Upper Tract Urothelial Carcinoma - Clinical Case Management – Jonathan Coleman
May 24, 2021
Jonathan Coleman, MD, Professor, Department of Surgery and Urology, Memorial Sloan Kettering Cancer Center, New York, NY
Sam S. Chang, M.D., M.B.A. Patricia and Rodes Hart Endowed Chair of Urologic Surgery Professor Department of Urology at Vanderbilt University Medical Center, Nashville, TN
Sam Chang: Hello, everyone. I have the great honor, actually, to introduce Dr. Jonathan Coleman from Memorial Sloan Kettering Cancer Center. He's a full professor there. I've had the honor and pleasure to know Jonathan for many years. He kept me out of trouble when I was a fellow there at Sloan Kettering, now a couple of decades ago. He's going to enlighten us today on actually... I asked him to look at a case of kind of higher risk upper urinary tract urothelial carcinoma. What he thinks about what he does, why he does what he does, and after he goes through the case, we'll actually talk about different situations and reasons why he made choices that he did. Dr. Coleman, thank you again for participating in this and enlightening us. Thanks.
Jonathan Coleman: Well, thank you Dr. Chang, and thank you to the organization for putting this together. This is I think, a nice opportunity to be able to share some thoughts on a fairly rare disease, but one that we're seeing much more of and that's upper tract urothelial carcinoma. This is a disease process that in the kidney world is sort of the bladder cancer of the kidney world. These are often fairly sick patients and requires sort of a very methodical approach, I think, to manage this disease, especially as we're trying to learn more about this cancer and how it may affect certain unique populations of patients. With that, I'm going to share some slides on a case that I've been managing or that I've managed in the past and just highlight some of the aspects of this unique cancer.
So, this was a woman, 71 year old woman, who I saw who had microhematuria. She was a prior smoker. She had undergone a VATS procedure for lung cancer in the past and also had a history of breast cancer. She had an exam that was otherwise unremarkable and a C-T scan that showed this. So, her workup is we normally would do, would be to get a C-T scan, a C-T urogram for upper tract imaging and do a cystoscopy. Under C-T scan, we identified this mass in the right renal pelvis. Her cystoscopy was unremarkable.
So, in the course of her therapy, she underwent a diagnostic ureteroscopy. I just want to make the distinction here between a diagnostic ureteroscopic procedure and a therapeutic endoscopic procedure. During that diagnostic ureteroscopy, which is defined really as an examination designed to provide some information about the extent of disease and to do a biopsy for histology and grade in the upper urinary tract. It is not really intended as a therapeutic intervention. So, this is a procedure that's meant to be fairly brief, to provide very limited information about what we're dealing with. As result of that procedure, we found that she had low-grade upper tract urothelial carcinoma. Again, we biopsied the tumor at that time a then we provided a brief laser fulguration to the area just to prevent and stop any bleeding. One of the things you want to avoid during this part of the procedure is something that could potentially lead to tumor spillage outside the kidneys.
This is designed to be a very gentle procedure. After that result came back, we then had a discussion with her about options for management. She elected to proceed with a therapeutic ureteroscopic procedure, in which at which point we were planning to eradicate all sites of tumor inside the renal pelvis. That's a slightly different operation, which is now intended to go into the kidney, remove all evidence of tumor and to be a therapeutic intervention for the patient. That was completed with a laser fulguration.
As far as following guidelines or management for this approach, these have been fairly well delineated by the E-A-U in terms of identifying those patients who may be eligible for a Pro Trac endoscopy management. This really defines a group of low-risk patients. Those that have to have a unifocal disease, ideally. Those with a fairly small amount of tumor burden, a low-grade on a biopsy and on cytology, and really no evidence of invasion on imaging. Those features that are associated with high-risk are delineated here. Those include any of these features. Any features such as hydronephrosis, tumor size actually should be greater than two centimeters in size, high-grade cytology, multifocality, high-grade on biopsy. They note that patients with previous radical cystectomy for bladder cancer is included in their guideline. This again, is the European guidelines, but we generally follow these similar guidelines here in the U-S.
I'm just going to show a couple of quick pictures of what her tumor looked like. This was a white light version of what the tumor appeared to be. This is a papillary lesion. We knew that coming across very well and this is something that we deal with sometimes. That is, not really well, being able to see the borders of a tumor like this. Sometimes we'll use something called narrow band imaging. There's been a number of publications about now band imaging and how it may help you to identify other sites of disease in the renal pelvis. We use a laser then to eradicate all sites of disease with a multitude of different imaging modalities and techniques that we use during the procedure.
