Ashish Kamat: A warm welcome to all of you from the UroToday studios. I'm Ashish Kamat, Professor of Urologic Oncology at M.D. Anderson Cancer Center in Houston, Texas. And we're live in Las Vegas at AUA 2025.

It's a pleasure to welcome to the UroToday studios two eminent personalities in GU oncology, Professor Jonathan Rosenberg and Petros Grivas. Welcome, gentlemen.

Jonathan Rosenberg: Thank you.

Petros Grivas: Thank you for having us.

Ashish Kamat: So you guys are doing a lot of stuff here, but one of the things that we have you doing at the AUA-- and thank you for doing this-- is debating at the IBCG-AUA forum. You're presenting a case, you're debating. But we're talking about ctDNA and its role in personalizing therapy for bladder cancer patients. So shed a little bit of light on, where are we today in 2025, Jonathan, when it comes to the use of ctDNA in management of patients?

Jonathan Rosenberg: So we have a lot of very good retrospective data that suggests that ctDNA predicts relapse for patients and that patients who are ctDNA negative after radical surgery will do better on average. And we have some data that suggests that if you give ctDNA-positive patients immunotherapy, they will be less likely to relapse.

And that it may not help ctDNA negative patients, although it was in the context of a trial that was negative with a drug that is no longer approved in bladder cancer, and a subset analysis not preplanned. So all sorts of problems with the data. But the promise is there to be a very robust biomarker.

Ashish Kamat: It's interesting you say that, and I completely agree. These are subset analyses. And we also have to take that with a large cherry and a dollop of cream on top.

But studies are being planned, and studies are ongoing. And you have other studies-- for example, the TOMBOLA that was presented at the EAU. So shed a little bit of light-- and maybe I'll ask you this, Petros-- on your sense on where the ongoing studies are going to help inform the field.

Petros Grivas: Great discussion. And I agree with Jonathan that the data so far have been promising. And I think the take-home message so far, it seems to me that ctDNA, as Jonathan says, seems prognostic. So higher chance of recurrence and death if the DNA is detectable positive, as opposed to being negative and undetectable.

And there have been a number of studies that Jonathan quoted. IMvigor010 is what Jonathan said. And then we had data from TOMBOLA trial. We had data from ABACUS and NABUCCO trials, retrospective analysis of prospective trials with the caveats that Jonathan quoted.

The TOMBOLA, for example, saw the very low chance of recurrence if you're ctDNA-negative. But of course, there was no randomization in those patients. It was description of the outcome if you're ctDNA-negative.

And I think going forward, the ongoing trials will be very useful for us to understand the clinical utility of ctDNA, whether it has impact on the patient outcome if you use it. And two of those trials-- IMvigor011, led by Tom Powles and other colleagues, mainly enrolling in Europe, which is patients who are ctDNA-positive get randomized to atezolizumab or placebo. And those DNA-negative, they remain on surveillance.

Those DNA-negative cohort, we saw some data already, I think, in the previous meetings. These patients do fairly well. About 10% chance of recurrence, only if you are ctDNA-negative. So we're waiting for a ctDNA positive cohort to understand the benefit or not with immunotherapy.

And the other trial, of course, is the MODERN trial with Alliance and Matt Galsky, and Jonathan is involved. Has two randomizations, ctDNA positive patients. They get an escalated immunotherapy, nivolumab plus relatlimab, anti-LAG-3, versus nivolumab alone. In ctDNA negative, they get nivo or observation. So two different questions, an escalation and a de-escalation question.

Jonathan Rosenberg: And then, two points. In the de-escalation question, the patients are treated at ctDNA conversion. And it's a non-inferiority to trial to see if it's safe enough, you can just wait and watch those patients.

The data from IMvigor011 is interesting as it's been presented. It really took the people who were negative and stayed negative for a year. And so the people who do well do well. And it was not particularly shocking that if you're negative a year after surgery, you're going to do very well. But when you look at the trial from IMvigor010, about a third of patients, maybe a little less than that, relapsed and died of their cancer, in fact.

And so we don't really know exactly what to do with these negative patients. And do you feel comfortable enough not treating them in 2025, based on the existing data? And I'm not quite there yet for the most part, except on a clinical trial.

