Pembrolizumab's Impact on Cystectomy-Free Survival in BCG Unresponsive Bladder Cancer from the KEYNOTE-057 Trial - Girish Kulkarni

May 19, 2023

In an informative discussion, Ashish Kamat welcomes Girish Kulkarni to discuss his work around cystectomy-free survival data and the impact of pembrolizumab monotherapy for BCG-unresponsive CIS, as presented in the KEYNOTE-057 trial. Kulkarni highlights the promising results from the trial, stating that patients who responded to the therapy had a median cystectomy-free survival time of 58 months, while non-responders' median time was at 18 months. They also highlight that only five out of 43 patients who underwent cystectomy were upstaged. Kulkarni discusses the timing of treatment to cystectomy and how it affects the upstaging rate. Further discussion revolves around the future of treatment sequencing, the role of molecular testing, and how to educate patients on the importance of timely cystectomy to prevent metastatic disease.


Girish Kulkarni, MD, Ph.D., FRCSC, Urologic Surgeon, Department of Surgery, University of Toronto

Ashish Kamat, MD, MBBS, Professor, Department of Urology, Division of Surgery, University of Texas MD Anderson Cancer Center, President, International Bladder Cancer Group (IBCG), Houston, Texas

Read the Full Video Transcript

Ashish Kamat: Welcome again, Girish, and thank you for taking time off your busy AUA schedule and spending with us here at UroToday. You have several things going on at the AUA, and one of the things is to present the cystectomy-free survival data and the implications of pembrolizumab in the KEYNOTE-057 trial. So tell us what you're going to be talking about.

Girish Kulkarni: For sure.

I guess just as a primer, I think many people in the audience do know about the KEYNOTE-057 trial of monotherapy pembrolizumab for BCG-unresponsive CIS has FDA approval, has Health Canada approval, as well. So it's approved within North America. The data have been published.

So what we are going to present at the AUA this year, what I'm presenting, is the cystectomy-free survival. So, in the cohort A, that's the CIS patient population, there were 96 patients enrolled, 43 eventually went on to cystectomy. We've looked at the cystectomy-free survival in those who responded and those who didn't. So in those who responded, the cystectomy-free survival time, the median, was 58 months, which is excellent. Non-responders, as one would expect, was lower at 18 months.

I think the most important part of the poster and the subgroup analysis that we've done in the 43 patients is the fact that we only have five patients that were upstaged of the 43 that had cystectomy. That's in keeping with what we see in all of our non-muscle invasive cystectomy literature. So I think we are not harming patients by trying one round at least of therapy in the BCG-unresponsive setting.

Ashish Kamat: Yeah, I think that's a very important message. Because obviously when this trial was first initiated, and it was one of the earlier trials that used the current FDA recommendation for the single-arm study, a lot of people, including investigators were worried that, "Are we harming patients?" And this is very reassuring data for patients that you're not hurting yourself if you try one round of pembrolizumab.

Now, based on the data that you're presenting, any insight into the timing, from the time of the treatment to cystectomy, and how that affects the upstaging rate?

Girish Kulkarni: Oh, absolutely.

So we do have a table demonstrating that if you have a short time to cystectomy, then the risk of upstaging and upstaging locally or even nodal disease is actually quite low. The patients who had T3 or T4 on their path staging or node-positive actually were the ones who were the delayers, the refusers of cystectomy. The enrollment in the trial was either patient refusal for cystectomy or ineligible for cystectomy. So the people who refused cystectomy upfront tended to do that after, as well. So we have a few patients where their median was 12 months or even up to 30 months to cystectomy, and all of those had very advanced path findings. So we can do one treatment for BCG-unresponsive, I think that's demonstrated well in this study. Can you do two? I don't know if we have that information yet.

But clearly the time to cystectomy is an important variable, which we would assume was the case based on the biology of urothelial carcinoma.

Ashish Kamat: Yeah. I think what this data and other emerging data along these lines suggests is that it's up to us, really, to continuously educate our patients on, "Yes, there is a window in which we can quickly try to save the bladder, but beyond that, you're actually risking your life because you can get metastatic disease."

A few general thoughts from you on not just this specifically, but with all the work that you're doing in this area, how do you envision the appropriate sequencing of different therapies coming about when let's say we have four drugs approved. Do you think molecular testing will be useful? Do you think just clinical data? Just brief thoughts there.

Girish Kulkarni: Yeah, if there are biomarkers that can help us select if a patient would respond to an immune checkpoint inhibitor versus gene therapy or an oncolytic virus, that would be great, or even our photodynamic therapy. Right now, we don't have those data, obviously, but all the phase two trials are collecting urine, tissue. So hopefully those types of correlative studies will come to fruition.

In terms of sequencing, it's a tough question, as well. Can we actually add another therapy or two? I get worried, and I'm the same as yourself, where if a patient fails once I'm really on them to consider a cystectomy, especially if they're eligible. Most of the patients in KEYNOTE-057, were ECOG 0, so they were eligible, they're just refusing. So I really try to stress to them that we have a window of opportunity for cure, and we may lose that.

As to which item or which drug or therapy we're going to be sequencing, I think that's future trials. It's not that different than the prostate cancer landscape where we're going from single agents to multiple agents, dual therapy as opposed to monotherapy and trying to figure out timing.

Ashish Kamat: Yeah, it's a great problem to have, right?

Girish Kulkarni: Yeah, yeah.

Ashish Kamat: So once again, thank you for taking the time and spending it with us today.

Girish Kulkarni: Thank you, pleasure.