Then, follow up afterwards. This is the pathway that we use at Memorial. I wouldn't say that this is a standardized pathway that all centers use, but generally we follow patients for the first...after their first treatment, we follow them every four to eight weeks with repeated endoscopies to look for any evidence of recurrence in the upper tract. Then, while we're undergoing that process, if it requires two or three sessions to clear out the kidney, we also want to make sure that we're following with C-T endoscopy or C-T urograms at the same time. We don't want to forget that there can also be other areas of disease to worry about. Once we get to the point where a patient is N-E-D, that means we can't find any evidence of disease on endoscopy, we then switch them over to a less aggressive strategy. That is typically following them up with endoscopy every three to four months and a C-T scan. Again, every six months. Once they get past another year with no recurrences, we then move to an even more abbreviated up for the patients afterwards.
This is based on our clinical experience. Just to describe what patients can expect from management for...with low-grade disease in the upper tract with endoscopy, this is a paper from Mike Grasso's group, just showing in a fairly large cohort with long-term follow-up that these patients can do, similarly, as far as cancer specific survival, when compared to patients who've undergone radical nephrectomy or nephroureterectomy for also a low-grade disease. Then this was a review paper that Mark Cutress and Dave Tolley put together from the European experience, showing that these approaches can be successful in eradicating tumor in the upper track or controlling tumor in the upper track. It's important to note that recurrences are highly likely, and that is, that over 50% of patients will experience recurrences in the upper tract during endoscopic management follow-up. So, just to get back to her case, we did her procedure in over a 12 month period of follow-up.
There was no recurrence in her upper tract. Her endoscopic management appeared to show a clearance of the disease and she was doing well. Just to summarize the approach for endoscopic management for upper tract patients, this may be appropriate in highly selected individuals. Those with low-grade tumors, small volume cancers, and especially those with other risk factors, such as Lynch syndrome, a solitary renal unit, or those who are otherwise poor surgical candidates.
It's notable that these patients are at very high-risk for recurrence in the upper tract, so they need to be followed very closely. There should be very well-defined thresholds for converting these patients to nephroureterectomy, including patients who progress to high-grade disease or show a high index of proliferation clinically. It's really important to make sure that your patients are prepared for this, so we need to focus on patient education and make sure that their expectations are met. We followed her clinically for three years and follow-up on her C-T scan showed that she now had a recurrence in her contralateral kidney. In here, you can see in the upper pole of her left kidney, now there was a new mass that was picked up on her C-T scan. She had never developed any tumor in the bladder, in the interim.
We performed a ureteroscopic procedure on her, now at this point in time, and we also performed genetic testing because she had contralateral disease and there was a concern as to why she was forming these tumors. What this biopsy now showed is that she had high-grade tumor in her contralateral kidney. While it was low volume, the presence of high-grade cancer was a concern. We also discovered that she had Lynch syndrome and this is something that is not unusual to find in our patients. Now that we're looking with genomic testing more and more, we're finding that a large proportion of patients with upper tract disease have Lynch syndrome. How does this change the prognosis for this patient? Well, just the finding of high-grade disease alone is associated with a greater than 50% five-year mortality rate, regardless of patient's stage. Now, certainly higher stage cancers are associated with a higher risk of progression, but high-grade disease by itself defines a high-risk cohort of patients.
After discussing options with her in terms of next steps, she obviously was very concerned about the risk for disease recurrence down the road, as we were as well. Another concern in this patient population is the risk that comes with nephrectomy that leads to chronic kidney disease. This is a study showing that the impact of chronic kidney disease that may be impacting these patients in terms of their risks of long-term survival, including cancer-specific survival and overall survival. Functional renal loss is a major issue for these patients.
After discussing with her the next best steps, we explained that there was a good chance that if her cancer progressed after we had removed her kidney, that it'd be unlikely we'd be able to provide her with a meaningful form of treatment with chemotherapy agents in the post-operative setting. For that reason, she elected to go on to our clinical trial, which was a prospective study evaluating the use of chemotherapy in the neoadjuvant setting. She received Gemcitabine/Cisplatin times four doses. She underwent an uncomplicated robotic assisted nephroureterectomy. Her pathologic stage after chemotherapy was P-T-A high-grade non-invasive and zero disease. At four-years post-operative follow-up, she's been doing well with no recurrence.
