Urinary Minimal Residual Disease Detection Predicts Recurrence in BCG-Unresponsive NIMBC and Quantifies Molecular Response to Nadofaragene Firadenovec - Vikram Narayan

February 20, 2024

Sam Chang hosts Vikram Narayan for a discussion on the potential of urinary markers in predicting responses to nadofaragene, a treatment for BCG-unresponsive bladder cancer. Dr. Narayan discusses findings from a study utilizing the UroAmp test for assessing urinary minimal residual disease (MRD) in patients treated with nadofaragene. This next-generation sequencing approach reveals the mutation profiles in urine-derived DNA before and after treatment, offering a new method for identifying early responders and potentially guiding treatment decisions. Their conversation highlights the evolving landscape of bladder cancer treatment, emphasizing the importance of precision medicine in improving patient outcomes.


Vikram Narayan, MD, Urologist, Emory University School of Medicine, Grady Memorial Hospital, Atlanta, GA

Sam S. Chang, MD, MBA, Urologist, Vanderbilt University Medical Center, Nashville, TN

Read the Full Video Transcript

Sam Chang: Hi, my name is Sam Chang. I'm a urologic surgeon in Nashville, Tennessee, at Vanderbilt University. And I have the pleasure of discussing many topics when it comes to urologic cancers.

But to be able to have a future star in Vikram Narayan who is at Emory University is a real pleasure and honor for me. Dr. Narayan is an assistant professor at Emory, and recently presented, and was the first author of an abstract at ASCO GU 2024 that focused on actually the use of urinary markers, and perhaps its predictive capability when it comes to the use of nadofaragene.

So Vikram, thank you so much for spending some time with us, and look forward to hearing some of the work that you did in terms of this presentation.

Vikram Narayan: Thanks a lot, Sam, for that gracious introduction and for the opportunity to share some of this work. So yeah, as you said, what we wanted to do... First of all, I want to give a shout-out to the SUO Clinical Trials Consortium, the SUO-CTC, which designed this study originally.

And I think, as the audience hopefully knows, led to the phase three study that was published a couple of years ago now, and ultimately led to the FDA approval of nadofaragene firadenovec, which is a mouthful to say.

But it remains, at the moment, the first intravesical gene therapy that's available for patients with BCG unresponsive disease. So what we did in this particular project was... As folks are aware, nadofaragene offers about a 53%, 54% complete response rate, at least in the phase three data by around three months.

But if you look at all comers in the study out to a year, the durability was only around 24%, 25%. And obviously, that's been a stubborn number across a number of different agents that have been looked at in the space.

And one of the questions that we have, of course, as we're trying to think about how best to take care of patients with this difficult disease, is how do we potentially identify responders early, people who are doing well with therapy so we can pivot to other treatments, etc.

And so one of the tools that has started to emerge and was of interest to us is this concept of urinary minimal residual disease, which essentially uses next-generation sequencing to look at mutations in the urine.

So we're actually looking... The assay is made by a company called Convergent Genomics. It's called the UroAmp test. And what it does is it basically looks at somatic tumor mutations in urine-derived DNA.

And there's been a lot... Historically, there have been a lot of urinary tests. What's different about this is it actually... Although folks often wonder how reliable are urinary tests, because there's a lot of environmental factors, in terms of hydration or patient statuses...

You can actually see the ratios of the mutations conserved within the same patient. And so it makes it an interesting test to use. And so what we did was this particular study looked at the phase two nadofaragene trial, so it's a smaller cohort of patients.

And what we did was these patients on trial had urine collected before they received nadofaragene and at their three-month evaluation. So we had two time points, pre- and post-treatment. We had urine from each of those.

We essentially ran the UroAmp assay, did DNA extraction, next-gen sequencing, and then we were able to deliver what UroAmp calls their MRD or minimal residual disease status.

So in terms of actual numbers, again, numbers are very small. In this particular dataset, we had about 35 patients with valuable data that we were able to run this test on.

And what we found, the first take-home message was that when you look at the mutation profiles in pretreatment urine, you actually found very variable profiles.

