Arlene Siefker-Radtke:
Thank you all for joining us today to discuss about the recent abstracts presented at GU ASCO 2021. I am Arlene Siefker-Radtke, a Professor of genital urinary medical oncology at the University of Texas, MD Anderson Cancer Center in Houston, Texas. And I'm joined today by several stellar colleagues who are really doing amazing work in the field of genital urinary tumors. We bring to you from across the pond or the ocean as it is, Dr. Alison Birtle, who's a Professor at the Lancashire Teaching Hospital in Preston United Kingdom. We have Dr. Andrea Apolo from the National Cancer Institute in Bethesda, Maryland and Dr. Thomas Powles from the Bart's Cancer Institute in London, United Kingdom. Thank all of you for joining us today.

Alison Birtle:
Delighted, Arlene.

Thomas Powles:
Thank you for [inaudible].

Arlene Siefker-Radtke:
Thank you so much. And I know that this year's GU ASCO in the field of bladder cancer, we had several Phase III trials showing level one evidence or early evidence for activity of different agents. Tom, you were involved enfortumab vedotin trial, why don't you tell us a little bit about your work?

Thomas Powles:
Well, it's kind of you Arlene. It's a global effort and many different people, hundreds of sites around the world took part and investigators. Essentially, we've got established in urothelial cancer platinum based therapy, and immune checkpoint inhibitors are established. And we performed a randomized Phase III study of the enfortumab vedotin versus chemotherapy in those individuals who previously received or progressed on the two lines of therapy, immune therapy and chemotherapy. The randomised trial was of enfortumab vedotin which is an antibody drug conjugate targeting nectin, it looks really active in urothelial cancer.

And on top of that almost all urothelial cancers overexpress nectin so it's a really strong rationale. The control arm was standard chemotherapy: paclitaxel, docetaxel, vinflunine. The study met its primary endpoint of overall survival with a hazard ratio of 0.70, which was statistically significant. 13 months overall survival in the study arm, it also hit PFS and response rates endpoints. It worked across broad subgroups of patients and the toxicity profile was in line with expectations, adverse events and special interest, skin toxicity, neuro toxicity, peripheral neuropathy and also hyperglycemia.

And actually, just as with chemotherapy and immune therapy, when they came along, we learned how to use those drugs. We're going to need to learn to use enfortumab vedotin from a global perspective as well. And I'm really keen to take part in that educational process. I think it's practice changing. My current standard of care will be frontline chemotherapy, maintenance avelumab, and EV at first progression. So that would be how I would treat patients. I know there are other people who feel differently about that.

Arlene Siefker-Radtke:
Oh, that's excellent data that you presented and certainly suggests that level one evidence that enfortumab vedotin is here to stay and will be part of our long term strategy for urothelial cancer. I know there have been questions about some of these specific toxicities. I wonder if other members of the panel would speak about their experience and what they use to monitor and manage these patients receiving enfortumab.

Andrea Apolo:
I can make a comment. I think that the randomized data looks terrific. And one of the remarkable things about the response rate which was 40% is that we saw it in the Phase I, we saw it in the Phase II and now we see it in the Phase III. So not like a lot of other agents and actually is maintaining these responses in the larger trials. I think toxicity is definitely something that is a learning curve with this agent. People don't think of it as a chemotherapy because it's a antibody drug conjugate, but it's a chemo and it has chemo toxicities. The alopecia the cytopenias and it does have some unique toxicities, the hyperglycemia is unique.

The rashes are unique, but generally they're pretty well managed. And just like we dose reduce, we hold doses within a clinical trial, you can't do that. But once you're managing these patients in the clinic, you can hold doses and dose reducing manage them and learn how to manage them. It's a weekly dose, three weeks on one week off. So I think there's a lot of room for holding and managing our doses depending on adverse events.

Alison Birtle:
I just want to say that, Thomas, it's all going to be about education, isn't it really? Because we didn't know how to manage immune toxicity to start off with, but we're pretty good at doing it now and educating our patients. And because this is going to be very different in terms of the hyperglycemia which we still don't really know the mechanism for, it's going to be just educating the acute oncology teams, the chemotherapy nurses, the patients and the oncologists outside of the main centers about what to look out for as this drug becomes used more in combination and more widespread when licensing in Europe is achieved.

