Read the Full Video Transcript

Alicia Morgans: Hi, my name is Alicia Morgans, and I'm a GU Medical Oncologist and an Associate Professor of Medicine at Northwestern University in Chicago. I'm so excited to have here with me today a good friend and colleague, Dr. Matt Galsky, a Professor of Medicine at Mount Sinai in New York and also a really invested investigator in urothelial carcinoma. We are here to talk with him today about a recent presentation at ASCO 2021, where he really looked at using an approach to treating muscle-invasive bladder cancer that spares patients from undergoing cystectomy or a directed, definitive local therapy. Thank you so much for talking with me today, Dr. Galsky.

Matthew Galsky: Thank you. Thanks for having me.

Alicia Morgans: Wonderful. Can you tell me a little bit about your approach to this patient population? What was your study design? What did you investigate?

Matthew Galsky: Oh, of course. Neoadjuvant chemotherapy followed by cystectomy is a standard treatment for muscle-invasive bladder cancer for years. And for years, it's been known that a subset of patients will achieve a complete pathological response that is associated with a very good prognosis. We get very excited about that, but unfortunately, we tell patients they have had a pathological complete response after their bladder has already been removed. And so the concept is whether or not there is a subset of patients who can be treated definitively with systemic therapy plus TURBT and not undergo cystectomy. This is not a new concept. This concept has been around since the early days of MVAC, where there have been observational studies published looking at patients who declined cystectomy. And we know that a subset of patients will enjoy long-term, durable, bladder-intact, disease-free survival. And so we know that this is possible. And the question is, can we define which patients can achieve this approach?

So we designed this study in 2015, and we designed it with two real hypotheses in mind. One was that ... I guess three, maybe. One was that this concept has not been tested rigorously, prospectively. In fact, there was a similarly designed SWOG study, which used carboplatin-based chemotherapy, but not cisplatin-based. We think that cisplatin-based therapy is really a standard in this disease. There have been a few other studies with variations on the theme, but very few prospective studies. That was one.

Number two is that we believe that clinical response in this setting is a biomarker. We talk about the rigorous development of biomarkers all the time, but this is a biomarker, and so it needs to be defined consistently, and it needs to be measured consistently, and its performance characteristics need to be assessed. And when you have different studies testing different approaches, defining things in different ways, then the literature is all over the place and you can't make sense of it. A second hypothesis that we had was rigorously defined clinical response may be a biomarker that might help aid in selection.

And then the third hypothesis that we have I think gets at some of the concerns about this approach that have been raised, and rightly so, but there is this notion that there's a disconnect between clinical response and pathological response, and that is why maybe this approach shouldn't be pursued. And one, that's clinical response being inconsistently measured and defined, but two, that undermines the potential, I think, for delayed cystectomy at the time of local recurrence, to be able to quote-unquote salvage outcomes in patients. And so can a risk-adapted approach be as good as a cystectomy in everyone upfront approach? It's a hypothesis, but it's not a hypothesis that has been well tested, and salvage surgery is a hallmark of organ sparing approaches for other diseases that are treated in oncology. That's the long-winded background.

The study design involved giving four cycles of gemcitabine, cisplatin, and nivolumab in patients with clinically localized muscle-invasive urothelial cancer of the bladder. After that, patients underwent a restaging evaluation that included an MRI of the bladder, unless that was contraindicated. Then they had a CT, urine cytology and a repeat cystoscopy with mapping biopsies of the bladder.  If there was a visible tumor, of course, a visible tumor biopsy. If there was not, then their prior scar site was biopsied in addition to other areas in the bladder. If all of those tests were negative, that defined a clinical complete response, and if patients achieved a clinical complete response, they had the option to proceed without cystectomy and receive an additional four months of nivolumab or to proceed with cystectomy, recognizing that that is still the standard of care in that situation. If they didn't achieve a clinical complete response, then cystectomy was recommended.

Alicia Morgans: Really, really interesting. And before we get to your results, what you found in this trial, I wonder if this gave you an opportunity for those few patients who did not decide to proceed with continued immunotherapy treatment but really went to cystectomy, were you able to confirm in those patients that they truly had a complete response pathologically, if you did have the bladder specimen to review that?

Matthew Galsky: So I think patients voted with their feet in terms of this approach in two ways, and I think that really speaks to a need to balance the efficacy of our treatment with the potential impact of our treatments on quality of life. And I say that because the trial accrued incredibly quickly. In fact, I think it's one of the few trials that I've led that it wasn't an issue in terms of making sure that we were getting the trial accrued. The second was that only one patient who achieved a clinical complete response opted to proceed with immediate cystectomy. Of note, that patient had a low-grade papillary tumor on pathology.

Alicia Morgans: Wow. Wow. Well that, in itself, as you mentioned, is striking. It really clearly ... the patient's voice could not be heard more clearly in this setting. People want to keep their bladders, which is not a surprise. But to the point where they are willing to continue on this monitoring and take a chance. So very, very important. What did you find as you monitored these patients on study?

Matthew Galsky: So, seventy-six patients were enrolled in this study. At the time of the data lock for ASCO, 64 patients had reached the post-cycle for restaging evaluation. Of those patients, 31 had a complete clinical response, so a clinical complete response rate of 48% to date. Of those patients, as mentioned, one patient opted to undergo an immediate cystectomy.  So we have a cohort of 30 patients so far who opted not to undergo cystectomy and proceed with additional nivolumab followed by observation.

