Read the Full Video Transcript

Ashish Kamat: Hello, and welcome to UroToday's Bladder Cancer Center of Excellence. I'm Ashish Kamat from MD Anderson Cancer Center in Houston, Texas, and it's my distinct pleasure to welcome today, Dr. Roger Li from the Moffitt Cancer Center. Dr. Li is an assistant member of the GU Oncology Department and Assistant Professor in Urologic Oncology, and Dr. Li has been a thought leader in the space of BCG unresponsive disease ever since he was here at MD Anderson several years ago. He's taken that initial spark and done a lot of studies and deep dive into the topic, and it's my pleasure to welcome him here today, to share with you, the audience, his thoughts on the BCG unresponsive disease space and some of the conundrums that we and the FDA face when evaluating and designing clinical trials. With that, Dr. Li, the stage is yours.

Roger Li: Thank you so much, Dr. Kamat, and thanks so much for having me. It is truly a pleasure of mine, distinct honor of mine, to share my thoughts in this stage on a topic that is near and dear to my heart, that you had actually piqued my interest in originally during my fellowship and had guided me through all along. So, thank you very much for your guidance as well.

I'd like to talk to you guys a little bit about the whole entity of BCG unresponsive bladder cancer today, and as you know, this is a relatively new topic. We all see patients in the clinic who BCG has failed, who had recurrent non-muscle-invasive bladder cancer after BCG treatments. And whenever we see some of these patients, there are several questions that we need to consider. Does the timing of the recurrence matter in BCG treatment? As we all know, BCG is a very convoluted treatment protocol. How many courses of BCG has the patient had prior to us claiming that the therapy has failed? And finally, does the grade of the recurrence matter?

We will consider each of those different points, but I just wanted to point everyone's attention to this paper, as well as an earlier paper put together by a consensus group that was assembled at the GU ASCO 5 years ago now, looking at this disease entity and trying to define what exactly BCG unresponsive disease is. And in this paper, by the IBCG, led by Dr. Kamat, the BCG unresponsive disease was, again, defined into two different entities, the refractory disease, which is a persistent high-grade disease at 6 months, versus the relapsing disease in which patients actually have good response at 6 months and then develop recurrence thereafter. There are a lot of caveats, as you imagine, on the number of cycles of BCG that the patients need to have prior to developing disease recurrence, and also the timing of not only the treatment but also the timing of the disease recurrence in relation to their last BCG dose.

So, with that being said, we'll delve a little bit more into the data that is behind some of the criteria that the FDA has been adhering to. This first paper came out of the University of Iowa, in which the authors looked at a clinical trial that was done in the mid-2000s, using a combination of BCG and interferon therapy to treat those patients who have recurrent non-muscle-invasive bladder cancer after they've had BCG in the past. What they found was that the timing of the disease recurrence after BCG made a huge difference. And in those patients with the recurrent papillary disease, that, really, failure within the first 6 months, portended a higher propensity to fail additional BCG interferon treatments in the future. Whereas for those with recurrent CIS, failure within the first 12 months portended a higher rate of failure to combination treatment. And hence, for those patients with recurrent CIS after BCG, it's really thought that their recurrence has to happen within the first 12 months of treatment for them to be deemed to be BCG unresponsive.

What about how many courses before claiming BCG failures? So, from that same paper out of the University of Iowa, the other factor that they found portended to a higher propensity for failure, is patients who are treated with more than two previous induction courses. We also looked at our MD Anderson data, myself and Dr. Kamat, and others in our group, and what we did was we compared patients with high-grade recurrence, either after induction BCG alone, versus induction BCG plus at least an additional course, whether that be a maintenance course or a repeat induction. What we found was very dramatic, in that the recurrence-free survival, progression-free survival, and cystectomy-free survival were all much worse in the patients who were truly BCG unresponsive. In other words, in those patients who had not only induction BCG, but also an additional treatment. So that kind of validated the requirement for the two, either induction courses or induction plus maintenance, before claiming BCG has failed.

