Read the Full Video Transcript

Sam Chang: My name is Sam Chang. I'm a urologist in Nashville, Tennessee, at Vanderbilt University Medical Center. And I am fortunate to be joined with one of my all-time favorite people—not favorite physicians or medical oncologists—one of my all-time favorite people, Dr. Jeannie Hoffman-Censits, who is an associate professor at the Johns Hopkins University Medical Center and works at the Sidney Kimmel Cancer Center.

We are quite fortunate to have her actually give us an abbreviated but highlight version of a talk that she gave at the SUO 2024, looking at adjuvant systemic therapies for urothelial carcinoma. So I know how busy she is. We are very, very fortunate to have her spend some time with us.

And so, Jeannie, thanks so much for spending some time with us and educating us as always. I’ll turn it over to you.

Jean Hoffman-Censits: Thanks, Sam. Pleasure’s mine. Love getting the chance to work with you and your team at UroToday. Thanks so much for the kind invitation to give another overview about the talk that I gave at SUO.

All right. So we’re going to talk about an overview of adjuvant systemic therapy in urothelial cancer—where are we today? A little different than where we might be tomorrow.

As a medical oncologist, my specialty is only seeing patients with urothelial cancer. So Sam and I work with a lot of urologists, seeing patients in the perioperative setting—patients who may or may not have gotten preoperative or neoadjuvant chemotherapy for their high-risk bladder cancer or upper tract urothelial cancer. And we’re going to focus the conversation today on what kind of conversations we, as medical oncologists, have with patients in the postoperative setting.

So these are patients that have high-risk tumors. That’s going to be defined as those who, if they had neoadjuvant chemotherapy, have residual chemorefractory disease from T2 to T4 or node-positive disease, or patients who may have not been offered or were not a candidate for neoadjuvant therapy. These are going to be the higher-stage patients, T3 or above. And, in particular, for the studies that we’ll talk about, these patients had to either not be a candidate or declined adjuvant therapy. So in this day and age, who might those patients be? They could be those with a long-standing BCG-refractory tumor that looked like it was a noninvasive tumor and wound up being upstaged to maybe a node-positive tumor, or a patient with a high-risk upper tract tumor who was not a candidate for chemo.

Of all the conversations we have—and I know talking about advanced disease is a tough one too—but from a conversation standpoint, thinking about the potential benefit and the potential risks, this is a group of patients that just went through a very major surgery, found that they had a cancer diagnosis, may or may not have gone through chemotherapy. And now, we’re talking about next steps.

For many patients, they are taken aback a bit at this point in the conversation, feeling like, “I thought my surgeon got it all out. And if doing therapy would have been a good idea, why didn’t I come here first? And if my scans are negative, why do I need this?” So these are just some of the questions that we get asked in clinic.

I think one of the things that we’ve done as a multidisciplinary team is really have a discussion that this is going to be a continuum of care. We’ll see patients in the postoperative setting. We’ll talk about their path report, and then we may or may not offer more therapy in the postoperative setting based on what the pathology report tells us.

So this is just a highlight summary slide of the three large-scale, randomized, prospective clinical trials that have been completed to date. The first is with nivolumab. That’s been published, and that’s actually a guideline-endorsed adjuvant therapy that we use and have been using in the postoperative setting. There’s a study of pembrolizumab that was also positive. Both of those studies were looking for DFS at the date of the first recurrence—either a local recurrence outside the urothelial tract, distant recurrence, or death.

So both of those studies were positive. The nivolumab study was a year of adjuvant treatment with nivolumab, once every two weeks, versus placebo—so slightly different design. That was a very positive study.

The second was with pembrolizumab, which was more recently published. That one was an open-label trial versus observation—again, an improvement in disease-free survival, which was the primary endpoint for both of those studies.

