Read the Full Video Transcript

Ashish Kamat: A warm welcome to all of you from the UroToday studios. I'm Ashish Kamat, urologic oncologist at MD Anderson Cancer Center in Houston, and we're live in San Francisco at ASCO GU 25. It's a great pleasure to welcome to the studio, Professor Matt Galsky. Matt, you've been here with us many times. Welcome once again.

Matthew Galsky: Thank you.

Ashish Kamat: So obviously at a meeting like this, you've got a lot of stuff going on and you're talking about NIAGARA. You're talking about CheckMate. You've got other things as well. Right now let's focus on the nivo story. So what's happening? What's new? What are you going to share with us?

Matthew Galsky: So CheckMate 274 is, as you know, a study of adjuvant nivolumab in patients with high-risk muscle invasive urothelial cancer after radical surgery. The initial study design—the overall study design—included patients with upper tract urothelial cancer and bladder cancer. And, as you know, in some of the early analysis of the data on subgroup analysis, there was this suggestion of maybe an enrichment of the effect in patients with bladder cancer.

Whether or not that's real or not, I think we can debate that and we can talk about that more. But there is this opportunity, because the bladder population is larger, to really narrow in on the bladder population. The other reason to do that, of course, is because lots of neoadjuvant studies are coming out, and those neoadjuvant studies are limited to patients with bladder cancer.

And so it's a little bit more of a contextualization in terms of the neoadjuvant approach versus the adjuvant approach and the pros and cons of each. So this analysis really focuses on the patients with bladder cancer, and looks at outcomes but includes not only disease-free survival, the primary endpoint of the study, but also overall survival. Overall survival still at this point is an exploratory analysis—it's an interim analysis.

The final analysis has not yet been done. It's an event-driven analysis. We expect that into the not too distant future. When we look at patients with bladder cancer and look at disease-free survival and overall survival, then there are a number of subsets that are of interest, given the landscape of current therapy, including does it matter if patients got neoadjuvant therapy or not? Does it matter what their PD-L1 status is, etc. So all of those subsets are presented in the context of bladder primary tumors.

Ashish Kamat: So what's the message in those contexts? Because as you said, even though the trial wasn't designed for that, we're all looking for clues in the current context of what's coming up.

Matthew Galsky: So the big picture is that there's an improvement on exploratory analysis in the bladder cancer–focused population, an improvement in disease-free survival with adjuvant nivo versus placebo. That effect size seems to be a little bit larger than the overall study population. So that's take-home message number one.

Take-home number two is that effect size even enriches more in patients with PD-L1–high expressing tumors. As you know, ex-US, some of the labels are restricted to that population, not within the United States. On subset analysis of the initial study on the forest plots, there was this suggestion that maybe patients who got neoadjuvant chemotherapy benefit more from adjuvant nivolumab.

When you look at the subgroups presented for bladder only, that doesn't seem to be the case. The effect size is pretty similar whether or not you got neoadjuvant therapy or not. What is different is, as you can imagine, your risk of recurrence. Patients who had neoadjuvant chemotherapy and have residual disease have probably worse biology than patients who are treatment naive. And so the outcomes overall for that group are a bit worse. But the effect of adding nivo in the adjuvant setting doesn't differ if you receive neoadjuvant therapy or not.

Ashish Kamat: And from all the analyses you've done—because we're always looking how to de-intensify treatment as well—any clues as to which patient you might not want to recommend adjuvant nivo to?

Matthew Galsky: Not based on this data set. There's really not a population of patients, based on their eligibility for this study, that do so well on the placebo arm that you can say that patient shouldn't get adjuvant treatment. Of course, what we hope might offer an opportunity for de-escalation is the use of MRD testing, the use of ctDNA in the post-operative setting. We haven't seen ctDNA data from CheckMate 274.

We've seen it from IMvigor10, of course, a similarly designed study. And that offers a hypothesis that perhaps there's a population of patients for whom we can de-escalate treatment. And that's being tested prospectively in the MODERN study.

