How to Define Treatment Failure After BCG Treatment for Nonmuscle-Invasive Bladder Cancer - Peter Black

October 14, 2020

Co-chair of the Bacillus Calmette Guerin (BCG)-unresponsive non-muscle-invasive bladder cancer session during the 2020 Virtual Bladder Cancer Advocacy Network Think Tank (BCANTT) meeting, Peter Black, MD joins Ashish Kamat, MD, MBBS providing a succinct overview of this session highlighting the clinical and the translational panel which addresses defining treatment failure after BCG treatment for nonmuscle-invasive bladder cancer (NMIBC) patients. Currently, there are guidelines that tell us that these patients should have radical cystectomy but for patients who are medically unfit or who would prefer not to commit to cystectomy immediately, there has been a push to look for novel treatment options. In summarizing this BCAN TT session, Dr. Black highlights three therapies in this disease space, one currently federally approved in the United States and two others, pembrolizumab, oportuzumab and nadofaragene firadenovec. He focuses on the clinical aspects of what is taking place with the new developments in BCG-unresponsive nonmuscle-invasive disease, laboratory advances, how best to sequence and design the next generation of trials, and what are we going to do to understand the biology better so that we can use that to guide treatment decisions for these patients.


Peter Black, MD, Senior Research Scientist, Vancouver Prostate Centre, Professor, Department of Urologic Sciences, University of British Columbia, Vancouver Prostate Center

Ashish Kamat, MD, MBBS, President, International Bladder Cancer Group (IBCG), Professor of Urology & Cancer Research, MD Anderson Cancer Center, Houston, Texas

Read the Full Video Transcript

Ashish Kamat:  Welcome to UroToday's Bladder Cancer Center of Excellence. I'm Ashish Kamat from MD Anderson Cancer Center. It's again, a pleasure to welcome Peter Black from Vancouver, whom everybody knows really well. Peter is joining us today to share with us his thoughts on BCG-unresponsive bladder cancer from a slightly different perspective this time around. He's going to summarize the thoughts he shared with us at the Virtual Think Tank this year, which was held earlier last month. Peter, the stage is yours.

Peter Black:  Thank you very much, Ashish. And thanks for the opportunity. I'm going to frame this talk really around the Think Tank, which means it is not only my opinion, but I'm going to try and relay the opinion of the people who were speaking at the meeting. This was a virtual meeting that we did in early August, and it was all about BCG-unresponsive non-muscle invasive disease. I think the audience is probably very familiar with the Think Tank. This has become the meeting in North America for bladder cancer experts, both the clinical experts, the physicians, as well as the researchers, whether in the lab or clinical epidemiology or whatever it might be. There's also a strong component of patient representation and advocates. So it has very special chemistry that works well.

This year's meeting was planned for Denver. It was going to be a two-day meeting and Matt Galsky and Sima Porten were the chairs. But of course, because of COVID, it was reduced to a two-and-a-half-hour virtual meeting on August 7th. Colin Dinney and I were lucky to be the chairs of this part of it under the leadership of Matt and Sima. You can see here we had two panels. We had a clinical panel, as well as a translational panel. You can see some recognizable names on the list here. I would highlight that we had Karen Sachse, who is a representative of BCAN, so a patient advocate, and we were happy to have that input into some of the topics of discussion.

So let me jump right into what we talked about in the clinical panel. We basically had Arjun Balar, a medical oncologist in New York talking about pembro in BCG-unresponsive disease. We had Rian Dickstein talking about oportuzumab and of course, Colin Dinney talking about nadofaragene.

But I'll give just a little bit of background. So the context, of course, is BCG and what happens if BCG stops working. BCG has a really remarkable track record in urology. We've been using it for more than 40 years, and it is still standard of care and we still get good results with it. The SWOG protocol of induction and maintenance is still the standard after the publication of this paper way back in 2000. But at the same time, we are still faced with the usual problem. What happens if BCG stops working? There has been a series of papers that culminated in this guidance document from the FDA on how to define treatment failure after BCG. This is particularly important for clinical trials, as suggested by the title of this document, but it is also important for routine practice and when we know to move on from BCG and do something different.

So what is BCG-unresponsive disease? I think again, the audience will be very familiar with this, but in a nutshell, it is a Ta tumor or carcinoma in situ, so high-grade Ta or carcinoma in situ that has not disappeared by six months, or a T1 that has not disappeared by three months of BCG, of course. So in the first, it is induction and one round of maintenance or two rounds of induction. For T1, it's after a single round of induction. Or it's a relapse after a patient has reached a CR. They then recur within six to 12 months of the last dose depending on whether it is papillary or CIS.