And then the third was atezolizumab versus observation, and this was a negative study. So we have two positive studies, one negative study, and so far one FDA-approved agent that’s guideline-approved by the NCCN in the postoperative setting. This is just a similar schema going over really who are these patients that we’re seeing in the postoperative setting, and how am I thinking about it, and who are we really offering this treatment to?

Again, we’re talking about patients who may or may not have been candidates or had the opportunity to get preoperative cisplatin-based chemotherapy. Patients will go on to get bladder surgery—any kind of diversion was fine—but they did have to have a radical cystectomy or cystoprostatectomy, no partials there. And it was really focusing on those high-risk patients and also focusing on patients who had bladder cancer. So all of the studies pretty much capped or limited the enrollment of patients to those who had bladder cancer. Really, the majority of patients had bladder cancer, and about 20% in each of these studies had upper tract urothelial cancer.

I have the nivolumab study highlighted in dark green because, again, that’s a positive study, FDA approved, guideline endorsed. Pembrolizumab was also a positive study, but it was reported out later—FDA-approved drug, but guidelines are still pending. And the atezolizumab study was negative, and we have some thoughts about why that might be the case.

For patients that don’t have as high-risk disease, we can offer surveillance to those patients. I think this can be an individualized decision as to whether or not maybe a younger, fitter patient may or may not want or want to have a conversation about therapy. But certainly, for those patients with higher-risk disease who didn’t get preoperative chemotherapy, we would first offer them platinum, and if they were not a candidate for or refused that, then we can consider adjuvant checkpoint inhibitor.

When I think about which I would give in the adjuvant setting, I really go back to how the trials were designed. And that’s why I put this on the slide and reiterate that if someone has a surprise high-risk tumor—node-positive tumor—that I would first offer them or think about how to optimize them for adjuvant chemotherapy. And if they weren’t eligible, then consider adjuvant immunotherapy as another option.

Importantly, also from a timing perspective, those trials really captured patients in the postoperative setting. They had to be at least four weeks and no longer than 120 days post-op. So there really is this finite period of time where decisions should be made, patients should be prepped, scans should be done, so you make sure that you understand what disease state the patient is in before offering them adjuvant therapy.

For those patients who have a hard time or some bumps in the road where they really feel like they need longer than that to make a decision, there’s not an absolute time where someone’s getting adjuvant therapy. But if it takes them six months to make a decision, then I think the decision is that they’re getting surveillance instead.

Just to put these trials side by side—not necessarily to compare them to one another, but just to show the Kaplan-Meier curves at the high level—that both in the nivolumab study and the pembrolizumab study, the intention-to-treat population was positive in terms of improvement in median disease-free survival. These were both statistically significant improvements.

I think one of the interesting things, as a medical oncologist, thinking about why potentially could patients get better outcomes—not only from the drug that they got in the study but also drugs that they may have received in the future. Looking at the patients on the right-hand side who were in the observation arm in the adjuvant pembrolizumab study, if they had recurrence, the majority of those patients who went on to get subsequent therapy were treated with a checkpoint inhibitor, and they’re a perfectly reasonable class of drugs. But compared to antibody-drug conjugates like enfortumab vedotin or an FGFR3 inhibitor, both of which have response rates upwards of 40% to 50%, those patients proved to be, if they did have progression either during the study or shortly thereafter, to have disease that was also refractory to a checkpoint inhibitor, and thus maybe they had time to—and were healthy enough to—receive additional agents that may have had a higher chance of response.

We saw that, too, in the adjuvant nivolumab study. This was called a PFS2, or not just looking at the PFS in the study, but the progression-free survival that happened if the patient was treated with subsequent therapies. This is a really cool endpoint that Galsky and colleagues looked at. I don’t think we’ve ever really seen this before, but I think probably the same is true—that those patients were well enough to get subsequent therapy, and we already knew that a checkpoint inhibitor, at least alone, was probably not going to be the best choice in terms of the treatment for those patients.

