Management Trends & Outcomes of Patients Undergoing Radical Cystectomy for Urothelial Carcinoma, the USC Experience - Anirban Mitra

July 4, 2022

Anirban P Mitra joins Ashish Kamat in highlighting the University of Southern California's experience in the management trends and outcomes of patients undergoing radical cystectomy for urothelial carcinoma of the bladder. This has been updated to reflect current practice patterns, including neoadjuvant, adjuvant therapy, and stage migration. This study looked at clinical-pathological as well as outcome trends of patients undergoing radical cystectomy for urothelial cancer at USC following a consistent surgical philosophy in the face of changing management paradigms over time with the aim being to compare and contrast both historical as well as contemporary cohorts in order to evaluate how this temporal evolution of high-risk urothelial carcinoma management has impacted outcomes following radical cystectomy.


Anirban Mitra, MD, Ph.D., Urologic Oncology Fellow, Department of Urology, The University of Texas MD Anderson Cancer Center, Houston, TX

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 and Cancer Research at MD Anderson Cancer Center in Houston, Texas. It's a great pleasure to welcome today, Professor Anirban Mitra, who is way ahead of his years as far as maturity and experiences concerning bladder cancer. I have to introduce him as a current fellow at MD Anderson, but this is work that he has performed when he was a resident at USC in Southern California talking about the management trends and outcomes of patients undergoing radical cystectomy for urothelial carcinoma of the bladder.

As we all know, the USC experience has for many years, it has been sort of the benchmark for patients undergoing radical cystectomy without additional therapy. Anirban, you actually updated this series to reflect current practice patterns, including neoadjuvant, adjuvant therapy, stage migration, et cetera. So it's a pleasure to welcome you, and the stage is yours.

Anirban Mitra: Thanks so much for the very kind introduction Dr. Kamat, and thank you to UroToday for this opportunity to promote our work. As Dr. Kamat mentioned, I will be discussing some salient aspects from a recently published report looking at the evolution of the USC radical cystectomy experience for urothelial carcinoma of the bladder, really, as it broadly pertains to management trends as well as outcomes.

I have no relevant disclosures. The UroToday audience is certainly aware that radical cystectomy is the gold standard of treatment for high-risk, invasive urothelial carcinoma of the bladder. Our study looked at the clinical-pathological as well as outcome trends of patients undergoing radical cystectomy for urothelial cancer at USC following a consistent surgical philosophy in the face of changing management paradigms over time with really the aim being to juxtapose both historical as well as contemporary cohorts in order to evaluate how this temporal evolution of high-risk urothelial carcinoma management for the bladder has impacted outcomes following radical cystectomy.

To this end, we looked at consecutive patients who underwent radical cystectomy for bladder cancer with intent to cure between 1971 and 2018 at USC. The cohort included 3,347 patients, 80% of whom were males with a median age of 68 years, and a median follow-up of 10.1 years during which 31% of patients recurred with a five-year recurrence-free survival probability of 65%.  Approximately 57% of patients died with a median five-year overall survival probability of 55%. Clinical course perioperative management and pathological characteristics were evaluated using uni-variable as well as multi-variable models.

When we looked at overall trends, we noted an increasing trend for robot-assisted radical cystectomy since 2009, although the open approach was still used in over 70% of patients in the last decade. The median postoperative hospital stay has decreased from nine days since the implementation of ERAS protocols that we are really familiar with in 2012 to currently being a median of five days. And while there was a decreasing trend towards the proportion of ileal conduit urinary diversion, as well as an increasing proportion of orthotopic neobladders until about the early 2000s or so, the last decade really, again, witnessed an increase in ileal conduits associated with a higher percentage of patients who were both older and had significantly more comorbidities as seen across most academic centers.

In addition, while our extended pelvic lymphadenectomy template has essentially remained consistent since the 1980s, specimen submission was changed from unblocked to anatomically-defined nodal packets in 2002, and this was met with a consequent increase in node yield. You can see from the red curve that neoadjuvant chemotherapy administration has increased over time with a concomitant decrease in adjuvant chemotherapy shown in the blue curve, and this was especially notable in the last decade. This really also corresponded well with the uptick of patients with pT0N0M0 disease or essentially those experiencing complete pathological response at the time of radical cystectomy, although there was really no change in positive margin status as seen over here in the orange curve.

