Christof Vulsteke: Thank you so much for having me. And indeed, I'm also very excited when I first saw the data results. I had already the feeling during the study, I saw these results in my patients, but it's always so exciting when you see the real data. So what we presented at the presenting session in ESMO was the phase three KEYNOTE-905 study of perioperative Enfortumab vedotin plus pembrolizumab in participants with muscle-invasive bladder cancer who were Cisplatin-ineligible. And just to give you a bit of a background, patients who are not amenable for systemic therapy and are Cisplatin-ineligible, so what do we do in muscle-invasive bladder cancer? We can give Cis/Gem, we can give MVAC dose-dense with or without preoperative durvalumab.
We can give adjuvant nivolumab for high-risk disease if there's still after neoadjuvant chemotherapy, a lot of disease left and high-risk features. But when you are Cisplatin-ineligible, we go for upfront surgery. And these are the patients that are normally older, you're frail, you have more comorbidities. And if you look in the literature about what's the prognosis of these patients, it looks very poor. But there is not that good data. No prior phase three trial has also shown a benefit from some kind of preoperative therapy in this population. And based on what we have achieved in the EV-302 study, so we presented that our lead author presented that in 2023 in a metastatic setting, the first-line setting. There we doubled the progression-free survival, we doubled overall survival. We had a 29% complete responders, of this complete responders, 75% was still in a complete response after two years. So maybe we have cured some patients with liver metastasis or lung metastasis. And why don't just bring this fabulous regimen into the peri-op setting? And that was the aim of this study, the KEYNOTE-905 Cisplatin-ineligible situation. But it started at the two-arm study. It started with pembro-monotherapy versus the control.
It was only when these results in the metastatic setting looks so promising in the EV-103 and then EV-302 that we added that third arm, Enfortumab pembrolizumab arm. And then it was a three-arm study. There was a period of 1:1:1 randomization, but in the end of the study, it was 1:1 only for the Enfortumab pembrolizumab pre-op. So three cycles preoperative versus the control. And then after the surgery, we had six cycles of Enfortumab vedotin, and we have the remaining 14 cycles of pembrolizumab because mostly in all the trials we give pembrolizumab for one year. And here we chose for the Enfortumab vedotin for nine cycles. So, three before, six after. What was the rationality of that? Well, in the metastatic setting, it seemed that nine cycles is the median number of cycles. And we wanted to optimize the curative treatment here. So we chose for nine, three, four, and six offers. So, primary endpoint was event-free survival for all patients centrally assessed event-free survival. All patients concurrently randomized to Enfortumab vedotin and pembrolizumab versus the control and key secondary endpoints overall survival and pathological complete response. So it was that what we presented; that the control versus that interventional arm with EV and pembro at ESMO.
And this is a bit of timeline of how it evolved. And I think I have to just have to stress there that in 2022, also nivolumab adjuvant because they're the FDA approval was there. EMA approval was there, but restricted Europe for PD-L1 positives but you could also give adjuvant nivolumab for investigative discretion. And all the rest I mentioned already a bit. But also at that moment, the cisplatin declining patients could also enter the trial because there are these patients that don't want that 5% overall survival of a cisplatin-based chemotherapy. But this was a minority of the patients included, but they were allowed from 2022. So the statistical analysis for sake of time may be short, but to say that if we could reject the null hypothesis of event-free survival over survival and PCR, pathologic complete treatment, rate only then we could test for a pembro versus control. So this is the first interim analysis of the primary endpoint. And like we showed, there was met. And before I show the real result, just to go to the participant disposition. So 170 patients with EV + pembro arm. 174 to the control.
The vast majority completed the neoadjuvant phase with Enfortumab vedotin and pembrolizumab. And the patient not undergoing surgery were quite similar in both arms. So 21 in the EV+P arm and 18 in the control arm. And 100 patients started adjuvant phase. So you lose some patients indeed after surgery that only 67% started adjuvant phase, but it was the protocol that we started adjuvant phase. So the different reason why they didn't start adjuvant phase, there's a reason that you inform the patient about a pT0N0 that patients say, "Why I don't want the adjuvant portion." Don't forget, surgery was mandated in the study. So it's a very invasive procedure and a lot of patients still have the ongoing adverse event of surgery and don't want in this frail comorbid population adjuvant phase. And there was also... So, the patient withdrawal. So multifactorial reason but only 67% started adjuvant phase, but it was mandated for protocol to start if the patient wanted. So this is the baseline characteristics. A bit older than the other in the VESPER trial and in the NIAGARA trial, it was around 63, 68 years. Here, the median age was 73 in ITT population. ECOG performance 2 was allowed and it's not like in other trials. Mostly they are excluded. The majority is cis-ineligible, like it was a cis-ineligible trial. It was only at the end that we allowed the cis-eligible. And the main reason for cis-ineligibility was a renal impairment. And this is something striking that the investigator, in 75% of patients, the investigator says it's a clinical T2, but we did hear central assessment.
