Read the Full Video Transcript

Zachary Klaassen: Hi. My name is Zach Klaassen. I'm a urologic oncologist at the Georgia Cancer Center in Augusta, Georgia. I am delighted to be joined today by Dr. Tom Powles, medical oncologist from Barts Cancer Institute in London in the UK. We're going to be talking today about ESMO 2024 data, practice-changing data from NIAGARA. Tom, thanks very much for taking some time to join us today.

Thomas Powles: Zach, thank you for inviting me. It's a real pleasure to be here.

Zachary Klaassen: Now, we're excited to go through this data. It's absolutely fabulous, so I'm going to pull up the slides and I'd love for you to present some of the data that you presented.

Thomas Powles: So, Zach, NIAGARA was presented at ESMO this year. It was a huge team of people. It's the largest trial we've done in muscle-invasive bladder cancer. It's 1,000 patients. And you'll remember from before that we really struggled to do these neoadjuvant trials. We didn't recruit well, and actually a lot of the trials that we did before, well, we didn't actually answer some of the questions as robustly as we might've liked, particularly in the adjuvant chemotherapy setting. And so when we undertook this study, we thought, "Let's do a big study and see if we can get both progression-free, but also overall survival," because overall survival is a key endpoint.

Of course, you'll be aware that it's a big problem. Half a million urothelial cancer deaths. T2 bladder cancer is not like T2 prostate cancer. No. Half the patients with T2 bladder cancer or muscle-invasive bladder cancer, they go on and relapse and die from their disease, and it's a quick disease as well. It's not like my experience with kidney and prostate cancer. You don't want to leave your best therapies for third and fourth line in urothelial cancer. I don't think personally you need to do that for any cancer, but I think urothelial cancer is a good example of that. And the principal question that we're asking is actually quite a simple one. It is when you see patients with aggressive disease, T2 to T4a muscle-invasive bladder cancer, are you better off starting immune therapy straight away? Is that a good approach?

The way we did the trial is we designed it in a large randomized phase III, but it's this perioperative approach which is important. It's not just neoadjuvant therapy; it's four neoadjuvant cycles with chemotherapy, and it's also eight adjuvant cycles with the cystectomy in the middle. So it's this sort of sandwich effect. Now, the comparator arm is the standard of care. Cisplatin-eligible patients, but we allowed creatinine clearance down to 45 to split the dose. And I think with hindsight that was the right thing to do. I don't think biologically there's any difference between patients with good renal function and slightly less good renal function in this setting. So the biology is the same.

So I don't think this subgroup is different, but the cisplatin-based chemotherapy has been a bit exclusive in the past. Not all patients can get there. It's the classic eligible population that you would be familiar with. Obviously, they needed to have T2 to T4a disease, 5% of patients were lymph node positive, and we allowed histology outside of pure urothelial cancer. So components of other cancers were allowed. Event-free survival, path CR as the primary endpoint, and overall survival a key secondary endpoint, and we were powered for that.

The patient characteristics, as you expect in a 1,000-patient trial, well-balanced in both groups. The T2 patients were capped at 40%. That was in line with the SWOG study, the original sort of MVAC study that we kind of benchmarked the trial off, and that's how we made our statistical assumptions from there. And the PD-L1 status—PD-L1 is not a great biomarker in urothelial cancer. It didn't work here either. I'm keen we move on from that. The event-free survival hazard ratio of 0.68 was statistically significant, clinically meaningful. The curves go apart and they stay apart. One caveat I'd like to mention on event-free survival: it's different from what you might've seen before, which is disease-free survival, in that disease-free survival is designed by the death of the disease coming back. Event-free survival also has this extra group. Those patients that did not have surgery were considered an event.

And actually when you looked at our patient population, we actually did really well in that 85% of patients essentially in both arms had surgery. Slightly more in the study, in the Durvalumab arm, had surgery, number one. Number two is 70% of patients went on to receive the adjuvant period and 50% of patients completed that adjuvant period. When you look at the CONSORT diagram, it's striking. Also, a similar number of patients completed the neoadjuvant chemotherapy. It didn't look like Durvalumab made the surgery more complicated, and therefore this curve is a clinically meaningful curve. We've looked at event-free survival in different ways using different methodologies, and it hasn't made any difference. The forest plot analysis, you can see all the dots on the left-hand side of one. So it's broadly favoring Durvalumab. I'm not a great believer in overinterpretation of forest plot analysis. Inevitably, the confidence intervals are going to get wider because the subgroups are underpowered.