Thomas Powles:
One of my observations is the drug's being moved quite quickly earlier in the disease testing. And we're not sure about how those trials are going to pan out. But I know that there is... We saw some data at ASCO, which I think Andrea is going to talk about looking at response rates in individuals who had only progressed on immune therapy. We've previously seen in enfortumab vedotin plus pembrolizumab with 74% response rates, which are really, really high. Remember, GemCarbo's response rates are significantly behind there.

So I think this drugs can establish itself really early in this disease process, I think it's going to be really important in the future. And for me, the outstanding issue is that at last, we've got a new class of active agents which are good at getting in control of disease. We've really struggled with that to some extent in the past. And so I think this is a really important step for bladder cancer.

Arlene Siefker-Radtke:
I certainly agree with all of you. The ability to use a non-platinum agent has been a largely unmet need for patients with urothelial cancer. And we also saw a presentation at ASCO about the use of enfortumab vedotin in these patients with poor kidney function or who were otherwise ineligible for platinum-based therapy. Alison, what are your thoughts? Is there a higher risk of using this agent in poor kidney function or people who are not eligible for the typical standards of care?

Alison Birtle:
Look, I think these patients are going to die of their bladder cancer. And at the moment, we've had no agents really, that we've been able to use, particularly if they aren't suitable for carboplatin. And so it is very exciting to be able to offer a novel agent for these patients. Arlene, I'd like to ask you a question if I may, there was one abstract looking at sequencing of giving erdafitinib in patients who are FGFR receptor positive, and EV, which I know that you were involved with, and I thought that data was potentially very interesting, although we can't use erdafitinib, certainly in the UK at the moment. I just wondered if you wanted to share your experiences of that with us?

Arlene Siefker-Radtke:
Well, I think the use of sequential novel agents is certainly important since we don't yet have curative data in this setting. And trying to determine should an FGFR altered patient, do they benefit as much from immunotherapy? Do they benefit as much from chemotherapy? And do they benefit as much from other targeted agents such as enfortumab? I can tell you my own clinical sense has been that we do see activity with these additional novel agents, especially with agents like enfortumab vedotin.

So I think the Nectin-4 expression may overcome some potential biology, if it is indeed associated with the FGF altered tumor. I think there still is some debate out there about FGF altered tumors and how well they respond to immunotherapy. With retrospective data sets suggesting they may respond as well and others suggesting they may not respond as well. And designing trials that help us answer these questions are going to be essential as we move the field forward. Will bladder cancer become like a leukemia where in certain biomarkers you have a different strategy of agents or sequence of agents that you use to a patient's benefit?

I think it's still too early to tell. But I think the future may be heading toward that personalized strategy where each bladder tumor requires its own essential treatment. They're not all created equal and we even see that with agents such as immune checkpoint inhibitors. Not every patient responds and being able to predict those that benefit might help enhance so patient selection. And even thinking about immunotherapy we had another great abstract at GU ASCO, looking at adjuvant nivolumab and its role in the treatment of urothelial cancer. Andrea, I know you've done some work in this field, would you care to describe the trial and the early findings?

Andrea Apolo:
Yeah, of course. And I was very grateful to be a moderator for this abstract. This is the CheckMate 274 study. It was presented by Dr. Dean Bajorin. It's a randomized Phase III trial of adjuvant nivolumab versus placebo in high risk muscle-invasive bladder cancer patients. They randomized 700 patients and the primary endpoint was disease free survival in the intent-to-treat population, and in the PDL positive population using the nivolumab as a 28 A with 1% or greater as PD-L1 positive. And the trial was positive at both ends.

In the intent to treat population, the median disease free survival was 21 months for the patients receiving nivolumab versus 11 months for the patients receiving placebo. So this was quite significant, the hazard ratio was 0.7. And in the PD-L1 positive, this was even more remarkable with the hazard ratio of 0.5. So very remarkable data. Overall survival data was not presented and this is because the number of events had not yet been reached. And there was a question about whether this changes standard of care.

