Read the Full Video Transcript

Ashish Kamat: Hello, everybody, and welcome to UroToday's bladder Cancer Center of Excellence. I'm Ashish Kamat, urologic oncologist at M.D. Anderson Cancer Center. And it's a distinct pleasure to welcome once again, Professor Yair Lotan. He's a urologic oncologist at UT Southwestern in Dallas.

Yair, thank you so much for taking the time today. We're really looking forward to hearing the work that you've done, and then gaining some pearls of wisdom on this whole BCG-unresponsive space that you've dedicated a lot of time and effort into. So looking forward to it.

Yair Lotan: Great. Thank you so much. I appreciate the opportunity to discuss these papers. First of all, I want to thank all my collaborators, because this was a large, multicenter, retrospective effort. And the paper that we're going to discuss right now is one of several papers that were published from this collaboration.

And this paper focused on oncologic outcomes of sequential intravesical gemcitabine and docetaxel, compared to BCG in patients with BCG-unresponsive non-muscle invasive bladder cancer.

These are my disclosures.

So we all recognize that the guidelines currently recommend radical cystectomy as the highest chance of cure for patients with BCG-unresponsive non-muscle invasive bladder cancer. But there are several other potential treatments that are approved, including pembrolizumab, nadofaragene, and Anktiva, as well as clinical trials and other treatments such as gemcitabine and docetaxel in patients who decline radical cystectomy.

And we all know that radical cystectomy has lifelong quality of life implications, whether or not they impact urinary function, sexual function, potentially bowel function. And it is a complicated surgery with a long recovery. So many patients who have noninvasive disease decline cystectomy and prefer a bladder-sparing approach.

Now, depending on therapy availability and provider practice, some patients don't have access to some of the newer agents and may receive additional BCG after they are deemed unresponsive. There's also an option to get intravesical gemcitabine and docetaxel, which has shown promising one- and two-year high-grade recurrence-free survival.

So the purpose of this study was to compare oncologic outcomes in patients with BCG-unresponsive disease who either received additional BCG or intravesical gemcitabine and docetaxel.

This was a study that incorporated a large cohort of patients over a 20-year stretch. And so a lot of the patients received treatment before any of these newer agents were available. This is why many of the patients were those who received BCG. And then there were patients who received gemcitabine/docetaxel, which tended to be more recent.

The median follow-up was actually quite long—47 months for the BCG group and 39 months for the gem/doce group—and of 299 patients, 68% received additional BCG, and about a third received gemcitabine and docetaxel. There was a lot of similarity in terms of age and gender, but we did note that there was a higher percentage of patients with T1 disease who received additional BCG—about 46% versus 31% for gem/doce—and it's an important consideration when looking at the data.

So I think the first thing you can look at is just high-grade recurrence-free survival. And you can see that even though there appears to be a benefit for gem/doce over BCG, there is overlap in the curves and the p-value is 0.17. So we didn't find a significant difference between the two. But when you look at progression-free survival, there was clearly a benefit for gem/doce compared to BCG. And granted, these are associations—these are not randomized trials—but still, you can see a separation of the curves that definitely favors gem/doce in this data set.

When we look at cancer-specific survival, again, similar to progression-free survival, the group with gem/doce had a better cancer-specific survival, which was statistically significant. And the overall survival was close but not statistically significant, with the p-value of 0.058. But again, it's possible that with more patients, you would see that become clinically significant.

So we did a multivariate analysis to try to control for stage. As we noted, there was a difference in stage between the two groups. And what you see is that both getting BCG and the stage were independent predictors of progression, but in opposite directions. So BCG conferred a 2.8 or 280% higher risk of progression, and having Ta or CIS compared to T1 was about a 50% lower risk of progression.

We also found that patients treated with BCG were more likely to undergo cystectomy, even though we recognize that there's some subjectivity in terms of what additional options were available. Some of these were older patients where there were no other available drugs other than to remove the bladder.

We did two sensitivity analyses to try to compare the groups a little bit more accurately. And since gem/doce was rarely used prior to 2013, we analyzed patients who received salvage therapy after 2013. That included about 70% of the BCG patients and 96% of the gem/doce patients. And the gem/doce group still had improved progression-free survival, metastasis-free survival, cystectomy-free survival, and cancer-specific survival, and there was a trend to improved overall survival. So when you compare most of the patients in treatment, you see a similar trend.

