Read the Full Video Transcript

Ashish Kamat: Hello, everyone, and welcome to your today's Bladder Cancer Center of Excellence. I'm Ashish Kamat, urology oncologist at MD Anderson Cancer Center, and I'm delighted to be joined today by Dr. Jacob Taylor, who is currently at Urology Clinics of North Texas in Dallas, Texas. So, Jacob, thank you so much for taking the time today and joining us to share with us and our audience the work that you did, along with our collaborative efforts with Dr. Lotan in the international bladder cancer group in this multicenter database, and really looking forward to what you have to share with us as far as your insights into the long-term outcomes of bladder-sparing therapies compared to upfront radical cystectomy in these patients.

Jacob Taylor: Thank you for that introduction. And yes, it's my pleasure to be here with you today. This was part of my fellowship work last year at UT Southwestern, where, as you said, we were looking at long-term outcomes of bladder-sparing therapy compared to upfront surgery in patients with BCG-unresponsive bladder cancer. So we'll jump into it here.

A lot of authors on the paper, so a few conflicts. Mine are at the top here. A brief outline of my talk will be a brief introduction on this group, our methods, and endpoints that we looked at. We'll briefly look at the results and then some conclusions and takeaways.

So as many of us who work in this space know, up to 40% of patients with high-risk non-muscle invasive bladder cancer recur after adequate intravesical BCG therapy. However, there's limited data on oncologic outcomes in this patient cohort in a real-world setting. Most of the data comes from single institutions. Most of the data is retrospective. So we sought to add to this literature gap by forming a collaborative from 10 institutions around the US, Canada, and France, looking at outcomes in patients with BCG-unresponsive bladder cancer.

And specifically, our primary endpoint was overall survival in patients undergoing upfront cystectomy compared to bladder-sparing therapy, with our other secondary oncologic outcomes including rates of recurrence, progression, metastasis, cancer-specific, cystectomy-free survival. So as I said, we included 10 institutions, mostly academic centers, between 2005 and 2022 for patients with BCG-unresponsive bladder cancer.

And BCG-unresponsive disease was defined by FDA criteria, so this included any patient with high-grade T1 on the first assessment following induction therapy; any high-grade non-muscle invasive bladder cancer within six months of adequate BCG defined by the 5 plus 2 definition; and then finally, any CIS with or without papillary disease within 12 months of adequate BCG therapy. We then categorized patients into those undergoing upfront surgery and compared them to patients who underwent bladder-sparing therapy.

Some brief characteristics of our cohort: We included 578 patients, about a quarter of whom underwent upfront radical cystectomy, and 72% (416 patients) underwent bladder-sparing therapy. Seventy percent of these patients were between 2013 and 2020, so that was before the approval of pembrolizumab and nadofaragene. The median follow-up was fairly good at 15 months in the whole cohort—16 months in those undergoing upfront surgery, and 47 months for those undergoing bladder-sparing therapy.

On average, patients were about two years younger if they underwent upfront surgery compared to patients with bladder-sparing therapy. There was no difference in sex, BMI, or comorbidity status. And just a brief breakdown on the right here of the institutions that were involved in this collaborative—we're appreciative that they shared their patients with us.

Some brief characteristics, again, between the two groups, mainly highlighting that patients who underwent upfront surgery were more likely to have invasive disease—so more likely to have T1 disease, both at the primary disease point before BCG therapy, as well as at the unresponsive disease classification point after BCG. They were more likely to undergo upfront surgery, which makes sense. Clinical stage, again, whether it was Ta or CIS, was more likely to be higher in the upfront surgery group.

So this table outlines the first-line bladder-sparing therapies for patients that underwent bladder-sparing therapy. As part of this follow-up, we tracked patients that had 2, 3, 4, 5 different lines of salvage bladder-sparing therapy, but this is the first line. So in the 416 patients, a fair number—38%—underwent continued BCG, with an additional 36% of patients undergoing an alternative intravesical agent. This included gemcitabine, docetaxel, single-agent chemotherapy, and occasionally a clinical trial, with about 7% (29 patients) undergoing an immune checkpoint inhibitor (pembrolizumab or durvalumab on clinical trial).

Our primary outcomes are in these graphs here, and on the top left, you can see that although not statistically significant, there was a trend for improved survival in patients undergoing upfront surgery compared to bladder-sparing therapy. But again, this wasn't statistically significant, although you do see that the lines are separate. But this is far out after a median follow-up—closer to 6, 8, 9 years—they do separate out but not a statistically significant difference between the two groups.

In regards to metastasis and cancer-specific survival, the lines are pretty equivalent and show no statistically significant difference whether patients underwent upfront surgery versus bladder-sparing therapy.

So this table shows our Kaplan-Meier estimates for patients undergoing bladder-sparing therapy and their oncologic outcomes over time. I want to highlight a few points here at the bottom, boxed in red. Importantly, for patients that underwent bladder-sparing therapy, oncologic outcomes were fairly good at 12 months. There was a low rate of cystectomy at 12%, and low rates of metastasis and death from bladder cancer at 2% and 1%, respectively.