As one might expect, bladder cancer is a very heterogeneous entity. And these are BCG unresponsive patients, so many of them were pretreated with several rounds of BCG.

But interesting data, I think this figure here, top left, what it shows you... I'll just try and walk you through because it's a little bit busy, is you can see recurrent status of the patient, depending on whether it's gray or green.

And you can see the total mutations increase, more likely you are to have recurrence. And the UroAmp status is also listed here. One of the things we looked at was, can we predict recurrence-free survival in these patients using urine at the pre-treatment time point, and then also at the post-treatment time point?

So you can actually see that the curves separate out. This MRD status here is an algorithm that essentially looks at... It's enriched for several different mutations. And basically, we actually had the company work with us to develop a BCG unresponsive algorithm. So we looked at that as well.

And you can see that there is relatively good discrimination between those who were positive for MRD and who were negative for MRD. But actually, the most interesting stuff was post-induction.

So if you look at the patients who were after their induction treatment at their three-month check, recurrence-free survival was 100% for MRD negative patients in this, again, small cohort, which really is interesting.

Because it suggests that, one, patients who are responders... They have fewer mutations that are being picked up in their urine, and you could potentially use that to actually stratify patients who are likely to respond down the line.

And that may be true, I should say, Sam. Not just for nadofaragene, but perhaps for other treatments as well. So we obviously looked specifically at nadofaragene. But one of the things that we found, that was most interesting in this study, is you can actually classify patients by their MRD status itself.

So if you were to look at, for example, the types of mutations they had pre versus post, you can see that some of them had not only a reduction in the types of mutations but also the absence of those. And these we would call complete responders.

And then in other patients, you would have partial responders where there were still some mutations present, but then you would see a reduction in it.

And then there were some patients where the mutations actually were very similar pre and post urine. And in others, where the mutations seemed to increase.

So one of the, I think, real take-home messages of this project is that when you're designing trials such as investigating nadofaragene or other agents, currently we do...

The biggest thing we use, of course, is histology and clinical parameters. We say, "Hey, this person is BCG unresponsive," or we say, "This person has CIS. This person has high-grade papillary disease."

But as you know, Sam, and as our audience likely knows, not every CIS is the same. We've all been there, where we've looked in the bladder and the bladder looks horrible. There's red spots everywhere, and there's tons of CIS.

And then there's your other patient who has a small amount of CIS in one location. We know that those patients are different, but yet they're currently classified the same in clinical trials.

What we hope to use UroAmp for in the future, in part based on this project, is really can we use the UroAmp MRD status to stratify patients as...? If you were to assess a patient longitudinally, can you look at them and say, "Okay. This person has an unresponsive profile, in terms of the molecular residual disease they have?"

"Do they have an expanding profile, perhaps suggestive of recurrence? Or do they have a complete response profile?" Such that you have more information than just your urine cytology and your cystoscopy, both of which are a bit more subjective.

So these are some of our take-home messages, which are that, one, this urinary molecular residual disease test does allow for quantitative assessment of molecular response to drug treatment.

You can potentially do balanced stratification of control and intervention arms. I think that's our hope in the future. And importantly, at least for nadofaragene, it seemed to be highly predictive of recurrence.

And again, the caveat here is that this is from the phase two data. We do have urine from the phase three data, and we're working on doing this with that larger cohort of patients to see if that holds true.

But those are the take-home messages. And I do want to give a shout-out to Dr. Dinny, Dr. McConkey, and the entire Convergent team as well, which helped us put all of this data together, as well as the team at MD Anderson.

Sam Chang: Vikram, that is fantastic. Before we drop your slides, can you click back a couple of slides? I just want to ask a couple of questions about them.

Vikram Narayan: Sure.

Sam Chang: Let's look at this one because it's basically full of data. So the UroAmp test is an off-the-shelf test. Correct?

Vikram Narayan: Yep.

Sam Chang: And I'm assuming it's a proprietary panel of mutations, whatever deletions, whatever that they've accumulated to make this?