This is the result from that trial that were presented at the A-U-A last year and are now in press looking at the end points and the outcomes of a neoadjuvant trial of Gemcitabine/Cisplatin. We noted that our response rate was well over 60% in patients who are undergoing therapy. A small number of patients were non-responders; in other words, those who had an outcome of greater than pT1 disease. It was notable in this series that up to 19% of patients actually had pT0 at the time of final pathology. As far as the end point of pT1 disease after neoadjuvant chemotherapy, what we saw is that in terms of progression-free survival and overall survival, those patients who had less than T2 disease after neoadjuvant chemotherapy, actually did extremely well in terms of survival outcomes. This matches, or at least is consistent with data from the Powell trial that was published two years ago, showing that the use of adjuvant chemotherapy after nephroureterectomy can also improve overall survival and disease-free survival in patients who have had an upper tract urothelial carcinoma.
The problem in this group, obviously, is that a large number of patients who undergo nephroureterectomy are unable to receive chemotherapy after their surgery because of poor kidney function. That was shown very well here in terms of the effects of the use of effective forms of chemotherapy, such as Gemcitabine/Cisplatin when compared to ineffective forms such as Gem-Carbo. Really, the ideal time to give chemotherapy to these patients is when they can receive Gemcitabine/Cisplatin. That may be in the neoadjuvant setting for patients with poor underlying existing C-K-D. To summarize, I think some of the key salient end points of a case like this, that is that it should be recognized that upper tract diseases is a complex disease, requiring risk stratified approaches to patient management. These really should be evidence-based, based on standardized management strategies and a number of guidelines have now been developed to help lead clinicians through that process.
As far as organ-sparing endoscopic treatment, that certainly can be appropriate for selected patients. Those should really be in patients with low-grade tumors and low-risk features. There are evolving techniques and new therapeutics being developed in this space, including new agents that can be used as chemo-ablative agents in the upper tract. This is something I think we'll hear more about in the future as these techniques evolve.
It should be recognized that organ-sparing endoscopic management should also take into account patient's underlying risk factors, such as Lynch syndrome. We're finding that Lynch syndrome is present as many as 10 to 15% of patients with upper tract disease. It needs to be looked at more carefully. As far as identifying or defining a high-grade or high-risk cohort, any patient with high-grade cancer is by definition, a high-risk cohort. In this group of patients, combination strategies need to be considered, whether in the neoadjuvant versus adjuvant setting with chemotherapy. Effective surgical therapies and treatments can certainly be provided with both robotic and open techniques, whatever clinicians feel they're most comfortable with. Then the role of lymph node dissection in that setting is one that's being explored.
Sam Chang: Fantastic presentation, Jonathan. I just want to make sure that the audience knows, he was too modest to mention this, but Jonathan is the chair of the newly formed A-U-A guidelines committee for upper tract urothelial carcinoma. So, you really heard it from a real expert in terms of what's being evaluated in terms of his last side, try to summarize the importance of evidence, risk stratification and treatment. I'm a nitty gritty kind of person in the trenches, as many of the clinicians are. So going through the case, we've got a few minutes. I want to just ask you some specific questions regarding care as I was thinking about things. Let's think about earlier in her scenario with low-risk disease, how do you biopsy this tumor? Give me just a quick idea of how you do it and your technique.
Jonathan Coleman: That's a great question. One of the biggest problems we have in the upper tract is providing good data regarding grade and stage of cancers. In fact, we really can't provide very good staging information from the Pro Trac based on endoscopy. Unlike in the bladder, where we can do a good T-U-R, we can have get tissue from deeper along the lining of the urothelium. We really can't do that in the renal pelvis. I know there are different techniques out there. Some people like to use something called the BIPopsy forcep. I'm not a big fan of it because I think it's hard to fit through a scope atraumatically, and I think it does have some risks. One way that I've found to be able to do effective biopsies is to use a basket. So a basket can be used to encircle the tumor.
The key is not to pull the basket fully in. You want to close it most of the way, but not all the way. Once you've got it pulled or closed most of the way, you then just start to kind of tug on the tumor and eventually you'll get some sample that comes off. Then you can withdraw that fairly atraumatically. Another way to do it is if you have a small cup biopsy forcep, is that instead of grabbing the tissue like you normally would with a cystoscope and pulling abruptly; instead of pulling, you want to push and lift. You grab it with the tip of the cup and then you push away a little bit and you lift and as you lift, you'll feel it pull right off and then you've got a good sample that way and you can take that out through the scope. Using those two approaches, we've been able to get good samples.
