Guidelines Aim to Harmonize and Optimize Urothelial Carcinoma Clinical Trials - Ashish Kamat and Matthew Galsky

March 13, 2024

Sam Chang leads a discussion with Ashish Kamat and Matt Galsky on their co-authored article, focusing on new trial parameters for urothelial carcinoma. This collaboration, involving SITC, IBCG, FDA stakeholders, and patient advocates, responds to the urgent need for detailed clinical trial guidelines due to the dynamic landscape of bladder cancer treatment. The dialogue highlights the transition towards control arms in clinical trials, especially for non-muscle invasive bladder cancer, and emphasizes the necessity of categorizing patients by risk level more accurately. For advanced diseases, the conversation underscores updating comparator arms to reflect the latest standards of care. The discussion also points out the critical role of ctDNA in future trial designs and patient management. This initiative promises regular updates through SITC's dynamic online platform to ensure the guidelines remain relevant and comprehensive.


Ashish Kamat, MD, MBBS, Professor of Urology and Wayne B. Duddleston Professor of Cancer Research, University of Texas, MD Anderson Cancer Center, Houston, TX

Matthew Galsky, MD, Medical Oncologist, Director of GU Medical Oncology, The Tisch Cancer Institute, Mount Sinai, New York, NY

Sam S. Chang, MD, MBA, Urologist, Vanderbilt University Medical Center, Nashville, TN

Read the Full Video Transcript

Sam Chang: Hello, everyone. Thanks for joining us. My name is Sam Chang. I'm a urologist in Nashville, Tennessee, and we are very fortunate to have two of the leading experts in urothelial carcinoma. We have Professor Matt Galsky, who is at Mount Sinai, and Professor Ashish Kamat, who is at MD Anderson. We're going to speak on an article that they have actually been kind of the lead authors in bringing a group of experts together that are looking at setting up new trial parameters for patients with urothelial carcinoma and the trials we're designing. Ashish, I'm going to start with you. How did this all come about, and tell us a little bit about the background?

Ashish Kamat: Thanks so much for having us here, Sam. As you know, and because you've been involved in guidelines and directing bladder cancer for the AUA for many years, we need specific guidelines in different areas, in the clinic and of course when it comes to the conduct of clinical trials. Back in 2016, obviously in collaboration with the GU ASCO group, with AUA, with the IBCG, the FDA put out a guidance document that was then matured in 2018 that pretty much led to an explosion of trials in the BCG unresponsive space.

Through SITC, Dr. Galsky and I were developing the guidelines for immunotherapy, a question came up, "Hey, why don't we get the two groups together to develop clinical trial guidelines that would inform investigators, pharma, anybody at every disease state in bladder cancer, which then could be used by the FDA to say, this trial follows these specific guidelines and hence it's a trial that, at least in the design, meets certain base minimum criteria?" So this effort was actually a joint effort between SITC, Society for Immunotherapy of Cancer, IBCG, International Bladder Cancer Group. But we had stakeholders from the FDA involved in public discussions. We had patient advocates, so BCAN was heavily involved in this. Some of the preliminary work actually happened at the virtual BCAN because that was during the time of the pandemic. So that was the genesis of this.

Sam Chang: Let's talk about certain highlights. Ashish, let's start with you. What about some key points that you want to kind of advocate when it comes to non-muscle invasive bladder cancer?

Ashish Kamat: So much has changed, Sam, when it comes to non-muscle invasive bladder cancer, right? Starting from 2017, '18 till now, 2023, when the paper was published in '24. The biggest change has been the approval of two agents in BCG unresponsive disease and many more coming down the road. One of the big things that we try to advocate for here, and again in a public forum, is the fact that now that we have a standard of care, which is gemcitabine-docetaxel de facto standard of care and two approved drugs in this BCG unresponsive space, we should now, moving forward, have control arms. No longer just rely on single-arm studies where you don't really know what's happening. Is the improved response rates we're seeing over time because patient selection is better or because the drugs are better? We don't know. So having some sort of control arm built-in is important.