But as you go out to 24 months and five years (60 months), rates do begin to rise. About 40% of patients who underwent some bladder-sparing therapy ultimately got cystectomy by five years after the BCG-unresponsive diagnosis, with a 14% metastasis rate, 14% rate of death from bladder cancer, and roughly a 30% all-cause mortality rate. That's important, and we'll get back to why it's important a little later.

In regards to pathologic outcomes, there was no statistically significant difference in regards to stage, and these are patients that both underwent upfront cystectomy or underwent cystectomy after either one or two lines of bladder-sparing therapy. However, I will highlight there was a trend—this was statistically significant—toward worsening nodal status in patients undergoing cystectomy after bladder-sparing therapy. Why this is is not entirely clear. Some of it may be due to selection bias, some of it may be due to just low numbers, or some of it may be inherently due to the fact that if we're delaying cystectomy in these patients, we're putting them at risk for a higher rate of node positivity and invasive disease.

So in conclusion, there's no statistically significant difference in overall survival, cancer-specific, or metastasis-free survival between patients undergoing bladder-sparing therapy and upfront cystectomy. But as I mentioned before, we need close vigilance in this patient group, as patients with BCG-unresponsive disease who delay cystectomy and undergo bladder-sparing therapy have a high risk of recurrence and progression over time, even though those rates within the first year or so are low.

However, it's important to note that approximately 32% of patients who do undergo bladder-sparing therapy will eventually undergo cystectomy, and these patients again have higher rates of node positivity and increasing rates of non-organ-confined disease. So further prospective evaluation is needed and critical, especially with the rapidly growing choices from a clinical trial perspective as well as an approval perspective of different agents that are now available to patients. A prospective collaboration is underway with partners we worked on this project with.

So I'm deeply grateful for including yourself and your group down there at MD Anderson and many other collaborators which made this project possible. Thank you for that, and definitely curious about your thoughts, as you are one of the leaders in the field.

Ashish Kamat: Thanks so much, Jacob, for presenting that paper. I think it's very important to recognize—when we first proposed the definition at the AUA and then, of course, GU ASCO and the AUA collaborative with the FDA, et cetera, everyone was skeptical: are these patients going to have a good outcome or not? In the clinical trial field, it clearly exploded, and now we have multiple drugs available—off-label use of Gem/Doce, as you showed.

And it's very, very important for people who take care of patients with BCG-unresponsive disease to recognize that the definition for a single-arm study was purely for clinical trials. It was never meant to inform the clinical decision-making, but it does inform the clinical decision-making because people have hung on that definition. So data such as what you presented is reassuring in some ways, if you recognize that in very highly selected patients—because, as you know, the database is retrospective, and we all select patients.

So the patients that we selected for bladder-sparing therapy, even though you can do comparisons, you can't quantify the intangibles. But in a very selected population, it appears that bladder-sparing therapy is not as dangerous as might have been thought in the past. It's not a replacement for radical cystectomy, but it's not as dangerous, so long as you follow certain criteria.

And I want to bring you back to the point that you made—and I'm glad you made that—with increasing lines of therapy, you said there is a perceived increased risk of the patient obviously losing their bladder because they're recurrent, but also having metastatic disease. Could you share with us a little bit more of your insights? Because I know you did a lot of analyses based on your analyses and your knowing the data that you presented. Do you have—and if you do, what is that number, a cutoff, where you say, "OK, after so many lines of therapy, it appears that it might be unsafe to try more bladder-sparing therapy"?

Jacob Taylor: So that's a good question, and I totally echo your statements on this as a highly select patient population. In regards to what the cutoff is, we don't know for sure. But I think some of it comes down to tumor stage. We did some other work with other data that I think is being presented shortly, in regards to the importance of tumor stage.

So if you're a Ta patient that recurs after one line of salvage bladder-sparing therapy, I think your risk of metastasis is for sure lower than if you recur T1 after salvage Gem/Doce as a second-line therapy. So some of it depends on that. I think we know that the risk of metastasis in general with T1 plus or minus CIS is higher than patients with pure CIS or patients with pure Ta. So I think that's important to note.

In our data, we did find that patients who underwent three or more bladder-sparing therapies had a risk of death over time—at about four years—of 28%, compared to a risk of death of about 14% for the entire cohort of bladder-sparing therapy. So whether that cutoff is two, we can't do two and a half—we're not really sure. I think the importance is, giving one bladder-sparing therapy is probably safe in the short term, as long as you follow them closely and have good imaging.

And depending on what happens at that recurrence after the first bladder-sparing therapy, then you make the decision whether you continue with a second line of bladder-sparing therapy.

Ashish Kamat: I'm glad you made that point because each time the patient has a recurrence and you're considering radical cystectomy versus a bladder-sparing option, it's up to us. And this is something that we've emphasized a fair amount in the BCG consensus document that just came out. You have to restage these patients thoroughly. You don't want to be missing occult T1 disease. You clearly don't want to be missing occult metastatic disease or muscle-invasive disease. And these patients have to be really restaged really well.

And if it appears that it's safe or relatively safe—Ta disease or CIS, but not T1b for sure, or even T1—then it might be safe to try round number two or number three, but it clearly is shared decision-making. So I'm glad you emphasized those points. And again, in the interest of time, we'll stop now. But thank you so much for taking the time and spending it with us.

Jacob Taylor: My pleasure. Thank you so much.