I think it's a little tricky right now and confusing because we have the IMvigor010 data that Tom presented, that did not meet its primary endpoint of disease free survival, with atezolizumab and the adjuvant setting versus observation. Although he did present some beautiful ctDNA data showing that patients that have positive ctDNA actually do benefit from a adjuvant atezolizumab.

This was a retrospective exploratory analysis. So now where we have this great DFS data with adjuvant nivolumab negative, atezolizumab data possibly positive with ctDNA positive data. And really, when you look at the hazard ratios of the CheckMate 274 data, it looks like primarily the patients that are PD-L1 positive are benefiting. So I think we need to wait for survival to see what is the outcome in these patients. That being said, we are running the AMBASSADOR trial, which is adjuvant pembrolizumab, in again, the same population, high risk muscle-invasive bladder cancer patients.

And we're almost done with the pool and I think it's important to see this data and how everything looks together in terms of what is the benefit of adjuvant immunotherapy in this high risk muscle-invasive bladder patient population that already got chemo. Half of them had already received chemotherapy in the neoadjuvant setting, or are cisplatin ineligible and now are postsurgery and really have no options that they can be put in an observation arm.

Arlene Siefker-Radtke:
So this was really compelling early data. But you pointed out very correctly that we have one positive trial and one negative trial. Tom, what do you think accounts for these differences between two trials and two very similar agents? Was it just luck of the draw? Or is there something fundamentally different going on?

Thomas Powles:
So I spent quite a lot of time with different trials in urothelial cancer. And we think the randomized Phase threes are the whole truth and nothing but the truth, but there is probability and chance involved in these trials. And that's why we have confidence intervals. So the first thing is that that may have played a little bit of a role, number one. Number two is, there may be subtle differences between nivolumab and atezolizumab although I'm not convinced by that, as you know, the avelumab study, a PD-L1 study, has the lowest hazard ratio of any of the trials. And that's a PD-L1 inhibitor and many people felt that was the least active of the drugs.

I don't think that's the evidence that we have in front of us. So I'm not convinced that in bladder cancer, there's a big difference between the drugs. The third explanation, which actually I'm a little bit more into right now is that the control arm was different. One was placebo, one was open label. For disease free survival that's relevant, because the dropout rate in the atezolizumab arm was 20%, only 10% in the nivolumab arm. And you can see a patient being ran, that's in the control arm, you can see a high risk patient being told they have a very high chance of relapse, being randomized to best supportive care and dropping out to go and find some other form of therapy, be it adjuvant chemotherapy or radiotherapy or something.

And I can see that process occurring well, of course, that's less likely with placebo. If you look at the outcome of the control arm... Sorry, the study arm, the median survival is 21 months for both in the nivo and atezo arm. But in the control arm, atezolizumab arm is much longer than the nivolumab control arm. And so that's a possible explanation for the findings. Remember, again, I agree with Andrea, overall survival is gonna turn out to be really important in this. And that will tell us exactly what's happened. And with one negative trial, with the uncertainty, or the positive trial with the DFS endpoint, it probably is premature to be treating unselected patients, whether you want to take a risk with the biomarker positive patients. That's a really interesting discussion and I'm not sure I know the answer to that.

Arlene Siefker-Radtke:
So would anyone on the panel see a potential early approval? Perhaps in that PD-L1 high cohort, there did seem to be a separation in the plateau of the curves rather than just a shift in the curves. And if you think of cure, which I think most of us in the neoadjuvant and adjuvant space, are trying to increase the cure fraction. Do you think there's a potential path toward an early or accelerated approval in PD-L1 high tumors?

Alison Birtle:
I personally think that there is likely to be in that patient group. I think the one thing we haven't pulled out in the discussion so far is that this CheckMate 274 was negative for upper tract tumors, which were about 20% of the population. And neither CheckMate or IMvigor 130 were... Sorry, 010 we're powered for that additional upper track group of patients, Arlene, because they were sort of added on afterwards when they were struggling to recruit the post cystectomy patients. Whereas, of course, we do have level one data from the updated POUT analysis that was presented in terms of adjuvant chemotherapy for the upper tract. And I think it just shows you that these are two completely different disease entities.

