The Clinical Impact and Molecular Drivers of ‘Secondary’ Muscle-Invasive Bladder Cancer - BCAN New Discoveries Young Investigator Award - Eugene J. Pietzak

July 20, 2021

Eugene Pietzak, MD, Assistant Attending Surgeon of Urologic Oncology at Memorial Sloan Kettering Cancer Center in New York City received one of the five 2021 Bladder Cancer Advocacy Network (BCAN) New Discoveries Young Investigator Awards for his work in “Defining the Clinical Impact and Molecular Drivers of ‘Secondary’ Muscle-Invasive Bladder Cancer.” There are relatively well-established molecular subtypes for muscle-invasive bladder cancer, whereas subtypes in non-muscle-invasive bladder cancer are less well-defined. The clinical utility of these subtypes is unclear when factoring in disease grade and stage. In this conversation with Ashish Kamat, MD MBBS, Eugene shares the background of this data and study in identifying predictors and classifying the secondary muscle-invasive bladder cancer after progression from non-muscle invasive bladder cancer. Research that supports how to better individualize and personalize treatment for patients with bladder cancer.


Eugene J. Pietzak, MD, Assistant Attending Surgeon, Memorial Sloan Kettering Cancer Center, Department of Surgery, Urologic Oncology Service, Assistant Professor, Weill Cornell Medicine, New York, NY

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: Hello, and welcome to UroToday's Bladder Cancer Center of Excellence. I'm Ashish Kamat, Professor of Urologic Oncology at MD Anderson Cancer Center in Houston, Texas. And it's my pleasure to welcome today, Dr. Eugene Pietzak, from Memorial Sloan Kettering. He's an Assistant Professor at Memorial Sloan Kettering and is the recipient of the BCAN Young Investigator Award.

Dr. Pietzak is going to present to us today, his background data and study in identifying predictors, and classifying the secondary muscle-invasive bladder cancer, after progression from non-muscle invasive bladder cancer.

So with that, congratulations, Eugene, and the stage is yours.

Eugene Pietzak: Okay. Thank you very much, Dr. Kamat, and thank you UroToday for the invitation. And of course, thank you to the Bladder Cancer Advocacy Network for this Young Investigator Award.

Here are my disclosures. None of them are relevant for this particular presentation, other than the research funding, which again, is highlighting a recent award from the Bladder Cancer Advocacy Network.

So as most probably listening to this presentation are aware, the treatment paradigm for patients with muscle-invasive bladder cancers, is generally, to receive cisplatin-based neoadjuvant chemotherapy, for those that are eligible for it, based on the 5% to 7% survival advantage, over proceeding with surgery alone. However, this survival advantage is really only seen in patients responding to chemotherapy. And there is real concern that patients that are not responding to chemotherapy may be harmed by delaying curative surgery.

There has been a lot of work, both in the molecular subtypes, as well as based on DNA damage repair genes, and other investigations, to identify biomarkers of cisplatin chemotherapy response. Just highlighting the work from Elie Van Allen, as well as Jonathan Rosenberg, a collaboration between the group at MSK, as well as Dana-Farber, where they did whole-exome sequencing of 25 responders and 25 non-responders. And what popped up as being particularly enriched in patients responding to cisplatin-based chemotherapy is this DNA damage repair gene, ERCC-2.

This is unique within the bladder cancer TCGA, where it's mutated at about 12% of all tumors, but it is particularly enriched in patients responding to cisplatin-based chemotherapy, and it is most commonly mutated in bladder cancer. So it is this unique mutation within this gene.

So as a fellow, when I was in the David Solit lab, I was working with my mentors, Dr. David Solit, as well as Bernie Bochner. And at that point, we were performing targeted exome sequencing of patients with newly diagnosed non-muscle invasive bladder cancer to try to describe the genomic landscape of that disease space. And what really popped out to us was that ERCC-2, the same gene that's particularly sensitizing to cisplatin-based chemotherapy, was also highly mutated in high-grade non-muscle invasive bladder cancer and that this was associated with a higher tumor mutational burden.

This was further validated more recently, by the group at Dana-Farber, again, finding a high rate of ERCC-2, missense mutations in high grade non-muscle invasive bladder cancer. And again, this corresponds to a higher tumor mutational burden. And this is of particular interest because there are compelling data that a higher tumor mutational burden correlates with improved response to an immune checkpoint inhibitor, as seen here. And that could lead to potentially this so-called long tail of response. And we know, BCG as probably the original cancer immunotherapy, also has a long tail for those responding, was noted as early as 1991.

And then the Dana-Farber group also recently showed that in patients with high-grade T1 non-muscle invasive bladder cancer, that ERCC-2 was particularly enriched in patients who described it as having a good outcome, which they defined as no recurrence with greater than four years of follow-up. And that good outcome was also associated with a higher tumor mutational burden.

