A Debate on The Management of BCG Unresponsive - Cystectomy Ineligible Bladder Cancer Patients: Pembrolizumab Vs. Nadofaragene Firadenovec - Arjun Balar & Peter Black
June 8, 2020
Arjun V. Balar, MD, Associate Professor, Department of Medicine and Director of the genitourinary medical oncology program at NYU Langone’s Perlmutter Cancer Center. NYU Langone Health, New York, New York
Peter Black, MD, Senior Research Scientist, Vancouver Prostate Centre Professor, Department of Urologic Sciences, University of British Columbia, Vancouver Prostate Center
Ashish Kamat, MD, MBBS Professor of Urology and Wayne B. Duddleston Professor of Cancer Research at MD Anderson Cancer Center in Houston, Texas.
Ashish Kamat: Hello and welcome to UroToday's Bladder Cancer Center of Excellence. I'm Ashish Kamat from MD Anderson Cancer Center in Houston. And I'm joined by two experts in the field of bladder cancer, Dr. Arjun Balar from New York and Dr. Peter Black from Vancouver. Thank you guys so much for taking the time to speak with me today.
Arjun Balar: Thanks for having us.
Ashish Kamat: For the purpose of today's meeting, I've essentially assigned each of you a side and you'll present your arguments to support it. Clearly, this is a side that you've been given, and you will make your strong arguments, but afterward, we'll have an open discussion so our audience can learn how they can counsel patients, and what your true way of counseling your patients is.
So, the patient that's being presented is a real-world patient, but also hypothetical in the sense that a cystectomy is completely off the table, for whatever reason. The patient may refuse it or you may not be able to do it in your hospital, say, it's the COVID situation going on right now.
So, with that in mind, the situation essentially is a very healthy male patient who is in his 60s that has T1 high-grade disease. Initially undergoes BCG therapy, gets an induction course, has an evaluation, everything's fine. Has another course of BCG as maintenance therapy and again, has no disease in the bladder at evaluation. At the next maintenance therapy of BCG, he is evaluated and found to have CIS in the bladder along with a papillary tumor that's TA high-grade. So it's not T1 high-grade, but it is TA high-grade papillary plus CIS. So for this patient, what would your recommendation be? Arjun, if I could have you start.
Arjun Balar: All right. So, I'll take the baton, and I'll make an argument for checkpoint blockade with pembrolizumab. A lot of evidence to support it. First, beginning with the first Phase I studies in advanced disease showing durable responses in close to a third or more patients. And then we tested this in the first-line setting in advanced bladder cancer showing response rates up to as high as 29%.
What's most notable here is that systemic checkpoint blockade leads to durable responses, leads to long-term group survival and you can see this in both Kaplan-Meier plots where you have long-term tails to the curve, anywhere between 30 and 40% of patients. So this was reason to test it in the first-line setting in BCG unresponsive high-risk NMIBC. This was KEYNOTE-057, a very important study that looked at two disease cohorts. Cohort A were patients with CIS with or without papillary disease. Cohort B were patients with papillary disease only and this focused on high-risk NMIBC patients who were considered BCG-unresponsive based on the most recent FDA definition.
Patients treated with pembrolizumab went on 200 milligrams every three weeks. They had mandated cystoscopies at every 12 weeks with urine cytology. This is a very rigidly conducted study with clear endpoints. And all patients must have BCG-unresponsive disease. Responses were noteworthy, but what's more important honestly is the patient characteristics of those who went in this trial. Highly representative. Those who had bladder cancer, the median age was 73, male predominant. The median number of instillations of BCG was 12. So this is a patient population that was clearly pretreated heavily. And the majority of patients had CIS alone; however, up to 12% of patients actually had high-grade T1 disease at the time of enrollment.
