Read the Full Video Transcript

Shilpa Gupta: Hello, everyone. I'm Shilpa Gupta. I'm a genitourinary medical oncologist at the Cleveland Clinic and really honored to be joined by Dr. Tom Powles from London, from the Barts Cancer Center. Tom is very well known to everyone who follows the kidney and bladder cancer world. So Tom, congratulations on a great ESMO meeting that we saw new data, which you presented. I want to talk to you about the NIAGARA study, which was, again, the first in the space to see chemo IO in the perioperative space in bladder cancer. If you can give us the high-level overview and where you think the field is going with this first regimen in this space.

Thomas Powles: Well, thanks, Shilpa, and thanks for asking me along today. I think NIAGARA was a huge undertaking, a thousand patients, the largest trial we've done, 200 sites around the world, a massive global effort, which was successful. It was the first study in urothelial cancer exploring an immune therapy, but in fact any immune therapy, which happened to be durvalumab in this case, PD-1, PD-L1 inhibitor, not just in the adjuvant setting but in the neoadjuvant setting as well. So neoadjuvant, platinum-based chemotherapy, four cycles with durvalumab, cystectomy, and then eight cycles of adjuvant durvalumab making a total of a year of therapy. It's for muscle-invasive urothelial cancer, T2 to T4a, N1 disease. Patients needed to be fit for cystectomy, obviously.

And we haven't done a study like this before in urothelial cancer. We've struggled to recruit to studies like this before in urothelial cancer, so I was really pleased when it came out and was positive. I think it's a practice-changing study, and this is why.

So when we look at the event-free survival, the hazard ratio of 0.68, when we look at the overall survival, the hazard ratio of 0.75, the curves go apart, and they stay apart. The path CR rate is 37% versus 27%, 10% bounce in path CR. A similar proportion of patients, in fact, more patients, the durvalumab arm ended up getting surgery. There were no safety signals associated with the neoadjuvant or the adjuvant components suggesting it was unsafe. Grade three or four treatment-related toxicity, 41% in both arms. The forest plot analysis showed benefit consistent for associated with the durvalumab arm. And so I think in summary, it's fair to say that this is the first positive study of any immune therapy in muscle-invasive bladder cancer with a significant overall survival signal.

I think it also has a DFS or an EFS signal, which was statistically significant. You were giving immune therapy first. I kind of feel that saving immune therapy for third-line therapy wasn't the right thing to do. We ended up with EV Pembro first line and metastatic, but I think deep down immune therapy is well tolerated and for patients at high risk of death and relapse from the disease, which includes muscle-invasive urothelial cancer, I think early immune therapy makes logical sense. And so I see this as being kind of attractive for patients and practice-changing. It's the first of a number of studies in this space, and I suspect those trials, other trials, will be positive too.

Shilpa Gupta: Yeah, thank you, Tom. This is definitely the first in the space and practice-changing, and I thought it was really amazing that we saw the OS signal at the first analysis because many times we have to wait so long. What do you think is the reason for that? I know this trial completed a few years ago.

Thomas Powles: So I mean 43 months was the median follow-up, which is actually now beginning to four years, and most relapses occur within two years. And when relapse occurs with urothelial cancer, and it can be pretty devastating when you look at the shapes of the curves. They go apart quite early on. So I think two things. Firstly, that urothelial cancer is an aggressive disease, and it doesn't surprise me that we get OS signals very different from prostate cancer. It doesn't surprise me we get OS signals, well, relatively early on, number one.

Number two is that urothelial cancer grows fast and front-line metastatic, second-line metastatic, third-line metastatic disease, not all patients get there. And so I think that when you're faced with an individual with an aggressive urothelial cancer, I'm not surprised by giving the best drugs at the beginning. It's better than saving good drugs in the locker for later.

And in my opinion, that includes the adjuvant setting too. The problem with the adjuvant setting is that you've got to get through the neoadjuvant phase, you've got to get through the cystectomy, come out of the side, be fit, and that's often four months down the line from when you start, and during all that time, the cancer's got the opportunity to develop micrometastatic disease. That's kind of where ctDNA I think comes in. And again, it just reinforces that issue for me, which is this translates, in my opinion, to giving immune therapy early before the operation has advantages because you're not waiting. And in the end, if you could look at cystectomy, neoadjuvant chemotherapy, immune therapies or PD-1, PD-L1 therapy is the best tolerated of those three. So I see this as an important intervention, and this early intervention associated with survival makes a lot of sense to me.

Shilpa Gupta: Absolutely, Tom. And Tom, what would you think about the path CR rates? Were you surprised with what you saw both in the control arm and the durva combination arm, and is it even a useful endpoint that we should be chasing anymore?

Thomas Powles: Yeah, I've spent a lot of time since the presentation with path CR, and I've read all of the publications, and there may be about 15 key publications, the classic CMV, MVAC, chemotherapy trials, the first generation of immune checkpoint inhibition trials, PURE-1, ABACUS, even some ADC studies. We've got SG EV, we've even got some DV PD-1 combination data. When you look at those trials, the variability around the clinical CR rate is huge, and I don't think we're staging these patients very well.

Some have 95%, the SPARC trial, 95% T2 disease, some as low as 30% T2 disease. The SWOG study, the one we benchmarked NIAGARA off, that's why we capped it at 40% T2, and I think that's probably about right. And then you look at path CR associated with that, and we know that T2 disease has likely to have more path CR from TURBT. There's an issue around re-TURBT whether the trials have that or not. And then of course with path CR, there seems to be, we've got into the habit of reporting some trials quite early, and then there's an issue about the denominator in the path CR because some patients haven't got to surgery, and in some trials, those patients who don't get to surgery, they're not included in the denominator.

