Renal Cell Carcinoma — Thomas Powles, MBBS, MRCP, MD

PROF POWLES: I’m Tom Powles. I’m an oncologist in London and Director of Barts Cancer Center. I’m going to talk today about advances in metastatic renal cancer over the last 12 months.

KEYNOTE-564 is an important trial, probably the most important data that’s come out this year. I think it’s important because to date, we haven’t had any positive adjuvant data in renal cancer.

We’ve done a number of trials in that space. They have all, but S-TRAC, showed no progression-free survival advantage. And none of them have shown even a trend towards overall survival.

ASSURE had a significant arm in it with no progression-free survival advantage. And we know VEGF targeted therapy was not associated with long-term complete responses in the metastatic space.

And so, we do know that the immune checkpoint inhibitors are different. We know, for example, with ipi/nivo, there’s 30% of patients with advanced disease with long-term durable remission. And therefore, in my opinion, the hypothesis that immune therapy or immune checkpoint inhibition may be associated with cure, reduction of risk, better survival, is stronger than VEGF-TKI therapy.

In 564, it was presented as a Plenary session at ASCO.

I’m actually going to show you the updated data. This updated data comes from Toni Choueiri at ASCO GU 2022. And essentially, it’s a randomized Phase III trial. It’s a 1,000 patients. It’s intermediate high-risk clear cell renal cancer. Now the definitions are slightly different from previous scoring systems such as the Leibovich scoring system, and so we’re going to have to learn this new system. There’s also 6% of patients had M1 NED. And what that means is they had surgery for metastasis, particularly lung and lymph node metastasis, and then had a year of adjuvant therapy.

The primary endpoint was disease-free survival, with overall survival as the secondary endpoint.

Here’s the updated data. You can see the curves go apart nicely and they stay apart. You can see the zigzag shape is consistent with the timing of the CT scans. You can see a reduction in the risk of relapse which is now 37% which is statistically significant. And you can see the curves go apart and they start apart, which is important.

It’s also worth looking at this, because that was on the left-hand side was the original 24-month follow-up, this is now 6 months of further follow-up. Now don’t get me wrong, this is not yet mature data. We want 5- and 10-year data ultimately. But this data is extremely encouraging and it seems to be getting a little bit, dare I say it, a little better and not worse which is terrific noise.

One of the important issues around these data is actually they secondary endpoint of overall survival. You’ll see that the overall survival hazard ratios are actually very low, but the number of events are also low, and the number of events compared to those needed remains low.

So initially, at the first analysis at 24 months, we have 51 events. We’ve got about 20 more events in that now. And you can see here the hazard ratio of 0.52 looks terrific, but it is immature and is not yet statistically significant and that’s because of alpha allocation. Those numbers are less than 0.05, but the statistical design means you need to have a lower number than that for it to be statistically significant.

Nevertheless, what we have is a significant delay in progression and we have an early trend towards overall survival. When you put these together, this is attractive for patients. And this is an option that should be discussed with patients.

Two caveats to that. The first is that there is adverse events, or there are adverse events associated with these drugs. The important issue around adverse event profiling is some of the adverse events are lifechanging, such as diabetes, hypothyroidism, but also you can get pneumonitis. There is neurology and even cardiac damage associated with these drugs. Now when you add all of these up, it comes to not an insignificant number of patients. And we do know that only about 30 or maybe 40% of patients will relapse. And therefore inevitably, without biomarkers, we are overtreating a significant proportion of patients.

Different guidelines’ committees have met. The one that I’m involved with, the ESMO Guidelines Committee, has given as a 1C recommendation. And that essentially is a recommendation to say that it’s strong evidence, but it’s not yet absolutely conclusive without that survival signal. And this is something that should be discussed with patients. But remember, patients don’t get a second opportunity adjuvant therapy. And many patients will find these data with a 37% reduction in the risk of relapse and a trend towards a reduction in, or an improvement in survival as attractive, particularly those with high-risk disease.

Here is the adverse event profile for this drug.

And I’ve gone over the issue on the left, fatigue and pruritus perhaps less relevant. Hypothyroidism replaceable with tablets. Diarrhea and rash are not too significant. Arthralgia, yes, of course, and nausea. It’s actually the long-term irreversible side effects which are really important. You can also see here the timing of these adverse events and when they occur on the right-hand side. We haven’t seen this data previously. And essentially, the vast majority are happening well within the first 3 or 4 months of therapy. And essentially, in my experience, the majority of these adverse events occur in the first 6 months. However, even in the metastatic space I’ve seen adverse events occur in 1 or 2 years after therapy, and even I’ve seen events occur after the drug is stopped. So as a group of oncologists, we need to be very prudent regarding this.

The next data I wanted to share with you is the ipilimumab and nivolumab data. This data, I think, is extremely important because I believe that what we are seeing now is, for the first time, long-term, durable remissions in advanced renal cancer. We’ve not seen in advanced renal cancer these remissions with sunitinib or pazopanib. You’ll note the Kaplan Meier curves all come down, progression-free curves all come down to zero. And here we see the 5-year data. and I think there are 2 outstanding data points. The first, on the left, is about a 50% 5-year overall survival for these individuals treated with ipi/nivo with intermediate and poor-risk disease. And if you said to me 10 years ago, or even 5 years ago when we started this study, that we’d have 50% 5-year survival, I would say that is a terrific advance. And the second is the 30% progression-free advantage. Remember, this looks like these patients will go on into a durable longer-term remission. And if that were the case, who knows, perhaps 30% of patients of patients will have long-term remission. And how much of this is being generated by CTLA4.

You’ll know that I’ve always been nervous that ipi/nivo is not great at getting initial control of disease. The response rate is modest, but 20- to 30% of patients progress straight through therapy and they do poorly. And that’s one of the reasons why I’ve been supportive of using VEGF-TKI/IO combinations because we know that’s much better at getting initial control of disease. However, from a patient’s perspective, if this data continues and we can’t reproduce this 5-year data with a VEGF-TKI/IO combinations, it would be reasonable for doctors and patients to say, I’m prepared to take that early risk of losing control for the long-term durable benefit. And so the long-term survival data for these upcoming VEGF-TKI/IO combinations are crucial and ipi/nivo has set a very high bar. And many people in the community feel much of that is being driven by the ipilimumab, by the CTLA4 inhibitor.

Here's the data for the intermediate- and poor-risk patients. Again, I love the curve on the right. Look at the 31%. Look at that plateau. Who would have guessed that intermediate- and poor-risk disease would have this at this point.

I wanted to talk very briefly about the favorable-risk patient. It’s probably the most popular debating topic amongst my colleagues at the moment because actually there are 3 or 4 options. Number 1 is surveillance for these patients, many of them it’s attractive. Brian Rini had an excellent publication along those lines. Number 2 is, currently Level IA evidence is for len/pem and cabo/nivo, or axi/pembro. And that’s generated off those randomized Phase III trials. But you’ll be aware the most recent data, a subset analysis of the most recent data from all of those trials shows a hazard ratio compared to sunitinib of approximately 1. And here we have the ipi/nivo data.

You’ll be aware, and you can see on the right-hand side that sunitinib initially outperforming ipi/nivo. But what we’re now beginning to see is a crossing of these curves. And if the plateau of that survival curve stays at 60% beyond 5 years, which is possible, again, as I said before, it may be that the long-term data with CTLA4 is more attractive to patients. Most of the patients who I meet are interested in 5- and 10-year survival, not 6 months versus 12 months.