So what were the outcome data that we saw? The first thing that we look at is CR rates, and here what we see is a CR rate of 41% in patients, in the 96 evaluable patients. So this is a CR at three months of treatment. And what's most important also is that patients who didn't achieve a CR, if they had to discontinue treatment, they didn't progress to T2 or higher disease at the time of treatment discontinuation. That's one of the biggest concerns we have in these patients where we know cystectomy's curative, but we also don't want to lose the opportunity for cystectomy if the treatment is not successful. And that does not appear to be the case here per patients treated on this trial.
But after you achieve a CR, what's the most important after that? It's the durability of the CR. These are the data that we presented at the FDA ODAC in December 2019 which was the basis for FDA approval. 96 evaluable patients. Of the 39 who achieved a CR three months, what you can see is that the median duration of response. So if you're in a CR, how long does that CR last? On average, it lasts around 16.2 months. But if you look to the right of the curve, you can see that there's clearly a flattening, there's a number of censoring events that are occurring at 15, 18, 24 months and so on. But what you can see is there's clearly a plateau, and there may be a percentage of patients who have durable responses that could last for a long time, potentially many years. And I've treated patients on this trial, in fact, the most in the country, and we have several of these patients who are doing quite well years after having started treatment.
Safety is always a concern regarding any agent you use, and systemic checkpoint blockade has immune-related adverse events. The ones we worry about in particular are the ones that can be potentially permanent such as apophysitis, adrenal insufficiency, and Type 1 diabetes mellitus. But what you can see is you only have a single event out of all these otherwise toxicities and treatment was exceptionally well tolerated on this trial. And the safety events that we see here very much mimic what we see in patients treated with advanced disease.
What about quality of life? Several instruments were used, the FACT-BL score, which was relatively stable throughout treatment, but this is really a PRO score that's really for advanced bladder cancer patients. When we look at the quality of life CLSS score, this is the one that really focuses on lower urinary tract symptoms, and these are the symptoms that most patients with localized disease suffer from.
And what you can see is a suggestion that actually with pembrolizumab therapy, some of their local symptoms actually get better with treatment. Now, this is pelvic sometimes pain and discomfort, frequency and urgency and so forth. So the classic lower urinary tract symptoms could actually potentially improve with treatment. But certainly, it does not get any worse based on this scoring assessment.
So how do we put this all together? We know that CPI has revolutionized bladder cancer treatment and fundamentally changed how we think about the disease. Pembrolizumab induces CRs in BCG-unresponsive CIS. And what this suggests is that systemic activation of the immune system leads to anti-tumor immunity. And a CR in this context, in my opinion, is potentially more reliable compared to an intravesical agent, which acts mostly intraluminally and really doesn't treat subepithelial or disseminated disease, which is the primary concern.
The approval has redefined multidisciplinary care. I think the partnership between urology and medical oncology is going to be extremely important. Not that it wasn't already, but the expertise of both urologic oncology and medical oncology will be a requisite. But I think this is an important advancement and probably the best treatment for this particular patient.
Ashish Kamat: Thank you, Dr. Balar. Dr. Black, you're next.
Peter Black: Okay, so I would like to make the case for nadofaragene as first-choice treatment for this patient and patients like this with BCG-unresponsive carcinoma in situ.
So nadofaragene is a completely new drug. We haven't seen anything like this before in urology. It's basically gene therapy. It's an adenovirus that carries the interferon alpha-2B gene, delivers it into bladder cancer cells that then make interferon, which leads to cell death and eradication of bladder tumor.
It has been tested first of all in a Phase I trial that just was dose-finding and safety and then there was a small single-arm Phase II trial in patients with BCG-unresponsive disease, not only CIS, so there are papillary tumors in here as well. With very reasonable results in a cohort of 40 patients. The decisive data was a Phase III trial that was presented first at the SUO in 2019 and again more recently at GU ASCO.
This again is a single-arm Phase III trial done in the US and multiple centers coordinated through the SUO Clinical Trials Consortium. The drug's administered together with a detergent CIN 3 that promotes the uptake of the viral vector. It's administered every three months, and some patients on trial went well beyond a year with these q3 months dosing. The trial included a mandatory biopsy at 12 months in the CIS cohort, which is different from the KEYNOTE-057 trial that Dr. Balar just presented where there was no mandatory biopsy.