The other thing in the trial, in the manuscript, we sort of went through low-risk disease, intermediate-risk disease, how to stratify patients that are intermediate-risk bladder cancer, and at least for the trial design, focus on low-grade intermediate-bladder cancer patients separate from high-grade intermediate-bladder cancer patients because the endpoints are different. Too many companies and too many patients have invested their time in enrolling in trials that then don't read out the way we want them to because the inclusion criteria are so heterogeneous that you can't really make head or tail of them despite the powers of randomization, which really can't account for this sort of heterogeneity. So in low-risk patients, an appropriate control arm, just TURBT. Intermediate-risk patients have the low-grade and the high-grade patients separated out. And then clearly when it comes to BCG naive, exposed, and unresponsive have the appropriate control arms.

Sam Chang: Yeah, I think it's a really good point because the heterogeneity of the disease really compounds the difficulty when you're trying to compare apples to apples and oranges to oranges. How about in that BCG unresponsive group, Ashish, what should be the comparator arm for those individuals or for those clinical trials?

Ashish Kamat: Great question. Clearly radical cystectomy, which still is the guideline recommendation and should be our recommendation when we talk to patients, can't be the control arm. So what is the control arm? That's a difficult question because if you look at the approved agents, pembrolizumab or nadofaragene, the 12-month response rate is about 20-23%. And now to mandate patients going into a control arm with a drug where we have to tell them your response rate is only 1 in 4 is not fair.

So what we are recommending, and again, leaving it up to the trial designers, but in discussions with regulatory bodies, is to allow best practice, best practice at that center, and then give a choice. Either what I said, the de facto standard in the US now is gemcitabine and docetaxel, or also allow pembrolizumab or nadofaragene, to keep the generic names, and/or anything else that might be approved. That's sort of the soft recommendation.

Sam Chang: Yeah, I think that's a difficult area to kind of determine. To allow some flexibility definitely makes it more real-world. Obviously, every time you have some variability, you introduce some type of variation that may impact the outcomes. But I think the day and time, and I think you agree with this, of the single arm, just as you said, probably is no longer fair in this current situation.

Ashish Kamat: Exactly. I think it's not fair to our patients to just have them go into single-arm studies and not be able to make much of the results because the results are getting better, and we hope it's because the drugs are getting better, but we don't know.

Sam Chang: Matt, what about in the invasive world? When we have invasive disease, we have neoadjuvant types of thought processes, the adjuvant, and obviously the more advanced disease with metastatic disease. Tell me some key highlights when you look at clinical designs for those disease settings.

Matthew Galsky: I would say the key highlights are thinking about harmonizing eligibility across those disease states a bit more and, to the points you were just making, harmonization of eligibility, harmonization of endpoints really helps in a number of ways. One, it helps in terms of cross-trial comparisons, which we try not to make, but if you're going to do them, it's best to do them when you're actually comparing apples to apples somewhat.

The other thing that it helps facilitate is meta-analyses. We're always taught that we're not supposed to rely on one randomized trial to change practice. In oncology, we do it a lot because it's harder to generate multiple phase III trials. But when we do have those trials, the level of evidence gets higher, and having clinical trial data sets that match facilitates doing those meta-analyses, which is arguably the highest level of evidence. So I think those are some of the key highlights, really trying to think about not only naming things the same, event-free survival, for instance, but actually, what does event-free survival mean in one study versus another? And it's actually based on the same definition.

Sam Chang: Let's tackle the difficult situation with comparator arms. We've talked about BCG unresponsive. How about, let's go then now to the far end of advanced disease, metastatic disease. We've got so many big changes that have occurred. At this point, what do you think should be the comparator arm, Matt, with future trials looking at those patients with metastatic or advanced urothelial carcinoma?

Matthew Galsky: I think one of the points that you highlight is how quickly the field is moving, and I think that's great. It's great for patients to have a document like this somewhat outdated in terms of the control arms that are proposed within months of publishing the paper. I mean, that was unheard of in the past. And so I think that's really a testament to the field.

What I'm referring to specifically is that we talk about gemcitabine, we talk about platinum-based chemotherapy as a comparator for first-line metastatic urothelial cancer trials in the paper. Now we actually have two regimens that have improved survival compared to standard platinum-based chemotherapy. I think currently EV plus Pembro, arguably, is the comparator arm for patients with metastatic urothelial cancer. I think in the immediate term, it's somewhat complicated because when we think about trials, practice-changing trials, those are international trials and not all regions of the world have access to all of these treatments as rapidly as we do in the United States. And so I think we do have to be thoughtful about designing phase III trials that are actually feasible to enroll, at the same time recognizing that we also need to design those trials so that when we get the result, it's actually still relevant for that period of time.