So this is going to be a really important concept for us a oncologists in renal cancer, shifting away from a couple of months’ of progression advantage into what is sustainable remission. And indeed, Dave McDermott has been saying for some time should we be using landmark survival analysis rather than hazard ratios which, in the end, is important for patients.

The adverse event profile for ipi/nivo is shown here and is important because the addition of CTLA4 makes the first 12 weeks of therapy challenging. Yes, of course, things are more straightforward after that because you’re on single-agent nivo. But my experience of ipi/nivo is prudent and still is required during those first 12 weeks and most patients get some adverse events. And the ones that trouble me most of course, pneumonitis, transaminitis, bad skin rash, diarrhea. And the weird and wonderful, the cardiac, the neurology side effects, of all which we’ve seen.

PRISM was a study presented at ESMO.

This trial looked at giving ipi a different way. What about giving it instead of every 4 weeks, giving it every 12 weeks. The primary endpoint of the trial looked at the Grade 3/4 adverse events and giving it less frequently reduced the frequency of adverse events.

And you can see here the progression-free survival curve looks very similar at the moment. So it doesn’t appear that adjusting the ipi at this point, the hazard ratio’s exploratory which is why it’s not shown, but it’s very close to 1. The reality is at the moment, this ipi question has not translated into clear knowledge about whether we should give it all at the beginning or longer. And you’ll see here also, these PFS curves come down further, they come down at the 20s and 30s, or below the 30s, early 2 years. So this data hasn’t necessarily reproduced the data of 214.

DR LOVE: One of the things that’s kind of interesting was at the GI meeting there was a study of durva/treme in HCC that was positive. And the speculation was it might end up with treme being approved as another alternative for first-line therapy. Particularly one subset would be people who can’t get bev or atezo/bev first line.

So my question is, so maybe there’s going to be more interest in treme, and I know it’s been looked at with GU cancers, and I know they’ve used different schedules. And I was interested by what you just said about the different schedule with ipi/nivo. What do we know about toxicity with treme versus ipi? Is it really like the same drug? Is it very different? Is the efficacy — are they similar or different?

PROF POWLES: So they are slightly different. They seem to have different effect on T-reg cells. So it looks like tremelimumab may have slightly less effect on T-reg cells. But there’s also this really important issue around dosing. And we know that that ipi 3 is the melanoma dose and ipi 1, of course, is the kidney cancer dose. The durvalumab and the tremelimumab dose, the single-agent treme dose was given at much higher doses than we’re giving in the combination. And I’ve always in the back of my mind been inquisitive about whether we could have given higher doses of treme? And had we done that whether or not we got better results? You’ll be aware that we used durvalumab and tremelimumab in bladder cancer. And there’s also a trial called CALYPSO, which is looking at durvalumab and tremelimumab in kidney cancer, second line, randomized against durvalumab. A randomized, Phase II, only a 100 patients, 50 in each arm. And that data is expected at ASCO.

DR LOVE: Interesting. Okay, please continue.

PROF POWLES: I wanted to switch gears briefly and talk about the combination of VEGF-TKI, PD-1, and the reason why I wanted to do that is all of the 3 trials have had updated data.

We seem to be in what I think is an important habit of the pivotal trials giving data, as time goes by, to work out what’s happening in the future. And 426, of course, was the first trial to read out with axi/pembro. The original hazard ratio was 0.53. We saw response rate of 60%. And we showed a PFS hazard ratio of 0.7, which wasn’t the lowest hazard ratio for PFS. You’ll remember the lowest hazard ratio of a PFS is with lenvatinib and pembrolizumab, it was 0.39, that seemed extraordinarily low. And I’m going to talk about the CLEAR trial in a second.

But what have we learned from this trial? Well, what we’ve seen is dilution in the hazard ratio for survival from 0.53 to 0.73. This is relevant because this reduction in the hazard ratio has occurred across all 3 trials. And actually, it pretty much lands up exactly where ipi/nivo is now at 0.7 or so.

So we’re now in a position where all 4 of the first-line approved drugs with Level IA evidence have a survival advantage versus sunitinib in the 0.7s. Now when one looks at the landmark analyses of these, again they all look very similar. It’s apparent the CLEAR trial, the control arm, that was len/pem. The control arm appeared to perform best. That may be because it was the most recent trial. It had most access to subsequent therapies. But as it currently stands, when one talks to patients about overall survival, it’s very reasonable to talk about these regimens with equivalence.

Now the VEGF-TKI/IO combinations are better at getting initial control of disease. Only about 5- to 10% of patients had progression of disease on these combinations. Whereas with ipi/nivo it might be as high as 30%, which is obviously important, and particularly those patients with high risk of liver metastasis and high-volume disease where you really want to get control, VEGF-TKI/IO or PD-1 looks preferential. But it’s very hard to argue with any clinician who says actually, I use this regimen, rather than that regimen. But the thing which I’m talking to healthcare professionals about at the moment is the most important thing is how we give these combinations, not choosing between each. And so I always say, well, pick one and use it well. There is a real skill associated with giving these combinations. This combination is well tolerated. There’s personalized dosing associated with axitinib, of course. And importantly, there is transaminitis with this combination, which is perhaps a bit more frequent than we’ve seen with other VEGF-TKI/IOs. Nevertheless, axitinib has a shorter half-life and I think is a terrific drug.

The second area which requires attention, and this again applies for all 3 of the VEGF-TKI/PD-1 combinations, is that the axi/pembro curve, the PFS is now down at 30% at 36 months. And that’s pretty much exactly where ipi/nivo was in the intermediate- and poor-risk patients at 60 months at 5 years.

We now need or would like to see a plateau in that curve. Because if this curve goes all the way down to 5- or 10% and all the patients progress ultimately, it would be reasonable for others to say, well, is the CTLA4 associated with the long-term durable remission and is that relevant?

And so, we need to look carefully at these data as we follow up. And this will help us inform patients about long-term durability. The adverse event profile of these drugs are actually relatively straightforward. Axi is quite easy. You just stop the axi and the adverse events should get better quickly. And if they don’t, then you’ve got immune-related adverse events and require steroids.

CheckMate 9ER is cabozantinib and nivolumab. The cabozantinib is given at 40 mg. I think this is attractive. This is the oldest of the VEGF-TKI/IO studies to have positive quality-of-life data. David Cella and I have chatted about this and he liked this quality-of-life data. I struggle a bit with it because I can’t see how the quality of life of cabozantinib/nivo should be significantly different from axi/pembro, to be honest. I know axi/pembro is a well-tolerated regimen. Some people say to me is it because the cabo is given at 40 mg? And that’s plausible. Nevertheless, I think there are more similarities and differences, in my opinion.

The important issue here is we’ve seen updated overall survival data, and this was presented at ASCO GU. This is data actually at 24 months and minimum follow-up. Median follow-up was 30 months.

And you can see here that the hazard ratio at 0.70 is very similar to the 0.73 that we saw with axi/pembro.

And then we move onto the CLEAR trial. And like 3 identical twins, sorry, triplets, we see very similar data here. The one thing that’s different with this is the response rate is 73%, which seems higher, and the hazard ratio for progression-free survival of 0.39 seems very impressive.

You’ll be aware that there’s a third arm in this trial, len/everolimus. Realize you talk about this because the data is not as impressive as that for len/pem.

These updated overall survival data show the curves going apart, but there is a suggestion that they’re coming back together again. There’s a hazard ratio of 0.72 in this curve. Again, 0.72, 0.73, 0.70. They couldn’t be closer had you asked them to be. Which is why they are largely interchangeable from an efficacy perspective.