Even though the adjuvant atezolizumab trial was negative, one of the great gold mines of this study that I think has really opened up a new phase in urothelial cancer research is looking at the impact of circulating tumor DNA. So what this study did was, even though it was a negative study, it collected circulating tumor DNA on patients. And those who did not have any evidence pre- or post-op of circulating tumor DNA—on the left-hand side—their survival didn’t seem to be impacted as to whether or not they received adjuvant immunotherapy. But if the circulating tumor DNA was positive and they received immunotherapy, that’s where you’re seeing a real separation in the curves, which is really fascinating. That led to a significant improvement in overall survival, 25 versus 15 months. So from that negative trial, it really was very compelling in terms of thinking about the design of future studies in the adjuvant setting.

That led to this study, which has now actually completed accrual, believe it or not, from the time that IMvigor011 was negative until now. So this over-500-patient study is completed accrual, and we’re awaiting the outcome. This study is looking at adjuvant atezolizumab for patients with bladder cancer only who are ctDNA positive. Those patients are randomized to treatment versus placebo, versus those who are negative and who undergo surveillance.

There’s an open study now called the MODERN study. This is led by Matt Galsky. And this study is built very much in the same way, but it also includes patients who are ctDNA negative. So the top randomization—ctDNA positive, again, patients with bladder cancer in the postoperative setting—have nivolumab or nivolumab versus relatlimab. And then in the ctDNA negative, those patients can be randomized to nivolumab or surveillance, and if their surveillance scans eventually become positive or the ctDNA becomes detectable, then they would go on to get nivolumab in the adjuvant setting. Excuse me—not if their scans are positive, their ctDNA becomes positive.

So it’s a really well-designed study because, as medical oncologists and as patients too, one of the things that we really are concerned about is: are we overtreating patients in the adjuvant setting? Currently, the data that we have on ctDNA is mostly retrospective, but these two prospective studies I think will really help us understand how to use that tool and how to really best counsel patients in that postoperative setting as to what path they should take.

Perioperative therapy, I think, is going to be something that we’ll be seeing a lot more of—this is neoadjuvant chemoimmunotherapy followed by adjuvant immunotherapy. And this trial called the NIAGARA trial was published late last year, looking at patients again with invasive bladder cancer getting neoadjuvant chemotherapy of cisplatin and gemcitabine for four cycles or chemotherapy plus durvalumab. And then in the postoperative setting—no matter what the pathology report showed (complete response, partial response, T1, T2, or if they were upstaged)—every patient would go on to get adjuvant durvalumab. This was also a positive study in terms of event-free as well as overall survival, which is incredibly compelling.

As of now, I think we’re waiting for updates to guidelines for this to be incorporated more into standard of care. So I think there’ll be a lot of changes coming this year, when it comes to the perioperative treatment of urothelial cancer.

Sam, you have been so kind to let me talk to your audience in the past about upper tract urothelial cancer. This is a subset of urothelial cancer that’s more rare, less data, of course, and near and dear to both of our hearts that we think about a lot.

We still have, of course, the POUT study. Alison Birtle and our colleagues in Europe did this large, randomized trial looking at adjuvant chemotherapy—either cisplatin or carboplatin and gemcitabine—versus surveillance, again showing an improvement in DFS as well as metastasis-free survival. So for those upper tract patients who may have not had the opportunity to get pre-op chemotherapy, this is still a discussion that we have.

There was also a follow-up done in the JCO just a year ago, showing that there’s still that benefit that holds up. When we think about—and maybe question—why those newer adjuvant studies that I went over that are incorporating ctDNA, why are they not including patients with upper tract disease? Probably likely because of this subgroup analysis. And the subgroup analysis for the adjuvant nivolumab study actually looks quite similar to the one in the pembrolizumab study, where it looks like the majority of the benefit in these trials is coming from the patients that have bladder cancer.