Ashish Kamat: Right. Exactly. And again, congratulations on that study as well, because we really have answers in there—escalating therapy, de-escalating therapy, intensification—which will be answered. Let me pick your brains a little bit on a topic that's related to this, that is relevant right now: the NIAGARA study. Clearly it's a different paradigm where you're having IO plus chemo up front in these patients, but then you're also continuing the IO therapy later.

How does that fit in with the nivo paradigm in your practice? In other words, is the IO benefit upfront with the neoadjuvant chemotherapy or is it being driven post? I know we can't answer that based on the existing data, but using the two trials and everything that you know—and you know more about them than most people—what's your sense as to how we can use that to counsel our patients?

Matthew Galsky: So I would say that the introduction of immune checkpoint blockade into the muscle invasive bladder cancer setting overall has been a success. Remember, these are the first studies in muscle invasive bladder cancer that have improved outcomes beyond neoadjuvant chemotherapy and cystectomy in the past 30-plus years. So the use of immune checkpoint blockade early is a good thing from an efficacy standpoint.

And then you get into the weeds of, well, when do we give it and how should we give it? And there are some trade-offs there. I would say that treating everyone with everything upfront ensures that we're treating patients who are going to benefit, who we might have missed if we waited till later, but it might be overtreating some patients. Alternatively, waiting until after, you're probably missing out on some patients who could have benefited.

And so that really becomes a shared decision. Does a patient want to incur the potential risk of receiving immune checkpoint blockade and start out with everything at once, or do they want to proceed in a more risk-adapted fashion? And of course, without randomized studies between those approaches, we're making cross-trial comparisons. But I think the important thing is to get the immune checkpoint blockade in and have a discussion about when you get it in.

Ashish Kamat: Right. But I'm going to push back a little bit on that simply because I have you here in the chair. If you look at the durva CG data, it looks like the path CR patients still do well continuing on the treatment arm. But if you look at CheckMate, the path CR patients do not get adjuvant nivo. So how do you put those two in context?

Matthew Galsky: Yes. So that's sort of what I was getting at by some patients missing out. If you wait until the adjuvant treatment decision, you will miss out on some. And to counter your comment, if you look at the EFS data of path CR patients who didn't get neoadjuvant durvalumab, it's pretty good. It's better with the combination, but it's pretty good. So you will overtreat patients with the upfront approach, and you will undertreat patients with the post-op approach applied to everyone.

And this is what we struggle with in the perioperative setting—how do we deliver the minimum amount of treatment to any individual patient that gets them cured? And we don't have the tools to necessarily do that yet, but I think they're coming.

Ashish Kamat: And that's a perfect segue, again, since you're the leader of the MODERN study, could you share with the audience the design and what you hope to see from there?

Matthew Galsky: So the MODERN study built on CheckMate 274, similar patient population, patients with high-risk pathological features—T2 or higher if patients receive neoadjuvant chemotherapy, T3 or higher if they hadn't received neoadjuvant chemotherapy and are cisplatin-ineligible. But the difference is that after surgery, patients have ctDNA testing using Signatera.

If it's detectable, they're randomized to receive adjuvant nivo standard of care or adjuvant nivolumab plus relatlimab, the LAG-3 inhibitor—so doublet immune checkpoint blockade versus single agent. If ctDNA is undetectable, then patients are randomized to receive standard of care adjuvant nivolumab or observation, with the opportunity to receive nivolumab if they develop a conversion from undetectable to detectable ctDNA.

The ctDNA-detectable population—it's a seamless phase II/III design. So the phase II endpoint is ctDNA clearance. We want to see ctDNA clearing at a higher rate on the combo versus the single agent. If we do, then we'll expand. And that's an overall survival endpoint. The ctDNA-undetectable, the endpoint—of course, it needs to be different, and the endpoint there is noninferiority for disease-free survival between the two.

Ashish Kamat: Excellent. And that clearly is where we're going from doing everything to the patient in one paradigm to highly personalizing it in the other paradigm. Matt, always a pleasure to chat with you. Thank you for taking the time. Enjoy the rest of the meeting.

Matthew Galsky: Thank you.