The lenvatinib given at 20 mg is associated with toxicity and a significant proportion of patients require dose reduction. But remember, the majority of patients in 9ER also required a dose reduction. And so essentially, every time one looks at these datasets and tries to find one data point which is outstanding, you can always find a counterpoint in one of the other 2 studies.

I show this curve to highlight the point around the IMDC favorable risk patients. And you can see now, while the initial data suggested survival benefitted with len/pem in the good-risk patients, the most recent data puts it essentially at 1.

This is a similar position to sunitinib. And for those people who said I still give sunitinib to good-risk patients because nothing yet has been better for overall survival, it’s hard to be critical of that. I do put it to those individuals that we have higher response rates and we have higher progression-free survival in the good risk for the VEGF-TKI/IO combinations and therefore, I also say what about giving it the benefit of the doubt and looking at the long-term data which may be significant? But that’s speculative and I accept that point.

These are the ESMO guidelines and you can see they’re largely consistent with what I’ve said.

I’m probably going to change gear a little bit and talk about some second-line drugs in a minute. And the issue with the second-line drugs, you see the evidence goes from Level IA all the way down to Level IIIB and that’s a massive step. And essentially, we don’t have any good second-line randomized data and we need to see that data in the future.

Before I get there, I wanted to talk very briefly about the axitinib/avelumab data. And this is the most recent overall survival cut. You’ll notice that this has similar follow-up to axi/pembro. And while avelumab and axitinib has never had significant overall survival, and while the original sunitinib — axi/pembro data was 0.53 while this was more like 0.8, you can see here now we’ve got 0.79. These curves go apart and they stay apart.

So the avelumab is probably doing something here. And actually, when you look at the landmark of this, it’s not that different from the previous combinations. I’m not suggesting we should be giving this combination. It’s not got significant overall survival, but for those people who say, PD-L1 therapy has no activity, I would probably challenge that a little bit.

And here’s another PD-L1 combination in the front-line space, COSMIC-021, cabozantinib and atezolizumab. And I’ll just show you the response rate data here, but you can see again the majority of patients responding to therapy in line with what we’ve seen. Now, these are numbers are relatively small. And I think it’s important to — by the way, these are the cabo given at slightly different doses. I don’t think it makes a big difference between the two, 40 and 60 mg. Nevertheless, response rate over 60%. And we don’t know what the results with atezolizumab would have been, although we know the bevacizumab/atezolizumab data is not as impressive as we would like.

The progression-free survival, you can see here between 15 and 19 months is in line with what we’ve seen with other combinations, PD-1 and IO. So there’s nothing wrong with this combination. Whether it’ll be tested in the frontline space, I’m not sure anybody knows.

I’m going to switch gears completely now and talk about the HIF-2alpha inhibitor belzutifan. This data was published in The New England Journal of Medicine. These are patients with VHL. These patients have multiple primary renal tumors. You can see from the left-hand side and what looks like a spider, the left arms of the spider, or legs I guess is right, is the cancer which growing with time before the intervention of belzutifan. The right legs of the spider are the intervention of belzutifan and the effect. And you can see a triangular shape where the cancer is initially growing and then the cancers reducing.

This effect is associated with the belzutifan and this is important for two reasons. Firstly, this has got FDA-approval in this group of patients and prevents the need for multiple surgeries, which is terrific. But more importantly, it’s the proof of principle that targeting HIF can have a direct effect on clear cell renal cancer, particularly those proven strongly by VEGF.

This is completely the other edge of the spectrum and this is patients with heavily pretreated advanced renal cancer. And you can see here response rates in the region of about 25%, 15% of the patients getting disease control with belzutifan, is well tolerated. It causes some anemia and it causes some fatigue but is a really well-tolerated agent. And this is perhaps the most promising drug that we have in development. And there are studies, there are prospective studies, randomized Phase III trials of belzutifan versus everolimus in this exact heavily pretreated population. We’re looking forward to the results of that trial.

We can also give belzutifan in combination with cabozantinib. These data are much more difficult to interpret because we know cabozantinib probably has a 25% response rate in its own right in this population. And you can see here that perhaps the results are slightly more impressive than we saw for single-agent belzutifan, but at the same time how much of that is being driven by the cabozantinib?

The reason why I show these data, combination, because it’s hard to interpret, is because the majority of people in the community, including myself, feel ultimately belzutifan will be an agent that’s given in combination with other therapies. It’s well tolerated, with modest response rates. And therefore, these sorts of data I think will be the basis of randomized trials which are currently ongoing.

DR LOVE: Can you explain biologically what the difference is between these VHL patients and the typical clear cell patient in terms of their genomics and germline versus somatic? Like, just real basic.

PROF POWLES: Well, VHL families have inherited Von Hippel Lindau disease and those were 3 German doctors from the 100 or 200 years ago who noticed families who developed essential brain tumors and had bleeds in their brain, that was the most common abnormality. But on top of that, more recently, we found that this group of family — these families — which is an inherited process, also have a high incidence of clear cell renal cancer. And there is a strong link between VHL , HIF-1alpha, and the VHL mutations result in up regulation of HIF, which in turn results in the generation of clear cell renal cancers. And, therefore, by blocking that we can reduce the size of those tumors. These tumors are a very clear — the oncogenic process is very clear. Whereas clear cell renal cancer is more complicated than that. and actually, not all clear cell renal cancer patients you can identify VHL mutations. The majority do, don’t get me wrong. But there are many other mutations or a number of other mutations associated with clear cell renal cancer.

They’re sporadic and they’re more mutationally complex than VHL type tumors on the whole.

DR LOVE: And what actually do you see in the germline analysis? Do you see germline mutations in VHL, in Von Hippel Lindau?

PROF POWLES: Yes, you do. Where you don’t see that with clear cell renal cancer.

DR LOVE: And belzutifan, what type — is it a small molecule? What exactly is it?

PROF POWLES: It’s a transcription factor, so it blocks the — it’s basically a transcription factor, so it’s a novel type of agent that we’ve not used before. And Bill Kaelin who works at Dana-Farber, he won a Nobel prize for his work on HIF and the development of this transcription factor.

DR LOVE: And these patients with VHL, this genetic syndrome, the RCCs that they get, do they respond to treatment the same way as sporadic RCCs? You know, TKI/IO, that stuff?

PROF POWLES: Well, we know they respond to TKIs, but they also respond to HIF inhibition. They’ll respond really sensitively to any VEGF blocking agent. And we don’t know about their response to immune therapy, but we do know that certainly HIF inhibition is very sensitive. And I suspect drugs like axitinib and tivozanib, which are perhaps the most specific VEGF-TKIs, will work extremely well as will, of course, bev.

DR LOVE: And again, these VHL patients, what’s the usual natural history of their RCC?

PROF POWLES: Their natural history of their RCCs is they get multiple tumors and require multiple operations. And that’s why this is an important intervention because you can prevent — in the end, they run out of renal tissue to filtrate the blood. So keeping with the belzutifan is keeping nephron, so sparing nephrons from surgery.

The patients tend to run into trouble with other systemic effects such as brain tumors associated with — and bleeding associated with VHL. And renal cancer, it’s only 1 part of the disease. And it’s not actually the most lethal part of the disease.

DR LOVE: So the idea with belzutifan, I mean do you — if they do have metastatic disease, does belzutifan helpful there also?

PROF POWLES: We don’t know that yet. We haven’t looked at it. But I think belzutifan, the answer should be yes. And if you said to me, why do I think belzutifan is best placed? And I’ve said this before, I think it’s really early, good risk, metastatic disease, which is still quite VEGF-driven. Because later in the disease process, new mutations occur and it becomes more complex.