Important to note that cisplatin-based neoadjuvant combinations have essentially been the mainstay for both these neoadjuvant as well as adjuvant regimens, and we also noted that patients receiving neoadjuvant chemotherapy were significantly younger and had more advanced disease. When we look at the associations of clinical pathologic features and outcomes by multi-variable analysis, the usual suspects were noted to have significant associations. Specifically, outcomes were associated with advanced pathologic stage delayed to radical cystectomy, tumor upstaging, lymphovascular invasion, positive surgical margins, as well as administration of neoadjuvant and adjuvant chemotherapy.

When we examined outcome trends, we noted really no change in recurrence-free survival across the decades. For example, you can see that the median five-year recurrence-free survival probability was 64% during the 1990 to 1999 decade. This only changed to about 66% from 2010 to 2018. However, when we look at overall survival trends, this improved over time. For example, the median overall survival probability from 1990 to 1999 was 51%, and that five-year overall survival probability improved to 62% from 2010 to 2018.

We then assessed the trends in proportions of patients receiving neoadjuvant chemo and their associated pathological stages. As I mentioned previously, the neoadjuvant chemotherapy administration trend actually increased across eras and this corresponded well with increased pT0N0M0 rates. Among patients who did not receive neoadjuvant chemotherapy, the pathological stage was associated with worse recurrence-free and overall survival across strata. However, when we looked at patients who did receive neoadjuvant chemotherapy, there was really no significant difference in the recurrence-free and overall survival probabilities among patients with pT0N0M0 disease and as well as those with organ-confined disease. However, when we looked at the outcomes of these subgroups, they were significantly better than those of patients who received neoadjuvant chemo but unfortunately did not respond and had pathological extravesical and/or known metastatic disease at the time of radical cystectomy.

Basically, when we look at it in a different way, among patients with CT2 or greater disease, those who were upstaged at radical cystectomy following neoadjuvant chemotherapy administration had significantly worse recurrence-free and overall survival compared to those who directly underwent radical cystectomy. However, patients who were downstaged at radical cystectomy following neoadjuvant chemotherapy administration had comparable overall survival and minimally worse recurrence-free survival probabilities compared to those who directly underwent cystectomy.

We should note that our study has some limitations that were in consideration.  Specifically precise metrics of performance status, as well as comorbidities, were not accurately recorded on all patients given the study's retrospective nature and these were therefore excluded from the analysis. In addition, subcategories of variant histology were not examined in detail as they comprised a small minority of the population and this was really beyond the analytic scope. That said, we've separately reported on the outcomes of these variant subgroups previously.

Finally, and most importantly, this population also predates the use of immune checkpoint inhibitors and other targeted therapies for advanced bladder cancer management. In conclusion, we present a homogenous radical cystectomy series with a long follow-up where consistency of surgical philosophy permitted focused comparisons, and this adherence to oncologically-sound surgical principles, we believe, can potentially result in consistent outcomes for an otherwise aggressive cancer. We also do know that while there has been an improvement in overall survival, this trend, at least in our cohort, has not been recapitulated with a recurrence-free survival over time. But that said, our stable RFS rates do mirror contemporary outcomes from other large academic centers.

I would like to echo those patients who are downstaged after neoadjuvant chemotherapy do tend to derive a great benefit, but it's also critical at the same time to identify those who may not respond to conventional cisplatin-based neoadjuvant chemo because these will be the patients who may potentially benefit from novel therapeutic strategies or early radical cystectomy. With that, I would like to thank my co-authors for their support in getting this work published. The study is now in the Feb 2022 issue of The Journal of Urology. Thank you so much again for the opportunity to discuss our work.

Ashish Kamat: Thanks so much, Anirban. That was a very well done presentation summarizing the key points of your manuscript. In fact, you did such a great job summarizing the findings of the manuscript that it opens up our discussion and allows me to ask you some other questions because you have, during your residency, and of course, during your fellowship over here, establishing yourself as a content expert in many ways when it comes to bladder cancer. So I'm going to put you on the spot a little bit, okay? Are you ready?

Anirban Mitra: Yep.

Ashish Kamat: First off, tell me, was there anything about your findings that surprised you?