And some of the trials are small, like 30. This is a terrible answer, by the way. Some of the trials are small, and so when you look at it, it's just impossible to—some of the trials I went to, they've given a path CR rate. I've recalculated it the way I think they've tried, and I've come up with a different number. And so I think we need to be much more consistent in the way we report path CR in the future, and I think obviously a thousand-patient trial like the durvalumab study with four cycles of gemcitabine/cisplatin, the control arm with 27% as the control arm and 37% with the addition of durvalumab, I think that becomes a new benchmark. I think the previous trials are clearly informative in their own way, but it's very difficult to put together a cross-trial comparison of path CR or T2, so I really wouldn't go down that route. I think it's counterproductive.

Shilpa Gupta: Absolutely. And Tom, do you think this study was in the era of prior to CheckMate 274, there was no adjuvant durvalumab available. If, let's say, it is available, if it were available, do you think we would see this much difference in this? There's really no clear answer about adjuvant versus neoadjuvant right now.

Thomas Powles: Yeah, so I've not had the opportunity to have that debate yet with someone, but I'm looking forward to it, and perhaps we should do a podcast on it. I'd be up for that. Look, this is what I think. I think 274 is a great trial. I think it has a DFS advantage, 0.71 in the initial analysis about there. It's never hit significant overall survival. It's a higher-risk population. It's those patients with pathological, not clinical, stage at baseline. And you look at those curves, and they go down pretty quickly, so it's a higher-risk population, it's a more select population, and therefore cross-trial comparisons are not particularly useful. In fact, I think they're unhelpful.

My take on this, and I think there's some parallels between this and maybe the maintenance durvalumab trial, I think the issue is you've got to get through the chemotherapy, get through the cystectomy before you get into that immune therapy piece. And if you're an immune therapy believer and think that the immune therapy is the transformative drug for many patients, which I believe is the case, then I think you're better off giving that right at the beginning while patients are waiting to have their cystectomy. I think perioperative immune therapy, during that 16 weeks before the patient comes out the back of adjuvant therapy and says, "Right, I'm fit to have adjuvant treatment." I think if you're giving immune therapy, which has a long half-life and doesn't have that many side effects—serious side effects for some patients, but most patients not—I think it's better giving it earlier conceptually. I mean, that's where I am at the moment. So I think the higher path CR rate, the significant overall survival benefit, giving the treatment earlier, and the huge difference in the two populations means that my preference is the neoadjuvant space.

Shilpa Gupta: Absolutely. And Tom, I thought it was so refreshing to see that this study actually is very inclusive. They included the variant histology subtypes, creatinine clearance—I think this is the first time we've seen it as low as 40 mL—and patients did well with the split dose. So I think the Galsky criteria may not be a relevant criteria anymore. And in the real world anyways, our threshold was lower than 60. So what is your take on that, Tom, that do we really need to be thinking about the Galsky criteria or really pretty much most people can get cisplatin based on the kidney numbers?

Thomas Powles: Yeah, I think going down to 40 is advantageous. I always felt that 60 limit or 50, depending on who you were, it felt a bit exclusive. It felt a bit, yeah, it felt a bit exclusive for me. I think coming down is attractive, obviously. There are other trials, other immune therapy studies with EV Pembro and single-agent Pembro and EV Durva and EV durvalumab—they're around the corner. My gut feeling on that, Shilpa, and I don't know the results of these trials obviously, but my gut feeling on those trials is they're likely to be quite positive if chemotherapy durvalumab is positive.

And some people said to me, "Is there synergy, additivity, or antagonism between these?" I think EV Pembro is probably additive. Some people say it's synergistic because of the high path CR rate, and if it is synergistic, then we would expect terrific results in this space. I think chemotherapy and immune checkpoint inhibition together, I think it's only partially additive. I think the response we know from CheckMate 901, but also 361, that cisplatin appears to be a better dance partner than carboplatin. I was wrong about that after all. I've been a long time arguing the other way. I'm wrong about a lot of things. So yeah, I think overall, I think that my take on this piece is that this field is—NIAGARA, I think, is the first of a number of trials, and the transformative space here is going to end up, I think we might end up curing a lot of bladder cancer patients, and I think we might cure some of them without surgery.

Shilpa Gupta: That's a great point, Tom. And I think it's also encouraging to see that there was a lot of talk about, well, PD-L1 is not really panning out in bladder cancer with the TESO adjuvant trials. And this has really debunked that notion as well.

Thomas Powles: This is a long discussion, Shilpa, but each of the drugs has had a good day and a bad day, and that includes pembrolizumab, 361 frontline pembrolizumab plus chemotherapy, negative trial. And even in the monotherapy arm, and in the monotherapy arm, the PD-L1 biomarker did better for the chemotherapy. So unlike kidney cancer, maybe lung cancer, in bladder cancer, the PD-L1 biomarker is unhelpful. I'm pretty clear about that, number one. And number two is I think at the moment there are more similarities than differences between these drugs.

Shilpa Gupta: Well, I totally agree, Tom. Thank you again for leading this trial and setting the pace forward, and I hope we see the approval soon and can start using it for our patients.

Thomas Powles: Thank you for inviting me, Shilpa.

Shilpa Gupta: Of course, Tom.