But is disease free survival important? I certainly think it is because at the moment if somebody's got recurrent disease in either the upper tract or muscle-invasive bladder cancer, they're potentially looking at six cycles of chemotherapy and then maintenance switch with nivolumab. So in terms of resources is going to be far more and keeping somebody disease free as we've seen in prostate cancer in an early disease state, less time in the metastatic setting is far more economically viable. So I personally think it is an important endpoint. And I think it's potentially something to look at in those patients with the biomarker positive.

Arlene Siefker-Radtke:
So that-

Thomas Powles:
I mean, I don't want to dominate this discussion at all, but there was the adjuvant study, Cora Sternberg's EORTC study, that achieved a disease free survival hazard ratio of 0.48. And then an OS of 0.78, and didn't change practice. And that was a study that didn't meet statistically significant OS. And so I think most... And when you ask colleagues around the world, it does seem to be this consensus, as Andrea says is that without survival, it tends not to be pursued. So I think that the regulators might say yes for that biomarker positive, whether it will get reimbursed is a different issue in Europe, which is complicated.

I know Matt Kowsky and we chatted about this on many occasions, and Matt says, actually the patients are going to drive this, because many of them will want it. And I totally understand that. And that comes back some of what Alison said before is when you look at the shape of those curves, they look attractive in the biomarker positives, and patients will pursue that. I think, particularly in the US, a lot of patients will get it, I think.

Arlene Siefker-Radtke:
So that's a really great point about the use of disease free survival compared to overall survival. And I believe the trial that you're referring to from Cora Sternberg was her adjuvant trial, where patients received either adjuvant chemotherapy or chemotherapy at progression. And with the use of an adjuvant strategy, it really enhances that disease free survival endpoint. But if people do not have a cure from therapy, and it's only delaying the inevitable then we may not see that long term survival, which could account for some of the lack of survival, when patients receive the same treatment at progression of disease. So looking at survival endpoints become very important. And I know there's other trials that have looked at disease free survival with long term data coming on overall survival, including the POUT trial for upper tract tumors. So Alison, now, you've done work on that topic.

Alison Birtle:
Yes, thank you Arlene. So and Tom is one of the other trial management group members on POUT. We originally presented they first data three years ago at ASCO GU and this year presented an update. And POUT was a study of initially intending to recruit 346 patients who had a completely resected upper tract tumor, PT2 to PT4 and they could be node positive as long as any visible nodes had been resected at the time of surgery, and then they had a negative post operative CT scan. We didn't mandate a formal lymph node dissection, because when POUT was first being designed, we undertook an audit across the UK. And there was no enthusiasm for doing a formal lymph node dissection.

And indeed, there's still no level one data to support that approach. And the primary endpoint of the study was three year disease free survival, I guess, partly because in [Cora Sternberg's] 30994 study of adjuvant chemotherapy and muscle-invasive bladder cancer that had been identified as a key points of interest. And also because these patients do have a high rate of relapse in the first three years. Secondary endpoints were overall survival in the usual toxicity endpoints and looking at bladder recurrence and metastasis free survival. And so three years ago, the trial actually shut prematurely because it met its primary endpoint early and showed a significant improvement in disease free survival of about 17% at three years.

Alison Birtle:
And this year we presented the mature data, which shows that at three years, there's... Sorry, at three years, and at five years, there's an improvement in both disease free and metastasis free survival in favor of adjuvant chemotherapy. And that had to be initiated within 90 days of surgery and was gemcitabine with either carboplatin or cisplatin based on the GFR alone. And the difference is 21%, at three years in favor of chemotherapy and four cycles of adjuvant chemotherapy, and then 17% difference at five years. We did present the metastasis free survival, which was very similar at three and five years 19% difference between the two groups, again, in favor of adjuvant treatments. And the overall survival data was 12%, but was not statistically significant.