Anirban Mitra: Yeah. I think we were really hoping in our heart of hearts, if you will, of some improvement in recurrence-free survival assuming that's one of the key parameters, if you will, of oncologic outcomes. That said, we've looked at and analyzed trends from our cohort, more recently in an internal analysis as well, and did not see any RFS differences. That did turn out to be the case. We, and again, I haven't gone through all the rigorous statistical analyses that we've actually laid down in both the paper as well as the supplementary data, but no matter whichever way you look at it, and no matter how you try to balance these variables out by multi-variable analysis, there is no improvement even with an increase in neoadjuvant chemotherapy administration. That said, again, there are some potential explanations for it, and I'm sure we will discuss that going forward, too. But this is with the caveat that we have not looked at targeted therapies that are more prevalent with both neoadjuvant as well as adjuvant therapy use nowadays in the realm of advanced UCP.

Ashish Kamat: Actually, that's a perfect segue. Again, based on your read of the data, believe the side that you don't have any statistically-significant analysis to talk about, what do you think are some of the reasons for that?

Anirban Mitra: I think to a certain extent it's got to do also with patient selection, the way in which both at USC and at many other academic centers across the country and across the world, really, over the last decade, patients have been selected for receipt of, in this case, cisplatin-based neoadjuvant chemotherapy based on very guided and well-informed but nevertheless, clinical pathologic-based data. Things like CT2 or greater disease, potential hydronephrosis, and things of that nature, which may be okay from a broad perspective, but it is really not personalizing the candidacy of an individual patient. This probably drills down to the issue of us being able to better identify and define patients who may be most responsive to traditionally neoadjuvant chemotherapy while also identifying those bad actors that may not unfortunately respond, and at the same time may have a delay in radical cystectomy and may in fact benefit from either other types of therapy or probably even early cystectomy.

Ashish Kamat: Those are obviously valid points that you make, and that's sort of like in some ways, stage migration, either up or down or however that happens, which in some ways you alluded to in your presentation. But clearly, you need to have huge numbers, not that you don't have large numbers, but huge numbers to really make those statistical inferences. Let me ask you another question now. In your experience during your stint at USC, and now with your two years of fellowship here at MD Anderson, have you developed a sort of a sense as to the nuances that go into radical cystectomy? I asked that question knowing that you have, but I want you to answer it. And things that you would want to share with, say, some of the trainees that are listening in to this broadcast, as to how to improve outcomes of patients undergoing radical cystectomy?

Anirban Mitra: Sure. The first thing, starting right off at the bat, I would think is very appropriate, and accurate clinical staging is the key. That begins probably at the time of either the presentation to surgeons' practice, be that at the time of the initial TURBT, or probably even at re-staging TURBT, essentially making sure as to what the clinical tumor stage of that patient is, making sure that there is a muscle in the specimen, things of that nature, using enhanced cystoscopy techniques as and where available and indicated.

In addition, when the stage has been otherwise accurately established and patients have gotten to the stage of radical cystectomy, meaning they are either not candidates for bladder preservation or have failed bladder preservation, then in addition to performing a thorough oncologic operation, also ensuring that the node of dissection is performed well. By well, I mean it's not necessarily a numbers game, as much as it's strictly adhering to the lymphadenectomy template. Again, one may debate between extended versus super-extended, and that's the reason why the SWOG clinical trial is still out there and we are definitely, we are waiting, everybody is waiting for the results. But that said, having a very thorough lymphadenectomy template and making sure that all those nodes are cleared out would also help with the oncologic efficacy as part of that operation.

Ashish Kamat: Do you have any thoughts on the migration of patients as far as the continent versus incontinent diversions and how that might have tracked the adoption of, say, robotic cystectomies at USC?

Anirban Mitra: Firstly, pretty much all the diversions are intra-corporeal or, at least during my time, I did not see any extra-corporeal robotic-based diversions. That said, we will do both, ileal conduits, as well as neobladders. That said, there is certainly a trend one way or the other in terms of the types of diversions that may be offered based on both patient comorbidities as well as patient choice. We have typically seen over the last decade, both for robotic cystectomies as well as for open cystectomies that these patients typically are getting older, more frail, definitely more comorbidities. And that with a variety of other things, again, notwithstanding the absolute contraindications to a neobladder, one, we did see that there's been that slight uptick if you will, of ileal conduits. Now, whether that essentially tracks with a type of surgery, the way in which patients are counseled, that is clearly dependent from center to center. But it definitely from an objective perspective, does track some of those patient characteristics that I just mentioned.