The grading of the recurrent tumor is also very important. While the title of this article that we also wrote a couple of years ago says that the low-grade recurrences are not benign, because they are not, there is about a 12% progression rate if you were to have a low-grade recurrence after BCG treatment to muscle-invasive or metastatic disease. But, if you compare the patients who have low-grade recurrences versus those with high-grade occurrences, you do see that there is a dramatic difference, again, in recurrence-free survival, progression-free survival, as well as cystectomy-free survival.

Finally, in addition to all of those different parameters, we've also put out a recent statement, just from our experiences accruing patients to the ongoing BCG unresponsive trials. Several specific clinical scenarios in which the patients, while they do not exactly fit the BCG unresponsive criteria, but nevertheless, in our opinions, they still are unresponsive to treatments. And this pertains, first of all, to the timing of the course of BCG. So, in the original IBCG paper, it was actually specified that the two courses of BCG treatment have to happen within a 6 to the 9-month window, and we also advise that clinical trialists who are accruing patients to the BCG unresponsive trials adhere to this timeline.

The second issue has to do with the global shortage of BCG, and so in response to that, a lot of the urologists, not only in the United States but also in other parts of the world, are reducing the dose of the BCG down to one-third. And, as we all know, there are EORTC trials in the past that have shown somewhat equivalent efficacy of the one-third dosing to the full dose in some patients with non-muscle invasive bladder cancer.

Another mantra that Dr. Kamat and others had started was, "Once BCG unresponsive, always BCG unresponsive." And this pertains to the patients who were deemed to be BCG unresponsive, then were started on some additional treatments, whether it be a clinical trial or intravesical chemo, and because of that treatment, the salvage treatment, fall out of the window that was specified by the FDA to call it the BCG unresponsive disease. However, if you look at the disease patterns of progression and recurrence in these patients, because they were deemed to be BCG unresponsive once, they are unlikely to respond to additional BCG even after, even if they were to fall out of a strict window that's specified by the FDA and other governing bodies.

And finally, because of the SWOG 1602 trial that is ongoing in the United States, the Tokyo strain BCG has been used throughout the United States, and of course, in other parts of the world.  In Europe and other countries, there are different BCG strains that are used. But we did some preliminary scans through the literature, and we think that at least the Tokyo BCG strain should be deemed to be as efficacious as the TICE strain. And so for patients who are treated with a Tokyo BCG and fail after that, they should be deemed to be BCG unresponsive as well.

So just a couple of minutes about the treatment and the clinical trial endpoints for BCG unresponsive disease. As you know, radical cystectomy and pelvic lymph node dissection remain the gold standard for BCG unresponsive disease, but given that this disease typically happens in an older and frail population, a lot of the patients are not able to tolerate this procedure with the very high morbidity that it contains, and hence the need for conservative treatment strategies. We knew this.

These were the criteria that were deemed appropriate by the IBCG and the original statement paper, again, where for BCG unresponsive CIS, it was thought that clinically a reasonable mark to hit for the 6-month point was 50% complete response, and at 12 months, for that to be at around 30% durable response. Of course, we know that some of the clinical trials that have reported out since then, including KEYNOTE-057 and the nadofaragene trial, did not quite hit those marks. But, compared to historical data in which we did a systemic review for all of the therapeutic trials that were done after patients that received BCG, and as you can see here, in the CIS-containing tumors that were found after BCG treatments, the median 12-month complete response rates were about 17%.

So, I think that's a reasonable margin for us to deem as a threshold to surpass. And, indeed, that has been surpassed in the KEYNOTE-057 and the nadofaragene trial. As we all know, pembrolizumab was approved as an agent for BCG unresponsive CIS, as a result of KEYNOTE-057, and we are still waiting on the FDA's adjudication on the efficacy for the nadofaragene trial. So with that, I'd like to thank you for your attention, and I'm happy to take any questions.

Ashish Kamat: That was great, Roger. Thanks for taking us through a quick and succinct summary of the whole BCG unresponsive disease state. Let me just ask you a couple of questions. First off, I just want to clarify for the sake of our audience, that when you said that the Tokyo strain of BCG would be acceptable, I presume you are basing that on the potential results of the SWOG study. That if it shows that it is the same as TICE, then it should be allowed in the United States. Is that correct?