Ashish Kamat: Yeah, I think that's the next question I was going to ask you. So if you have a patient that walks into your clinic and has the ctDNA test done-- or actually, let me back up. Do you get ctDNA testing outside of the context of a clinical trial in your patients post cystectomy?

Jonathan Rosenberg: I have. I'm not going to sugarcoat that. I think-- I've done it, but it's a complex discussion. When it comes back positive, it's easy. You say to the patient, you need treatment. And this is going to come back. There's a 3% to 5% chance you may clear it, and you may be fine.

But the rest-- but if you're negative, we don't know what to do with you at the moment. And the standard-- and if your pathology stage is high, should we go ahead and treat you? And I think the answer at the moment is yes. For a node-positive patient, for a T4 patient, these are people who are exceedingly high risk for relapse. And if they're ctDNA-negative, I'm not sure I feel comfortable enough saying yet that those patients are OK.

We also haven't seen really good breakdown of the ctDNA categories based on pathologic stage from those trials. And so trying to understand the matrix of what our understanding of risk is now versus what our understanding of risk will be adding another factor, I think, is going to be critically important.

Ashish Kamat: Yeah, I agree, and I look at it the same way. Petros, do you have any differing viewpoint there?

Petros Grivas: I agree with Jonathan. I think we're waiting ideally for trials that show clinical utility. I'm not routinely using the test right now in the adjuvant setting, mainly because I don't know what to do with the result in terms of the impact on decision-making.

Of course, I can envision scenarios, as Jonathan said, that the patient who's on the fence and may not want to do it may have comorbidities. And there may be another data point that ctDNA may offer to help inform the dialogue, with the caveat, again, that the data are not prospective from a randomized trial to show clinical utility.

So I can see the potential reasoning to use it. I just have not incorporated routinely yet for the caveats we discussed. I think we need this data, and I think IMvigor011 and the MODERN trial will inform the dialogue.

To your point, though, Ashish, and Jonathan's patients sometimes come to me with already having the test done. And of course I will look at it, no doubt about it. I will discuss it with the patient, and I will outline the pros and cons.

And I think it comes down to what was discussed-- the pathologic states, risk of recurrence, comorbidities of the patient, goals, wishes, preferences. Anxiety sometimes play a role, and also insurance coverage. All of those factors may impact the decision-making. And the ctDNA testing, if available, may add to that dialogue.

Ashish Kamat: Yeah, and I apologize because I should have clarified for the audience, not so much for you. But we're talking about tumor-informed ctDNA. Do either of you get non-tumor-informed ctDNA at all, outside of clinical trial?

Jonathan Rosenberg: I have not been, no

Petros Grivas: Me neither.

Ashish Kamat: OK.

Petros Grivas: We're waiting for data.

Ashish Kamat: So we're talking about tumor informed ctDNA. Moving into the earlier space-- and this is not the topic of the debate-- but in the earlier space, because we also see patients who come in and have a ctDNA test with a bladder tumor in place and say, it's negative. I don't want to get chemotherapy before I have my radical cystectomy. To that patient or family, how would you counsel that question?

Jonathan Rosenberg: It's a very tough question, because there is some data to suggest that those patients do exceedingly well. But we don't have prospective randomized data to know.

And so I think we're going to see some data in the next six months that starts to answer that question about whether those patients need preoperative therapy. And I think the answers are going to surprise us a little bit.

And so I think that's a discussion that we'll start to get some real prospective data on in 2025. And hopefully we'll start to settle the debate on that.

Petros Grivas: I agree. I'm on the same side with Jonathan. The data so far with neoadjuvant cisplatin-based therapy have not relied on ctDNA. It has been agnostic to that.

I think we need this granularity of data to see who needs therapy and who doesn't. I don't think we're there yet. I think ongoing trials that utilize ctDNA may help inform us who are the patients who may benefit or not in terms of the predictive question.

Because we know ctDNA is prognostic. But the question is, is it predictive of benefit to chemotherapy or immunotherapy? And we're trying to get this data, scratching the surface, but we're not there yet.