DR LOVE: There is evidence that belzutifan works in sporadic RCC?

PROF POWLES: Yes. And in heavily pretreated patients, it has response rate of about 25%. And when you combine it with cabozantinib in heavily pretreated patients, that goes up to about 35%.

DR LOVE: And I assume that the genetic patients are far, far outnumbered by sporadic. It’s just a tiny number? What fraction are patients —

PROF POWLES: Yeah, 99 to 1.

DR LOVE: Okay. Great. So that’s interesting.

In terms of tolerability issues of the belzutifan, any issues at all?

PROF POWLES: No. It’s well tolerated. Associated with a bit of anemia. Maybe some shortness of breath. A little bit of hypoxia.

DR LOVE: Any reason to think belzutifan would be useful in other solid tumors? There’s a lot of anti-VEGF strategies used in medicine — in oncology. Or, because of its biology, it only works in renal?

PROF POWLES: I think renal is probably the most pure VEGF-driven tumor. And, therefore, I think it’s probably — well, I can’t guarantee this to you, but I would not be exploring too widely outside of renal cancer.

DR LOVE: Interesting though. Okay, please continue.

PROF POWLES: Well, I want to just switch gears now, if I may, into the papillary renal cancer. As you know, non-clear cell is a family of different cancers, the most common of which is papillary renal cancer. There’s papillary type 1, and the papillary type 2. The papillary type 1 tends to be driven by met mutations; whereas papillary type 2 by FH mutations. Although, I think that is somewhat of an oversimplification.

This is a study, that Monty Pal published in Lancet recently and was presented at GU ASCO last year. Essentially, it compares 4 drugs, cabozantinib, quizartinib, savolitinib, and sunitinib. Sunitinib is the classic VEGF-TKI. We all know about that drug. But the other 3 have a component of MET inhibition to them. Savolitinib is a MET inhibitor in its own right. And cabozantinib is actually a broad spectrum VEGF-TKI with MET inhibition also.

I want to draw your attention to the light blue left-hand curve, the PFS curve was positive for cabozantinib. Cabozantinib, therefore, has become the preferred agent with PFS advantage. It’s reasonable to pursue this without an overall survival advantage because the numbers are so modest with only 40 in each arm, it’s not reasonable to expect OS.

Could we have done this differently and recruited patients —more patients? Yes, of course. And is it easy to recruit hundreds of patients in an investigator-initiated trial into — with papillary renal cancer? It’s very hard to do actually. And so it’s reasonable at this point to say that cabozantinib has superseded sunitinib as a preferred standard of care but without a clear survival advantage. And so there are some people who still use sunitinib, it’s hard to be overcritical about it. But cabozantinib appears an attractive agent.

In the same disease, papillary renal cancer, here we’re talking about actually non-clear cell renal cancer, so we’re including those other cancers, oncocytoma, chromophobe, et cetera.

And this is 427 and here we’re looking at pembrolizumab in this setting. The color code show you papillary, chromophobe and unclassified cancers. And you can see here, response rates, which are very reasonable for this setting. And actually, the overall survival of this drug, pembrolizumab, is in the region longer than we’ve seen for other agents.

Now we have a series of single-arm trials for different agents ranging from pazopanib, everolimus and sunitinib. We’ve got a little bit of randomized data comparing sunitinib and everolimus. And there’s not much in it, but sunitinib is probably a little bit better. And now we’ve got the cabozantinib, which has a progression-free survival advantage over sunitinib.

Nevertheless, when you look at this whole dataset, there are no trials with many more than 100 patients. And actually, this data set with pembrolizumab is probably as impressive as any of the others that we’ve seen.

The last area I wanted to go into was some of the weird and wonderful. And I talk about TIVO-3 as part of the weird and wonderful. Tivozanib is a drug which is not widely used in the United States. It’s not had FDA type approval because of some of the issues with TIVO-1 and the frontline trial.

Here is a randomized trial of tivozanib versus sorafenib. These patients got previously treated advanced renal cancer. It includes patients who have previously received immune checkpoint inhibition. And while there was a significant progression-free survival advantage at 4 versus 6 months with a hazard ratio of 0.73. And indeed, tivozanib was well tolerated and it’s a specific VEGF-TKI, there is still no significant survival advantage of tivozanib over sorafenib.

You may say, what does tivozanib add to the party? And there are 2 things in this dataset. The first is, it’s a well-tolerated VEGF-TKI and it may be good for combining for the future. The second important issue is underlying the point that even axitinib failed to get a survival advantage in the AXIS trial versus sorafenib. And none of the drugs, the VEGF-TKIs, seem to be able to beat each other for overall survival, while they do beat — manage to beat each other for progression-free survival. And so again, there are more similarities than differences between these drugs from an OS perspective.

And the last thing is many patients in this study had previously received immune checkpoint inhibitors. And actually, this is one of the more robust datasets of patients who’ve previously received immune checkpoint inhibition.

Here’s some long-term landmark PFS data, and you can see the advantage of tivozanib versus sorafenib.

I’m coming towards the end now. And this is a study looking at some biomarkers. They’re actually circulating biomarkers. It’s a study I think which is of interest. And it’s from the METEOR trial which is cabozantinib versus everolimus. And we’re looking at classic biomarkers such as VEGF, AXL. We looked at MET. But METs also related, of course, to HGF. And we can see here perhaps the most striking feature is HGF appears to be quite powerfully prognostic. And that was one of the conclusions of this particular paper. The reason I show this slide is there’s still a lack of biomarker research in clear cell renal cancer. I’m aware that there are RNA signatures from the tissue which looked promising. But these signatures really are something which remain exploratory. And it’s unusual in clear cell kidney cancer to have so many treatment options for our patients, but no direction in which to select these patients.

DR LOVE: I’m curious what happens when you add an IO to sunitinib? Pazopanib? Or TIVO?

PROF POWLES: So the answer — the original trials, the original Phase I/II trials were done with sunitinib/pazopanib and axitinib. The original trials showed that pazopanib was associated with an increased frequency of transaminitis. You’ll be aware that pazopanib is associated with high transaminitis as a single agent. And then combining with sunitinib was really challenging at 50 mg. There may be issues around dosing, 4 weeks on and 2 weeks off. And it may be why we gave cabozantinib at 40 mg rather than 60 mg in some of the trials.

So combining these drugs at full doses, we struggled a bit. We gave it with axitinib, 5 mg BID, and Mike Atkins presented that data with 60% response rates and spectacular PFS advantage. And really, we just left behind the difficult to tolerate combinations.

DR LOVE: Any big trials that you’re looking forward to see; maybe that are going to come up over the next 6 months or year, particularly in the adjuvant setting? Any neoadjuvant studies being done?

PROF POWLES: Well, at ASCO there was a neoadjuvant trial, Axel Bex presented it, on a handful of patients, 25 patients. Neoadjuvant axi/avelumab. Response rates, 30% reduction in the size of the primary tumor of only about 30%. So it’s not a spectacular activity in the neoadjuvant space. But there are other adjuvant trials. And there are 3 that were notable. The first is there’s an atezolizumab study and we’re looking forward to the results of that soon, I hope. There’s also an ipi/nivo trial. And again, I hope that’s not too far down the road. And then you mentioned durvalumab and tremelimumab. There’s actually a trial with durvalumab versus durvalumab/tremelimumab versus placebo, which is also around the corner. So that’s the perioperative space.

