we had to do this different statistical analysis. There is a strong trend to overall survival in this study but it didn’t meet statistical significance. I guess the thing to say is the reason it was not an OS primary endpoint is that would have needed a humungous study in terms of numbers because it’s a rare tumour type. So anything with a rare tumour type will require global participation if you want an OS endpoint. At the time the study was set up, which started recruiting back in 2012, no matter what I did, nobody wanted to do this study globally with us at the time. So that’s why it was run from the UK, delivered from the UK.
ID: We didn’t know each other at the time or I would have done it with you. You know that. Can you defend in 30 seconds, I know this is going against me because I’m eating my own time, but can you defend in 30 seconds carboplatin? Sometimes I get the discussion in the tumour board about the carboplatin and in the POUT.
AB: Yes I can because we’ve seen it in the initial and then in the subsequent five year follow-up data, that data has been strengthened. The carboplatin arm the confidence limits are always going to be wider because they were fewer numbers but you do a tester interaction in these subgroups. That side has shown this in the STAMPEDE population as an elegant thing, a publication that it’s not all about p-values. People often get focussed on that and they don’t realise about the test for interaction and if that is negative it shows you that the primary endpoint is the same in your subgroups. That was further strengthened in the updated analysis from JCO this year. So patients should have carboplatin, not just cisplatin. But we’ve also done a data looking at all of this as a publication with myself and Rick Bryant and a couple of other colleagues from 18 months ago looking at how when we look at the strength of the carboplatin data in other studies it has probably been underestimated.
ID: Thank you. That was a little bit more than 30 seconds but I loved it. Okay, so we’re going to move on, I promise we won’t take too much extra time. I’m going to move to the last presentation.
Advanced disease, variant histologies, FGFR, further perioperative
ID: I’m going to try to be really quick and this is a bit of a, we say in the Spanish pot pourri, I don’t know if these words exist in English but it’s a mix of different things. Because I am going to talk about perioperative studies, I’m going to talk a little bit about advanced disease and then about practical situations that we have in our daily clinic such as the variant histologies and some tumours with particular alterations. So let me see if I can do that in ten minutes and get everybody still awake.
I’d like to start with this work from Antonio Cigliola from Milan working with Dr Andrea Necchi. This study was quite interesting to me. The study is interesting because they are testing the value of perioperative sacituzumab. We spoke about sacituzumab before, it’s a drug that these days is a bit controversial, but they were brave enough to design a study in the neoadjuvant context where they wanted to test what was the activity of sacituzumab govitecan, this antibody-drug conjugate, anti Trop-2, that is a topoisomerase inhibitor is the payload of this drug. So we all know about topoisomerase inhibitors. So they said, okay, why don’t we test the activity of sacituzumab in patients with muscle-invasive bladder cancer, both as a single agent or in combination with pembrolizumab.
So they designed this study called SURE-01 and 02. Antonio was brave enough to present at this ASCO the results, the preliminary results, of SURE-01 and I’m happy to share those with you. So this is basically the design, nothing really fancy other than these guys do PET scan and bladder MRI which is not accessible to everybody around the world. They also do a UGT1A1 test which is something that we should start considering doing if we’re going to be using this drug in the future. It’s something doable if you talk to your genetics department. So basically a pretty standard population of patients with muscle-invasive bladder cancer, T2-T4, good performance status. Patients either refused, and that was the majority, they refused neoadjuvant cisplatin-based chemotherapy or they were ineligible for that.
As I said, we’re going to present the SURE-01 which is four cycles of SG at the usual doses of 10mg/kg on days 1 and 8 followed by cystectomy and then post-surgery management according to local guidelines. These guys thought, okay, our primary endpoint is going to be pCR, which makes a lot of sense, and then they added some other interesting correlative studies. So this the big picture of the study, as you can see small numbers – 31 patients were actually treated. Probably here what calls your attention is three patients died during the adjuvant treatment. If you look at the table, I really want to bring your attention to the last lines where you can see that there was a significant number of patients with variant histologies and histologies as complicated as squamous cell, plasmacytoid, micropapillary. So I think we need to take that into account. But let me show you one thing. What happened with this study is they treated eight patients and this is what they saw – grade 3-4 neutropenia, they saw grade 3-4 diarrhoea in percentages as high as 75% and 50% respectively. So unacceptable. And they saw some deaths and one of the deaths was considered treatment related, it was due to sepsis in the context of severe neutropenia. They had an early death considered treatment related.
So that will probably call your attention and say, okay, what’s going on with this regime. So what they did is they adjusted, they amended the protocol to 7.5mg/kg, they introduced something that now is going to become very common with this drug, which is GCSF, and they excluded patients that were at high risk for febrile neutropenia. The thing has completely changed because when they dose reduced, if you look at the last column here, you can see a lot of zeros. So those are the known toxicities that they saw when they dose reduced to 7.5mg/kg. So this is an important message and the message is really 10mg/kg in this context is not the right dose and when you dose reduce toxicities are pretty manageable. The other thing they saw is that UGT1A1, which is actually the enzyme that metabolises this compound, it seems to matter. So something also to take into consideration.