There will always be an issue of sample error in terms of where you're sampling from. Obviously, we're sampling from the superficial areas of the tumor and not the deep areas. I always recommend getting a post-biopsy cytology. When you do that post-biopsy cytology, you should try and grab some of the floating tissue or floating cells that may be in the renal pelvis. When you suction that out, you'll get additional tissue that way. That may help to improve your cytology. Any evidence of a positive high-grade cytology in the upper tract, even if you have a low-grade biopsy, should be an indication that you're actually dealing with high-grade cancer and your sample was just an error from one part of the tumor. High-grade cytology with low-grade pathology in our system is consistent with high-grade cancer, not, not with low-grade cancer. They often co-exist.
Sam Chang: To an incredibly helpful point, I think Jonathan, well three; the basket, which I use, the biopsy technique, which I haven't used, which I employ, well now started employing and then getting the cytology after the attempted biopsy. I really liked that idea. Along those lines, if some patients have a carpeting, you don't actually ever see a filling defect on the scan. You've gone up in your ureteroscopy and you see this carpeting and it's really difficult to get a sampling. For you, is it enough at that point, visualization of this carpeting and your cytology comes back positive for high-grade carcinoma? Is that enough? Or do you still need tissue?
Jonathan Coleman: Yeah, in our system that's enough. So, we actually included those patients as being high-grade in our trial by definition. It's likely you're dealing with a situation like C-I-S in that case. By definition, in our series, that's a high-grade cancer,
Sam Chang: Last question on this lower-grade lesion. Laser...holmium or this new thulium...what do you use and what's your initial set that you use?
Jonathan Coleman: Yep, so we have a holmium laser at our center and we just upgraded to the new MOSES laser, although we used a lower energy laser for years before that successfully. I think the key to this is really using low energy, but high frequency. Not really going with something that's likely to perforate through the wall. This is unlike stone disease where you can move a stone where you want it to be and treat it. Here, we're really getting the laser where we want it to go. I think the problem is we can't really shoot around corners with a laser. Where I find that the MOSES setting helps is actually because you get a bubble blast off the end of the fiber with the MOSES technique. That does give you a little bit of ability to treat off to the side of the laser and not what's directly in front of you.
I typically will treat first on the presets, which are the ablation, the tumor ablation settings, which I believe is an energy of 0.5 and a rate, that they give you to start is around eight. I turn it up to around 15 and I try and treat very superficially to start. As we get closer to the wall, I'll switch over to MOSES laser settings on ours, just to give us a slightly different bubble effect. You have to make sure you're not perforating these patients. I think it's very important to avoid perforation. The thulium laser, I don't have experience with, sorry.
Sam Chang: Yeah, no. We've trialed it. I tell you, it is definitely, and the biomechanics behind it, I'll be the first to say I have no idea, but it definitely was more hemostatic and seemed to ablate. The visualization was great. I tend to use a digital scope for these procedures as opposed to our normal scope. The visualization is so much better with the digital scope. The last question from a nitty gritty standpoint, your follow-up protocol as this patient went out...what do you do with the stents or what do you do with stents? Does it vary for each patient or obviously when you know you're going to come back, are you concerned that there's residual disease and you're playing comeback? In that four to eight week window, you probably would stent, but I don't want to speak for you. What about the one that you've lasered? Do you do it like a stone and take it out on a couple of weeks? Do you leave it in? What do you tend to do with stents? I'm sure it's more individualized, but in general, is there a practice?
Jonathan Coleman: Yeah, I think as you pointed out after the first procedure, when I normally do a diagnostic procedure, I will typically leave a stent behind. I think it does help if you're going to go back on that patient, you think you might be going back. We all know how the ureter can get thickened after just having a stent in place. I think having a thicker ureter, that's maybe a little bit more resilient to be able to handle say, if you're going to use an access sheath, I personally don't use a lot of access sheaths, but I know people who like to use them. I think having a slightly thicker ureter may help in that regard because I worry about the ureter splitting or having problems with placing access sheaths in a virgin ureter. I will leave one in temporarily, initially.
However, once I'm starting to manage a patient over time, I typically will start to do these procedures without leaving stents afterwards if I can. I thought there was an interesting paper that was recently out. I don't know what your practice is, but I typically, if I see a tumor in the bladder and I'm seeing tumor in the upper tract, I always like to treat the bladder first before I go to the upper tract to make sure that everything in the bladder is treated. There was a recent paper showing that if you don't treat the bladder first and you end up leaving a stent in patients afterwards, the concern might be that you would see tumor from the bladder into the upper tract through a stent. And now this was from cystectomy literature and so forth from people getting bladder...bladder tumors getting resections and then having upper tract endoscopy done afterwards. I do worry about the seeding and the implantation issue from the lower track to the upper tract with stents.