We did together with pathology specimen. Everything was sent to central. And here 75% is a clinical T3. So you see that in that all that staging in a lot of trials, it's quite messy. So I think clinical staging should be standardized, and in my opinion, centralized to have this good clinical staging in a clinical trial. So most of them were pure tumor carcinoma. There was variant histology features were allowed if it was less than 50%. And these are the results. And I think you don't have to look in the room because this is very early separation of the curves, very sustained. And you see the event-free survival at 24 months in the control arm is 39%. So this is a population that does very poor. And this is a cis-ineligible population. So don't compare with the NIAGARA. There's a bit of an overlap, but there it was a cis-eligible population. So this is a patient population that does very poor. And you see with EV+P, it's 74%. So hazard ratio is 0.4 in favor of the EV+P arm. In all subgroups, it was there. You can slide and dice all subgroups, but it's there for all subgroups. And then the overall survival was also already hit with a hazard ratio of 0.5, already 63 versus 80%. So it's quite good. And you see here at the right side of my slide also that patients could receive subsequent therapy.
A criticism could be, "But the control arm didn't receive EV+P." But it was not there, so we could not give it. It was not available. It was not there, the results because it was already closed for recruitment. So I think this is just a very good result also for overall survival. And this is the first five of phase three trial that shows the overall survival in this patient population. You can also slice and dice here and the benefit is there for all of them. And for me, this paves the way, I think, for bladder sparing approach. Because if you see that in 57% of patients, you can say you have pT0N0, then at least in my center, my patients ask me, "Why did you remove my bladder?" And I say, "Because it's still the standard of care." But it should pave the way, I think, for bladder-sparing approaches. And the denominator we used here was ITT population. So all the patients that were pre-op informed about a fantastic response and they refused to undergo surgery, they will also count it as non-response. In fact, I think it will be even 63% if you... But to be honest here, it was 57% pCR. And this is also the highest pCR ever reported in a phase three trial. Safety, well, you see in the control arm that patients in this population undergoing such an invasive procedure have a lot of side effects. We see 64% Any grade TEAE in the control arm. So this is a comorbid population.
But if you look at the adverse events during the surgical phase. The grade three adverse events, the serious adverse events and adverse events leading to that, it was quite similar between both groups. So it doesn't seem that EV+P impacts or worsens surgical complication rate. You see, for example, treatment-emergent adverse events leading to death in the surgical phase only were four in the EV+P arm and nine in the control. The safety, what we could expect, but maybe also for the sake of time, I can go to the side effects of special interest that will be more visible. This is during the surgical phase. It's mostly anemia, prostate cancer, similar between both arms. Prostate cancer you can expect in the context of a cystoprostatectomy and urinary tract infection. But regarding the safety of the products, I think this is more important is what we can expect with Enfortumab vedotin. Skin tox in the first two months, you have to be experienced also as an oncologist, a medical oncologist, to see that the red flags of bullous, of fever, painful rash. Then you have to intervene if you know that it's very manageable. First two months, skin has to be watched on. And if you go to the fourth, fifth cycle, think about peripheral neuropathy. And if your patient is getting disabled, I think it's also a bit of fault of the doctor. You have to stop when it's necessary, give drug holidays, reintroduce when it's less than a grade one.
So polyneuropathy and skin tox is what you have to remember. And for pembrolizumab, we all use this in lots of cancer. So this is what we could expect for pembrolizumab. So in conclusion, I think a hit for event-free survival, a hit for overall survival, a hit for pCR. So I can only dream that this result will be replicated in the upcoming B-15 trial in the cis-eligible population, but I'm very hopeful. So with that, I'm open for questions or discussion.