But look at this – 47.6% of ypT0. Many studies in the perioperative setting cheat a little bit because they include the pTis, the pT1, these guys, this is ypT0. So real pCRs, 47.6% including weird histologies. So to me the message is once you dose adjust, this drug has some activity that probably deserves at least some consideration. This is what they did with a number of patients that achieved what they called clinical complete response which is something that is not formally defined worldwide. But they considered patients with a good VI-RADS, cystoscopy negative, ctDNA negative and pCR and they said, okay, maybe you don’t need to be operated. They did that with six patients and the patients are still doing okay. So this is another interesting thing about histology and biomarkers.
And the conclusion I already mentioned. We need to at least look at this with some curiosity. This drug needs to be adjusted in terms of dosing but this drug probably has some level of activity in this group of patients. It’s interesting to me the ctDNA data and it’s interesting to me that had activity in patients with variant histologies. So an interesting study in the perioperative.
What is the next study that I want to share with you is EV pebrolizumab versus chemotherapy. Probably all of you have heard a lot about this study already and this is one more squeeze of data. That was presented by Jens Bedke in the EAU 2024 and what he presented was a sub-analysis of the whole data. So these graphs have been shown many times and we all understand very well that EV pembrolizumab really makes a difference in first line when compared with chemotherapy, a hazard ratio of 0.45 in PFS, 0.47 in overall survival.
What he presented was new data, it was, okay, what about upper tract compared with lower tract. What he showed is it works, it is active and the numbers are impressive, both in PFS and OS. So probably nothing unexpected but it's also data to give us more confidence about the use of EV pembrolizumab. We may discuss later whatever pros and cons and obviously if anyone is interested in this field and the toxicities associated with EV.
Let’s move to the last study which is this abstract about the efficacy of erdafitinib. This is a study which is an analysis from UNITE. UNITE is a big effort in the United States where a number of centres are collecting information for real-world data analysis. So here what they did is collect patients that presented with some FGFR2 and 3 alterations, some of them canonical, so typical, some of them non-canonical, so less frequent. Also they included two other alterations – CDK2 and MTAP. Basically the message is these alterations that are not so frequent in cancer patients with urothelial carcinoma, probably 10-15% in real life, I don’t know if 20% is an overestimation, at least in my clinic I don’t see so many. We know from previous trials that erdafitinib in this population works. This work is very timely because erdafitinib is probably going to get reimbursement in Europe coming up relatively soon and we’re hoping that the EMA is going to give a positive answer, probably after the summer. Then we may have this drug at least being discussed in several European countries. Its place right now is after platinum and checkpoint inhibitors.
So what they presented here in this study was this series of patients from North America with different alterations in FGFR2 and different alterations in CDK2 and MTAP. What they saw was really activity, I’m not going to read through this table of the messages. Erdafitinib works regardless of canonical or non-canonical and it works regardless of CDK2 or MTAP loss. So, of course, this is only a small picture of the whole data but I think that’s the key message. Probably the message that I’d like to send to the audience is we need to start thinking that there is also personalised medicine in bladder cancer. We need to start doing FGFR. It’s quite impressive that this is not as implemented as one would think, so many hospitals do not have access still to FGFR testing. I think it's something we should start doing from the very first line because bladder cancer doesn’t give you a lot of time to make decisions.
The last one has to do with histological subtypes and this is the data from JAVELIN. We know JAVELIN very well and this piece of work that was presented by Yohan Loriot has to do with a sub-analysis, again real life. So how do patients perform with avelumab maintenance if they don’t have urothelial carcinoma? Small numbers – we’re looking at 44 and 57 patients in each group but the message is that the data is quite comparable to the overall analysis. The hazard ratio is a little bit higher, 0.74 versus 0.69, I believe, in the whole study, but the data is actually quite consistent – 19 months versus 14 months. So what I would say is we should not exclude patients from receiving avelumab maintenance only based on the histological subtype.
Panel Discussion
ID: So, with this, I think I’d like to stop here and open these presentations for discussion. So Alison, Patrizia, I’m happy to answer any questions. Thank you.
AB: I’d like to just go back and look at the sacituzumab data because I think I started off by saying that I’m old enough to remember irinotecan when it first came out when I was a resident. These patients used to come in with dreadful diarrhoea, dreadful dehydration, and then once we adjusted the dose and realised that actually the dose that we were using from the phase III study probably wasn’t the best, we then never had a problem. I just wonder whether sacituzumab has got a bad name because the MTD that was used was a bit too high for whatever reason. You can see as soon as they’ve done the dose reductions that things look significantly better, don’t they? So I was really interested in that data, to be honest, because obviously it’s always tragic if you’ve got a neoadjuvant death, isn’t it? Any death is tragic but something in a curable cancer is just even worse. So I was particularl
