Upper Tract Urothelial Carcinoma - Clinical Case Management in High-Risk Disease - Sima Porten

October 16, 2021

In this patient case evaluation discussion, Sima Porten joins Sam Chang presenting a case of a 64-year-old man with high-risk urothelial carcinoma. He presented with gross hematuria. His workup initially was started with a cystoscopy that was done by a urologist close to home, which was normal. 


Sima P. Porten, MD, MPH, Assistant Professor, Department of Urology, UCSF Medical Center

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, Department of Urology

Read the Full Video Transcript

Sam Chang: Hello everyone. I am Sam Chang. I am a Urologist at Vanderbilt University in Nashville, Tennessee and I have the great honor to actually spend some time with Dr. Sima Porten. Sima is an Associate Professor of Urology at UCSF. And I think all of you are familiar with her. She is no longer the future rising superstar, she is a superstar and I think her national recognition has only increased as she has led multiple national meetings focusing on urothelial carcinoma.

And so we have asked her today to actually give us a case, kind of example, of someone who has high-risk disease, how she evaluated the patient, how the patient was diagnosed, any treatment questions, any issues that came up. And so she has been kind enough to share a case with us. And so Sima, we will turn it over to you and thank you again for giving us and aligning us with some information on upper tract disease.

Sima Porten: Thanks for that kind introduction, So, Sam. I picked a case that I thought was kind of real-world, so it's not perfect, which is great. And so this is a 64-year-old gentleman who presented with gross hematuria. His workup initially was started with a cystoscopy that was done by a urologist close to home, and that was completely normal. He is really healthy. He has some hypertension, no real risk factors, he was never a smoker, no family history of cancer. His creatinine was slightly impaired at 1.3 with a lower GFR. Although it's not too terrible, he did have some low hemoglobin about 12.9. His baseline in the past has been around 15 to 16 and so this was fairly significant hematuria he was having. And cytology was sent at the time of that cystoscopy that appeared normal and was what we get sometimes, which is atypical urothelial cells or Paris 3.

At UCSF, and in many centers around California, our cytopathologists have moved to the Paris system to try to align how we speak about cytology. Paris 3 sort of designates a pretty large window of the possibility of having any kind of urothelial carcinoma. Anywhere between 10% to 40% is what is stated.  It comes with this supporting paragraph that says, "Cannot rule out the presence of high-grade cancer." It is present in that clinical situation of 10% to 40%, which is not the most useful. But in this case, we are suspicious of upper tract disease, so he had a CT scan.

And as you can see in both these pictures, they're slightly different shots of the right kidney and this is the urogram phase as you could see a pretty large what appears to be a tumor, that's obstructing the right collecting system and it does extend kind of throughout the calyces and on the left-hand side of these two pictures, you can sort of see it is causing the renal pelvis to expand. It's difficult to tell in this imaging specifically and our radiologist did have some troubles deciding, do they actually think particularly on that shot on the right, if this was infiltrating into the parenchyma, but in the end, they concluded they felt like it was just a long-standing obstruction that was sort of filling the calyces instead. When you look around at the lymph nodes on both these areas around the hilum, it looks a little shoddy. There is not really any big lymphadenopathy.

However, as you move down, there are some inner aorta cable nodes and you can see that more clearly on the CT scan on the right side, where you see a little, third-round circle in between the aorta and the vena cava. And so this made us suspicious of, well, is this a node-positive urothelial carcinoma? This is a pretty large tumor. We're really worried a little bit about that lymph node. And so what are the next steps?

We have a lot of suspicions. Your gut tells you, this is a urothelial, upper tract urothelial carcinoma, but how do you get a tissue diagnosis? So with this, that lymph node biopsy, when you kind of look at the window, it's difficult. And so in speaking with our interventional radiologist, they said, "Hey, you really should try endoscopically. We'll do it if we have to, but it is a little risky and it's a little difficult and there is a chance of it being non-diagnostic."

