Ashish Kamat: Hello and welcome to UroToday's Bladder Cancer Center for Excellence. I'm Ashish Kamat, professor of urologic oncology at MD Anderson Cancer Center, and it's a distinct pleasure to welcome again Pat Hensley, who is here today to talk to us about his award that he received from the Bladder Cancer Advocacy Network, the 2022 John Quale Travel Award to present his outstanding work. As many of you know, Pat spent two years with us here at MD Anderson, did a phenomenal job. Unfortunately, he didn't stay back, but he went back to the University of Kentucky where he is now an assistant professor in the Department of Urology and Pathology. And I have no doubt that you'll continue to rise to even greater heights. Pat, the stage is all yours, enlighten us.

Patrick J. Hensley: Thank you very much, Dr. Kamat. It's a pleasure to be here with you today. Certainly an honor to receive this travel award and I'm excited to share with you some of our data. So current guidelines strongly support the use of cisplatin-based neoadjuvant chemotherapy prior to radical cystectomy for muscle-invasive bladder cancer. And the reason for this strong guideline support is primarily related to level one evidence from two prospective trials, which indicate that cisplatin-based neoadjuvant chemotherapy confers a modest survival advantage. A 5 to 6% reduction in death at five years compared to upfront radical cystectomy. However, in these trials, over a third of patients experienced at least a grade four toxicity. So neoadjuvant chemotherapy does come at a cost.

It's clear we're in need of a better way to select patients for neoadjuvant chemotherapy who can both A, tolerate it and B, exhibit a meaningful tumor response. This so-called risk-stratified approach to neoadjuvant chemotherapy could be refined with clinical predictors, or predictive biomarkers of neoadjuvant chemotherapeutic response. There's some retrospective data that suggests that histologic features such as variant histology, lymphovascular invasion have limited independent predictive potential for neoadjuvant chemotherapy, therapeutic response. And additionally, molecular subtyping using RNA sequencing techniques, it's cumbersome, expensive and not really readily available clinically, which really limits its utility as a predictive biomarker.

So in this study, we investigated whether a simplified immunohistochemical signature and pre-treatment TURBT specimens could predict response to neoadjuvant chemotherapy, prior to radical cystectomy, with the ultimate goal of using tumor biology to risk-stratify patients for neoadjuvant chemotherapy. Our primary aim was to develop an immunohistochemical panel which predicts responses to cisplatin-based neoadjuvant chemotherapy in clinical T2 patients undergoing radical cystectomy. Our secondary aim was to correlate our IHC biomarker expression with both molecular subtype as well as clinical outcomes.

We did so by generating a tissue microarray with TURBT specimens, with 68 cores from patients that were treated with clinical T2 disease with cisplatin-based neoadjuvant chemotherapy, all of these patients received gemcitabine cisplatin. We also obtained corresponding tissue cores for RNA expression analysis using molecular subtyping and based on some of our prior work as well as predictive biomarkers in the literature, we identified a panel of 17 candidate immunohistochemical biomarkers for study. And these involved a variety of cellular processes, including the epithelial-mesenchymal transition as well as the DNA damage response repair pathway. First, we identified immunohistochemical predictors of a complete pathologic response.

Tumors which retained expression of RB-1 and exhibited low expression of the epithelial-mesenchymal transition marker, N-cadherin had higher rates of complete pathologic response with 100% of RB-1 deficient tumors failing to achieve a complete response. Complete response rates were also significantly higher in tumors with low proliferation index as measured by Ki-67 and high expression of the DNA damage response gene ERCC2. Taken together, we developed a four-marker panel, highly predictive of complete pathologic response with complete response rates ranging from 0% in tumors expressing zero out of the four favorable markers and up to 64% complete response in tumors that expressed all four favorable markers. And many of these markers were redundant predictors of a meaningful partial response to chemotherapy as well, not just a complete pathologic response.

Next, we investigated the predictive capacity of these markers to associate with survival outcomes and there are again, four markers which associated with improved cancer-specific survival, including low Ki-67, intact RB-1 expression, low N-cadherin expression and absence of CK-5/6. You can see that on Kaplan-Meier survival analysis for progression-free and cancer-specific survival that tumors that expressed three or more of these favorable markers exhibited excellent survival, while the tumors that expressed zero out of the four favorable markers exhibited universally poor survival outcomes.

Lastly, we sequenced cores from 37 matched TURBT specimens to evaluate the predictive capacity, of molecular subtyping and correlate IHC expression with molecular subtype. You can see that based on the MD Anderson Cancer Center Molecular Subtype classification, that pathologic response and survival were well stratified with p53-like tumors exhibiting the worst outcomes and luminal tumors universally exhibiting the most favorable survival outcomes. However, unfortunately, none of the individual immunohistochemical markers were independently associated with molecular subtype, which would suggest that our immunohistochemical biomarkers cannot be used as a surrogate for molecular subtype.

In conclusion, we've identified IHC biomarker panels that were both predictive of neoadjuvant chemotherapeutic responses as well as survival outcomes after cystectomy. As expected, these panels of candidate markers consistently outperformed individual markers.

Moving forward, I think this work clearly needs external validation, but I truly believe the future of management of surgically resectable muscle-invasive disease is going to be tailored to the patient and the biology of the patient's tumor rather than a one-size-fits-all approach for neoadjuvant chemotherapy for all patients, followed by cystectomy. Our data have identified markers of inherent cisplatin resistance, for instance, and these could be used potentially to identify patients for a timely upfront cystectomy, or an alternative neoadjuvant approach rather than the traditional cisplatin-based neoadjuvant approach. And then on the other hand, we identified markers of cisplatin sensitivity, which could be used in bladder preservation protocols as an adjunct to some of the work being done in clinical trials with DNA damage response gene alterations.