So of course, the guidelines tell us that these patients should have a radical cystectomy, yet a lot of us believe that that might be overkill, and certainly patients often also believe that that might be overkill, especially for a high-grade Ta and carcinoma in situ. For T1, I think there's some agreement that yes, we should remove forward with radical cystectomy in the BCG-unresponsive state.

But still, there are a lot of patients who are medically unfit or who would prefer not to commit to cystectomy immediately. So it's for these patients that we are looking for novel treatments. We have a bunch of options. I'm not going to go into any of these in detail. But overall, they do not work very well. The best that we have is not really good enough. We see here, for example, on the left of the screen, this is a SWOG trial testing gemcitabine in BCG refractory patients. They are not truly unresponsive, but they are fairly similar. And you can see at one year it is only a 28% recurrence-free survival rate. In two years, it's 20%. So these are not very promising numbers if you have a fit patient in front of you who really is a good candidate for cystectomy. More recent retrospective data, using the popular combination gemcitabine, docetaxel looks a little bit more favorable with higher recurrence-free rates. But I think we still need to do better. This has all the caveats of course, of retrospective data.

So our best second-line treatment is not good enough, and there has been a big push to look at novel therapies. This is where we then had our experts jump in, as I mentioned, and talked about these three drugs in the context of BCG-unresponsive disease. The trial results that they presented, they all pertain to carcinoma in situ more so than a papillary disease. For nadofaragene, we do have some evidence in the papillary disease, but the focus just in case it gets lost in subsequent slides, is BCG-unresponsive carcinoma in situ. Pembrolizumab, an immune checkpoint inhibitor that we are familiar with from the metastatic setting. Oportuzumab monatox, which is like an antibody-drug conjugate. It's actually a recombinant fusion protein, and we will get to that in a minute. And then nadofaragene firadenovec. I'm intentionally sticking to the generic names here. I think we have to get used to those. Nadofaragene is a virus-based adenovirus gene therapy.

Arjun Balar was up first. He talked about Keynote-057, which was a single-arm phase two trial looking just at the CIS population, so cohort A. They had pembrolizumab IV infusion every three weeks for up to two years. The primary endpoint was complete response rate at three months. What Arjun didn't talk about, but it did come out in the discussion, and I was the chair in this trial, so I'm sneaking it in here, but 1605 was a very similar trial, a SWOG trial, cooperative group trial, looking at atezolizumab for one year in patients with BCG-unresponsive disease. We presented the complete response rate at six months, so different endpoints at ASCO. In this study, we had a mandatory biopsy at six months, which was not in the other study.

You could see here just the results side by side are fairly similar. So Keynote, first of all, the three-month complete response rate in CIS patients is 41%. That's the primary endpoint. It goes down to 31% at six months, about 20% at 15 months. 15 months is important because it's one year after this. So we talk about the durability of one year and it's about 20%. Grade three treatment-related adverse events about 13%. If you compare it to the SWOG trial, a somewhat smaller trial, a very similar number at three months. Even though this wasn't the primary endpoint, we went back and looked at that just to compare it to Keynote-057. At six months it was 27% CR, again noting that there's a mandatory biopsy in this trial. Durability is pending. So it's really only valuable if we see some durability of response and a similar rate of toxicity.

I should add that the pembrolizumab, of course, is FDA-approved as of the beginning of the year, and so this is in clinical practice, which was very important than in the discussion part of the session. Oportuzumab, I said is like an antibody-drug conjugate, but it is basically an antibody that recognizes EpCAM on the cell surface. It is fused to the exo-toxin of pseudomonas, which is protein. So that is why it's a fusion protein. It binds to EpCAM, taken up into the cell, the exo-toxin is released and it disrupts protein synthesis and causes tumor cell death. So this is a very interesting drug. It is also administered intravesically, and it's relatively treatment intense. So it is twice weekly for six weeks, and then once a week for an additional six weeks. That's induction. Then there is maintenance treatment for up to two years, and it's every other week.

So a lot of installations and you can see the three-month CR rate in the CIS population is very similar at 40%, and you get a similar drop-off also around 27% at six months. It's down to 17% at 12 months. This is under review by the FDA for BCG-unresponsive CIS. I should add that was Rien Dickstein who presented that data. He was one of the lead investigators.