We also did a propensity-weighted analysis. And again, progression-free survival and cancer-specific survival were worse in the BCG group compared to gem/doce. There were no differences, though, in cystectomy and overall survival, as well as in high-grade recurrence-free survival, so fairly similar outcomes.

So in summary, the intravesical gem/doce was associated with lower rates of cystectomy, progression, and cancer-specific mortality relative to further BCG therapy. And clinicians should carefully consider gem/doce as an option when treating patients with BCG-unresponsive disease. Thank you.

Ashish Kamat: Thanks so much, Yair. The way I look at this data—and again, kudos to you for putting together this coalition. I remember, you brought it up at one of the IBCG meetings, and then you ran with it. And now it's been this treasure trove of information, which really helps to inform people's questions.

The way I look at the data, essentially, in the BCG-unresponsive population, gem/doce has an advantage over BCG. I mean, you can look at that, and that's fairly clear. And as you said, even when you did the multivariable analysis and you accounted for other factors, gem/doce still has an advantage.

What's striking though, and I'd love to hear your thoughts on this, is that in both treatment arms—the BCG arm and the gem/doce arm—the results were actually better than you would expect for a BCG-unresponsive defined population based on the clinical trial data that we're seeing with the newer agents that you mentioned. Any thoughts into this, for example, low progression rate that we're seeing in patients in this cohort, compared to the reported clinical trial data? I mean, essentially, it looks the same as these expensive drugs.

Yair Lotan: Yeah, so this is one of the challenges with interpretation of retrospective data. Historically, a lot of people had looked at old Memorial data and some other series, and the data suggested that only about 20% of patients who were BCG-unresponsive would respond to more BCG.

And here, you're looking at these data sets, and even after four or five years, you have about almost half the patients still are recurrence-free, and about 75% are progression-free. And you ask yourself, why aren't we just giving more BCG to some of these patients, if we could select them?

There are a couple of problems with retrospective data. Some of these are patients from 10-15 years ago, and if you had a patient who you had no other treatment to give, and yet they refused cystectomy, would you be inclined to biopsy them every three to six months if you saw a red lesion? What would be the purpose?

They weren't going to let you take out the bladder. Maybe they were asymptomatic. They're not bleeding. We know a lot of patients with CIS are not particularly symptomatic. Maybe they have some irritative symptoms. But there was no protocol that had mandatory biopsies or said, "You really need to treat these patients." And if the patient wasn't going to accept the only other treatment you had, which was cystectomy, maybe you just weren't inclined to biopsy them.

And so it appeared like they were recurrence-free when, in fact, maybe they still had persistent disease. Less likely for T1 disease, as we noted—and we'll talk about in a moment with the STAGE paper in a separate analysis—but the fact is that you have to take this data with a grain of salt. I think what you can look at more accurately is, did they die of cancer? And the truth is, they don't die of cancer, so you can't hide that. You can hide a recurrence, and maybe a progression if you don't biopsy the patient, but if they died of disease, it's really hard to say they didn't die.

And the fact is that it's still rare for those patients to die of disease, even if they didn't get additional treatments, and maybe that's the only one that I could say, this is much stronger data. But you're absolutely right—right now, we're in a situation where we have a lot of single-arm trials. They're not randomized against more BCG. They're not randomized against gem/doce. And so we have to try to take into consideration all the factors that go into the prospective data, compare it to the retrospective data, and then decide whether or not we do need additional trials randomizing against other treatments.

Unfortunately, we don't have a standard of care, and the FDA currently is not requiring that. But maybe in the future, there will be room for some randomized trials to try to really compare two different treatment approaches.

Ashish Kamat: Yeah, I think that's an important point that you bring up, right? I mean, you can't hide from death, and the cancer-specific survival graph clearly speaks for itself. And you know this, of course—we've published our M.D. Anderson highly selected group of patients treated with more BCG after being BCG-unresponsive, and if you select patients correctly, we get about an 85% CR rate, and it's durable for up to five years. But it's a very highly selected group of patients.

And I want to just emphasize for the authors, I don't think you're putting out this message—and I certainly don't want to put out the message—that we should be offering more BCG to patients when they become BCG-unresponsive. We're just showing the data and saying, "Hey, this is the response that we saw when we looked at the retrospective series." But please, enroll patients in clinical trials. Please support all the development and everything that's ongoing at the moment.

Thank you, Yair, for sharing your time and your expertise with us on this paper. And like you said, we'll discuss a little bit more in the next video that you'll see on UroToday's Cancer Center of Excellence.

Yair Lotan: Good. Thank you so much.