And so why did we really care about this lymph node? Because in my opinion, this would change management completely in terms of being something that is clinically node-positive versus something that is likely organ confined. And this would mean that this patient should get systemic chemotherapy to start. And if there's a response, then you would proceed to surgery. We know that when you do upfront nephroureterectomy with lymph node dissection in clinically node-positive patients, the outcomes are poor. Recurrence-free survival at somewhere around one to two years is 25% to 33%, depending upon which data set you look at, and so that's not particularly fantastic. You also know that if someone's kidney function is not awful, but not perfect, and so if you take away those nephrons on the right side, even though they may not be as robust as the left, you worry that you would lose your window for a cisplatinum based chemotherapy. We said, "Okay, let's do this endoscopic biopsy."

So we went to the ureteroscopy. The picture on the right, I think is my iPhone taking a picture of our screen because we couldn't get the imaging to work, which is definitely real life. We couldn't get the image capture to do that, but mostly it appeared to be a primarily papillary tumor that was pretty much filling the renal pelvis upper tract and funneling into the proximal ureter. We got six biopsies. Some of them were a fairly decent size. I use a piranha forcep. I did not put an access sheath or try to use a larger biopsy forcep, the BIGopsy, which you have to backload, but I feel like I can get some decent biopsies with that. And we got very good tissue. However, the result that came back from our pathologist was a bland papillary urothelial neoplasm. And their differential was PUNLMP versus low-grade Ta. We still had this atypical urothelial cell cytology from the upper tract.

I was hoping something would kind of make a reconcile of our two findings. I was really hoping that we would at least get cytology that was high grade from up there, or a biopsy that was high grade to sort of confirm that lymphadenopathy. And the main reason is that the bulky nodes were [inaudible 00:07:12] inner aortic cables. There was some shoddy lymphadenopathy in the hilum and paracaval, although that is possible, when you look at data, looking at how many diseases can you capture based on your lymph node templates? For a right-sided proximal tumor, if you just do the paracaval and hilar, and retrocaval nodes, you get about 80%. And this is from data from Dr. Matin down at MD Anderson and also our colleagues in Japan, Dr. Kondo. When you extend that to the inner aorta cable for a right-sided template of dissection, you'll catch about 96% of the disease. You do not get that extra bump up on the left side as much, but you do definitely on the right. And so again, we are still perseverating a little bit over what is this lymph node?

And so we are really, our sort of conundrum was, is this clinical node-positive with under-sampling of the primary? This was a huge tumor. Did I only get the papillary bits that were the low-grade part? Was this a mixed tumor and the deeper stuff was the more awful tumor? Or is this a high-volume low-grade Ta with an alternative explanation for the lymph node? This gentleman is in his late sixties. In our experience with even prostatectomies and no dissections, could it be lymphoma? Or could it be something else? And so we had a couple of thoughts and we discussed this in our tumor board. Should we go with surgery upfront? Should we go ahead and try and do the percutaneous biopsy of the lymph node? Do you try to resample that primary from the other way with a perc biopsy to see if you can prove there is high-grade disease? Or do you use PET because you don't know what else to do, to see if that lymph node is hot or not?

We took out that last bit about using PET scan, because if it was the next most common thing, which would be possibly a lymphoma that would light up on PET too. We didn't think that would be as useful and so we did ask our interventional radiologist to biopsy this.

Sam Chang: That's exactly what I would have done too.

Sima Porten: That's what you would have done? Okay, great.

Sam Chang: I would have done that too because I was looking at that slide and I was going to give you some grief about PET, because we here at UCSF, we get a PET of our pets because we got to get a PET. But I'm so glad you guys decided not to get a PET because of all of those things that you raised. But your decision tree, I think is so important. Just like you said, everybody worries this is bad cancer. But your tissue diagnosis doesn't confirm that, but before doing an [inaudible 00:10:02], I would have done everything I could to try to prove that this disease is as bad as we think. Because I want to give this patient systemic therapy. I'm thinking. I'm not sure, but so I would have gone to a different interventional radiologist, honesty, and said, "Hey, look at these films, what do you think?" And our most experienced one because some, as you know would be more aggressive than others. You guys did biopsy it, and I will let you keep going.