And then for nadofaragene, Colin Dinney presented this data and he, of course, has been involved in the development of this drug from the beginning. He presented the trial data from their phase three trial. Phase three is still a single-arm trial, but since there's been a prior phase two, it is being called phase three. Similar for the oportuzumab, this is a virus, adenovirus, that carries the interferon-alpha-2b gene. The drug is prepared with CIN 3, which is essentially a detergent that helps the virus get into the urothelium, and the bladder cancer cells will actually express interferon and that can lead to cell death.

You see here for the CIS population a complete response rate of 53% at three months, gradually dropping off to 24% at 12 months. Higher in the papillary cohort, which we didn't really focus on too much here. This is anticipated. I note here just at the bottom, that eight patients had progression to muscle-invasive disease. We do not know what that number is for oportuzumab. We do know that no patient progressed on Keynote-057 while on treatment, but some did progress subsequently on cystectomy. So the numbers are similar, and this is not necessarily a red flag at all.

So we lined up the three drugs against each other and had a good discussion of some of the pros and cons and how this might be rolled out in clinical practice.

For pembrolizumab clearly, it is the first across the line. It is FDA approved, and it's great for patients that we have this option. The downside is the potential for immune-related adverse events. There is also the necessity for infusion in an infusion center, which is often referred to as chair time. And then the cost. Certainly expensive, although we don't know how the other drugs, how much they will cost in comparison, but the cost is a consideration.

For oportuzumab, the big issue is just the frequency of intravesical instillation. So having to catheterize a patient for intravesical administration is already an issue, but to do it frequently, of course, is a heavy burden on the patient. Otherwise, it is very well tolerated, as is nadofaragene. Nadofaragene, I didn't mention it in passing, but it is given every three months. So it's an infrequent intravesical administration, but nonetheless, it is intravesical. So I think intravesical, generally, we think of as being a good thing, but from a patient perspective, we do have to remember, that, also is invasive.

So some of the discussion points. First of all, Arjun Balar rightly pointed to the fact that pembro has redefined multidisciplinary care of nonmuscle invasive disease. We now need to co-manage these patients together with medical oncologists and urologists, and it's not just a matter of handing the patient off and being done with it. The patients still need cystoscopic surveillance, for example, from the urologist, even if the medical oncologist is managing the systemic therapy. I think there are educational gaps where the medical oncologist needs to learn a bit more about the nonmuscle invasive disease, and we, as urologists need to learn a little bit more about immunotherapy and managing adverse effects.

The question came up, another discussion point, is how with these three new drugs, two of them potentially nearing FDA approval, one of them already approved, what do we do with standard non-approved therapy, such as intravesical gemcitabine/docetaxel, which is popular and very well tolerated. It doesn't cost very much. It looks good in a retrospective series. The big caveat, of course, is that we do not have a high level of data, and it remains uncertain how this will fit into the treatment algorithm.

If we do have all three drugs approved, how will we pick the first drug or match a drug to the patient, and how will we sequence them assuming that a patient may decide, even if one of them doesn't work, that they still don't want a cystectomy and they want to go onto the next one. So there's some discussion around this.

This is very importantly where Karen from BCAN came into the discussion and offered the perspective of the patient and how this should be important for treatment where the patient prefers systemic versus intravesical administration. What is the potential for side effects for each agent? What is the enthusiasm for the patients for one agent versus the other? Sometimes patients are very excited for example, about immunotherapy. Karen also pointed to two important studies. One is a CISTO study that Angie Smith and John Gore are leading. It's a PCORI-funded trial that will compare additional intravesical therapy after BCG failure versus cystectomy. Patients are not randomized in any way, but it's an observational trial of patients undergoing standard of care. BCAN itself has done a survey to see how patients balance the risks and benefits of different treatments in this setting. That will all help us decide in moving forward.

Finally, for clinical trials we talked a little bit about, are we ready for comparative trials in BCG-unresponsive disease? For example, if pembro is approved, can we compare to pembro? I think there is consensus that it is not yet required by the FDA, but eventually, we will move in that direction.

