A roundtable discussion, moderated by Thomas Powles, MBBS, MRCP, MD, covered the latest updates in bladder cancer treatment and research, including recent data from ESMO 2023. Dr. Powles was joined by Shilpa Gupta, MD; Sia Daneshmand, MD; and Petros Grivas, MD.
In the first segment of the roundtable series, the panel kicked off with conversations about data released at ESMO, including EV-302 and CheckMate 901.
—
Dr. Powles: Today we’re going to talk about some of the changes in urothelial cancer. We’re going to talk initially about advanced disease and we’re going to talk about first-line metastatic [disease]. We’re also then going to dive in and talk a little bit about erdafitinib and ADC [antibody-drug conjugate] doublets. Then we’re also going to discuss a little bit some of the changes we’re seeing in non-muscle invasive bladder cancer, which is I think super exciting. We’ll do it in that order. We’ll talk about some toxicity issues too, of course. Who wants to kick off with what happened at ESMO? Shilpa, would you like to start?
Dr. Gupta: Yes, of course. It was my first ESMO in person, and I thought it was the best ESMO ever. The frontline bladder cancer presentations and the standing ovation that you got for the EV-302 [trial] was really mind-blowing. I think for the first time we’ve seen in 3 decades that something beat platinum, and it was really great to see that, as well as the CheckMate 901 of nivolumab/gemcitabine/cisplatin. Then we also for the first time have seen some data on sequencing strategies in metastatic disease with the THOR trial being negative and really perplexing as to why erdafitinib did not do as well as pembrolizumab in FGFR selected patients. I thought the entire breadth of the data presented was really great.
Dr. Powles: Shilpa, you were on that EV-302 piece, and we’re going to talk about that in a second. But maybe Petros we could start with you because Michiel Van der Heijden, MD, did an amazing job when he described the data looking at cisplatin-based chemotherapy with nivolumab and maintenance nivolumab versus first-line standard gemcitabine/cisplatin, gemcitabine/carboplatin chemotherapy. That trial was a positive study. In the past, we know that there have been studies which have had marginal results. The chemotherapy atezolizumab study had a positive PFS [progression-free survival]. Actually, it’s P value was 0.05 and just missed OS [overall survival] by tiny amount. Then there was the 361 study, which didn’t show PFS or OS, but actually the numbers were in the zero point sevens and zero point eights. With hindsight, I was kind of involved, the statistical analysis plan probably wasn’t perfect, but I can say that now. The counseling helped.
Petros, talk to me about the results of that study. What did it show? Did it surprise you? The key question, and this is really important for you too and Sia you might want to it dive in, is it better than maintenance nivolumab and should we be giving chemotherapy and immune therapy together?
Dr. Grivas: Great questions, Tom, as always. Let’s start by the historical context. As you mentioned IMvigor130 chemotherapy/atezolizumab versus chemotherapy and also KEYNOTE-361, which was chemotherapy/pembrolizumab versus chemotherapy, those trials did not change practice. There was no substantial difference in overall survival. As you mentioned, the IMvigor130 had a very marginal modest PFS benefit but did not change practice. In our mind, the chemotherapy plus concurrent checkpoint inhibition combination did not have enough value. We definitely had the very significant and impressive OS benefit with avelumab maintenance in the switch maintenance setting in the frontline. We all thought that the sequential approach is the way to go, not the concurrent approach.
Now I think 1 caveat or thing to keep in mind is that IMvigor130 and KEYNOTE-361 lumped together cisplatin and carboplatin, and this may have potentially impacted the performance of the control group there. Also if you think about there may be some differences in the immunomodulatory effects from cisplatin versus carboplatin, maybe cisplatin may be a better partner for immunotherapy. That may be a hypothesis why the gemcitabine/cisplatin/nivolumab versus gemcitabine/cisplatin was a positive trial. If you look at the response rate specifically at a certain point, you saw that the gemcitabine/cisplatin/nivolumab was in the mid-fifties—the response rate, and the gemcitabine/cisplatin was about 43%, slightly lower compared to what we saw in the performance of the control group in the other studies. But in the ballpark in the forties area, slightly numerically lower, but still I think that the gemcitabine/cisplatin performed pretty well because median OS was about 16 months in the control group.