And I think there's a number of reasons behind that. One, it's a secondary endpoint. Two, the study shut prematurely because we've met the primary endpoints. And at that point, there were 261 patients randomized, compared with the 346 we'd intended to recruit for which study was powered for overall survival. And also, if we look at when the first patients were recruited, it was in May 2012 and shutting in November 2017. So a lot of the patients that would have received adjuvant chemotherapy, this was in the pre IO era.

So the treatment they would have gotten and relapse would have been second line chemotherapy with something like taxol and we know that that's not particularly effective. So I think there's a number of reasons why the overall survival is not statistically significant. But I stand by the discussions we've had about why I think disease free survival is an important endpoint for this patient group and it is different to muscle-invasive bladder cancer.

Arlene Siefker-Radtke:
Mm-hmm (affirmative). So I'd like to ask Andre, and Tom, we've had a bit of this DFS/OS discussion. But what do you feel? If we have to wait for nivo survival data, do you think the POUT adjuvant data should we wait for overall survival data for guidelines approvals or the routine use of an adjuvant strategy and upper tract?

Andrea Apolo:
I think that's a difficult question to answer. I would like to see survival benefit if I'm giving a therapy early. And I think I would more likely give an immunotherapy because it has less toxicity than the chemotherapy in that setting, especially if their survival data with DFS data, I think I would have a really transparent discussion with patients about the data and see how aggressive they would want to be.

Alison Birtle:
Can I just say, though, I mean it's great comments, but we know that the nivo for DFS in the CheckMate 274 study didn't improve survival in the upper tract group. And POUT adjuvant chemotherapy has already been incorporated into EAU guidelines. And in terms of quality of life, we've published the two year quality of life data for the study, which shows that you do get a temporary dip, which coincides with cycle three out of the four cycles of adjuvant chemotherapy, but then by six months returns to normal. And at both 12 months and 24 months, patients in the treatment arm had better quality of life than those in the surveillance arm, presumably because that was tying in with some progressions within the surveillance arm.

Andrea Apolo:
I would say it's difficult to interpret the upper track data in CheckMate 274 because they only allowed 20% of upper track patients to enroll. And so you're looking at a subset analysis in terms of the DFS. So I think right now, it's still too early and we may not be able to answer that question, particularly with the CheckMate 274 with such a small subset.

Arlene Siefker-Radtke:
So what are the panel's question or panel's thoughts? If you have an upper track tumor, do you give nivo adjuvant treatment because there has been data suggesting, in mostly single institution studies, including those at Anderson and some small series that there's benefit from a neoadjuvant approach by single arm trial data? Do you prefer to wait in all patients and only give chemotherapy at the time post surgery? Anyone have any or want to share their own practice thoughts?

Alison Birtle:
I mean, obviously, I stand by the data from the only Phase III randomized control trial of systemic therapy that we have in this population of 261 randomized patients from across 75 UK centers showing a benefit in disease free survival. In terms of neoadjuvant, they're all smaller studies, and they've shown PCR rather than anything else.

Andrea Apolo:
I still have a discussion with the patient about neoadjuvant chemotherapy, extrapolating from bladder and like you said there is some smaller data sets showing benefit for neoadjuvant chemotherapy, but I think we really struggle with the staging. And generally, you would be talking to a high grade patient that's thought to be a T1. And sometimes it's not clear whether they are a T2, and you rely on imaging for that. So that can be the difficult part of it. If it's a known T2, I think it's a lot easier than if it's a presumed T2.

Thomas Powles:
Yeah, I think it's an interesting discussion. We had the ESMO guidelines meeting, I won't talk about the detail of what happened, but there's no consensus on this. And I think a lot of it comes back to patients. And I suspect if the nivo trial is positive in the ITT population, a lot of patients in the US will get adjuvant nivo. Clearly many people feel survival is really important irrespective of the setting. And that will continue to be the case. And certainly from a guidelines perspective, it's difficult to insist survival in one area and not the other, which is an area that we really struggle with. So it's interesting, the debate will go on and that's why we're here.

