Read the Full Video Transcript

Ashish Kamat: Hello and welcome to UroToday's Bladder Cancer Center of Excellence. I'm Ashish Kamat, Professor of Urologic Oncology and Cancer Research at MD Anderson Cancer Center in Houston. And it's a pleasure to welcome back to this platform, Professor Peter Black, who really needs no introduction. He's the Khosrowshahi Family Chair, Professor of the Department of Urological Sciences at the University of British Columbia, Editor-in-Chief of the SIU Journal, and pretty much an expert in all things bladder cancer and he has recently published under the auspices of the International Bladder Cancer Group, the IBCG, a new proposed definition called, BCG-Exposed in high-risk non-muscle-invasive bladder cancer. And this is a very important definition for clinicians, regulatory agencies, et cetera, who do a lot of work in non-muscle-invasive bladder cancer. And for that reason, Peter, we thought we'd have you come here today and talk to us a little bit about your publication and your thoughts on this new space, so to speak. With that, the stage is yours.

Peter Black: Perfect. Thank you, Ashish. It's a pleasure to be back. I think these UroToday sessions are really valuable, especially during the pandemic and I think getting a lot of attention. I think this is a very important topic. I think it's a relatively new topic. We are taking a gray area, non-muscle-invasive bladder cancer, and trying to define it and remove some of the grayness.

I have a bunch of disclosures I don't think are relevant today.

But let me just put things in context a little bit. We are talking about high-risk non-muscle invasive bladder cancer when we're talking about the different disease states related to BCG. And so we're focusing on carcinoma in situ, TA tumors, and T1. And of course, there are different risk stratification tools out there. They are really quite similar when it comes to defining high risk. And I don't think it matters too much which one you use, whether you use a European tool or the AUA risk groups. In the end, especially if you're thinking of trial design, they will come out the same. But in general, high-grade T1, high-grade TA, and carcinoma in situ are what we are talking about.

The importance of it being high risk is that high risk is always treated if the patient is eligible with BCG. That remains the standard of care after now more than 45 years since the original publication and it is given with induction and maintenance as per the SWOG protocol, which is now also almost 22 years old. This remains a standard of care for high-risk non-muscle-invasive bladder cancer and it's really the use of BCG that defines the disease states within high-risk non-muscle-invasive bladder cancer.

We, of course, are very familiar with BCG-naive. This would be the patient who's embarking on a treatment path with high-risk disease. Some will end up being BCG-intolerant. I'm not going to talk about them. We've become familiar with the definition of BCG-unresponsive, which we will go into a little bit of detail and today we want to focus on BCG-exposed.

And so the BCG-unresponsive disease state has received a lot of attention over the past several years. A panel from the AUA, as well as the FDA, got together the first time in 2013 and developed some definitions in clinical trial design rules essentially, which culminated in this document in 2018. This is a guidance document from the FDA that is steered towards industry and doing clinical trials. But it's also very relevant for general clinical practice because it defines a patient population that has really exhausted BCG as a treatment option and these patients should move on to something other than BCG.

And so BCG-unresponsive includes patients who meet the more traditional criteria of BCG refractory plus also those who've responded to BCG and have an early relapse. And so it's the Ta/CIS patients who've had induction plus one round of maintenance or a second round of induction BCG, high-grade T1 after induction only, and then those who recur within six to 12 months after the last dose of BCG.

And due to the attention that has been put on this disease state and the clarity around trial design and definitions, we've seen a huge amount of activity in the clinical trial landscape with multiple different agents being tested. Pembrolizumab has been the first of all these agents to be approved but other agents are almost certain to follow in the near future. You see also, interesting how we have systemic therapies for non-muscle invasive treatment, as well as the more usual intravesical therapy. There's a lot going on. And the question is, how can we transfer some of this activity now into earlier disease states?

With that in mind, the International Bladder Cancer Group got together, a broad international panel of experts in bladder cancer and we wanted to address how we can define this disease state and then also suggest parameters for clinical trial design.

One of the first questions that came up was what is the name of this entity? BCG failure is actually a term that came out of the original FDA, AUA discussions but it's a fairly broad generic term that is used for different things in routine practice so we didn't think that it was specific enough. BCG experience is used by some. Ultimately we developed a consensus around the term, BCG-exposed.