We have a medium follow-up of about a year. We had co-primary endpoints of this study, one was to define the clinical complete response rate, and the second was to determine the performance characteristics of clinical complete response for predicting quote-unquote treatment benefit. And we defined benefit as a composite endpoint because of this issue that patients could undergo cystectomy immediately versus not. So benefit was defined as a pathological complete response if patients opted for immediate cystectomy, or being two years metastasis-free if patients opted for no cystectomy. We do not have long enough follow-up for that second co-primary endpoint.

That said, of the patients that have been followed up, of those 30 patients, eight have had a local recurrence. We've had no patients who have had metastatic recurrence before local recurrence. We've had one patient who has had a metastatic recurrence after local recurrence and cystectomy, and we have a number of patients out beyond one year from the start of treatment now who are disease-free with an intact bladder.

One of the issues in terms of this concept of whether or not delayed cystectomy might be a reasonable, risk-adapted approach ... Of course, in the absence of a randomized dataset, we can't answer that question definitively, but one bit of information that might help one sort that out is the surgical pathology at the time of delayed cystectomy. Right? And so of the patients who had delayed cystectomy, about half of the patients had carcinoma in situ or less, and then we had two patients who had T2N0 disease, and we did have one patient who had T4N1 disease. That is the patient who ended up developing metastatic disease. So the majority of those patients have had the low-stage disease at the time of delayed cystectomy.

Alicia Morgans: Really interesting because as you mentioned before, I think one of the dogmas of muscle-invasive bladder cancer is that we really cannot trust our clinical staging, that we will under- and over-stage our patients both pre-and after neoadjuvant chemotherapy, and that if we are not proceeding rapidly to this organ directed surgical therapy or even chemo, chemoradiation if we are going more in a bladder sparing approach, that we are not serving our patients well. That does not seem to be what you found in this trial for the most part. Nothing is perfect, of course. And I'd love to hear how you think ... how has your surveillance strategy, maybe the use of MRI, how has that improved your ability to understand the clinical stage in a way that seems to be actually quite effective in this cohort of patients?

Matthew Galsky: I think we need longer follow-up, of course, to know that the patients who have opted to have their bladders left intact, to make sure that that is indeed a viable approach for those patients. That said, I think based on the data so far, clearly, clinical complete response defined the way that we've defined it is prognostic. I don't think that there's probably much of a question about that. We know that in part by the pathological stage of disease for the patients who didn't achieve a complete clinical response, and those patients have a much higher pathological stage of the disease. It's definitely prognostic. And I think that the rigorous definitions and rigorous assessments that are done are helping with the predictive or prognostic nature of that assessment.

Alicia Morgans: I would agree. I do think it's interesting, though, particularly the MRIs. I don't know how standardly people are using MRIs as they are staging bladder cancer patients. And I wonder if this is going to evolve, particularly if we are using this clinical complete response in the future as one of our metrics of determining how we move forward with care. I'm excited to see how that particular piece of it continues to contribute to our staging, to help us understand better who these patients are. And sort of along those lines, I assume, though, I know you don't have the data yet, but will you be able to look at correlatives, look at biomarkers from maybe tissue-based biomarkers or other biomarkers suggestive of response within this cohort of patients?

Matthew Galsky: Absolutely. There's a suite of pretreatment biomarkers that are being assessed on the TURBT tissue that might help inform which patients who have achieved a complete clinical response are really the ideal patients for this approach. And then there's the addition of these post-treatment biomarkers to add to the clinical assessments that we are already doing. In terms of pretreatment biomarkers, that was actually integrated prospectively as a key secondary endpoint of the study. As you well know, there are a series of studies seeking to determine whether or not a similar approach can be employed in patients with tumors harboring alterations in DNA damage response genes.

We felt that we really needed to lock in the performance of the clinical biomarker before we layered on top of that genomic biomarkers. But we did want to validate those prospectively as a secondary endpoint.  So we have a panel of four DNA damage response genes that are a key secondary endpoint in terms of whether or not those add to the predictive or prognostic nature of clinical restaging. That data has been presented at ASCO but correlated with clinical complete response rate, which is not the right endpoint. The right endpoint is to correlate those alterations with long-term outcomes, so that will be coming in the future.

One of the key post-treatment assessments that we already have data on but has not been analyzed yet is radiomic data from the MRIs that are done post-treatment, so certainly we're hoping that there are features that can be extracted beyond what can be seen with the naked eye that helps inform whether or not this approach might be reasonable.

Alicia Morgans: I think that's fantastic. Great planning, great work, and I'd love, as we finish up here, to just hear your bottom line summary for those who are listening who are really looking for this as a potential way to move forward in the future, of course, after a little more study or a lot more study. What is your bottom line from this trial?

Matthew Galsky: This is not a standard of care approach as of yet. That's the key take home. We need longer follow-up from this dataset. We need the results and longer follow-up of complementary datasets, the RETAIN study from the group at Fox Chase, the RETAIN II study that they are doing now. [inaudible 00:14:43] alliance study. I think all of these datasets looked at together will help inform whether or not there is sufficient data that we feel confident that at least a subset of our patients can pursue this approach as part of a standard management algorithm for muscle-invasive bladder cancer.

Alicia Morgans: Fantastic. Congratulations to you and to your team, and thank you so much to the patients who participated in this work. I really appreciate your time and your expertise today.

Matthew Galsky: Thank you.