In the metastatic space, I’m super excited about COSMIC-313, which is cabo/ipi/nivo versus ipi/nivo. So that’s our 3 strongest weapons, VEGF-TKI, CTLA4, PD-1 versus CTLA4, PD-1. I’m really excited about that study.

There’s also the belzutifan study of HIF versus mTOR. So, belzutifan versus everolimus. I’m excited about that trial because we’re expecting a survival signal and therefore, belzutifan getting globally approved. Belzutifan, remember, is that HIF-2alpha inhibitor, which was initially designed for those VHL patients, but it also has activity in sporadic clear cell renal cancer with a 26% response rate in heavily pretreated clear cell renal cancer.

DR LOVE: What about TIVO and IO? Has that been looked at?

PROF POWLES: Yeah. So that combination has been tested. And there’s a study looking at tivozanib with nivolumab versus tivozanib in pretreated patients. I think it’s an exciting study. It’s a very worthy study. It’s not as perhaps as exciting as studies like COSMIC-313 or adjuvant trials. It’s basically saying do you need to keep going with immune therapy in those patients who’ve previously had immune therapy?

It’s hard to see how rechallenge with immune therapy would cure patients, but it may delay progression, it may improve survival. And, therefore, I think is entirely reasonable.

Urothelial Bladder Cancer (UBC) — Elizabeth R Plimack, MD, MS

DR PLIMACK: So thank you so much for inviting me to speak on urothelial carcinoma, 2021 Year in Review. It was quite a year, as it seems most of the years are these days with a lot of new data. So we’ll start walking through it.

So here’s the agenda. We’re going to cover localized non-muscle-invasive bladder cancer. Neoadjuvant approaches. Adjuvant. First-line metastatic. And then Next Line therapy considerations.

So let’s start with non-muscle-invasive bladder cancer. The KEYNOTE 57 trials has been previously presented and published, but an update was recently published in Lancet Oncology in 2021 and this is looking at Cohort A — I’ll remind folks the trial had Cohort A and Cohort B. Cohort B, we have not yet seen results on. That’s your more typical T1 urothelial. This is BCG-unresponsive non-muscle-invasive bladder cancer with CIS, so a very specific pathologic criteria. Patients were treated for 1 year with pembrolizumab and 96 patients in the trial were evaluable for response.

So one thing to note about this disease, this clinical condition, is a low-risk situation for metastases. These patients don’t usually metastasize. And it is curable with cystectomy, with a pretty high cure rate. So, again, the bar for systemic therapy in non-muscle-invasive BCG unresponsive disease with CIS is pretty high. I think in my mind we have to cure at least some people and prevent cystectomy.

So are we doing that?

Well, here is the group of patients who responded and the swimmer plot to the left. And you can see some are still swimming past the 24-month mark, but if you take everyone that was treated in the trial, which is 96 evaluable patients, 11 of them are still potential to be cured of this. That’s about an 11% maximum durable, complete response rate. And again, for their follow-up will be needed to show us if it indeed is durable.

If we’re just delaying more intravesical therapy, you can see the green dots in the swimmers plot include the moment in time where those patients were then routed towards additional treatment for a recurrence.

The good news is, I don’t believe anyone has had metastatic disease in this trial, at least it’s not been reported. But I think this is part of the conversation we have with patients who come for systemic therapy for non-muscle-invasive. We make sure they know they can be cured with surgery if that’s what they choose to do. and that the durable CR rate is maximum about 10%.

Okay, moving on to muscle-invasive bladder cancer, this is a clinical condition where the bladder cancer is localized to the bladder but invading into the muscle. We know that the standard of care here is either trimodality therapy with chemoradiation or, as is preferred at our site and most, neoadjuvant chemotherapy with cystectomy.

So 2021 saw at ESMO presented results of a randomized trial of 2 different versions of chemotherapy we have been using in the neoadjuvant setting for decades but have never been prospectively evaluated in a comparative fashion. So, it was 12 weeks of chemotherapy for all patients. Gemcitabine and cisplatin was given over 3-week cycles and dose-dense-MVAC was given over 2-week cycles. And so the study compared 4 cycles of gem/cis to 6 cycles of dose-dense MVAC. The choice of different numbers of cycles is one that’s been critiqued bitterly as part of this study, but I think we can still learn a lot from it and I’ll walk you through that.

So this study enrolled 437 patients over 5 years in France. They were randomized 1:1 to get 1 version or the other.

What you can see here is, in terms of complete response, which gives one the best chance of long-term survival, we saw 36% rate with gem/cis and a 42% rate with dose-dense MVAC, which turned out to be statistically significant in favor of the dose-dense MVAC. Similarly, statistically significant results were shown for non-muscle-invasive disease after treatment and organ-confined disease after treatment. So here with sort of the short-term outcomes, dose-dense MVAC was superior.

And then they looked at long-term outcomes progression-free survival and overall survival and both showed a benefit to the dose-dense MVAC arm.

Based on these results, the NCCN Guidelines have been updated to prefer dose-dense MVAC as the standard of care in this space. I’m not showing you dose delivery or toxicity data. There was slightly more pancytopenia in the neoadjuvant dose-dense MVAC arm, but no significant difference in long-term adverse outcomes. And remember, this is a group we’re treating for cure with a short course of chemotherapy, either 4 or 6 cycles. I think the ultimate number of cycles is debated. There’s multiple Phase II studies showing smaller numbers of cycles 3 or even 4 for dose-dense MVAC are probably adequate. And so that’s where the uncertainty lies. But I think, in general, for most patients, the dose-dense MVAC approach shows in randomized fashion superior results.

So here is a more exploratory analysis, and this is also in the neoadjuvant space but it’s a very different population of patients.

The group in the VESPER trial were cisplatin-eligible by definition because those regimens contain cisplatin. This is a study look at neoadjuvant enfortumab vedotin in patients who are not candidates for cisplatin who had muscle-invasive bladder cancer but were candidates for cystectomy.

So before we go into the results, I’ll say this. In our practice, this is a very narrow population of patients. We feel comfortable giving cisplatin to patients with creatinine clearances above 50, that’s our bar. Most patients we feel can get it safely. And those that can’t typically have multiple comorbidities that also make them a risky candidate for surgery. And I’ll get to one of the abstracts presented at GU ASCO in a minute, showing that for those patients trimodality therapy with a scrapping of the bladder, TURBT, followed by chemotherapy, concurrent with radiation therapy, is probably a better approach. But this study looked at that narrow group of patients and treated them with enfortumab vedotin.

So 22 patients were treated. Overall, I’m showing here the adverse event rate. Three patients died on study, again related or unrelated. It is unusual for a patient to die of a disease that’s curable. Some of them, it was related to their surgery indicating maybe a group at high risk for surgery. And then the side effects here are noted in terms of EV. So Grade 3 EV related AEs included asthenia, dehydration, skin rashes, which are an issue with EV.

And so this study, there was another on a poster at GU ASCO looking at ipi and nivo in this space. I think what we’re seeing is, these are really tough treatments. They have toxicities. The group of patients who can’t get cisplatin may be at higher risk for surgery and may not really be good surgical candidates.

And there was really nice retrospective review by Dr Zlotta from Toronto, looking at case control comparison of chemotherapy with radiation, trimodality therapy, compared to neoadjuvant chemotherapy followed by cystectomy with equivalent results. And so this may be, or at least in our practice here, we’ve concluded that this type of patient is probably best served by trimodality therapy and not a novel neoadjuvant compound.

DR LOVE: What is known about combining chemotherapy, or for that matter antibody drug conjugate like EV, with an IO in the neoadjuvant setting? Of course, that’s being done in lung cancer, a lot of other tumors. What do we know about it in this situation?