And this slide tries to define exactly what we're looking at here. And this is a little bit complex, so I'll spend a couple of minutes on this. But if we go all the way over to the left side of this graph, we have patients with high-risk non-muscle invasive disease. If they get BCG induction only without maintenance and have a high-grade Ta or carcinoma in situ recurrences at the three-month time point, so at the time point of the first evaluation, typically those patients would still go on to receive more BCG.  So they are not BCG-unresponsive, the old term is BCG-resistant. And so we would include them in the BCG-exposed. They are not BCG- naive, they are not BCG-unresponsive, they are in that in between, which is BCG-exposed.

Then we have the patients in dark blue who actually meet the criteria for BCG-unresponsive. I won't go over that again.

And on the right, we have sort of two different populations, there are patients who receive adequate BCG, which means they really are getting induction and maintenance and potentially the full three years of maintenance but then recur or there are patients who get induction only or some variation thereof but not the classic SWOG protocol, who have a high-grade recurrence. And so both of these patient populations have had some BCG but they do not meet the criteria for BCG-unresponsive. And as long as they recur within 24 months of the last dose of BCG, we would consider them BCG-exposed.

And so to summarize that definition, it is really high grade persistent or recurrent non-muscle-invasive bladder cancer within 24 months of the last BCG dose but not meeting the definition of BCG-unresponsive, especially the time component and the adequacy of BCG will distinguish the two entities. The International Bladder Cancer Group panel thought that a high-grade recurrence after 24 months, after the last dose of BCG was a good cutoff for eligibility because beyond that, you would expect the outcome of a re-challenge with BCG to be similar to a BCG-naive patient so that we would want to concentrate trial activity in a higher risk patient population. We excluded low-grade recurrences since they have different biology and different clinical behavior. And we also excluded BCG-intolerant patients because as I'm going to mention in a slide to come, the standard treatment for BCG-exposed patients would be additional BCG. And of course, a BCG-intolerant patient could not be randomized to that. We'd also expect that BCG-intolerant patients might have a better outcome because they have not yet demonstrated any treatment resistance.

One of the hardest questions in this disease state is trying to define the anticipated outcomes that you'd expect for a patient with BCG-exposed non-muscle invasive bladder cancer since you want to define your endpoints and power your trial according to expected outcomes. We really have no good quality benchmarks for this. If we go back to literature, there are individual small reports, for example, of patients who recur after BCG induction only, but it's really hard to sort out exactly what the anticipated outcome should be. In the table here, we've indicated some numbers that are an approximation, especially for patients with a delayed relapse after induction and maintenance BCG. Some of the trials that have already been started in this disease state may actually inform future trials because they will give us a better indication of outcomes.

So additional points on trial design, the panel thought that only randomized controlled trials are valid to test new therapies in the BCG exposed population. The control arm would be BCG induction plus maintenance since that is what we would consider standard for these patients currently. There was controversy and indeed we did not arrive at a consensus on whether or not the CIS patient should be separated from the Tat-related patients and treated differently. There have been some signals from the FDA that they would like these two populations managed differently in a trial. If they are managed differently, then it would increase the sample size quite dramatically, which may cause problems with feasibility. If they are grouped together, we could, of course, stratify to try and accommodate for the different biology.

The primary endpoint would of course depend on that question of one group versus two groups for carcinoma in situ. It would make sense to look at a complete response at three or six months. And if you're looking at the papillary tumors or if you put papillary and CIS together, you would want an event-free survival with a specific definition of an event. And if you're including CIS, then the persistence of CIS would be an event. A recurrence in the upper tract or urethra, so outside of the bladder but in the GU tract, would be considered an event if the experimental treatment is systemic but not if it's intravesical. A lot of these types of nuances and details are addressed in the manuscript.

And so this is my last slide, which just shows that there are trials underway in this disease space, BCG-exposed, high-risk non-muscle-invasive bladder cancer. These trials were of course already launched long before this publication came out. It to some degree shows that we need more activity in this space. We do have a lot of patients who are receiving BCG, and in a way that leads to a BCG-exposed disease state. And I think there is a lot of room for improvement here. Again, the idea behind this manuscript was really to set out some ground rules to improve the consistency in trials and encourage people to design trials for patients with BCG-exposed non-muscle invasive disease.

And I will end there. Thank you.