So some are bound to cross one on the one hand, and of course on the other, if you shake the dice 20 times, you'll get two sixes in the end. And that's essentially what we're doing here, but actually there's nothing to see here particularly. The path CR rate is, I think, really interesting, and I could talk about this for a really long time. I'm not going to. I'm going to focus on the right-hand side. The right-hand side was all of the patients. The left-hand side was the statistical analysis, which excluded about 59 of the patients. It was done early. The ITT analysis shows only a 27% response rate for platinum-based chemotherapy. Some people say to me that's low. I've actually spent a lot of time on this issue over the last six months, and path CR in previous trials is actually a real moving feat.

So using the denominator, all of the patients included, not just those that had surgery—that's the first thing. Some trials do that. Other trials look at clinical CR. Some trials include TIS or even these early non-quite path CRs, but are not T1s. And actually, every trial seems to have done it differently. I think this is the definitive chemotherapy analysis, and it's 500 patients, it's centrally reviewed, and it includes the whole denominator that's included. The other issue is there are some patients that get path CR with TURBT alone, particularly those T2 patients. And there have been trials in the past, for example the VESPER studies, got over 90% of patients with clinical T2 stage at baseline. So actually when you look at the data in detail, there are lots of moving parts associated with neoadjuvant and path CR issue. And cross-trial comparison in this setting I don't think is very helpful.

The overall survival was statistically significant with a 25% reduction in the risk of death. The curves go apart and stay apart. 43 months of median follow-up is actually quite—it's getting towards, well, it's four years, it's getting towards five years. I think these results will continue along these lines. And if you move on to the forest plot analysis, you'll see that actually there's no subgroup that isn't benefiting in the same as the PFS analysis. The adverse event profile, as I said before, and I think it's worthwhile saying this as a caveat also, while there appear to be no differences between these two arms, which I think is great, and there doesn't appear to be any surgical complications or any trouble getting to surgery, which is great, and the treatment-related deaths are really, really low, which is great, the CTCAE criteria designed for chemotherapy, not immune therapy, and inevitably the addition of Durvalumab is associated with some toxicity—you can get diabetes and these endocrine disorders, some patients with some cardiac disease; these are all less than 1%.

And now encouragingly, these didn't delay or disrupt surgery, but I think we need to look at different ways of assessing immune-related adverse events in the context of these studies because it's important how we communicate with patients. What are my chances of life-changing toxicity? And we haven't got that right yet. This just shows chemotherapy-related toxicity mainly. And of course, nausea, vomiting, etc. But Zach, one of the things which I've been thinking about recently conceptualizing this is that when I look at neoadjuvant chemotherapy and I look at cystectomy and I look at Durvalumab as three interventions, it's pretty clear that cystectomy is the most challenging. For someone in the eighth decade of their life, losing their bladder and having a big operation is a really big deal.

I think second behind that is cisplatin-based chemotherapy. That's the second most challenging thing to do in this setting. You can see by this toxicity, it's all chemotherapy toxicity. And actually Durvalumab is probably, of those three, is the easiest of the three interventions. So in summary, I think we're in a position at the moment where we've, as a big group of people, delivered this big randomized phase III study. It's shown an event-free survival advantage, it's shown a 10% bounce in pathological CR, it's shown a 25% reduction in the risk of death, it's not shown increased surgical toxicity or complications, and I think it's also worthwhile saying that it's the first time we've seen overall survival for immune therapy either in the perioperative or indeed the adjuvant setting, and I think that's telling.

And then the other piece around it is that there are a series of other trials coming through in the future, and I think those trials are also going to be positive. So I see it as opening a new chapter, and I think in the future we'll be giving immune therapy with chemotherapy or other approaches to these patients before doing surgery, and dare I say it, we may even be doing less surgery as that path CR rate goes up with better and better therapies.

Zachary Klaassen: No, that's great, Tom. I know you've heard this many times since ESMO. Congratulations to you and the NIAGARA team. I mean, this is just, as you mentioned, a huge trial, huge undertaking, and really moving the needle, and we could certainly spend 30 minutes dissecting a lot of these nuances, but I want to focus on just a couple of talking points for our listeners. From a surgeon's standpoint, I could imagine a patient—they get their cisplatin, Durvalumab, they have the cystectomy. You mentioned the cystectomy is the hardest part of this whole process. They're sitting there in your office, they've got a path CR—you mentioned in the trial 70% of patients started adjuvant Durvalumab, 50% finished. In the real world, pending regulation approval and everything, what's your thoughts on what that uptake will be for that eight cycles after cystectomy?