DR PLIMACK: So there are trials looking at just that. So enfortumab vedotin with pembrolizumab showed extraordinary efficacy in the metastatic setting.

I believe the best trial design, and I know there are a couple floating around out there, is to test EV plus pembro in a group eligible for cisplatin, this sort of more robust, healthier group of patients, for a couple of reasons. One is the combination is not necessarily benign. EV has toxicity and pembro can have toxicity. Again, these are curable patients. And the results were so good in the metastatic setting they rivaled cisplatin. For the first time we saw results that rivaled cisplatin in terms of short-term outcomes with response rate.

So, stay tuned.

DR LOVE: Can you give dose-dense MVAC plus an IO, incidentally?

DR PLIMACK: Ah, great question. We are testing that here at Fox Chase as part of RETAIN-2 trial. There is data out of Europe that looked at that question as well.

The answer is, the safety looks good for that so far, but the efficacy may not be additive. And so that’s again consistent with what we saw in the metastatic trials, which I’ll go through in a minute, which looked at chemotherapy plus IO. We saw a benefit but not enough over chemotherapy to justify combination versus sequential. And so we’ll see as those studies evolve. But unfortunately, most of the randomized Phase III trials did use a gem/cis backbone. We now know that’s not quite as effective as dose-dense MVAC. And we’re also there’s a knowledge gap as to how dose-dense MVAC interacts with immunotherapy. So, we’ll learn.

DR LOVE: Interesting.

In terms, again of neoadjuvant chemo/IO, one of the things they talk about is, they call this like major pathologic response, but it’s not a path CR. Like they see stuff but it looks dead or something. So they have this different classification. Is that seen in bladder also?

DR PLIMACK: Yeah, it is. Let me just go back a slide to show you how we look at it. So here, if you look at the non-muscle-invasive, that bar in the middle, you can see less than T2, that’s like a downstaging. And then greater than T2 with the node-positive, that’s sort of an upstaging. So we do quantify response in different ways. In the top bar, you can see we separate it from T0, which is what I focused on in my discussion, but also sort of folks who progressed through and those who had essentially stable disease with some response.

Part of the problem with staging, Neil, is that it’s just really inaccurate in bladder cancer. So knowing where someone started is a guesstimate, and then knowing where they ended pathologically is a definite. And so that comparison is a little bit flawed in terms of calling it a response. But yes, we are looking at responses of all kinds in complete but also partial.

DR LOVE: Tonight we’re talking about this neoadjuvant nivo study, nivo plus chemo, that use in — they presented at lung — at ASCO and they had a major — different surgery, less pneumonectomies, for example. It was randomized. So it’ll be interesting in bladder, that would be awesome if you could modify the kind of the surgery you needed. I don’t know if that’s possible.

DR PLIMACK: So we’re doing that. We’re doing that actually here at Fox Chase and some other centers. Matt Galsky has a trial. And there’s a trial through the Intergroups looking at major response to systemic therapy, in our case dose-dense MVAC, and then we have a follow-up study, dose-dense MVAC with nivolumab, to see if pathologic complete response can be achieved, assessed appropriately, and those patients can keep their bladders. So thank you for that comment.

DR LOVE: No, it’s very exciting trying to think about what’s coming up next. One thing about bladder cancer it’s a much bigger operation than most solid tumors. It’s big operation.

DR PLIMACK: It sure is. Yeah, it’s life-altering. Usually there’s an ostomy. And so patients really enrolled enthusiastically in this trial, even with all the unknowns about whether this will work or be effective for them because there’s close monitoring and there’s a chance of keeping their native bladder which they value.

DR LOVE: I mean there’s been a lot of attention in breast cancer avoiding mastectomy. But I mean avoiding cystectomy from a quality-of-life point of view is gigantic.

DR PLIMACK: And you can see here the response rate, 36.4%, pathologic complete response rate, again those have the best outcomes. We were just talking about downstaging, and that’s what you see here, 50% downstaging rate.

This is almost identical to what is seen with gemcitabine and cisplatin. So it is, while impressive that we’re achieving this in a group that can’t get cisplatin, it’s not quite up to par with what we would expect with dose-dense MVAC. But again, these are different patients.

I just caveat this with 3 deaths in 22 patients is concerning, we know EV can be hard on the patients. And the real question is, would these patients be better served with trimodality therapy? I think we’re leaning towards yes, at least in our practice.

DR LOVE: And when you say trimodality, you mean chemo and IO, for example? Or just chemo?

DR PLIMACK: No. So trimodality therapy, TMT, has been a term used for having a complete TURBT of the bladder, scraping all the visible tumor, debulking, followed by radiation with low-dose chemotherapy as a radiosensitizer. It’s also sometimes called chemoradiation or CRT, but TMT is probably more accurate because you really do need that TURBT beforehand. And it is an equivalent treatment to neoadjuvant chemotherapy with radical cystectomy.

There are pros and cons to both, which are probably beyond the scope of this discussion. We tend to favor cystectomy for those who can achieve it, in part because a bladder that’s intact, whether from one of our RETAIN trials, for instance, or even from trimodality therapy, is prone to tumors down the road. So that’s one consideration. But for this group of patients who are cisplatin-ineligible, I think we may be overestimating their cystectomy eligibility, given the outcomes of this particular trial, and maybe TMT would be best for them.

DR LOVE: I certainly have heard about that therapy but I haven’t heard it called trimodality therapy. But again, that’s maybe just me.

DR PLIMACK: I think they’re trying to rebrand it and I’m all for it. It’s more interpretive.

So moving to the adjuvant space. These are patients who had perhaps neoadjuvant chemotherapy prior to surgery, perhaps no neoadjuvant chemotherapy. But after radical cystectomy had residual disease that’s high risk, as noted here. T2 to T4a, or node-positive. And the criteria a little bit different whether patients got previous neoadjuvant chemo or not.

This is the first trial evaluating adjuvant checkpoint inhibitor therapy in this space, it’s atezolizumab versus observation. And this showed no benefit in terms of disease-free survival or overall survival.

Subsequent studies, which I didn’t include here, show that circulating tumor DNA may be able to separate a group of patients who might benefit in this otherwise negative trial. Those data are compelling but remain to be validated. And ongoing trials are looking at that.

So this trial was negative for both endpoints, disease-free and overall survival.

However, hot on its heels, came this trial. This is the same design, high-risk patients, randomized — almost the same design — randomized 1:1 to receive nivolumab adjuvant, or placebo in this case. And this trial reported a disease-free survival benefit, which is shown here, for nivolumab over placebo. You can see the hazard ratio is 0.72. However, they did not report overall survival data, saying it’s not mature. And so the real question here is, what does this mean for our patients? This is now nivolumab, FDA-approved adjuvant therapy for high-risk muscle-invasive bladder cancer after resection. So it is a tool we have in clinic that we talk to patients about. But we have 1 negative trial, 1 positive trial. We’ve been here before with adjuvant renal cell carcinoma where we only had 1 positive trial, multiple negative trials. So other trials will read out to help flesh out the story.

However, what do we recommend for our patient? And I think the key is that immunotherapy is not without risk but it’s generally safe and well tolerated. And would a patient want this to delay recurrence even if we weren’t necessarily increasing their chance of cure? Whether we are increasing their chance of cure is unknown. But likely, if we are, it’s on the magnitude of what we see here in terms of the delta between the curves. So, maybe 10% or so.

So this is how we frame it to patients. We quantify their risk, their risk of recurrence, and we go through a discussion on whether this makes sense given the uncertainty at the moment.