Sima Porten: We were able to biopsy it and they were able to get some cells, but they had difficulty getting true cores or substantial cores because it was a fairly tricky location. And the read back came as atypical cells, which is the bane of my existence and that was suspicious, but not conclusive. And so in the end, when we discussed as a team with my colleague, Dr. Koshkin in medical oncology and others, I am extremely lucky in that I have experts like Dr. Meng, too and we can discuss cases altogether and sort of really work through this. We did believe he was high-risk because of the node, the size of the tumor, and some of the imaging characteristics. And we agreed that the biopsy was likely secondary to under-sampling and our medical oncologist decided to proceed with systemic chemotherapy.

In the meantime, we did do our in-house sequencing because we have been using this for the past four to five years, particularly in cases of primary unknowns, but it can be helpful in some of these other cases where a traditional IHC, and in other ways of determining well, what are these atypical cancer cells not revealing. We've been turning to our in-house sequencing. And so, although we sort of, we are already moving toward the decision of we're going to treat this with systemic chemotherapy and then proceed to surgery if there is a good response, the sequencing results then eventually popped up right before he had started his first cycle.

Sam Chang: Sima, so for this sequencing, does insurance cover this? Or is this just a UCSF research project? Or how do you do the sequencing?

Sima Porten: Yeah, so it started with our molecular diagnostics group and initially it was just an in-house sequencing platform, similar to Memorial Sloan Kettering's IMPACT. And some of the other commercially available platforms like Foundation, which is covered. Over the past years, it has gone through the process of being a CLIA certified test in the lab, and also now we are able to get coverage. However, there is a pre-authorization process that needs to happen and so now, yes it is covered.

Sam Chang: Got it. Got it. And this is basically on, this is not germline. This is basically somatic on tissue that you all have biopsied?

Sima Porten: Correct. You can use many different biopsy tissues. There is a recommendation to also have germline sequencing along with it so you are able to call different DNA-based alterations more accurately. And so that actually is part of the UCSF sequencing platform in that it can be done on the majority of paraffin or even fresh tissue, but paraffin-based tissue. And it looks at 500 commonly altered genes in all cancers,and it is DNA-based sequencing. It is not an RNA expression test. And they do recommend germline either from blood or you can do saliva or a buccal swab as well.

Sam Chang: Got it. And so for this patient, did the lymph node biopsy cells, did they have any microscopic kind of correlation with your endoscopic biopsy? Did they look kind of similar? Or could they not say anything regarding that?

Sima Porten: They couldn't tell. When they put them side by side, they just couldn't see anything that they felt would be conclusive. To them, this was suspicious that this was a, they said a carcinoma. And I said, "Okay, well, can you narrow it down? Is it a urothelial carcinoma? Which carcinoma?" And they said that is what they were very suspicious for, but they couldn't really tell from what they had available to sort of figure it out. And so we were able to get the sequencing done and there was a TERT promoter mutation in both samples, and an FGFR 3 mutation that was thought to lead, in terms of an activating mutation for FGFR 3 in both was similar. And so we felt a lot more confident that this was indeed a node-positive upper tract urothelial carcinoma and that our biopsies did undersample the primary.

And so this is the end of the case. He got four cycles of gem-cis. There was a discussion about should we do six cycles because it was a node-positive patient? And so the plan was to do four, see how he was doing, and image. And there was near CR of the node and shrinkage of that primary tumor. However, he was having a lot more difficulty with chemotherapy after that fourth cycle. There needed to be a little break in there due to some toxicity dehydration. He was needing a break, particularly with that last cycle. And so we decided to proceed with surgery.

We did a robotic-assisted nephroureterectomy and a retroperitoneal lymph node dissection, including the hilar, paracaval, retrocaval, and interaortocaval nodes. The final pathology was Ta of the primary tumor N0 of 21 nodes. It was a high-grade urothelial cancer, primarily a high-grade tumor involving the renal pelvis. There was a lot of treatment effect throughout the right upper tract and there was a treatment effect present in many of the lymph nodes taken. The largest inter-aortocaval packet was necrotic at the center of the node. And so it was our assumption on our diagnosis in terms of before we started the treatment plan, we think was correct and he has had no evidence of disease for about two years now.

Sam Chang: Oh, it's fantastic. I think this case highlights just like we talked about, if you do find a metastatic disease or lymph node-positive disease, I think that is the key step because it really does alter what you do. And I've had similar patients that have had not dissimilar presentations in terms of difficult location or the biopsy shows atypical cells, not definitive. We've erred towards let's treat this patient systemically because he or she smells like a more advanced disease and we were worried that we were going to do them a disservice with the procedure.