The second part, or the last third, I guess, of this session was dedicated to the translational panel. We had three talks. For the sake of time, I won't go into detail here, but Max Kates from Johns Hopkins talked about PD-L1 expression in BCG resistance. In particular, he identified a subset of patients, about one-quarter of patients who are non-responders of IPD-L1 expression prior to BCG therapy and points to those patients as perhaps ones who would benefit from the first-line combination. Woonyoung Choi is a research scientist also at Johns Hopkins. She talked about a transcriptomic analysis of BCG resistance. Started especially with subtypes, but ultimately focused on the fact that in resistant tumors, there are two groups. There are the immune excluded and the immune desert. The immune excluded have immune cells around the edge of the tumor and a very high stromal or fibroblast signature with the idea that the fibroblasts are actively excluding the immune cells.

She talked about targeting this stromal element, and for example, targeting TGF beta as a key regulator versus an immune desert, where there are just no immune cells and that is associated with FGFR3 alterations, although I think the clinical data suggesting that targeting FGFR3 is maybe not going to make much difference in this context.

Then finally, Alex Wyatt, who I know has given a talk on UroToday on specifically these issues previously. He went over some genomic analysis. He's done a very elegant analysis with a group in Basel, Switzerland, and a cohort of patients who underwent BCG and compared BCG responders versus those who recurred and those who progressed. Had just over 90 patients and looked at some of the changes. It's hard to find very distinct patterns. Some things came out. For example, that p53 alterations were more frequent in patients who recurred. Also, ARID1A mutations might be a predictor of recurrence.

So there's a lot more work to be done there. I think especially there's a need for integration across the different platforms of profiling these tissues. Discussion amongst the panel, which included also David McConkey and Rob Svatek, was looking at how we can use our better biological understanding to guide combination therapy because otherwise, we don't really have a good idea of how we should sequence or combine these new treatments. As I alluded to, Woonyoung highlighted at the targeting of TGF beta or FGFR in combination with IO, and of course, there was some talk about how nadofaragene/oportuzumab can both be combined and how the biology does support that. Dave McConkey was highlighting how we really need to work on identifying good biomarkers now with the data sets that we have so that we can validate them in prospective trials, such as SWOG 1602, which is comparing the TICE strain versus the Tokyo strain.  And then the third arm has a Tokyo strain plus vaccine. If we had a candidate marker now, we could validate it in that prospective trial.

Then Colin Dinney and Rob Svatek also spent some time talking about the need for better biomarkers. So the tissue specimens are so small with nonmuscle-invasive disease, especially carcinoma in situ, that it is very hard to get good RNA DNA from tissue, and Colin is especially keen on looking at urine, as is Dave McConkey. There is talk about, for example, cell-free DNA in urine, micro RNA panels in urine, and then looking at the cells in urine with CyTOF or flow cytometry. That would give you a good picture of the tumor and the tumor heterogeneity potentially and it would be a real-time marker that could be used longitudinally as a patient goes through different treatment options.

So just to wrap it up, this is my last slide. It was a very productive discussion of many of the clinical aspects of what is going on with the new developments in BCG-unresponsive nonmuscle invasive disease, but also a very fruitful discussion about some of the laboratory advances. Some of the things we are going to have to figure out moving forward are how best to select patients and treatments accordingly, how best to sequence or combine, how are we going to design the next generation of trials, and what are we going to do to understand the biology better so that we can use that to guide treatment. I will finish there. Thank you.

Ashish Kamat:  Thank you so much, Peter. That was great. We've had the other speakers from the session give expanded individual talks, and it was really nice to hear a summary from your perspective of all the talks put together. If any of our listeners want to get more details, they can clearly go to the other talks, which are full-length features of the individual talks per se.

From an overview perspective, Peter, if I could ask you a couple of questions. Our group and others have really tried to, and gotten the FDA to recognize the entity called BCG-unresponsive bladder cancer and opened up the door for pharmaceutical companies to come into this space. But from your perspective as someone who has done a lot of research and of course, takes care of patients all the time and interacts with them, do you think that this is the right space to be focused on going into the future?

Peter Black:  I think it makes a lot of sense when testing new drugs, especially, I think, especially for systemic immunotherapy.  It made a lot of sense since there are risks associated with it, the immune-related adverse events, and just the systemic toxicity. The patients are faced with major surgery as a standard of care. So I think it makes more sense in that context to take a risk with the systemic toxicity. Also, if you have a more aggressive tumor, it might need that extra boost of immunotherapy.