Ashish Kamat: Thank you very much, Christof, for that presentation. I mean, again, when the data came out at ESMO, everyone was completely floored by it. We were expecting, like you said, to see this phenomenal result, but to actually see the separation of curves, to see such a improvement in EFS and overall survival. And of course, in the pCR, really a huge benefit to our patients. You mentioned a couple of things, and I think it's very important. Obviously, we have to do work, but share with me some of your thoughts. Was there an assessment of clinical complete response prior to cystectomy? And if so, how did that correlate with the pathologic CR rates?
Christof Vulsteke: All patients received a baseline CT scan and a new CT scan before surgery, and they were informed just about results, but we have all the samples for circulating tumor DNA, but it's not analyzed yet. So they were not informed about this. So patients were asked to go for surgery, it was mandated by the protocol. So the patients that refused surgery was mostly because of patient withdrawal that they said, "If my scan as well, I will stop." Or because of other events. So it was not a trial designed to look for a clinical complete response and say you can avoid. Surgery was still mandated. So there's not that correlation yet of how do they do and what is the outcome. But what we can say from statistical point of view, that a patient that did not undergo surgery, he was also followed up for event analysis. And if he did have an event, he was counted as an event and he was not having an event, he was censored. So these patients that had a patient withdrawal, all because of being informed about a very good clinical response, they were all censored and they're all doing very well and don't have an event yet. So you can imagine that also these patients have had a pT0N0.
Ashish Kamat: Yeah, no, I think, and of course, like you said, the data exists and looking at some of the correlative output will be very important to help inform the bladder preservation paradigm that we are all very interested in. One of the things is from my perspective in academia and in your perspective, I think EV + pembro, it should be the standard of care, just an access thing. But I personally have done surgery on patients who received EV + pembro, but I do hear from colleagues in the community or elsewhere that sometimes the surgery is challenging, they can't do it properly, et cetera. I haven't sensed that in my own personal experience. I wanted to ask you, have you collected, do you have data on surgical complications related to EV + pembro? Any insight there that you can share with the audience?
Christof Vulsteke: Well, based on the safety issues, the problems during surgery and surgical phase were very similar. And so we don't have a hint there that the adverse events of any grade. Grade three or leading to that were different between the experimental arm of the upfront surgery. So based on this trial, it didn't seem to worsen surgical complication. Most of them received an open. Most was [inaudible 00:15:58], some were robotic assisted, but most was all open. It was around 242 sites. I think it's sometimes a bit higher than if you should have only centralized centers, some adverse events, but it was similar between two arms.
Ashish Kamat: And then lastly, I'm going to ask you a practical question because we are going to face this in the clinic. And again, I know the data has been collected and we don't have it yet, but in your experience, if a patient receives preoperative EV + pembro, has surgery and is now pCR, and let's assume you've done a ctDNA and ctDNA is negative, how will you use the data that you know and are aware of to counsel that patient on getting adjuvant EV + pembro? Any insights or pearls there from your practical experience?
Christof Vulsteke: Well, this is the perfect question that I think about every day because the trial is what it is. The trial says you need an adjuvant portion. We didn't have that arm without adjuvant portion. So I can only say scientifically the adjuvant portion is needed. And to answer that question, maybe also to have a trial without that adjuvant portion. But the IMvigor 011 inform us when to start, not when to stop. Here, it should inform you when to stop, but IMvigor 011 tells you when to start adjuvant therapy, not when to stop. But I think you're right that it will be like that if you have already a clearance of your circulating tumor DNA and you have pT0N0 and you have also maybe urinary.
The tumor DNA is negative and you have a negative cystoscopy and MRI. I think that in the real world, a lot of doctors will say to the patient, "I think I will be safe to omit further adjuvant therapy," but I think the patient should have to be informed. The science is that you need adjuvant portion. But in my personal experience, I would be very rigorously follow the patient, for example, neuropathy. And as soon as he has some, I will end earlier, extra drug delays. So with this information, if patient pT0N0, and he has, for example, after two cycles, adjuvant therapy already with some kind of neurotox, I will stop, extra delays, so not be that rigorous to complete all of them in the highest dose intensity.
Ashish Kamat: Absolutely. No, very, very practical consideration. Christof, we could talk forever, but of course we have to wrap it up in the interest of time. So congratulations once again, and thank you for taking the time and sharing your wisdom with us.
Christof Vulsteke: Thank you for having me.