Let's take the same patient without the lymph nodes and for low-grade disease, I think people would feel comfortable with that bulky disease with low grade to do a nephroureterectomy first. I don't want to put words in your mouth, but what about with this no lymph nodes seen, the chest is fine, this bulky mass that is high grade? This is the debate with the Pal trial data showing benefit with adjuvant. And we have just had a debate this morning at our tumor board regarding this type of patient. What do you advocate? Let's take the same patient, no lymph nodes, high-grade disease, looks like it's filling that renal pelvis, maybe even infiltrative like you were saying, some question. What therapy would a patient get recommended at UCSF?

Sima Porten: Yeah. So I think this is asking the question is there a role for neoadjuvant chemotherapy, similar to our paradigm for muscle-invasive bladder cancer? And the difficulty with upper tract disease is that you're not going to get that invasive or muscle-invasive or that correlative type of diagnosis easily because of our limitations with staging and also biopsy. It's difficult to even get anything deeper, denoting T1 disease, let alone T2 disease. Right, and so you can use some surrogates, so high grade, larger tumors, [inaudible 00:19:18] architecture. Do you think that it is T3 on imaging, the infiltration? And there are some nomograms that have been done, I think first it was published by Dr. Margulis, probably back in 2010 and there have been some reiterations and validations of that that includes hemoglobin. And it's one of those ones that you kind of line up the lines and you see where you're at with these different factors.

And it is sort of, it's trying to predict, well, do you really think you have a more advanced disease? And so that is where, should that patient get neoadjuvant chemotherapy? We did a study when I was a fellow at MD Anderson looking at this and there was a definite benefit to patients getting neoadjuvant chemotherapy, survival benefit and the conundrum with that is we had to use a historical cohort for comparison. It would be really great to see the ECOG trial that is looking at this, but it will take a bit of time for it to, I think, result for us. We tend to be believers in neoadjuvant chemotherapy.

So to answer your question, a patient like that, I would feel strongly about having them get neoadjuvant chemotherapy. However, if when they met with the medical oncologist and they had that discussion and they were not convinced, I would probably proceed with the surgery, but I would really, really, really, really like that neoadjuvant chemotherapy, particularly if I was suspicious of more aggressive disease. That's a hard case because you can't tell a patient there are randomized controlled trials or all of this data and this is what we think you should do. You say, that it is a rare cancer, there are trials ongoing. We do have data in the adjuvant setting. And so, but I really think you need to get it. And so in the end it ends up being a longer conversation, discussion with a medical oncologist, but I tend to favor neoadjuvant chemotherapy. I don't know. What about you?

Sam Chang: Sima, we were just talking about that difficult situation, where you sense for the different risk characteristics that you outlined in terms of a higher risk patient, the architecture, the multifocality, the bulk, the appearance on scans, the grade obviously I think, still to me is probably the most important factor. And you want to give that patient neoadjuvant chemotherapy and there are for me, we still do some neoadjuvant chemotherapy. And honestly, that's the surgeon's preference. Many of our medical oncologists here really focus on still, this is the data. The data says adjuvant. Despite the fact that they likely will lose some renal function or will lose some renal function.

To us, the tipping point is all the things we talked about, but then their renal function. If they already have some compromised renal function, then it's much easier to persuade our medical oncologists and the patients to proceed with upfront neoadjuvant. Our bias is that, and this is a question I'll ask you is, do you think, and this is all conjecture and this is obviously, but do you think there is a benefit to a gem-carbo, gemcitabine carboplatin combination in the adjuvant setting? And do you routinely or do your medical oncologists use that in patients who have high-risk disease following nephroureterectomy? Tough question, loaded question perhaps, but I was just wondering if you do or if you don't.