So this raises the question, if ERCC-2 mutations are sensitizing to BCG, does intravesical BCG potentially result in clonal selection for tumors that recur that would be resistant to subsequent chemotherapy? This hypothesis was first purported almost 20 years ago at this point, by this previous European urology publication, that basically suggested that patients with secondary muscle-invasive bladder cancer had significantly worse outcomes than patients with primary muscle-invasive bladder cancer. And I've noted about 40% of the patients received systemic chemotherapy.

So this is obviously, a testable hypothesis, but previous efforts have largely excluded patients treated with neoadjuvant chemotherapy. And so this led us to ask the question, is there a difference in response rate to neoadjuvant chemotherapy for patients with secondary muscle-invasive bladder cancer?

And so we performed a retrospective review of the Memorial Sloan Kettering experience, and we did in fact, find that patients with secondary muscle-invasive bladder cancer treated with cisplatin-based chemotherapy did have lower pathological response rates, as well as lower complete pathological response rates.

This held true, even when we adjusted, on multivariate analysis, for typical factors like clinical-stage, and this correlated with worse recurrence-free survival, bladder cancer-specific survival, as well as overall survival, for patients with secondary muscle-invasive bladder cancer, compared to primary muscle-invasive disease.

What was even more interesting, is that for patients with primary muscle-invasive bladder cancer, we saw the expected pathological downstaging, as well as survival benefits, for cisplatin-based chemotherapy, for which the guidelines are recommending it over cystectomy alone.

However, looking at patients with secondary muscle-invasive bladder cancer, we saw no difference in pathologic downstaging between those treated with neoadjuvant chemotherapy, and those not treated, with just cystectomy alone, and this actually corresponded to worse survival, actually, for patients treated with neoadjuvant chemotherapy, if they had secondary muscle-invasive disease. This may potentially be because of the increased delay in time to cystectomy for patients receiving predominantly gemcitabine cisplatin-based chemotherapy.

This is a retrospective study with some limitations. There is a selection bias in who is undergoing cystectomy. There's also a referral bias, as many of these patients were managed at an outside hospital, and they were only referred at the time of progression. So there could be issues with clinical under staging that we are missing here.

However, we subsequently also performed a genomic analysis, comparing pre-chemotherapy specimens for patients with primary, versus those with secondary muscle-invasive bladder cancer in the TCGA, and who had been previously sequenced as part of other research projects here at MSK. And what we found is that there was this enrichment in ERCC-2, for patients with primary muscle-invasive disease, potentially explaining why they had better response rates to cisplatinum-based chemotherapy.

We then further validated this in a prospective cohort of patients treated here at MSK. Again, seeing ERCC-2 missense mutations, almost exclusively in patients with primary muscle-invasive disease, and not with secondary.

And so this led us to speculate that potentially, as we are seeing high rates of ERCC-2 in high-risk non-muscle-invasive bladder cancer, but not seeing any really in the patients with secondary muscle-invasive bladder cancer, that there may be some clonal selection that's ongoing. And this difference may be explaining the differing responses to neoadjuvant chemotherapy.

So, at this point in time, based on this preliminary data, and the support from the Bladder Cancer Advocacy Network, we're seeking to prospectively validate these findings within an "intention to treat" cohort. We are also seeking to externally validate this with this sort of all-star team of collaborators, individuals at Johns Hopkins, as well as Columbia University, which we are very excited to be collaborating with further on another additional project.

This begs the question as can we identify clonal selection between the pre-BCG non-muscle invasive tumor and the post BCG specimens for those that subsequently developed secondary muscle-invasive bladder cancer? Or, are these tumors born to be bad? And can this be identified in a pre-treatment non-muscle invasive tumor, which could inform patient management to that point in time? Because of course, one of the most difficult decisions we have, or the tough conversations we have with patients, is when is the appropriate time to perform radical cystectomy, and the findings from this investigation may provide further insight into this.

Furthermore, we looked only at semantic differences between primary and secondary muscle-invasive bladder cancer, but there may be important germline differences as well.

And so more recently, Maria Carlo here at MSK performed an investigation, looking at pathogenic and likely pathogenic germline variants, and there were patients who had initial non-muscle invasive bladder cancer. However, at the analysis, these patients were clustered together with those with muscle-invasive bladder cancer.

Another group, from China, recently also reported, suggesting in their small cohort of patients, that patients with secondary muscle-invasive bladder cancer may actually have more pathogenic germline variants in DNA damage repair genes. The caveat being, that there wasn't any treatment detail about whether or not these patients actually received BCG.

There are germline polymorphisms studies that also suggest that ERCC-2, as well as other DNA damage repair gene polymorphisms, may be associated with differing responses to BCG. So this is something we are also actively investigating.

And then finally, we know that BCG as immunotherapy has an effect on the tumor microenvironment, and could this potentially also contribute to rates.

This is a figure summarizing some of the key cells that have been implicated in BCG response, based on a review that I'm currently working on with one of our incoming fellows, Dr. Chu. We know that the pre-treatment tumor microenvironment is important in response to BCG, and certainly, there are going to be differences in the post-treatment BCG tumor microenvironment, although this is less well-defined.