DR LOVE: Can I just ask you what your own hypothesis is about why one of the trials was negative and one was positive? Do you think it could be some kind of an events thing? I mean I would guess you would think that the treatment should be about the same? Do you think somehow about the design and the eligibility and the events and all that, that it’s that kind of thing?

DR PLIMACK: Yeah. So we’ve looked at this in some detail and the companies have looked at it in even more detail, one thing I know it’s not is that the prior study was randomized to observation and this one was randomized to placebo. And the modeling has been done to say placebo didn’t make this trial positive, there wasn’t enough — it’s not that dropout made the other study negative. So I just want to make that point because I’m generally outspoken against placebos. I don’t think they’re necessary in the adjuvant space. It’s a lot of saline infusions for our patients on the placebo arm. But that sort of soapbox aside, I think we don’t know the difference. I think there could be a difference in the drugs themselves. Atezolizumab and pembrolizumab have shown divergent results, with pembrolizumab generally showing benefit in cases where atezo has not in the past, but not always in that consistently. And I think these data may mature to show the real answer over time and we’ll just have to watch them.

DR LOVE: Are you generally offering this in your practice outside of trial?

DR PLIMACK: Yeah, that’s a great question. We are definitely discussing it with patients. But the way I frame is that we would love to have a tool to increase their chance of cure. We are hoping this does that. Right now we don’t know if it does or not. If it does, it’d probably be on the order of 5- to 10% increase in chance of cure. And then we go into the known risks of pembrolizumab therapy. And I generally approach this saying this is a true joint decision-making conversation with the patient, right. In other situations I can make a recommendation, and I do. And here it’s really making sure the patient understands the potential risk-benefits and the unknowns in the moment and some patients will say, yes, and some patients will say no, thank you.

So just to pull everything together for localized disease, which we just spoke about, KEYNOTE 57, pembrolizumab, is now approved for non-muscle-invasive bladder cancer with a CIS component but the potential for cure in this space is low, on the order of 10%. We’re still waiting to hear about Cohort B which represents the majority of our patients in this space. And I think the long-term results will define the clinical utility in this setting right now.

For muscle-invasive bladder cancer, neoadjuvant cisplatin-based chemotherapy is still the standard of care. Dose-dense MVAC has superior outcomes in terms of response rate and progression-free and overall survival compared to gem/cis.

Neoadjuvant EV, I put 3 stars there. I think we still need to see more, but in this particular cisplatin-ineligible, cystectomy-eligible population. It seemed dangerous and I don’t think is going to enter prime time at the moment.

For adjuvant high-risk resected urothelial carcinoma, this is the adjuvant discussion we just had, we have 2 different trials with different outcomes. The only data we have is for disease-free survival right now, no overall survival, and that’s with nivolumab. So that’s a shared decision-making conversation with patients at high risk for recurrence.

Okay, so moving on to metastatic urothelial carcinoma, it’s important I think to follow frequent updates of ongoing trials. This is not new news, that we’ve been watching these trials for years, but really what we talk about with immunotherapy in cancer are long-term, durable results and that’s what I think we’re seeing here in urothelial cancer.

So the panel on the left is KEYNOTE 52. It’s first-line pembrolizumab versus platinum-ineligible metastatic population. And you can see here now we’re out 4 years and 19% of patients are still alive. That’s not something we’re used to seeing in urothelial cancer, especially not in the cis-ineligible population. So I think this just goes to show that even though the label for first-line checkpoint has been revised and changed and we probably are going to use it for the most part in the maintenance space. The fact that we’re seeing some durable responses here is really encouraging.

The panel on the right is with pembrolizumab versus chemotherapy. This is part of the KEYNOTE 36 trial, which had another arm looking at pembro plus chemotherapy versus chemotherapy. And here, in the randomized setting, so the panel on the left, is non-randomized single agent. Here’s the same population, platinum-eligible, could be cis-ineligible or not, but these are also cis-eligible patients. And you can see, I think humblingly, the curves are both pretty good. At 2 years we have 40%, about, alive, which is good, but really no benefit to the pembro over chemotherapy. And I think that might be speaking to what we’re about to show you with the JAVELIN trial, which is that immunotherapy works best as maintenance after initial chemotherapy.

So here is metastatic urothelial carcinoma of the IMvigor130 study. Same sort of idea, but here we’re talking about atezolizumab. This particular study looked at atezo versus platinum/gem. It also had an arm with the combination which I told you had results that were not impressive. But the point I want to make here is that for atezolizumab in this setting, PD-L1 status really matters. And the panel on the left, you see the blue line, which is atezo, showing early deaths, right. If you go up to like 10 months here, way more patients die faster with pembro than chemo, and then the curves kind of start to coalescence. But I think we can agree here that patients are doing better on chemotherapy if they’re PD-L1 low than with atezolizumab in general.

When we look at the panel on the right, however, it switches. So here a PD-L1 high result shows a benefit to atezolizumab over platinum/gem combination. And I should say of all the trials that are similar like this looking at randomized checkpoint frontline compared to chemo, atezolizumab is the one that shows the curves separate. The others don’t.

So here, again I’m comparing PD-L1 high patients. On the left is the atezolizumab curves I just showed you, showing a benefit. The blue is atezolizumab. On the right is the pembrolizumab curve I just showed you, showing really no benefit to pembrolizumab over chemotherapy in this select PD-L1 high group.

So in our practice it’s very subtle, but I tend to favor atezo as frontline for these patients based on this cross-trial comparison.

Okay, Neil, I think you were asking about this, EV/pembro combination. This is in the metastatic setting. We were speaking before about whether it might move to the neoadjuvant setting, and I think it has in trials. But let’s look at the data that got us all excited about this combination in the first place.

So this is EV-103. It’s a basket study that looked, sort of expanded as we good results, and these data presented by Terry Friedlander at ASCO last year collect a total of 45 patients treated with EV plus pembro in the frontline setting. These are patients who are cis/platinum ineligible.

So here, this is one of the best figures of 2021 in my mind, this spider plot. So this is now updated with longer-term follow-up and you can see the blue bars are responders and the orange arrows are patients still in response. And so, while you can have progression and still have your blue line remain lower than your baseline with a new lesion, that’s probably not every blue line has arrows. I think we’re seeing some durability of response here now coming out to 2 years in patients who got this in the frontline. And again, this is a group of patients who historically don’t do as well, they’re not candidates for cisplatin.

And here are the longer-terms results are depicted differently. The median progression-free survival is 12 months. That in and of itself, isn’t necessarily better than platinum, but median overall survival of 26.1 months is really quite good. And so I think we’re seeing in this small cohort of 45 patients some very promising data that’s going to be followed up by — in multiple randomized trials, both in the muscle-invasive setting, in the metastatic setting, which is what we’re showing you here.

So here are the toxicities and we point this out only to say that EV has its own set of toxicities, pembro has its own set as well. And we did see them in this combination. So especially peripheral neuropathy, which is something we definitely see with EV, most were low grade, less than Grade 3. Skin reactions, very common. You see 20% Grade 3 and above skin reactions. Some of these can be severe. We’re watching skin tox very carefully with this combination, with EV in general. With pembrolizumab, we know there’s a risk of autoimmune rash. But whether they’re synergistic toxicity here I think is something trials like this will help us sort out.

And then glucose metabolism is altered by EV in ways we don’t quite understand, but here you can see greater than/or equal to Grade 3 hypoglycemia occurring in 9% of patients, usually manageable. Immune-mediated AEs are at the bottom. You can see 26.7%, a little higher than we’d expect with pembro alone, greater than Grade 3 immune-mediated events.