Sam Chang: Yeah, I think it's a real concern. I struggle with this. I don't know which to do first. If I resect and ablate, am I moving tumor around then and feeding it up, or is it better that I take the upper tract and I leave that tumor still in the bladder that I haven't disturbed? To be honest, I prioritize what's, to me, the more worse of disease. If I've had high-grade tumor in the bladder and I really have to take care of this, then I tend to do that first. If I don't know what's going on in the upper tract and I know I have a low-risk tumor in the bladder, I'm going to go up to treat the ureter first. I don't have a set policy because I struggle with that paper because I don't know what to think of it because I want to, in the majority of times, I would rather have a stent and not have a patient live with nephrostomy tubes.
I still do that preferentially. I want to have a little bit better data before I change my tendencies. I do worry about that. In the interest of time and I think this is just as important...again, nitty-gritty question: robotic or laparoscopic nephroureterectomy? I would say careers have been made how you take the distal cuff on your nephroureterectomy. I will be the first to say that I've probably done more than single digits, in terms of varieties of what I've done, and I continue to change. Can you tell our listeners here, how do you routinely take the cuff in a nephroureterectomy?
Jonathan Coleman: A couple of things. Number one is I think the seeding issue is real. Seeding from the upper tract to the bladder definitely does happen. One of the things that I think is important to do first is when you first start your case, I go right to the ureter. Very first step is I get the ureter exposed down by the by iliacs and I clip it or I tie it right there. I go down as low as I can and I put a tie or I'll put a clip around the ureter just to avoid any more urine getting from the kidney down. I don't know if it's how we manipulate the kidney or whatever, but I think there's clearly a seeding issue that occurs from the upper tract to the bladder. For some reason, it seems to be an issue during a robotic surgery or laparoscopic surgery.
I think it's because we manipulate the kidney a lot more. After the kidney has been ischemic, now we're playing around and we're getting down to the ureter and urine is washing down. Bottom line is I think, early ureteral clipping or control is important. I then go and do the nephrectomy and then as I go down to the distal ureterectomy, I do an extra vesicle approach. I'll put the patient, say in a robotic case, I put them in flex and they're up on their side in modified lateral decub, probably about 45 degrees and I'll actually put them head down. They're almost like a prostate in terms of their Trendelenburg positioning. There'll be pretty far down. What that does is that now allows the bladder to lie very laterally.
As soon as we empty the bladder with the catheter, as soon as you open the bladder to get the ureter out, so you go all the way down to do a formal bladder cuff. First, I place a purse-string suture around the bladder after getting it all the way down. I'll open the bladder and then the bladder will fill with air, fills with C-O-2 and there'll be no leakage at all. There's just no water that comes out, nothing comes from the bladder. We'll remove the ureter formally. I'll typically wrap the ureter and after I cauterize on wrap the ureter in SURGICEL, so that it's contained. Obviously, I clip the ureter right before I clip it again, right at the U-V-J before we take it. As soon as it's off, put it in SURGICEL, tie the purse-string very quickly, cinch it down, run another two layers on top of that, which is easy to do. It's very quick. Then we get the whole thing out. I really have not had problems with perivesical recurrences in my career. I've done hundreds of these cases. I've been satisfied with that approach and haven't had to worry about leakage issues.
Sam Chang: I'm gratified because I'm now doing quite similar to what you say. The V-Loc suture is what I use for my purse-string. I use that to hold up. That allows a very quick repair, maintains tension. You don't have that spillage. I've just recently started, although I'm not a big fan of it, I've been using the sealer just beyond my clip distally to seal and close. I've been doing that instead of wrapping it. I've been cutting distally towards the bladder beyond that seal to make sure. I love the idea of what you're saying in terms of the Trendelenburg. We don't have a bed that attaches to our robots so that we can move, but that makes sense. I need to keep that in mind. In the interest of time, Jonathan, thank you so much. I think everybody's learned quite a bit from you and we're looking forward to the guidelines coming out from the A-U-A that you'll be sharing. Thanks again for your time.
Jonathan Coleman: Thank you, Sam. Thanks to the organizers and the guideline... I think we've got plenty of data now from multiple different clinical trials and lots of large center experiences, multicenter experiences that the guideline I think we'll be able to create will be incredibly informative and helpful for patient care.