Ashish Kamat: Great. Again, if I were to ask you, based on your experience with radical cystectomies both during your residency and fellowship, what would your preferred recommendation to a patient be as far as diversion is concerned?

Anirban Mitra: I think it's really multi-factorial. Clearly, there are some patients who, for personal reasons or the other, would absolutely prefer one diversion technique versus the other. By this, I mean, one would consider incontinent versus continent; and continent, meaning, either a catheterized channel or an orthotopic neobladder. That said, beyond those absolute contraindications for an orthotopic neobladder, it's really got to do with shared decision making while at the same time also discussing with the patient the operative time, as well as some of their potential comorbidities.

That said, there definitely have been cases that I have seen, both operatively as well as from a long term post-op follow-up perspective, wherein patients are necessarily older but not necessarily that they have significantly more comorbidities and is at least in our limited experience at USC, we've not seen that in and of itself would be an indication or a contraindication, whichever way you want to put it, for an orthotopic neobladder. Again, it's so multi-factorial, but it's definitely based on those discussions between the surgeon and the patient.

Ashish Kamat: Again, I know that you know all this, and that's why I'm asking you these questions, but it's a very important point to make. It's just because one institution has a strong preference towards orthotopic, and maybe some of the institutions don't, is not a real reason to push patients one way or the other. It clearly depends upon patient preference. The other thing is that simply because a center or centers may be adopting minimally-invasive techniques does not mean that they should go away from oncologic principles, which most don't, but also the quality of life principles which diversions are clearly based on, the quality of life.

Anirban, you talked a little bit about nodule packaging and yield in the node counts. You talked a little bit about, again, the nuances of the technique. Again, just at a very high basic level, what would you recommend to people that are getting into this field, getting into performing radical cystectomy as far as your recommendations on open versus robotic, how to package the lymph nodes, et cetera?

Anirban Mitra: The key, I think, irrespective of whether it's open versus robotic, the key has to be thoroughness of the lymphadenectomy as opposed to whichever way one wants to do it, be it packaging, be it just sending all the nodes on the block. But really it has to be with respect to the defined templates if you will. Now whether that is standard, extended versus super-extended, again at USC, typically it's the super-extended template, just because that has been the case for a very long time. But that's also clearly the reason why we have the ongoing randomized trial, to look at the extent of lymphadenectomy. But the thoroughness is something that really cannot be compromised. This really should not be changing, be it in the robotic setting versus the open setting including everything, including the presacral, the commons, the internal and external iliac. And again, the superior extent can potentially be debated if one wants to go as far high up as the IMA. Again, for folks who are starting out, including me, thoroughness is absolutely the key.

Ashish Kamat: Just again, for the benefit of our listeners that may not be quite aware of this, it's important to emphasize that when we at MD Anderson and I'm sure at USC say standard lymph node dissection, it's not the [inaudible 00:21:11] operator of FASA only, right? The standard lymph node dissection here at MD Anderson goes all the way up to the common iliac where the ureters cross the vessel. In many places, that's actually considered extended, or in some places, I've even seen it being labeled as super-extended, which really doesn't make sense.

So, a standard lymph node dissection for patients that are undergoing radical cystectomy should include at least all the lymphatics around the hypogastric, the external iliac, and the common iliac, ideally to the bifurcation of the aorta, but at least to where the ureters cross into the true pelvis.

The other issue about packaging lymph nodes, again, in many ways, it's a cost issue. The more packets you send, the more cost there is to the system, maybe not to the patient specifically. The fewer packets you send, the less cost there is to the system, even though our patients do not see the cost. But again, that's something to keep in mind because the bladder cancer role is one of those that has a very high cost to the system and to the government per se.

Anirban, this was great. Again, just to remind folks, you are finishing your fellowship here at MD Anderson. You are going to be graduating in a few months. I have no doubt that wherever you end up, they will be lucky to have you. So congratulations on this fine work and thanks for taking the time and spending it with us.

Anirban Mitra: I greatly appreciate this. You've been very kind, Dr. Kamat and I appreciate the opportunity from UroToday to share our work.