And the ultimate goal of this work is to improve outcomes in patients with muscle-invasive bladder cancer, avoid overtreatment in patients who achieve a profound response to systemic therapy and move on to definitive surgical management in those patients who achieve little therapeutic benefit from systemic therapy.

Again, I really appreciate the opportunity to present our work here today. I want to thank my mentors and collaborators both at MD Anderson and at the University of Kentucky, especially you, Dr. Kamat, for your dedication and mentorship throughout this process. I'd like to thank specifically the Urology Care Foundation for their funding for this project as well, as obviously BCAN and their support through the John Quale Travel Fellowship. So thanks again for your time and the opportunity to present.

Ashish Kamat: Well, Pat, thanks so much for taking the time and sharing your work with us. That was excellent. As always, I'm going to spend the next few minutes asking you questions both about your work and also what the award means for you, because BCAN is a phenomenal organization, supports young investigators, but we all just like to hear from folks like yourself as to what it means for you. So let me start with your work first. Do you share that the IHC, which can be done easily in community pathology labs, has a fairly good predictive value when it comes to patients' response? What is your message for folks that are involved in this field, or up-and-coming folks and also established folks that are looking at molecular subtyping and also on the other hand have the IHC data available? How would you advise them?

Patrick J. Hensley: Well, that's a great question, because the impetus for this project was to identify clinically meaningful predictive markers that were ubiquitously available to both clinicians and researchers. And as you know, molecular subtyping requires a tremendous amount of expense and bioinformatics expertise. There's been additionally some recent conflicting data on the inherent cisplatin sensitivity of various molecular subtypes, and I think a lot of that has to do with just variability in the nomenclature of the subtypes, unfortunately, and as we move toward a consensus classification, I think that's a very important initial step for the bladder cancer research community to come to a consensus agreement on the nomenclature of these subtypes. I think the future of risk-stratified approaches to systemic therapy is likely molecular subtypes, but in the absence of a clinically meaningful, easily accessible, cheap molecular study, I think that the field is ripe for using these immunohistochemical phenotypes to help guide treatment decisions.

Ashish Kamat: Yeah, I think that's an important point that you make, because clearly subtypes are giving us insight into the molecular pathways and the biology of the tumors, but your panel, or panels such as yours, obviously once validated, could serve as surrogates once you match them and identify how well they predict for different subtypes, even as our understanding of the subtypes improves. And let me ask you this because again we, and I include you in the we, have tried to do similar work with subtyping when it comes to non-muscle invasive bladder cancer. One of the issues is that you need anywhere from 10 to 20 slides in order to get the RNA and DNA. And clearly in tumors such as small T1 disease, or especially CIS, that becomes really hard. And that's why it's only been a few folks, such as Dr. [inaudible 00:10:56] and the UROMOL study that have actually managed to give us informative data. Do you think, and what are your plans for looking at a similar panel of markers in non-muscle invasive disease, because that's, in some ways, a bigger unmet need?

Patrick J. Hensley: I agree, and I think a lot of these markers will probably serve as redundant predictive markers, because the markers we chose for this particular study are heavily mutated, heavily deranged, different proteins in various vitally important cell-signaling pathways that we see are both targets of therapeutic modalities that we have as well as common resistance mechanisms to various cytotoxic, or immunologic therapy. So I think that's a very important next step of this study for sure. One issue that I think with both molecular subtyping and protein-based profiling is, how do you quantify intratumoral heterogeneity and temporal expression changes that are induced by therapies? And I think the jury's still out. We don't have a great explanation for whether or not the protein expression and the RNA sequencing data, which of those is more reflective of the overall tumor biology.

Ashish Kamat: Yeah, I think it's a little mix of both, because with your protein expression, you can actually look at the effector with the RNA, DNA you're looking at the message and the trauma is really important, and I think work such as what you're doing is clearly very impactful and I would encourage you to continue the work that you're doing. Maybe also look at non-muscle invasive disease and see what it shows. Let's spend a few minutes now talking about the BCAN experience. So share with us in our audience what you felt and how the meeting affected you.

Patrick J. Hensley: Well, for any junior aspiring bladder cancer investigator, the think tank meeting, the annual BCAN Think Tank meeting, is just the absolute preeminent meeting for like-minded researchers, patients, patient advocates, industry partners, everybody to join together and talk about how to advance bladder cancer research in a clinically meaningful way. This was my first meeting. I think that the meeting has been going on, inaugurally, for the last 17 years, and it's just an absolute honor to be associated with this organization and be part of it and be supported by them. One thing that I really took away that was a unique experience for me, at a scientific meeting, was hearing the patient perspectives. And I think it grounds a lot of the researchers, and it was really nice. There was a patient impact panel, several of them that just really heard straight from the horse's mouth, what's important to patients, what we need to work on as researchers, and how to direct this field moving forward.

Ashish Kamat: Yeah, no, again, I've been involved with the Think Tank and BCAN for, gosh, now going on 16, 17 years, and it's amazing how the organization's grown from a small group of 12, 15 people to now over 200 people. And it's great to have folks such as yourself be part of it. The travel award that they do is excellent, and really it's great to see you getting more involved. So again, thank you for taking the time. Any closing thoughts for our audience?

Patrick J. Hensley: No, I just appreciate the time and I really look forward to this predictive biomarker field in the muscle-invasive disease setting. There's a lot of tremendous clinical trials that are currently looking at bladder preservation protocols, so certainly more to come in this field, and I really look forward to seeing the data mature. So thanks so much for your time and UroToday's time, and then the opportunity to present our data.

Ashish Kamat: Great. Take care.

Patrick J. Hensley: Thank you.