Dr. Seth Lerner presented “Quality Indicators in Radical Cystectomy” at the International Bladder Cancer Update meeting on Tuesday, January 24, 2017.


Keywords: bladder cancer, chemotherapy, radical cystectomy, lymph node metastasis, lymphadenectomy

How to cite: Lerner, Seth M. “Quality Indicators in Radical Cystectomy” January 24, 2017. Accessed Jan 2022. https://dev.grandroundsinurology.com/quality-indicators-radical-cystectomy


Quality Indicators in Radical Cystectomy

I don’t typically put it as a disclosure because it’s not industry funded, but I am principle investigator of a randomized trial of extended versus standard node dissection. It’s an NCI-funded trial. So, I inherently have some bias about equipoise regarding the need to do phase III trials, as Dan mentioned in his talk.[su_spacer]

These are what I’m going to, this is what I’m going to try to cover, in now about 17 minutes. So, I’ll go ahead. And I think that to me the key part, and I use this term of getting it right, and this is really what it’s all about. So, the initial TURBT, exam under anesthesia, all the data that goes into sort of processing and decision making regarding how to treat the patient with muscle invasive cancer is essentially here. What’s interesting to me is there is no data that I’m aware of that would suggest that a complete TURBT or a radical TURBT, particularly in the case of a big T3 tumor, somehow affects or has an independent effect on outcome. And as many of you know we talked about TUR biopsies of the prostatic urethra in the context of BCG unresponsive disease. I also do this in sort of anticipation of trying to figure out what to do with the retained urethra, if patients are a good candidate for an orthotopic neobladder. I’ve never really been comfortable with relying on intraoperative frozen sections, so I’m typically taking patients to the OR to restage them and including pelvic exam to look for T3 or T4 disease, and taking advantage of that opportunity to biopsy the prosthetic urethra. And the short version is that if that apical urethral margin is negative then it’s virtually impossible to have cancer at the true apical urethral margin. And we tell patients before they ever go to sleep that we’re going to do a neobladder. So, that’s just sort of my view of the world. Not everybody agrees with that or practices that way.[su_spacer]

One of the challenges that we have is that muscle invasive cancer we’re pretty bad at predicting occult metastatic disease and we’re pretty bad about under staging. So, we call a lot of things clinical T2, and at least half of those patients end up being P3 or P4, or node positive, and I’ve listed a number of things here that are associated with a higher probability of locally advanced disease. This is, and I’ve put a reference in there, Steve Culp, when he was a fellow at MD Anderson did this very nice work with the group there. And they showed that in patients without any of these risk factors, sort of the low-risk group, had a better outcome, acknowledging though that half of the patients with low-risk disease ended up being upstaged to high-risk disease on pathologic staging.[su_spacer]

We took a look at our own data. This paper is in press, yes in Bladder Cancer. It’s a much smaller group of patients, but we essentially show the same identical sort of features. So, you can risk stratify whether or not there’s, with the idea that well maybe I’ll take the low-risk patients straight to cystectomy and bypass neoadjuvant chemotherapy. But keep in mind that that’s not evidence based. It’s supported by some retrospective data.[su_spacer]

I think we already heard about the impact of positive surgical margins. So, a positive surgical margin is cancer at the inked margin, not ureteral carcinoma in situ, not apical urethral carcinoma in situ. These are true surgical margins. You can see that profound impact they have on outcome. They’re fortunately rare, but this is something that theoretically we have some element of control over as surgeons. So, wide unblocked resection, and getting a negative soft tissues margin, if that’s possible, is an imperative, and a quality measure for radical cystectomy.[su_spacer]

I’m going to talk a lot about, not surprisingly going to talk a lot about lymph nodes. And so, there is an obvious dramatic discrepancy in outcome between node-negative and node-positive patients, but the reality of it is that node-positive disease is curable, and particularly the patient with a single positive node in the true pelvis. So, if you take a patient who is PT2,N1 they have a 50% five-year survival probability. And this is just for your reference on the right side, the current node staging scheme, and this is in a bit more graphic detail. So, for those of you that are not familiar with this N1,N2 are positive nodes in the true pelvis, distal to the common iliac bifurcation and one being one node and two being multiple nodes. And then three is any positive nodes in the so-called extended template common iliac. That used to be M1. M1 node is now in the retroperitoneum. So, it has changed a little bit, but this is something that we’ve been using for quite some time. And the true pelvis includes what I’ve described here. and what’s interesting in the TNM system there’s an a priori requirement for 13 or more nodes in order to sort of accurately stage the patient, and Bernie Bochner has been on this committee and I asked him recently, you know, how did you guys come up with this? And it was a fairly thorough and extensive review of the literature, and as Sam pointed out probably in the realm of expert opinion, database expert opinion. Right Sam? Okay.[su_spacer]