So I would be cautious to not overread the data. If the patient is fit for cisplatin-based chemotherapy-- and these days, immunotherapy-- I would probably offer the standard of care. And of course, it goes to the gray zone. Spaces may have comorbidities. Maybe cisplatin ineligible borderline. I can see a potential again role of ctDNA in those gray zone scenarios that can help the dialogue. But for the majority of patients, I'm not using it yet in the neoadjuvant decision-making.

Ashish Kamat: Yeah, it sounds like both of you are using ctDNA in some ways, the way we used to use PET scans, in that gray zone where you can't quite tell, and you're trying to look for reasons not to do something or do something.

An interesting point that both of you raised, I just want to emphasize for the audience, is that ctDNA has been studied in the context of immunotherapy, for the most part, not in the context of chemo, not in the context of EV, for example. So we can't automatically extrapolate the results. Or can we? Am I misspeaking?

Jonathan Rosenberg: I think the prognostic part of it, we can. But what does it mean for the therapy? We don't know. And so in breast cancer, a lot of what they do in the adjuvant setting is based on prognostic markers. And if you're high-risk, you get treatment. And they've never had to make the link that that high risk means it's sensitive to that treatment.

And I don't think we should get caught in that trap. We need to see that treating high-risk patients makes a difference. And I think we're going to show that. And the question is not treating low-risk patients, does it still make a difference. And I think we're going to show that it probably does, with the right drugs.

Petros Grivas: I agree. I think the prognostic question, by definition, is regardless of treatment. And seeing a positivity of ctDNA confers higher risk of recurrence and death. And this probably across treatments.

The big question is exactly what we discussed. How does this translate into treatment decision-making? And that's a predictive question. We lack that data for now. And I think having data from ongoing trials, as Jonathan mentioned-- for example, EV/pembro trials, KEYNOTE-B15, KEYNOTE-905, the VOLGA trial, EV, durvalumab, plus/minus treme.

And of course, even the NIAGARA trial that saw data, I would love to see ctDNA data incorporated in the pathologic states. This will be very useful, I think, in our discussion with the patients and try to avoid overtreatment and, of course, undertreatment, because both of those are caveats in clinics. So I would like to see this data of ctDNA incorporated in the ongoing and future clinical trial designs.

Ashish Kamat: Yeah, you guys make great points. And we could talk about this forever. It's so easy to chat about this.

But in the interest of time, let's wrap it up and maybe a top-level message to our audience, remembering a lot of them are trainees as well. So Petros, top-level message?

Petros Grivas: CtDNA in bladder cancer, highly prognostic. At least in the localized bladder cancer setting, we have seen many data sets showing that. The big question is, again, the clinical utility, which means if you use it in clinical practice, does it improve the outcomes of patients? Does it inform your management to improve the outcomes?

We're lacking that second question. Ongoing trials-- IMvigor011, MODERN trial-- will help answer those questions. I highly always recommend accrual in trials. It's always good for the trainees to hear that. And of course can be, of course, also ctDNA right now in gray zone scenarios. There has to be a very thoughtful discussion with the patient and put the data in context so we do our best job possible to educate the patient and avoid overreading or underreading the results.

Jonathan Rosenberg: I don't have a lot to add to that. I would say that the trials are key here and that we need to advance the field to really understand our knowledge. And a lot of our preconceptions are wrong, and we just don't know which ones they are. And that's a general rule to live by as a physician and an oncologist.

And so just because we think that it works as a way to identify patients doesn't mean it actually does. And it doesn't mean what we're doing makes a difference. So we need to-- in particularly MODERN, we need to get that enrolled. And it will be amended, actually, to allow prior immunotherapy to future-proof it from all the other trials that are going to read out.

So if you've gotten any sort of immunotherapy or other treatment in the neoadjuvant setting, you can actually continue to be randomized to MODERN because still the de-escalation question is critically important. If you have a CR after cystectomy with EV/pembro, do you really need five more cycles of EV? Probably not, is my guess. But I don't know. So I'll put the patient on the trial.

So really want to encourage people to think about opening it. It's still going to be relevant in the coming years. And we need to support the National Clinical Trials Network sponsored by the NCI. It really is a jewel in the crown of cancer research in the United States.

Ashish Kamat: Great closing statements, both of you. Thank you so much.

Petros Grivas: Thank you.

Jonathan Rosenberg: Thank you.