Sam Chang: That feeds into Ashish's points regarding the BCG unresponsive population of the best standard of care for that individual area. To see the amount of rapid change that you just explained is really incredible compared to even 5 years ago, what we had stayed policies, used drugs that had been used for decades. It's really, I think, significant.

Let's go early in the disease process then, Matt, in the neoadjuvant group. In looking at the type of primary outcomes for the neoadjuvant invasive group, with everything that we're learning about prognostic as well as predictive biomarkers in terms of what really is disease that's present or absent, give us an idea of the next steps when it comes to clinical trial design for the neoadjuvant patients.

Matthew Galsky: For the neoadjuvant patients, we do make the case for Path CR and event-free survival as relevant endpoints, and certainly not using Path CR by itself. But I think this circles back to one of the questions that you asked earlier, which is that one of the values of doing this, to credential a surrogate endpoint such that one day it could be used for regulatory decision-making, requires meeting the criteria for surrogacy. And the only way you can meet the criteria for surrogacy is to have multiple randomized clinical trials to which the definitive endpoint is met, like survival, to then be able to show that changes in the surrogate completely explain changes in the outcome of interest.

So I think there are five, at least five, probably more, randomized phase III trials in the neoadjuvant setting right now, which all include Path CR as a co-primary endpoint. Think about the power of those data sets and being able to finally ask the question, is Path CR a surrogate for survival and can we use it in the future?

Sam Chang: Very, very good point. And that determination of pathologic CR, that's even a moving target as we attempt to really determine what's going on in the disease processes. While I've got two of the smartest and most experienced gentlemen, tell us, first of all, Ashish, this should be an iterative process. How often are you all planning on getting together to revisit this idea, this concept, because obviously this is a moving target?

Ashish Kamat: One of the beauties of partnering with someone like SITC is when the SITC guidelines come out, and I had the pleasure of chairing the SITC Guideline Panel for almost 6 years, what they do is they send out these reminders to the experts across all disease states saying, "Hey, do you think enough has changed that it warrants an online update in 3 months or do you need a full rewrite in 12 months?" So it's a very constantly iterative process, and that's pretty much what we plan to do here.

As Matt mentioned, in the advanced stages, now clearly EV-Pembro has to be considered. It may not be something you can mandate, but has to be considered. In the neoadjuvant setting, the FDA, so far, has said they're not going to consider Path CR at all, but that might change. And things may change in earlier stages of disease too. So I think this is something that, thankfully, through the auspices of SITC, we will be able to visit every 3 to 4 months and make changes in a dynamic fashion as and when needed, at least to the online version of the document.

Sam Chang: That process of being able to constantly get feedback from everyone, get that next step in terms of, "Hey, we're right on the precipice of making significant changes," and being able to do that so quickly and with so much agility really speaks to the timeliness of anything that's put out by this group, which is fantastic. Matt, I'll give you the last word. Tell us, what is, for you, the most exciting concept when it comes to clinical trial design for patients with bladder cancer? It could be non-muscle invasive, it could be invasive, but in terms of the concept of trial design, what excites you the most?

Matthew Galsky: I think ctDNA is likely going to be transformative in terms of how we think about trial design, trial endpoints, and ultimately how we manage patients in the clinic because it really hits on almost every topic that we spoke about. Eligibility. Eligibility for the adjuvant setting in patients who have detectable ctDNA versus not. Endpoints. Intermediate endpoints. Does clearance of ctDNA in a setting where you can't visualize radiographic evidence of disease correlate with longer-term outcomes, and is that basis to screen novel therapies more rapidly? So I think that particularly in the perioperative setting, I think that technology is going to have the biggest impact on what we do in the next 5 years.

Sam Chang: Well, Doctors Galsky and Kamat, gentlemen that you truly are, experts that you truly lead, we are incredibly lucky to have you for a few minutes and very much appreciate the insights in an ever-changing world when it comes to urothelial carcinoma and clinical trial design. We look forward to your future leads in not only trial design but the next steps in terms of bettering improvement for our patients with bladder cancer. Thanks so much.

Ashish Kamat: Thank you, Sam.

Matthew Galsky: Thank you.