Roger Li: Absolutely. Absolutely. So, ultimately, we will have to take a look at the results of the SWOG trial, but there has been some retrospective data in that realm, where the Tokyo strain has been compared to the Connaught, as you know, where it showed equivalent efficacy. And as we all know, that the Connaught actually is even better than TICE in some of the retrospective studies done in the past.

Ashish Kamat: Right. Right, right. It's kind of ironic because there is plenty of BCG in the rest of the world because it's used much more commonly as a vaccine against tuberculosis, and of course, many countries give it to newborns routinely. So there is plenty of BCG in the world. And if the SWOG study allows us to get the Tokyo strain, for example, in the United States, it would be great if it kept other strains, including the recombinant BCG that is being currently fast-tracked through the Canadian arm, essentially. So that would be great.

The other question I wanted to ask you was about the paper that you led on the different agents and the response rates. So, given what you found from the existing literature and the trials that were right out to date at the time of the publication, what would you say would be a realistic benchmark for people to design clinical trials and power their studies around?

Roger Li: That's a great question, Ashish. As you know, that kind of has been a floating target out there that nobody truly can pin down. It's very tough to say, because as you know, a lot of the studies done in the past, did not require patients to be BCG unresponsive, and hence, the accrual of the patients truly happened in a very heterogeneous population. It's very tough to compare trial against trial given that heterogeneity in the patient population. And also, the follow-up of those studies was also very variable. Some studies only had follow-ups for 6 months, some for about a year or so. And the definition of recurrence, some trials actually beam low-grade disease recurrences to be an event when it's after the treatment.

So, it was really tough for me to kind of decipher exactly what target or what the true average of all of the clinical trials that have been done in the past was, and I think we will have more data coming out as more of the BCG unresponsive trials come to fruition, as we already have the KEYNOTE and the nadofaragene, but it would be very interesting for us to conduct a meta-analysis, actually, of some of the clinical trials that are coming out because the study population is now much more homogeneous because of the BCG unresponsive definition, and also because of the parameters that have been set by the FDA around what institutes a recurrence event after additional salvage treatment.

Ashish Kamat: Yeah. There are so many questions I could ask you about this topic, but since we are sort of focused on the FDA conundrums, let me post one. With all the data that's coming out with the single-arm registration studies, and it appears like the 12-month CR rates for CIS patients are around 20%, 24%, so 20% plus or minus a little bit here and there. If you were to recommend to the FDA, would you recommend that they still allow trials, new trials, I'm not talking about existing trials that are already ongoing, but new trials to come in as single-arm trials? Or would you be recommending that one or all of these agents be considered as a control group?

Roger Li: That's also a great question. I think there's one camp where urologists and medical oncologists, we know that pembro has been approved and one to actually use the efficacy data behind pembro as the comparator arm. But, truth be told, I think we still don't understand that much about this disease entity, and the other factor to consider is also the quality of life that you buy the patients for delaying their cystectomy and what the benefits of just preventing progression is. As you know, there have been some studies that came out from the KEYNOTE-057 follow-up data that show that even though the intravesical recurrence rates were there, there is a relatively low progression rate.

So, those questions still need to be answered, and I think until we have those answers, I would say single-arm studies in this space, just because randomized control trials in this space are so hard, as you know, and patients are so topped to combine. Hence, we have so much discussion around the parameters that we have to set before accruing patients to these trials. So I think in these settings and fully keeping in mind the difficulties of accruing to these trials, I think single-arm studies should still be accepted.

Ashish Kamat: So Roger, once again, I want to thank you for taking the time and spending these 20 minutes with our audience. In closing, I want to leave you with a minute or so to share with us what you think are your closing thoughts on this topic.

Roger Li: I think we are really at the dawn of the BCG unresponsive age, where there are going to be a lot of new therapeutic agents that are coming online. And it's very exciting times, both for us as clinicians, and also for our patients, because all of us know how morbid radical cystectomy is and how much of a life changer it is. And so for us to be able to offer patients new options in retaining their bladders, but also to possibly prevent recurrence at a significant rate, is very, very exciting. I'm looking very much to the bright future ahead, and I look forward to working with you, Ashish, and others in the field to continue to push the field forward and continue to discover new therapeutics.

Ashish Kamat: It would be my pleasure, Roger. Once again, thank you for taking the time and stay safe and stay well.

Roger Li: You too.