I think that this nivolumab really prolongs overall survival as a positive trial. The big question is, is it a positive trial because of the concurrent administration of IO [immunotherapy], the maintenance component, or both? It’s extremely hard to tease this out. If you look at the curve, the curve separate relatively early and then stays separated. It’s very hard to compare between the CheckMat 901 and the JAVELIN Bladder 100 trial because of different designs. If you look at the reasons to use gemcitabine/cisplatin/nivolumab upfront will be the clinical complete response rate was about 22%, and the median duration of response was 37 months, almost 3 years. The primary progression rate, at least in the first scan, was less than 10% with gemcitabine/cisplatin/nivolumab; obviously many patients may have progressed later on.
Bottom line is it’s hard to argue to compare between the gemcitabine/cisplatin/nivolumab and the maintenance avelumab strategy. I would say it’s probably attractive based on the primary progression rate at least and the durable clinical responses that you may potentially keep more patients on treatment if you add the nivolumab upfront. In the absence of direct comparison and absence of a direct answer to your question, I think gemcitabine/cisplatin/nivolumab is a very reasonable approach in cisplatin-eligible patients in countries where pembrolizumab/EV [enfortumab vedotin] is not available, not reimbursed.
Dr. Powles: How much of the 0.78 overall survival hazard ratio, how much of that is being generated by the maintenance period? Because avelumab has a hazard ratio of 0.69; it’s only maybe half the patients who get it, maybe fewer. But if that’s sort of a 30% benefit for half the patients, that would generate a hazard ratio of 0.15, 0.85 would be maintenance therapy if that were true and here we’re 0.78. Is the survival being driven by the concurrent piece or is it just those responders? If it is the responders, who are those responders?
Dr. Grivas: Extremely good questions and impossible to answer this. The question is…
Dr. Powles: I’m going to ask impossible questions. My only goal is to ask impossible questions.
Dr. Grivas: But a very important one. The question is does the concurrent nivolumab during the chemotherapy administration add value and contribute to the survival benefit noted in the CheckMate 901 trial. As you mentioned, Tom, the hazard ratio 0.78 is not that impressive, I would argue, but it made it. It made a statistical significance. Again, if you look at the timing when the curve separate, it’s probably starting probably during chemotherapy, maybe on the end of the chemo.
Dr. Powles: Do you think that’s right?
Dr. Grivas: I think it’s probably…
Dr. Powles: I think they separated…
Dr. Gupta: That’s what I also thought, they had separate…
Dr. Grivas: At the end, after the end of the chemotherapy. Right?
Dr. Gupta: Almost in the maintenance period.
Dr. Grivas: Yeah. The question is that driven by the maintenance component? It’s possibly the case. It’s very hard to tease this out. I would argue that if you think about the maintenance avelumab, you save cost and toxicity with the concurrent administration of avelumab in the chemotherapy induction phase. But the question is we’ll never find the answer. The question is, again, which is not clear how many patients you do not lose with progression if you don’t use the IO upfront. But we don’t have the right answer right now.
Dr. Powles: I’m coming that. Sia, Shilpa, the question I guess patients are going to ask is if I don’t get the nivolumab now, there’s a chance that I’m never going to get immune therapy because if heaven forbid my cancer progresses and things are going wrong, it’s all growing too fast and we lose control. Do you say to patients, “Actually you’re probably right, let’s give the nivolumab now?” Or do you say to them the data doesn’t support that and therefore you don’t need it now and the 2 probably don’t work that well together? How do you take this data to your patients?
Dr. Gupta: That’s a great question, Tom. I think we have to put it in the context of the CheckMate 901. Hardly anybody got maintenance immunotherapy in the control arm. We are comparing it to the historical just chemotherapy arm.