This is a paper that we published quite a long time ago. It’s one of many node mapping studies. And this was an analysis of patients at the time in our cystectomy series that had node-positive disease, and we simply graphically display where all of the positive nodes are in a stage-specific manner. So, you see that these are otherwise organ-confined PT2 cancers, and PT2/3, PT4, and then the composite. And what you immediately take away from, and I’ll show you this again in another slide, is the fairly high percentage of patients, among those who have positive nodes who have node metastasis outside the true pelvis. So, you begin to think about okay well they’re there, I know they’re there, maybe taking those out with an extended node dissection may help these patients live longer, and that essentially is the key sort of scientific question around what should be the proper anatomic extent of a node dissection.[su_spacer]

This is a very nice study from Urs Studer’s group, Beat Roth who was the lead author actually just recently did a fellowship at Anderson and I believe went back to Bern. But they did a mapping study injecting a radionuclide in the bladder, and then using a camera to sort of outline where all of the lymph nodes are. You see this sort of curious absence of lymph nodes here, so how can that possibly be? You’ve got the common iliac vein here, you’ve got the pre-sacral space. Well, Urs Studer had defiantly said you shouldn’t be dissecting medial to the ureters otherwise you’re going to disturb the hypogastric nerves. We could have a debate about that, but there’s clearly lymph node metastasis in this area, and like Harry Herr they typically have not carried their so-called extended node dissection proximal to the area where the ureter crosses over the common iliac artery. But to be fair and to give them credit as you know in prostate cancer and in bladder they’ve made a big point about meticulously dissecting the hypogastric nodes here, and those are pretty easy to leave behind and you have to be pretty careful dissecting back there because the intra-iliac vein sits right there, and it’s pretty easy to get into, pretty easy to fix but you have to make a bit of extra attention to take out those lymph nodes. So, again some interesting work in the anatomic extent of the lymph node metastasis.[su_spacer]

Outcome in patients with node-positive disease is affected by a bunch of things; the number of lymph nodes, I use the term identified by the pathologist, because how we present the lymph node packets to the pathologist it drives how many lymph nodes they identify, what does a pathologist call a lymph node without going into those kinds of details. The number of nodes in both node negative and node positive disease drives outcome, number of positive nodes, the lymph node density, which is the number of positive nodes over the number removed. Pathologic stage of the primary tumor in each node stage drives outcome, and I already mentioned the last bullet point.[su_spacer]

How many lymph nodes does the pathologist need to look at in order to assure us that the patient is say node negative or the flipside of that node positive? And so, this is a study from a consortium that I put together a number of years ago, three institutions. Pierre Kharkevich [phonetic] was our, in Montreal, was our primary biostatistician. And what we basically determined was that you need to identify 40 lymph nodes, and this retrospective dataset in order to identify 90% of the patients that had node-positive disease. So, maybe another rationale for more is better because if I upstage you from say N0 to N plus I’m more likely to give you adjuvant chemotherapy for instance. We know it drives prognosis. If you’re N1 versus N2 or N3, you know, that might also help with decision making.[su_spacer]

There’s a huge cost to not doing a lymph node dissection. So, NX means the pathologist didn’t identify any lymph nodes in the cystectomy specimen. And you go well how can that be? All you guys are saying we’ve got to do a node dissection, we know it’s part of the operation. Well, this is the SEER database, and while it’s not as bad as perhaps it used to be we’re up to 40% of patients either had no or very little lymph nodes in the specimen. You can see that NX has a much worse outcome, and presumably because there’s a lot of node-positive disease in that, or there may have been other compelling, other intraoperative reasons, you know, maybe why the patient didn’t get a node dissection. And then you see it again looking at overall mortality, sort of midway between N0 and N plus, and just caveats here that the absence of a node dissection was associated with worse outcomes only in organ-confined disease.[su_spacer]

A little bit more about the rationale for an extended node dissection, and the surgeons will recognize the anatomy on the right, and the standard node dissection includes external, internal, and obturator lymph nodes, and we’ll identify about 95% of patients who have node-positive disease. So, that’s really important. Skip metastases are uncommon but they are reported up to about 7% of patients. What does that mean? Common iliac node metastasis in the absence of distal node metastasis. But if you extend the node dissection up you’re going to increase your node yields substantially, and a significant number of patients will be upstaged from N1 to N2 or N3. The summary of all of that is that if you have node-positive disease there’s a pretty high probability that you’ve got multiple positive nodes, and that you may have positive nodes outside the true pelvis. This is a very important component of local control. In other words, thorough node dissection as a profound impact on limiting the incidence of local pelvic recurrence.[su_spacer]

This is some work done at Penn using both SWOG data and their own data showing that if you have T3 and/or node-positive disease local relapse rate approaches almost 30%. A number of us have thought maybe that’s a bit high, but this is forming the basis of a new randomized clinical trial using of all things post-operative radiation therapy for these high-risk patients, P3, P4, and node positive, to try to reduce this incidence of local pelvic relapse.[su_spacer]

Does a more extensive node dissection improve survival? Here’s some SEER data published by Badri Konety suggesting that it might, and also a group of patients with node-positive disease the more lymph nodes removed were associated with improved outcomes. So again, some additional data points to suggest that it might provide a benefit, and then I think you already saw these data from SWOG 8710. This was the neoadjuvant trial showing the impact of the number of lymph nodes identified in both the node positive and node negative patients having a profound outcome, as well as impact on outcome as well as surgical margins. Again, emphasizing the importance of surgical quality, the things that we as urologists have some degree of control over.[su_spacer]