The primary endpoint was a complete response. It didn't define the time point for the complete response in CIS patients and then secondarily looked also at the durability of this response and then recurrence-free survival in patients without CIS. Enrolled 103 patients with CIS in the trial. Very similar to the pembro trial, as well as 48 with papillary disease only.
And these are the key outcome measures. So the complete response based on cytology and cystoscopy at three months was 53%. And then you see a drop off at six months, nine months. And then at 12 months, there is the mandatory biopsy and 24% of patients were without recurrence. Five percent of patients, so essentially five out of 103, progressed to muscle-invasive disease. Most of those on cystectomy, and we can compare this to the results that Dr. Balar just presented. You'll see that the sample size a little bit different the 97 versus 96 are just some of the nuances from the ODAC briefing document or the ODAC review. But here again, the three-month CR rate was 41% and the 12-month response is measured after the CR, so it's really a 15-month response. Dr. Balar showed 19%. I have it here it's 20% so comparable results at a year, although no mandatory biopsy in the pembro trial. And there were 3% of patients on pembro who did progress to muscle-invasive disease at the time of cystectomy after failure of the pembro.
The difference in toxicity I think is the biggest difference. The nadofaragene is very well tolerated. Seventy percent of patients had some degree of toxicity, which is mostly dysuria, lower urinary tract symptoms, maybe some fatigue, maybe some flu-like symptoms, but only a 2% serious adverse event rate. Four percent Grade 3-5, and 2% discontinued due to adverse events, which compares very favorably with pembrolizumab where you see, for example, the 29% Grade 3-5 AE rate and 10% discontinuation.
So I think overall if you have a patient with BCG-unresponsive CIS who is not suitable for cystectomy, we can look at efficacy and say it's more or less a wash. The trials were designed very similarly, so we are able to compare across trials. But again, the biopsy for example in the nadofaragene trial does change things a little bit.
But the real advantages are the absence of contraindications for the nadofaragene. There are few patients who cannot get it, whereas there are certainly patients who cannot get systemic immunotherapy. The ease of delivery, intravascular delivery once every three months versus systemic delivery every three or four weeks. Safety, of course, as I highlighted. And then cost, and we don't know the cost of nadofaragene yet, but it's hard to imagine that it would be anywhere near the cost of systemic immunotherapy. So I think the balance favors nadofaragene.
Of course, we are happy to have choices for our patients. I think that should be emphasized, and it leads us to think, "Well is it just a matter of which one first and the other one second." And we've made, for the sake of this debate, we've made the patient completely ineligible for surgery, but we should remember that the vast majority of patients on these trials are actually cystectomy eligible. And so if they have one of these treatments and have a recurrence, we really should be pushing them towards cystectomy rather than additional rounds of salvage therapy. But in the patient who is truly ineligible, of course, he or she will benefit from multiple treatment options.
So in summary, I just say that all relevant parameters favor the nadofaragene over pembrolizumab for treatment of BCG-unresponsive, actually, CIS not just high-risk and non-muscle invasive disease with rare exceptions of patients who cannot be catheterized or cannot hold an intravesical agent. Thank you.
Ashish Kamat: So, that was great talks from both of you. If I could turn to you, Arjun, since Peter did provide in his talk the pros and cons of each. From his viewpoint, it clearly looks like nado is the first choice for these patients. What would you say to that?