Vikram Narayan: Right. Yeah.

Sam Chang: Because I think all these urinary marker tests that are commercially available... And is Convergent commercially available at this time?

Vikram Narayan: Not yet.

Sam Chang: Not yet? So the ones that are currently commercially available will look at four changes in DNA markers. We'll look at three. We'll look at RNA change. We'll look at multiple different changes, but we've learned that it's more than just one or two, it's a panel. So this is a proprietary off-the-shelf looking at mutations within the urine. Correct?

Vikram Narayan: That's right. Yeah. So they do next-generation sequencing. They look at a 60-gene panel, and then they also do whole-genome aneuploidy. So they can profile the mutational burden that's in the urine.

Sam Chang: Okay. And so within their algorithm, they have figured out the ones that perhaps are most predictive. So is there one UroAmp...? In terms of the reading, it seems like there are two indications for this, a qualitative or rather a... Yes, almost a qualitative, "Yes. No. It's present or not."

Vikram Narayan: Yep.

Sam Chang: And clearly, when there's no minimal residual disease, at least molecular or DNA-based, our chance for recurrence-free rates and length of that time period looks great. Your 100% curve on the bottom right here looks fantastic.

Vikram Narayan: Right.

Sam Chang: And then your second slide, the next slide shows basically that there might be a quantitative component to this as well. Is that right?

Vikram Narayan: That's right.

Sam Chang: Okay.

Vikram Narayan: Yeah. That's exactly right, Sam.

Sam Chang: And the quantitative differences, we don't know yet if there are certain mutations that may trump others. Overall, if there's a decreased pattern, it seems to be, like you've got here, a partial response versus a complete response.

Vikram Narayan: Yes.

Sam Chang: So that's all great to me. A couple of things. One, you said this is in a BCG unresponsive cohort. So it's robust in terms of having patients that may be more likely, obviously, to have higher risk mutational changes that carry on increased chances of recurrence or progression.

I guess let's talk about the next steps. I want to focus on two. One, what your thoughts are regarding this test in the future? Next steps? And then vis-à-vis the use of nadofaragene, specifically with this.

Because you hypothesize that this test may predict how people do with say, intravesical chemotherapy or other forms of intravesical immunotherapy, or maybe even systemic IO.

Vikram Narayan: Sure.

Sam Chang: We don't know yet though for sure.

Vikram Narayan: Right.

Sam Chang: Okay. So tell me what your most exciting next trial is? Is it to look at the phase three data to confirm bigger numbers? Is it in the basically de novo untreated population, what the signals are between our classical low, intermediate, high risk?

To me, that would be fascinating, are these categories that we've basically lumped together and try to parse out, are they really that accurate? Maybe so. They seem to pan out.

Vikram Narayan: Yeah.

Sam Chang: But gosh, you would think... Tell me what's most exciting for you, from that standpoint?

Vikram Narayan: Yeah. I think you took some of the words right out of my mouth, Sam. In the sense that for me, the most exciting utilization of this tool is really to give us that additional missing piece that we currently have, when it comes to understanding why bladder cancer behaves the way it does.

And I would say that moving forward, the thing that I'm most excited about is to be able to, essentially, put together some trial concepts that allow us to utilize this test to check and see if a patient is responding to therapy and also when they respond to therapy.

And of course, there are some assumptions there that we have to validate. And so, to your point, we do intend to look at the phase three data and ensure that...

Because obviously, in a small cohort, how does this pan out in a much larger group? So that's going to be important. But assuming the assumptions are true, which is that you can see these MRD levels reduce, assuming a patient is responding to treatment, well, that's a potentially very powerful signal that you can use.

And perhaps most importantly, when it comes to these patients where you're not sure whether you should continue to give a patient a particular treatment, which I think we deal with that clinical scenario all the time in practice, where you've got a patient. Maybe there's a red spot. It's biopsied, it's negative. You're like, "Well, I was concerned about it. He's got some vague symptoms, he's BCG unresponsive. Should I take his bladder out, or should I try something new?"