Ashish Kamat: Thanks, Peter. As always, a very nice, succinct presentation. One of the reasons that the whole BCG-exposed paradigm is so important is also because there is a BCG shortage. Even though we have this paradigm through the FDA to do single-arm studies in BCG-unresponsive patients, in the real world, not many patients are actually getting that treatment and we are finding patients not eligible for clinical trials simply because they do not have access to BCG. And then the question is, well, what do we do with them? Having this well-defined disease space, I think is a huge benefit to anybody that's looking to design clinical studies in this space. It will give a standardized way of looking at trials.

And of course, we're not recommending that people do cross-trial comparisons, but at least if you're comparing apples and apples in different studies, you can make some inferences from the responses in different trials. A quick question to you, we've talked about this and my bias is that even in the BCG-unresponsive disease space, now that we have some agents that are approved in activity with some off-label agents, I feel that we should now move to randomized studies, even in the BCG-unresponsive space. What's your feeling on that topic? And also, how do you think that paradigm compares with the requirement for randomized studies in the BCG-exposed? And for our audience, why is one in your opinion, better than the other?

Peter Black: No, it's a good question. I think the difference between the two is that for the BCG-unresponsive state, up to now there hasn't really been a standard comparator so we wouldn't have known what to randomize to, since we can't really randomize to cystectomy. But for BCG-exposed, you have a standard treatment which would be additional BCG. It offers a feasible and easy comparator arm. I think that the time is definitely right for randomized trials in the BCG-unresponsive setting. I think that the FDA has come out publicly and stated that they're not requiring that. But I think that we as a field should be pushing for that because I think we will learn a lot more if we're comparing to treatments and it will allow us, for example, also to develop better biomarkers because you'll be able to see how the biomarkers perform in different arms of the trial.

And I think especially, combination treatments are promising. We have a lot of single-agent trials, single-arm, and single-agent, and now it will be nice to combine them. Take one of the intravesical treatments with one of the systemic immunotherapies, for example, and test that to either alone or different combinations. I think that is where we are headed. In the US, the single-arm model leads to approval and funding but I think in the rest of the world, it's not considered adequate. It may lead to approval. In Canada, pembrolizumab is approved but it's never going to be funded as a single-arm trial. And so a comparator trial that shows the benefit of one over another, I think would also lead to broader implementation.

Ashish Kamat: Yeah. And that workshop that you mentioned, and the FDA sort of left it up to us to advise them on whether we as a field felt that we needed randomization or single-arm studies. And we, as a field, the group of experts was divided. I think it was a 48/52 split, which way to move forward. I think we also need to come to a consensus amongst ourselves.

A question to you, and this is again for all BCG-related studies, but since we're talking about BCG exposure, how would you advise investigators, pharma, regulatory bodies to consider patients who've received different strains of BCG when they want to enroll in clinical studies of say, BCG-exposed manner? Accounting for randomization is one thing, but assuming you can't do that, do you have a preference for how you would design a clinical study with different strains of BCG being used as the entry point?

Peter Black: Yeah, it's another good question. And something that we discussed within this IBCG panel and I think ultimately the consensus was to ignore strains. That makes it too complex and we do not know well enough that it makes a difference, that we would even want to stratify by it in a randomized trial. My personal inclination would be the same, is that we don't really know that it makes a difference. And with the data that we have other than the one Swiss trial, we would think it probably doesn't make a big difference. And as long as the patient has had adequate BCG before, I think the differences in outcome from the next line treatment would be so small that I don't think we should really be worrying about it, would be my opinion.

Ashish Kamat: Any closing thoughts you want to leave our audience with, Peter, on this topic or any other topic since you have the mic now?

Peter Black: No, I'd just love to see us doing more trials in this space. We have some really good trials in BCG-naive and we've had all this activity in BCG-unresponsive and we're sort of preparing for what's next in that disease state but I think we really need to ramp up and deal with this BCG-exposed population. We see a lot of these patients, especially when we're screening for the BCG-unresponsive trials and we find out, well, no, that patient's not eligible, that patient's not eligible. They would be eligible for these trials. It would be nice to get some things going in this disease space.

Ashish Kamat: Yeah, absolutely. I agree with you. Having this well-defined disease space will hopefully spur more activity in there, more than what we already have. Thanks again, Peter, for taking the time. Stay safe, stay well, and hopefully, I will see you at the AUA now that the EAU has been postponed.

Peter Black: Yes. Thanks. You too.