Dr. Grivas: Actually 20% of patients on the chemotherapy control arm of CheckMate 901 got avelumab. Suboptimal, but it was about 20%.
Dr. Powles: But it’s less than you’d expect.
Dr. Gupta: Less what you would expect, right?
Dr. Grivas: Exactly.
Dr. Gupta: Then I think to tell the patient that you absolutely need nivolumab right now otherwise you’re in for really bad news won’t be accurate. I think we have to discuss the data and in my opinion, the curves did separate after the chemotherapy period. Are we going to lose some patients who may progress upfront? Obviously, if you’re seeing a patient with a lot of bulky disease, visceral metastases, you want to offer them nivolumab upfront, but I don’t think every patient needs to be offered.
Dr. Powles: Sia, what’s your take on this from a urology perspective? You must be looking and wondering, “What on earth?” Why are they obsessing with these numbers? They all look the same.
Dr. Daneshmand: We just want to take out their bladders. No. Let me just bring up GU 16-257, Matt Galsky, [MD]’s phase 2 trial gemcitabine/cisplatin/nivolumab given in the neoadjuvant setting. We’re seeing the same sort of benefit. It’s a phase 2, everyone is getting gemcitabine/cisplatin/nivolumab with the primary objective being the patients keeping their bladders, the complete response rate. We have a good number at more than 50% of the patients able to keep their bladders. We haven’t seen this kind of complete response rate with chemotherapy alone.
The typical CR rates are 22%, something like that, based on cystectomy. These are patients who got to keep their bladders. I have a several number of these patients in my practice right now who are several years out and are disease free. We thought maybe the DDR mutation patients are the ones who are responding well, and it hasn’t really panned out. We were not quite sure the selection of which of the patients. I kind of believe this data. I think like Shilpa said, the more volume disease you have, the more sort of visceral metastases and you think this patient, if they don’t respond to gemcitabine/cisplatin, perhaps it’s best to put that on initially and get the IO checkpoint inhibitor along with the chemotherapy from the start.
Dr. Powles: Sia, I’m going to ask questions kind of miles off, and I apologize. What is pathologic CR [complete response] as an endpoint in urothelial cancer to these neoadjuvant therapies, and what do we have to do to improve on it?
Dr. Daneshmand: Yeah, great question. I think it’s a double-edged sword. Pathologic CR right now, well the gold standard is a cystectomy, right? A pelvic lymph node dissection showing a complete response. But both in the bladder and in the lymph nodes, we and others have shown that it’s not just the bladder; 5% to 10% of patients who have CR in the bladder actually have positive lymph nodes, so the ypT0 N+, and we’ve published this. You have to look there as well; these are the patients who are later going to fail. Well, currently if you’re going to keep the bladder, what can we do to determine true a CR? It’s not a pathologic CR. We have to do the best sort of enhanced cystoscopy that we have; currently we use blue light if you have it. We have to do random-directed biopsies I suppose because you’ll miss CIS [carcinoma in situ] and carcinoma and cytoma and some invasive disease sometimes.
A bladder doesn’t look normal after neoadjuvant therapies. It’s not like we look in there and say, “Okay, there’s no disease.” We need better imaging. Unfortunately, MRI and CT and PET and all the current modern imaging really hasn’t been great in showing us. There’s always some thickening; there’s some abnormality that we just have to look. I would say proper staging in the operating room with biopsies, TRBT to really ascertain and cytology, just make sure you have no disease. The lymph nodes, you can’t do anything about; 3 mm lymph node positive disease you’re just not going to pick up unfortunately.
Dr. Powles: We’ve talked about the neoadjuvant nivolumab/cisplatin, we talked about some grayness about whether cisplatin was better as a dance partner than carboplatin. I’m not sure we’ve answered that. Sia, you’ve said actually in the neoadjuvant saying that might be the case. I’ve got some doubts in my mind as you know.
630 Madison Ave.
Manalapan, NJ 07726
© 2025 Mashup Media, LLC, a Formedics Property. All Rights Reserved.