Thomas Powles: So there are two parts to that, well, three parts that are important. The first is we have not in this trial—a shortcoming of this trial is we didn't have a third arm. We didn't have a neoadjuvant-only arm. Now, I have to tell you, I was in the room when the decisions were made, and the thought of doing a 1,500-patient trial was just something that—I know with hindsight everyone says, "Oh, you should have done that." It wasn't on the table. We were playing with numbers of 750 and saying, "Can we do it with," and the stats people were saying, "If you want OS you need 1,000." We were saying no, and there was even a debate about whether we should just go on EFS and not. So the answer is yes, we can answer that moving forward. I don't think we could have answered it in this study, but we haven't definitively answered that question.

I think we could look at that path CR population and work out what happened to that path CR population. They've done that in breast cancer and to see if the addition of the drug, continuing the drug, seems to make a difference, but again, that's a bit of a flawed analysis.

Zachary Klaassen: Sure.

Thomas Powles: My opinion in advanced disease, and I think muscle-invasive urothelial cancer is on the spectrum of advanced disease. These patients, the cancer comes back and heaven forbid they die quickly if we don't intervene at the right time. So in advanced disease, I don't think four months of immune therapy is long enough in my experience. We haven't been curing patients. Now melanoma, they may be doing that, but this isn't melanoma, and I'm not great on saying the melanoma data should apply in urothelial cancer. Those experiments didn't work back in the day. So I think the answer at the moment is I think four months isn't enough. I think those patients who have had a path CR, dare I say it, I would put it to them that they may still have MRD, minimal residual disease. And if it's winning, keep going.

I mean, you could actually bizarrely do it the other way around and say, "You haven't had a path CR and the drug's working, should we be doing something else?"

Zachary Klaassen: Yes.

Thomas Powles: So my take on this is yes, more to come on this space, number one. But number two, I don't think four months is long enough, and if the cancer comes back, heaven forbid, then you've got this issue—well, you'd have to re-challenge with immune therapy then if it was to come back because you haven't given enough therapy. And so I'm in a position where you say you get one good go at immune therapy, currently do it properly, and four months isn't long enough for me.

Zachary Klaassen: Excellent. I want to touch briefly on the other 50% of patients, the cisplatin-ineligible. What's going on with Durvalumab in that space of the patients that are cis-ineligible with muscle-invasive bladder cancer?

Thomas Powles: So in cisplatin-ineligible patients, and that's now a creatinine clearance below 40 because we can come down to 40 with a split dose, I think this is a really interesting field, and let me tell you why. Because we have two or three, I think, really important studies in this space. Antibody-drug conjugate, particularly Enfortumab Vedotin, is hepatically excreted, it's not renally toxic, and actually you can bring that creatinine clearance right down. And so there are two trials, two pivotal trials, that are ongoing. One study is with Enfortumab Vedotin, Pembrolizumab, Pembrolizumab alone, or cystectomy—that randomization. And I think that trial's going to be positive.

Zachary Klaassen: Yeah.

Thomas Powles: I think EV Pembro, it's got a 30% CR rate in advanced disease.

Zachary Klaassen: Yeah.

Thomas Powles: I think it's going to have a path CR rate of 50%.

Zachary Klaassen: Wow.

Thomas Powles: And then there's also Enfortumab Vedotin, Durvalumab, and Tremelimumab versus Enfortumab Vedotin, Durvalumab versus surgery in a second cisplatin-ineligible trial. But there's a third trial out there as well with Enfortumab Vedotin and Pembrolizumab versus platinum-based chemotherapy. And I think if that trial is positive, then people say, "Well, if it's beaten chemotherapy, it's better than doing nothing." And on top of that, why you wouldn't want to just exclusively limit that to creatinine clearance above 50? It's very likely to work in a lower creatinine clearance population as well. So we're going to have specific trials, we're going to have reinforcing trials, but the last bit, which I wanted to talk about, is we've also got a study called IMvigor011, and that's a circulating tumor DNA trial.

So for patients who have had a cystectomy—that cystectomy performed—the cancer comes back in about 50%, ctDNA seems discriminatory in predicting which patients are likely to relapse, and IMvigor011 focuses on that higher-risk population. So we're going to see a big change in this landscape over the next two years. NIAGARA is the first part of this journey, and that's one of the reasons, not just because we've got a survival advantage with this particular trial, but I think it opens a new chapter in the disease.

Zachary Klaassen: It's super exciting. I think we're seeing just an absolute multidisciplinary approach here. So again, from a surgical standpoint, medical oncology—we haven't even talked about potential bladder-sparing options if people get good results, and that may be another discussion for another day. But I appreciate you breaking down NIAGARA and the excellent discussion to follow. Thank you, Tom.

Thomas Powles: Oh, thank you, Zach.