This is one of the benefits with Per-Anders being in the audience of these nice collaborative reviews on a whole variety of subjects. This is just one that Derya Tilki led for lymphadenectomy. And the last bullet point indicates that currently the level of evidence is modest at best, low in terms of supporting routine use of an extended node dissection.[su_spacer]

The first sort of take-home message, and I’ve got to really speed up here, is it’s the anatomic extent of the node dissection, the completeness of the node dissection that really drives quality and also has a profound impact on the sensitivity for detecting lymph node metastases. So, you’ll see this diagram again, what is a surgical standard? Is it the true pelvis? Is it, yes, we do that in everybody bilaterally, or should we extend it like that? And it’s a real important question because if you look at other organ sites, and I’ll just throw them all up here, where there’s been a similar sort of body of retrospective data suggesting that more is better, when they did the randomized trial no, there was no evidence that a more extensive node dissection was associated with improved outcome in these three cancer types. The esophageal cancer literature has just recently been updated with a very nice paper in 2016, again showing the same thing.[su_spacer]

And let me just skip through that. So, there’s two trials. The first one was initiated by Juergen Gschwend in the German Oncology Group, and this is the consort diagram, and it’s a randomized trial of extended versus standard. They included T1 patients, no neoadjuvant chemotherapy, with an outcome measure of overall survival at five years. This is the distribution of, they have a very clever and very nice study about using molecular techniques to identify micro metastases in node-negative patients showing again the evidence for metastasis in the extended template. And Juergen presented the final data at ASCO 2016. Their manuscript is in preparation and essentially it was a negative trial. I’ve listed some reasons why it might be a negative trial, not the least of which is it is probably underpowered, and including a more favorable risk group may have driven that.[su_spacer]

We have a trial, this is the one that I mentioned to you, fairly similar design, although we do allow neoadjuvant chemotherapy. I’ll tell you that about half of the patients going into the trial have had neoadjuvant chemotherapy. You can look on the bottom left. We’re aiming for about 564 randomized patients. I’m very happy to say that we’re a patient or two away from completing accrual. It will be three years before we report the primary endpoint, which is three-year PFS, and we’re very excited about these two trials providing an opportunity to clarify the role of an extended node dissection.[su_spacer]

So, let me just kind of, there’s a point I want to get to. Yeah, so Peter and Sam have done a nice job of looking at some of the surgical issues and perioperative issues for reducing morbidity. I won’t belabor the point but except to say that the issue with robotic versus open radical prostatectomy if we learn anything from the p53 trial, from the extended versus standard node dissection studies in other organ sites is that the data to date is good, it’s instructive. The technical barriers have been overcome completely, which I think is an accomplishment in and of itself, but we need randomized trials to answer the question about is robotic cystectomy, or any kind of minimally invasive cystectomy, producing equivalent long-term cancer control? And that I think to a lot of us is really the fundamental issue, so many of these other barriers having been met. And fortunately, I think that we’re going to get some of those answers very shortly.[su_spacer]

Just to finish up on the blood loss; there is a bunch of data in a whole bunch of different organ sites that suggests that blood transfusions and the number of blood transfusions have an adverse impact on outcome. Some hypothesize that it might be an immunologic phenomenon. Here is some data in radical cystectomy, and I’ve just got the multivariate analysis there, and the association of receipt of any blood transfusions having a negative impact on outcome. What’s fascinating is that the American Association of Blood Banks have their revised guidelines. I think as surgeons we really need to be quite familiar with this, and I think it’s been a game changer. I mean we were transfusing patients at a much higher rate, really not following these guidelines, and at the end of the day surgeon judgment, anesthesiology judgement really trumps guidelines. But the truth of the matter is that we don’t need to transfuse a lot of patients that we’ve transfused in the past, and maybe there’s an adverse impact of transfusions.[su_spacer]

And then finally, this is another great study out of Bern. This guy Wuethrich is an anesthesiologist. This is a randomized trial of intraoperative alpha agonist and fluid restriction, and looking at a number of outcome measures, but notably blood loss and blood transfusions. And you can see with some of these fairly simple measures they were able to reduce blood loss and transfusion rate rather dramatically.[su_spacer]

And I’ll just go to, this is in your handout. This is our current algorithm. I think that if you’re doing any kind of volume of radical cystectomies that it’s really worthwhile to engage your anesthesiologist in an intraoperative sort of standardized pathway. They have all of the bells and whistles to be able to precisely measure intravascular volume throughout the operation, you know, given fluids where they’re needed, and do things to reduce blood loss, and again, you know, I don’t really ever have to look up the anesthesiologist with blood hanging and ask what’s going on? How come we’re not communicating? It just doesn’t happen anymore. And then this also translates to a much easier sort or pathway for perioperative pain management. So, that’s up there. Please feel free to take a look at it and thank you very much.