And the patient's like, "Well, you tell me. What do you think?" And of course, they don't want to take their bladder out. But you'll feel a lot better potentially if the studies and data bear it out, that you maybe get a urinary minimal residual disease test and say, "Okay. Actually, there's very little mutational burden in this."

And that's a good thing. That means that let's hold the course. On the other hand, if you don't see that, you can potentially distinguish between patients who just have bad inflammation versus actual disease, that maybe that's the patient that you do early cystectomy on. Those are the kind of questions that I think we need to answer.

The other thing that I think is useful here is, I think, the timing of therapy. So particularly for nadofaragene, one of the things that we don't know is among complete responders, when do you stop giving that patient additional therapy?

So the study that we have to date, the treating physician could continue to redose patients every three months, assuming they were complete responders. So do you do that forever? Seems unlikely. It is an expensive drug.

So then, do you stop at some time point? And we don't know. If you have an adjunct tool that gives you some additional information, I think that's going to be super powerful.

Sam Chang: No, I think you make excellent points regarding... As we enter this field of precision oncology, understanding, just as you said, when do we deescalate treatment? When do we change treatment?

When do we have the possibility of finding worse recurrent or progressive disease prior to our archaic look in the bladder, figure out what's going on, which we've done for more than a hundred years? And this is a real chance of finding disease early on and acting on it.

So, to segue that to nadofaragene, to me personally, as you started talking about it, I would love to see you guys be able to analyze the phase three data.

And I would love to look at the cohort of patients, obviously, who didn't respond. But to me, almost just as importantly, I would love to see the complete responders at three months.

It was at the three-month mark. If you had any long-term response, you had to respond by that three-month mark. And if it can predict in those patients who have the long-term...

So, I think to be able to then utilize that with a new agent that we're learning about, but at the same time learn even more about, to me is actually really exciting.

So, along those lines, have you all at Emory started treating patients with nadofaragene? Recently, in January of this year, they've opened up access, etc.

Vikram Narayan: Yeah.

Sam Chang: Tell me about your experience with that.

Vikram Narayan: Yeah. So we were part of the... As you know, there's an early access program as the company was ramping up availability. So we were one of the original early access sites.

Emory was actually one of the largest accruing sites on the phase three trials. So our nursing staff and clinical staff have a lot of experience with the drug.

We did just put our first patient on the commercial product, a patient with BCG unresponsive CIS. And so we'll see how... And this was just as recently as a few weeks ago. So, I think the jury's still out on how well he does.

Sam Chang: Sure. Sure.

Vikram Narayan: But I think we're really excited about having this as an option for patients. I think some considerations, of course, are that... And I think these are some of the questions we have to ask ourselves as urologists, uro-oncologists treating this.

There are a number of new agents that are emerging. And obviously, it's a great, great time because 5, 10 years ago, there were really no options for patients other than maybe Valrubicin or cystectomy and salvage agents.

Now we know about potential agents from CG Oncology, gemcitabine docetaxel, the immunity bioproduct. So there are several agents available. But one of the questions that I think is going to come up, that we have to be considerate about, is cost, sequencing of treatments.

And then where everything fits into place as we're understanding the trial data and how they compare and all of that. So I think there's a role here for us to be thoughtful, using tools like the MRD test to better inform the types of patients that we have as we think about who gets which treatment moving forward.

Sam Chang: Well, Vikram, thank you so much for your insights, and your experience, and having that ability, perhaps, to define those patient populations that are most likely to benefit or who are benefiting almost in real-time with the evaluation of a urinary biomarker.

I think that really obviously improves the personalization or precision of care for each of these patients. So I applaud you for your work, and I knew Emory was one of the largest accruing sites in terms of clinical trial participation with nadofaragene.

So really, really kudos to you all, to our friends and partners down south. And we look forward to future trials and future collaboration. And thanks so much for spending some time with us.

Vikram Narayan: Absolutely, Sam. It was a real pleasure, and thank you again for inviting us to speak about this.