I don’t think we have a really good explanation for why that is. I think everybody’s a bit surprised by it. I think if a patient walked into my office with metastatic upper tract urothelial cancer, I wouldn’t think twice—I wouldn’t think that immunotherapy would not be a right choice for that patient. But in the adjuvant setting, we’re just not really getting the same benefit.

So when I think about why this might be, I really think about the patients who walk into my office that have upper tract urothelial cancer who may have been treated or referred outside, versus the patients that have bladder cancer. There’s almost no heterogeneity in terms of the preoperative therapy for patients with bladder cancer. The majority of patients have had neoadjuvant chemotherapy. And if for whatever reason they did not qualify and they went to surgery, I know that they’re going to have the right surgery with a really wide lymph node dissection, and for men, a cystoprostatectomy. It’s very consistent that, across the board, patients are getting treated in that way.

For patients with upper tract urothelial cancer, I think there’s a lot of heterogeneity as to whether or not they got neoadjuvant chemotherapy—which I think a lot of us would at least have a conversation with patients about at this point—and whether or not they had a nephroureterectomy for high-grade disease, did they have a lymph node dissection, what was their postoperative kidney function? Again, we’re not seeing that same clear, defined expectation in the upper tract space. And I think it’s because it’s a rare disease. But when we look also at the subgroup analysis for the adjuvant nivolumab study, we do see that patients who didn’t have an adequate lymph node dissection, have an impaired creatinine clearance, didn’t get neoadjuvant chemotherapy—those patients also tended to do worse.

So I think that a lot of this is probably wrapped up in some of the heterogeneity of treatment, potentially a little bit with the biology of the upper tract disease, but it’s really not clearly defined. Nevertheless, we do not have large-scale adjuvant trials, including patients with upper tract disease that are in these modern eras.

We do have guidelines, though, Sam—we were both, I think, honored to be a part of this AUA upper tract guideline—that really helps guide surgeons and medical oncologists for this rare cancer.

And that’s all I have. Thank you so much for the opportunity to have a conversation with you and go over some of the slides from the SUO talk.

Sam Chang: Jeannie, thanks so much. That was so good in terms of the amount of material and key questions, key points, but then also key next areas of research and what we do.

So let’s start with the upper tract. Do you hesitate at all to consider adjuvant immunotherapy? You talked about the forest plots and the smaller number—about 20% in both—probably multifactorial, or maybe there isn’t a difference, and we just need to study it more. Do you personally hesitate to use immunotherapy in that adjuvant setting for patients that you think would benefit, as would be the case for our bladder cancer patients?

Jean Hoffman-Censits: Yeah. I think it’s a great question. For the most part, I would say no, I don’t, and especially if a patient had neoadjuvant chemotherapy and has a chemorefractory tumor.

So someone’s got what we think is the best kind of treatment that they can get in the pre-op setting. And then that path report comes back with a lymph node on it. I’m really, really worried about that patient. Whether or not they have bladder cancer or upper cancer, I’m worried.

So I’m glad to know that I have an FDA-approved shot on goal that I can give. We can have conversations about getting that follow-up staging and getting started. I think particularly for that situation, I do think it’s very important to understand what is that post-op staging—that you’re giving the right drug to the right patient at the right time—because I do worry that they may just be on their way to having a quick recurrence. And if that was the case, I might pivot to use a different combination of agents. But in that setting, I think I feel very comfortable that I’m doing something with the hope that I’m really decreasing the risk of the cancer coming back.

In someone who is not a great candidate for chemotherapy, wasn’t before surgery, is even a worse candidate—as we know, that can be the case for chemotherapy after surgery, especially if it’s regarding kidney function—

Sam Chang: Sure.

Jean Hoffman-Censits: I think, is there a question? There is a bit of a question as to whether or not anybody is the right candidate. In the adjuvant setting, we know that there is going to be some degree of overtreatment. But I think, importantly, all these studies were looking at the endpoints in an intent-to-treat population. The subgroup analysis is hypothesis generating. I think there’s important questions we’re probably not asking about the patients with upper tract tumors that I think is a missed opportunity. But in that setting, when you’re worried about what might happen to them, I think having a conversation about adjuvant immunotherapy based on the intent to treat is absolutely reasonable, and something I do on a regular basis.