Arjun Balar: I think it's a fair point. However, these trials were a little bit different in terms of the fact that the mandatory biopsy was required in the Instiladrin® study. One of the other issues in this particular comparison is in the pembrolizumab trial. Patients who didn't achieve a CR at three months, unfortunately, had to absolutely discontinue treatment. And what we know about systemic checkpoint blockade is that sometimes there are responses that can be delayed. Now, one of the limitations is that this trial was a little bit of ahead of its time and, in fact, when the study was initially designed in 2013 I want to say, 2012, it was a while, the FDA mandated that the treatment had to stop at three months. What we now know is that many patients who actually did not achieve a CR at three months are actually doing well and ultimately did achieve a CR in the bladder and what that number will actually look like, we'll need to look at other randomized studies to test this. But that's one of the limitations I'll draw about just direct comparisons in terms of efficacy between the two treatments.
The other issue is that yes, the CR rates if you were to take them at face value, were a bit less in the pembrolizumab study. Remember this is a systemic agent, and so if you're able to activate systemic anti-tumor immunity, I would argue that that CR in the bladder is probably a bit more reliable in terms of leading to does it represent truly eradication of a disease that you can be safely observed without cystectomy versus in something that's instilled in the bladder.
Ashish Kamat: Quick question to both of you, and I'll have you answer that first, Peter. When people were awaiting the results of both of these studies, patients obviously and we as the community were hoping to see numbers that would be higher than what was reported. Twenty, 24% is essentially one in five. Why do you suppose that is?
Peter Black: So I agree that the numbers are lower than ultimately what we hoped for. And I think especially for pembrolizumab if you take on all the extras of systemic therapy, toxicity and cost, it's really a marginal benefit. But in patients who are truly cystectomy ineligible, of course, I would see it as an advance. I think each drug is different. How it works is different. Arjun was just talking about systemic immune response, but this is carcinoma in situ, it's limited to the mucosa. It does not penetrate the basement membrane and so maybe the immune system doesn't have the same access to the tumor cells. This was something, as you know, I was a study chair for the [inaudible] trial with atezolizumab and there are a lot of questions up front about whether we can actually expect systemic immunotherapy to work in something that's limited to the mucosa. So there may be a component of that.
Ashish Kamat: To follow up on that question and Arjun, maybe I'll have you tackle that. In both of these studies, the patients had to be BCG-unresponsive and that's clearly because the FDA listening to their advisers, essentially all of you here on this call and others, allowed us to have single-arm studies in a particular cohort of patients that are BCG-unresponsive. But what is your feeling about, and I know studies are being done, but in the absence of data from studies, what is your feeling about these patients requiring to be primed with BCG before they have an immune response generated by systemic IO agents? Is that BCG priming necessary? Do they have to be BCG-unresponsive to demonstrate such outcomes?
Arjun Balar: So I think that's an excellent question, and I think that ultimately the trials will answer it. My feeling is that it's most likely not. I think we have to remember that the reason the initial studies were done in BCG-unresponsive disease was primarily because BCG is so effective upfront. CR rates as high as 70 or 80%, and we know it's a standard of care and some of these patients can potentially be cured with BCG. And so because of that, it only makes sense from a regulatory standpoint to investigate drugs in the post-BCG setting.
Now, we can also kind of revise history a little bit. We can take a look at pre- and post-specimens and see that BCG induces PD-L1 expression and that probably scientifically lends a bit more weight behind testing in the BCG-unresponsive setting. However, these frontline trials which are combining systemic checkpoint plus BCG as very first-line treatment in CIS and also after failure of induction BCG, I think those trials will likely sort all of that out.
Ashish Kamat: And speaking of trials that are looking at first-line IO or IO early in the first-line or initial failure, could you enlighten our viewers as to what those are and where they are in development and accrual?
Peter Black: Right. So pembrolizumab as part of their Phase III confirmatory studies they have KEYNOTE-676, which focuses on patients who have failed BCG induction and then are randomized to re-induction alone versus re-induction plus pembrolizumab. So that's one group.
There is the POTOMAC study, which is looking at durvalumab in the first-line setting. And then also there is the ALBIN study, which is also the randomized study that's looking at atezolizumab I believe. And then lastly, what I do remember, and I don't know the name of the study just yet, but Pfizer is also leading a randomized Phase III trial with their own subcutaneous PD-1 antibody in combination with BCG in high-risk CIS. So a number of different sponsors are running their studies in this setting that are all going to some degree answer that question is prior BCG really required for checkpoint blockade to work?