More recently in a collaboration with the UNC group, we also identified that the pre-treatment tumor microenvironment, a so-called hot tumor microenvironment, and immune high, was associated with a more favorable response to BCG. And then we subsequently validated this in the UROMOL cohort, with a small cohort of those patients that were treated with BCG. However, within the entire UROMOL cohort, there was no difference in this immune signature, suggesting that this could potentially be prognostic. The significance of this grant is that there may be a more immunosuppressed tumor microenvironment in patients who have been previously treated with BCG and then have progressed to secondary muscle-invasive disease.

There is some emerging data that the tumor microenvironment may have an effect on chemotherapy responses in muscle-invasive bladder cancer in some cohorts of patients, both treated with neoadjuvant chemotherapy, as well as adjuvant chemotherapy. The tumor microenvironment appears to potentially be predictive or prognostic of response.

And so this leads us to ask the question, could this cold tumor microenvironment in the setting of BCG failure, explain these observed differences? So this is something we are further investigating. We had previously looked at differences in molecular subtypes and did not see a difference. But we're also further investigating that at a larger cohort as well.

So thank you for your time and your attention. And of course, thank you to BCAN for this Young Investigator Award. And of course, for my mentorship team, who I'm incredibly grateful for, to have their support and their expertise, to help with this.

Ashish Kamat: Right. Thank you so much, Eugene, very well summarized and congratulations once again, on the award, very well deserved. When you listen to what you've been presenting, and obviously, it's critical science that needs to be investigated and clarified. But if we step back a little bit, one of the factors that have always confounded any of these other studies that look at either predicting response to BCG or sub selecting patients after BCG has not worked, has been the hodgepodge of regimens that have been used for BCG. So again, if you look at more recent publications from Dyrskjot's group and others, a very small percentage of patients actually got maintenance BCG, which is critical when it comes to truly determining if patients have responded, or not responded, to BCG. I'm sure you have thought of this, but how are you factoring that, into your studies?

Eugene Pietzak: Yeah, I think, that's an excellent question, and I think that's a great point. I think that is, in part, is why we are looking for external validation as well, with both our collaborators at Columbia as well as at Hopkins. And that will be something we will definitely investigate. I mean, obviously, there is a very strong relationship in maintenance BCG preventing recurrence, that relationship with progression to secondary and muscle-invasive disease. Although there's definitely a signal there, it's not nearly as robust as in preventing recurrence, but that is something we are definitely looking to further investigate as robustly as possible. So, that is something we are definitely looking into though.

Ashish Kamat: Okay. Great. Yeah. As you're aware, BCG has been one of the key things that I've been involved with for many years, and one of my fellows, was in my lab, is Justin T. Matulay, who has obviously graduated now, but the series that was probably published actually shows that in the current era, BCG maintenance, when given to patients, truly reduces progression to close to single digits. And then Pat Hensley, who's a current fellow in my lab, but now going on to the clinic, just had a publication that just came out a week or so ago in The Journal of Urology, that looked at the very question that you sort of looked at.

And when we only look at patients that have had induction BCG and maintenance therapy, in the so-called adequate amount, as the baseline defined by the FDA, their response to neoadjuvant chemotherapy, and the pathologic downstaging, et cetera, et cetera, sort of equalizes. So again, we do not know whether it is the early failures that are being weeded out based on biology, or is it truly a non-response to BCG. And I think the work that you have outlined will really help delineate that. And again, if you need other sites for collaboration, we'd be happy to help as well.

Eugene Pietzak: Great. No, thank you for that. Thank you for that very much. I probably will take you up on that. I'll definitely have to check out that publication. I wasn't familiar with that. I didn't see that published.

Ashish Kamat: Let me give you the last 30 seconds, a take-home message for our audience from what you presented today.

Eugene Pietzak: Sure. So thank you again for the invitation, as well as this award from BCAN. I think the important thing from this work that we are trying to garner is, just trying to figure out how to better individualize and personalize treatment for patients with bladder cancer. Certainly, the future is going to be further molecular profiling, whether that's molecular subtypes, or based on the DNA profile of patients, but what is potentially also relevant is the previous treatment exposures, as well.

And certainly, the tumor evolution, and the treatment selective pressures, always need to be considered, especially when we are dealing now with non-muscle invasive patients who are not just being treated with BCG anymore or intravesical chemotherapies, but there's a whole line of treatment options coming down from pembrolizumab, instiladrin, et cetera, all these additional lines of therapy. This is obviously an evolving space, but I think at the end of the day, we always need to keep in mind that there is this tumor evolutionary pressure that is ongoing on the tumor, and that has implications for subsequent treatments. And so, sort of doing a deeper dive into the underlying biology of the tumor, that is the present tumor, is going to be increasingly more important for patients with both high-risk non-muscle-invasive bladder cancer, but also obviously, muscle-invasive and metastatic bladder cancer as well.

Ashish Kamat: Right. Once again, thank you for taking the time, and congratulations, and keep up the outstanding work.

Eugene Pietzak: Great. Thank you very much.