Sima Porten: Yeah. So if like you were saying, if they don't get neoadjuvant chemotherapy and they have advanced disease on pathology, node-positive invasive, those types of high-risk features and now their kidney function cannot support cisplatin, I would say that I do send them to our medical oncologist for consideration of gem-carbo. There is usually a fairly robust discussion about where did the POUT trial really show its benefit? Was it in the patients who got gem-cis? Or gem-carbo? And was it primarily in patients who had T3-T4 or node-positive? Because I think those are slightly different. Right, and so they do tend to favor or are more in favor of adjuvant chemotherapy for those who are node-positive, whether it's gem-cis or gem-carbo. If it happens to be a more locally advanced pathology, like the high-risk features are T3-T4 essentially, there is less enthusiasm for using gem-carbo.

Sam Chang: Yeah, no, I think that's very similar to our thought process here, in terms of those patients, we tend to follow closely. We understand the data and there are always issues with subgroup analysis. But I think there's an innate bias that our medical oncologists have in terms of the relative ineffectiveness of carboplatin. Clearly, some patients respond, but the majority don't, and they tend to favor just as your group does of node-positive disease, clearly high-risk factors. T3-T4 can be high risk, there's no question, but we tend to follow those patients.

So, let's end with this question, with this patient that you're now seeing in follow-up. How do you do follow-up now in this patient? How often do you do cystos? How often do you do the upper tract evaluation on the left? I don't know how his renal function ended up. What do you do in terms of surveillance now for this high risk? Because he's NED for two years. Fantastic. And honestly, I think that portends a good prognosis down the line, but how do you do follow-up for this patient?

Sima Porten: Yeah, sure. I think this is also really a hard thing for patients in that I essentially have to say, "Well, we really need to monitor." Then they are like, "Why do we need to do more cystoscopies?" And we have this discussion that for patients who have primarily upper tract disease, that the recurrence rate, depending on data you look at is somewhere between 25% to 40%. And generally, again, recurrence has happened soon after surgery because you are concerned about seeding from the upper tract. We tend to use chemotherapy in the bladder. I do it after surgery, prior to discharge, before removing the catheter. In general, somewhere around two to three days after surgery with a cystogram, confirmed gemcitabine, catheter out, and to try to prevent a lower tract recurrence. I usually do cystos Q3 months for two years and then start extending it out, mostly trying to follow or mimic the NCCN guidelines for this.

For imaging, his creatinine ended up at 1.5. Our radiologists here actually do not mind giving a contrast in that setting. They will give a little bit of hydration and we will give contrast, they've sort of moved away from the contrast limits that were placed before with GFR. If they are unable to get CT scans with contrast-based imaging, I use MRIs with contrast. I have a really difficult time when patients can not get contrast because I feel like things are happening and you're not really seeing it. I may work in the PET in that situation because there's not much else you can do if they can't even get MRI contrast. That is pretty difficult. But I would say it's very difficult to see an intraabdominal or chest.

A chest CT usually is pretty good for looking at recurrence, either in the mediastinum or in the lung parenchyma, but for in the abdomen, I think it's really difficult to see that. You want some evaluation of the left upper tract as well because you worry about a tumor that could develop there. We are able to do MR urograms at UCSF or CT urograms as well to ensure no disease. And then, of course, I send cytology with the cystoscopy at the same time to look for that. But it's mostly having a fairly high suspicion that recurrence may happen and it could be either in one of the luminal organs or could be intraabdominal or chest.

Sam Chang: Right. No, I think that in the education process that you state, you're exactly right. It's taken me some time to figure out that it's better to tell a patient before you even take out the kidney and the ureter, you've already had some scopes and you are going to have more and I need to tell you upfront that you're going to have the majority of these and they are going to be in the clinic, you're awake, and et cetera, et cetera, because just like you say, their first thought is, well kidney and two ureters are out and there's no more disease. Great. I'm done. It's that continued education that's important.

Sima, thank you so much for spending some time with us. And I appreciate the real-world aspect of these high-risk cases because each patient presents differently. Some follow the classic pattern and make it obvious, high grade, big tumor, biopsy shows high grade, okay. But it's these types of tumors and presentations that actually, we all tend to face. And the discussion that you had, I think in a tumor board, I think emphasizes again, the importance of a multidisciplinary approach and weighing back and forth to the patients that we've got some evidence, we don't have all the evidence, and then coming up with a plan together. Thank you again for spending some time with us and enlightening us.

Sima Porten: Thanks for having me.