So is EV contributing to this? Are we having trouble attributing it because it’s a combination? Or is this just sort of what we’re going to see with this combo?

So not non-toxic, but maybe worth it because the efficacy is so good.

Okay, moving on. So this is now what I would say is the current standard of care in metastatic urothelial cancer and it’s sort of the JAVELIN approach. And we’ll talk through this study, there’s a follow-up slide in terms of follow-up data that will support this statement.

So patients with metastatic urothelial cancer should really first get gemcitabine and cisplatin or gemcitabine and carboplatin. If they progress through that, and that’s about 15- to 20% of our patients, then they go on to salvage therapy, usually with a checkpoint inhibitor such as pembrolizumab which had a positive Phase III trial in that second-line space. But for the 80- or so percent of patients who have a complete response, a partial response, or stable disease after this chemotherapy, in this trial they were randomized to maintenance avelumab versus best-supportive care alone. And these were results were really impressive.

We saw an overall survival benefit now, a hazard ratio of 0.76. You can see here at longer follow-up; I think the more stable curve — the curve is stable I would say out to 24 months now and you can see a difference of 10% in terms of overall survival. And then in terms of progression-free survival, here a couple of interesting things to note. On the right, you can see at 24 months 23.4% of patients who have not yet progressed. It was also interesting is that bottom curve, which is the percent of patients on best-supportive care alone, those patients are not getting any therapy. And you can see that at 24 months, 7.1% of them still have not progressed. So those are durable, complete responses. Again, that rate, we knew it existed. We didn’t know quite how to quantify it. But going out to 24 months, we think it’s about 7%.

What we recommend is upfront chemotherapy, cisplatin-based, followed by maintenance immunotherapy, and the Phase III data support avelumab here.

So now moving on to your patient who has already received their platinum-based chemotherapy for metastatic disease. They’ve already received their checkpoint inhibitor per JAVELIN or as salvage, then they land here and sort of the inclusion criteria for this study.

So this is a study of enfortumab vedotin, single agent, not in combination, as we discussed before, versus dealer’s choice chemotherapy, taxane or vinflunine. And you can see the hazard ratio is 0.7, benefit to enfortumab vedotin. So we were pleased to have a more effective agent in this space that — we saw it in the Phase II data. It was approved, accelerated — on an accelerated basis there, and now we’re happy to see this confirmed in Phase III.

Here is a different population. These are patients who received enfortumab vedotin without any cisplatin therapy. This is like the group of patients who got PD-1 or PD-L1 as first line. I’ll say, this is a narrowing population for the reasons we just discussed, that for patients frontline checkpoint is not the way to go. You want to go with chemo and then use the checkpoint as maintenance. But for your patients who did get first-line PD-1 or PD-L1, EV was tested in this space with pretty nice results. You can see most patients had shrinkage in the overall survival, with a lot of censoring, is really looking quite good for this — for this population of patients who generally do quite poorly, in general.

So moving on to TROPHY-U-01. So there were a lot of data on this new agent, it’s called sacituzumab govitecan. It’s an antibody drug conjugate to TROP2 with a topotecan type toxin attached. And they’re really interesting. Right now, we only have single-arm data for sacituzumab govitecan, and so we’re analyzing all these different cohorts. The randomized trial is ongoing. We haven’t seen results from that yet. They’ll be randomized to chemotherapy.

And so here you can see that we can achieve tumor shrinkage for patients who progress after platinum-based chemo and checkpoint inhibitors and here’s the overall survival probability. Again, I think most of us use EV before sacituzumab govitecan, based on the combination of efficacy, in terms of response rate, was a little better with EV. General familiarity, I’m sure. Sacituzumab govitecan is a little newer.

The toxicity is not nothing. There is some diarrhea and GI toxicity. This is definitely a tool in the toolbox, but generally given after EV.

So there was a small study of sacituzumab govitecan plus pembrolizumab presented at GU ASCO. The overall response rate was 34%, not really that much better to the agents alone. So pembro alone in this space, response rate is 21%. SG alone, response rate is 27%. And then compare it to the competitor, EV, where the overall response rate is 44%, I don’t think we’re going to be using this combination in clinical practice. Also because most patients I think deserve the opportunity for single-agent checkpoint in the maintenance setting because that can be very well tolerated for the long term, leading to a long period of patient benefit, with good quality of life.

Erdafitinib. So erdafitinib is a tool for a small subset of our patients, maybe about 20%, who have a qualifying FGFR alteration on NGS testing. And so here we have longer-term follow-up from Arlene Siefker-Radtke, looking at the 101 patients in the single-arm trial. You can see here their characteristics. Many of them had a number of prior therapies. And so here we have overall survival. You can see again where at 2 years about 30% still alive on this drug after multiple prior therapies. It’s really nice to see.

Progression-free survival looking a less optimistic in terms of durability of response. But again, it’s important to follow these long term so we know what the long-term outcomes are.

So having whirled through all of those data and updates in 2021, I summarize all our recommendations. This is what we do here at Fox Chase. There are certainly other appropriate ways to treat patients, but here, if were talking about frontline metastatic, usually we recommend dose-dense MVAC or gem/cis for 4 to 6 cycles, followed by avelumab or, in some cases, pembrolizumab maintenance. I didn’t show you those data, but there’s a small study suggesting pembrolizumab could work. And that is given every 6 weeks instead of avelumab every 2, which is its advantage.

In the cisplatin-ineligible cohort, those patients should get carboplatin and gemcitabine. As tempting as it may be to try EV or EV/pembro or pembro, those should be done really in the context of clinical trials and not without PD-L1 testing.

After they have a response, they go on to receive maintenance as above.

And then alternate therapies for cisplatin-ineligible patients, sometimes gemcitabine single agent can work if they haven’t had it before. And then if we’re going to use a frontline checkpoint inhibitor alone it would be only for a patient with PD-L1 positivity proven, and it would only be in my book, atezolizumab. Although some would say pembro is just as good.

And then going to subsequent therapy. So the box at the bottom, always consider a clinical trial. There are a lot of them out there. We’re really looking to improve this space, improve outcomes in this space. But right now with what’s approved now, according to the guidelines, post-first-line platinum recommend pembrolizumab either as maintenance or if they progress on their platinum, as salvage. Alternatives can be erdafitinib, or EV or different PD-1 or PD-L1 inhibitor, but generally it’s pretty standard to give avelumab if it’s in the maintenance setting.

Post-first-line PD-1 or PD-L1 inhibitor, enfortumab vedotin or gemcitabine/carboplatin. That’s if a patient is still cisplatin-ineligible, which we presume they would be if that’s what they got to begin with. Some alternatives are mentioned to the right. And then once they’ve had their platinum therapy and their checkpoint inhibitor, then we turn to enfortumab vedotin, sacituzumab govitecan. And for those who are FGFR altered, erdafitinib. And those can be discussed what order they’re given in. I think there’s no clear data to support one over the other, although we tend to go with EV and then erdafitinib, if possible, and the sacituzumab govitecan.

DR LOVE: There is HER2-positive bladder cancer, right?

DR PLIMACK: There is, and now that we’re sending NGS on metastatic patients to determine if they have an FGFR alteration or their PD-L1 status, I think we’re seeing some HER2. It’s not that common. We’ve targeted it before in bladder cancer with unimpressive results. And so, HER2-directed therapy has not hit the standard of care for bladder cancer yet. But I think these are trials are interesting. They’re novel antibody drug conjugates.