Sam Chang: Well, I think that makes great sense. Let’s throw a tougher situation at you, though. For the upper tract adjuvant setting, following a nephroureterectomy—not as common a scenario, but a scenario where you have a patient that did not get any neoadjuvant treatment, undergoes nephroureterectomy, a lymph node dissection. Things look good, except the patient has T3 disease but has a GFR after surgery of around 52, can get platinum, is younger, healthier, etc.

So the patient now in this scenario—you’ve got no head-to-head comparisons—but do you lean towards platinum-based chemotherapy for this patient in the adjuvant setting? Or do you consider IO? Tell me your thought process regarding that. I’ve given you the platinum eligible. And then I’ll throw out the scenario, same situation, but the GFR is 41, something like that.

Jean Hoffman-Censits: Sure. Well, I don’t know. You made it a little bit easier in the beginning because—

Sam Chang: Yeah. Sure. Of course. Yeah. I try to be nice here.

Jean Hoffman-Censits: I know. First was a softball. So in the 50s, which was your first question, and even a little bit below that, we have a few regimens now. So Jonathan Coleman’s neoadjuvant split-dose chemotherapy using cisplatin and gemcitabine, but cutting that full dose of cisplatin in half and giving half on day 1 and half on day 8, that’s really well tolerated, both from a GI toxicity perspective and for our patients where we’re worried about their renal function, which, as you know, is like half of them.

And so that is a regimen that I will extrapolate—there was a neoadjuvant regimen—and I’ll use that in the adjuvant setting. And we’re going to try and make that chemotherapy a bit more tolerable for those patients with impaired renal function. Interestingly, the NIAGARA study also allowed for that split dosing of platinum-based chemotherapy, which I think is really important because it’s speaking to the real population of patients that we’re all seeing in the clinic.

If they’re really not a platinum candidate—so we’re talking creatinine clearance of 30, an impaired functional status—I think that patient also might not be a great candidate for immunotherapy. The risk-benefit analysis in the post-op setting, where we’re thinking about the goal of decreasing the risk of the cancer coming back versus in the treatment setting when someone has active metastatic disease that is threatening their life—those are two different things.

So I think that, again, when I say it’s a hard conversation, it’s a hard conversation. And oftentimes, when medical oncologists meet patients, we can usually go over everything, come up with a treatment plan, and order treatment, and get it going within one conversation. Sometimes the adjuvant setting is two or three conversations before decisions are made, and I think that’s a bit of a harder one.

Sam Chang: Oh, no question. And it really is the balancing of unknowns in terms of true unknowns regarding risk and benefit, because there is a real benefit and a real possibility of overtreatment. So it’s difficult with nothing to measure, nothing to know if this is really being useful. And it is a difficult situation. But it’s one we honestly relish now. We never even had possible therapies that we could show benefit to prevention of disease in the future.

I think it’s really a step forward, although it makes the discussions actually more difficult. And it’s like the discussions we have, I say this classically with our prostate cancer patients—you’ve got to have good news and bad news. The good news is you’ve got a lot of choices. The bad news is you’ve got a lot of choices. In bladder cancer, we never really had choices.

So when you look at urothelial carcinoma and the different possibilities along the way, all these studies are coming out, even extending to non-muscle invasive disease. We really have started to make, I think, some real advances.

Jeannie, I’ve already taken too much of your time, but we appreciate so much you spending some time with us at UroToday. Always, always a pleasure. We always learn so much and really appreciate how much you’ve done for our urothelial cancer patients. And so we look forward to future contributions, as well as future times together.

Jean Hoffman-Censits: Well, thank you so much. I really enjoy talking to you and to your audience about this. Thank you so much.