But I think there is, if we have, and ultimately for some of the drugs, it's just the easiest path to registration. I think we need to recognize that. But for a drug like oportuzumab or for nadofaragene, there is no reason why those could not be very effective earlier on in the disease state, and it doesn't necessarily make sense to be testing it after BCG, other than the practicalities of registration pathway. So those two, for sure, I look forward to seeing how they do in earlier disease states. Of course, immunotherapies are ahead of everybody, and they are already being tested in the first-line in combination with BCG. I do wonder if that is not being a little bit aggressive, but the trials are underway and it will be very exciting to see what the results show.

Ashish Kamat:  From a patient perspective, do you think that there is a need or lack of an impetus to replace BCG?

Peter Black:  Yeah, that is a question that comes up frequently, and I'm always on one side of that. Of course, it is my perspective as a urologist. A lot of people think that we should be trying to get rid of BCG. I think that BCG works very well and we should try and build on it, build on the success of BCG. I think ultimately if patients recognize how effective it is and recognize that the toxicity is very manageable in the vast majority of patients, I think that would probably also reflect the patient's perspective. Of course, there is toxicity, there is a burden of requiring frequent catheterization for installation. If there was something easier that did the same, of course, that is to the patient's advantage. But I think it is a very effective treatment and that is ultimately what the patients want most. So I don't think we should be trying to get rid of it too quickly.

Ashish Kamat:  Yeah. I mean, again, I'm biased, but I agree with you there 100%. Last couple of questions. Let's just assume for the sake of discussion that all three are approved, and you are faced with a patient that fits the definition of BCG-unresponsive disease. Could you walk us or the audience through your thought process in counseling the patient on which drug you would recommend first, second, third?

Peter Black:  Yeah. So it's a good question. I think there are some variables in there that we need to consider. For example, how well a patient tolerates catheterization. I think in your usual patient who is fit for all three, all three are available and there is no reason not to give an IV drug, there is no reason not to pass a catheter for instillation, I would hold on to pembrolizumab as a backup due to the potential for systemic toxicity. I would probably use nadofaragene as first-line because it offers a good balance between the burden of treatment, So it is an instillation once every three months, with good efficacy. I think we have to view the efficacy as relatively equivalent across the three as best as we can tell.

But I think that the burden of treatment is less with nadofaragene, and it is well tolerated. Then I would put oportuzumab somewhere in the middle. I think I would make it a little bit dependent on the stage of the disease. So if a patient has BCG-unresponsive, high-grade T1, I would move pembro up in the sequence. If it is carcinoma in situ and Ta, I would probably use oportuzumab as a second choice, I think.

Ashish Kamat:  Yeah, I kind of agree with you there, even though it is not approved for that particular scenario, I think with the T1 high-grade disease. Worried more about micrometastases than with the Ta and CIS.  So using a drug that has some potential systemic antitumor effect might be beneficial. That sort of mirrors what I would do as well.

Now, coming back, and just as a finishing thought from you as a thought leader in this field, you showed the numbers for the different studies and they are right about, just take 20, 25% as a number to remember, which again puts us at 75, 80% of patients who will not respond to a particular agent. With all the work that you and your group have done in, and again in collaboration with our group and David McConkey and everybody essentially, what is your sense as to whether there ever will be a biomarker or panel biomarkers that will allow us to determine who should get what, and in what sequence?

Peter Black:  Well, we're certainly a long way away from that. I think one of the good things about all these trials with immunotherapy, from metastatic disease back to early disease is that there is always a good component of biomarker research incorporated into those trials. We learn something with every trial, even though there's still a lot to be learned. But with nadofaragene and oportuzumab, we really have no idea currently what patient population might benefit specifically from that.

So if we were going to use a high mutational burden, for example, to steer towards immunotherapy, we have no idea if that's also going to be good for nadofaragene or oportuzumab. I think, yes, I'm always optimistic that we will get there, but it's going to be a long time. I agree with some of the discussion at the meeting about how urine markers may be the way. You have to, as Alex said during the discussion, Alex Wyatt you have to deconvolute to all the information you get out of the urine and actually make sense of it. I think we will figure it out to some degree, but we are going to be making clinical decisions based on poor clinical parameters in the meantime for years.

Ashish Kamat:  So I just want to clarify. That sounded pessimistic, but I know you're actually being optimistic there. Right?

Peter Black:  Its long-term optimism.

Ashish Kamat:  Great. Well, hey Peter, once again, thank you so much for taking the time and always being willing to share your thoughts in research and others with the audience of UroToday. Really, truly appreciate it. Stay well and stay safe.

Peter Black:  My pleasure. Thanks very much for the opportunity.