I think in advanced disease we looked at chemotherapy and questioned whether that is required or not. I think a lot of that has been kind of put to rest. I think they're mechanistically very different, and I think we'll find very similar outcomes in non-muscle invasive disease as well.
Ashish Kamat: Peter, you know in the urology community there's been an explosion of people using combination chemotherapy, intravesically and the doublet that most people use and comes to mind as having been reported out the most is gemcitabine and docetaxel with fairly impressive results that you and I have discussed off and on. What is your sense as to how that data, which is not prospective, not randomized, even in single-arm studies compares to the intravesical gene therapy nadofaragene?
Peter Black: Yeah, I think we need to be careful with retrospective data. I agree that it looks encouraging, and the numbers would suggest that it's better than either of these two drugs and it's inexpensive, easy to give, well-tolerated. I think we need to be very, very careful with that data and really need to go with trial data. So, ultimately it'd be nice to have a comparison. I don't know if we will anytime soon. It's interesting to think how trial design is going to move forward now that we have one approved drug and potentially a second on its way and if we can start actually comparing these drugs to each other, or at least comparing new drugs and new combinations to one or the other of these. I think we have to go with trial data.
Ashish Kamat: Yeah, absolutely, I think it's good to have this baseline information, but that prospective studies are clearly needed. In closing, let me ask each one of you, now, let's assume that this one patient has the same recurrence that he has, but that cystectomy is something that he or she would be considering. And the patient is in your clinic sitting down with this information. How would you counsel this patient? Just in brief, and let me start with you, Peter.
Peter Black: Yeah, so I think my practice has evolved a little bit over the last 10 years. The last 13 years I guess since I trained with you guys at MD Anderson. I think there is time for patients with CIS and high-grade TA recurrence after appropriate BCG to try one additional salvage therapy before moving to cystectomy. So I will always offer cystectomy, but will also offer a salvage option. And up until now, it has usually been docetaxel plus gemcitabine. And here I would now add these to the list or maybe preferentially these since we have trial data. But I think we don't have nadofaragene yet available to us, but I think I would be primarily pushing in that direction before referring to a medical oncologist. And if that salvage therapy does not work, then they go onto systemic.
Ashish Kamat: So you would give them one attempt with salvage therapy and then strongly try to push the patient to select a radical cystectomy or would you try-
Peter Black: Yes.
Ashish Kamat: Two options? Okay, great. And Arjun, how would you counsel patients?
Arjun Balar: Right. So I mean, to be fair when patients are either referred to me or come to me for a second opinion it's generally because they don't want the cystectomy. Because as you know, Ashish and Peter, I certainly don't operate and don't intend to. And so the counseling I provide patients is relatively straightforward. I tell them number one, that their most curative option, permanently curative option in terms of bladder cancer, is a cystectomy. But with it comes the intermediate, short-term impact on the quality of life, perioperative morbidity, and mortality. And so for patients who are not willing to accept that reality, then I think alternatives to cystectomy are very much in consideration. Right now, my conversation would be toward pembrolizumab because it is an FDA-approved drug in this particular setting. And I've treated patients as a standard of care since the approval.
But whether it's that or Instiladrin®, really the issue is can we treat effectively and if we achieve a CR, that's great. If not, can we quickly go back to that curative option, which is cystectomy? And that's the hard part. And I tell patients that is that we can only see what we can see in the bladder. I don't know if we have data to say that only one salvage regimen versus two versus three, when is enough, enough or is too much, too much to avoid a cystectomy. And I think hopefully our trials will address that.
Ashish Kamat: So this has been really an excellent discussion. Very, very, very informative. Thank you both so much for taking the time to be part of this important educational activity. Stay safe and stay well.