Arlene Siefker-Radtke:
Mm-hmm (affirmative). So I have to agree it's a great debate and a lot of questions that haven't been answered. I personally would give adjuvant chemotherapy to a patient with an upper tract tumor who had not received neoadjuvant and was high risk at surgery. But I do have a preference to giving the chemotherapy neoadjuvant because it's much easier to give cisplatin-based chemotherapy when a person has two kidneys rather than one. So I think well, we'll wait further trials and further data to show us what strategy is optimal. Should we wait or should we treat early?

Arlene Siefker-Radtke:
And will we come up with an adjuvant strategy with other targeted agents that overcome some of the limitations of cisplatin? Maybe this will be future work for enfortumab or enfortumab plus immunotherapy as the field changes and moves toward targeted agents. And thinking towards those targeted agents, and I think, Andrea, you had some data about the CaboNivoIpi work that you've been doing in the setting of urothelial cancer.

Andrea Apolo:
Yes, thank you, Arlene. I presented the final results of the Phase I trial that included the dose escalation portion and seven expansion cohorts in GU all tumors, including bladder and kidney cancer and rare tumors with a doublet of cabozantinib-nivolumab, and the triplet of cabozantinib, nivolumab and ipilimumab. And in this very heterogeneous group of patients, we saw a really nice response. We saw 38% response in these patients with 11% complete response and the duration of response was 23 months. So that's really excellent.

In bladder cancer patients, the overall response rate was 42%, with 21% of patients having a complete response and in kidney cancer, the response was remarkable 63% of overall response rate, including sarcomatoid. And this data, really informed the safety and preliminary efficacy for larger trials that are ongoing and haven't even been reported including the CheckMate 9ER study, the Cosmic-313 study, the Alliance PDIGREE study for renal cell carcinoma and my current ongoing Alliance ICONIC study for rare GU tumors with the combination of CaboNivoIpi. And I wanted to also mention the 9ER data since we are talking about CaboNivo this would be a good transition.

We did have an update at this GU ASCO 2021 on the CheckMate 9ER data that was originally presented at ESMO 2020 that showed DFS benefit and overall survival benefit and an overall response rate benefit in kidney cancer patients in the first line setting, we had longer follow up. So initially, it was 18 months, now it was 24 months. And still, there was a DFS benefit with a great hazard ratio of 0.52. And there was also a survival benefit and still a great ratio of 0.66 comparing the combination of cabozantinib-nivolumab versus sunitinib. And looking specifically at the sarcomatoid arm, because these patients tend to have a worse overall outcome, there was a PFS and an OS benefit within this group, also. So great data and I think this data is important because of the CLEAR data that was presented at GU ASCO 2021. And Tom, would you want to talk about that?

Thomas Powles:
Thank you. I'm happy to fire away. The CLEAR trials is an interesting study, I think, because it's a three arm study. We've done lots of two arm studies. Again, sunitinib is the control arm, as always. But this time, it's lenvatinib and pembrolizumab. Lenvatinib is a VEGF targeted therapy. It also has some FGFR activity and as a single agent, and when combined with everolimus, it has shown a lot of activity and data with pembrolizumab at 20 milligrams looked pretty good. The middle arm of the trial had everolimus together with lenvatinib, and I'm going to talk about the len-pen, the lenvatinib-pembrolizumab arm almost exclusively, because it is practice changing in as much as it showed a disease free survival hazard ratio of 0.39, which is extraordinarily low.

The response response rate was 71%, which is really very high. And again, an OS hazard ratio in the mid 0.6s, sorry 0.66. And when you look at it together, it just looks like a very active regime. Is it more active than cabo-nivo and actually pembro? We don't know because we didn't do randomised trials against those. And actually, there are more similarities than differences when you look at efficacy signals between these. And each of the three combinations has a data point that's attractive. The original axi-pembro had that 0.53 hazard ratio that was eye catching. The cabo-nivo data's got the quality of life data, which looks eye catching. And now we've got the 0.39 DFS for len-pen, which is eye catching. And I suspect there'll be a big commercial discussion about which is best.

I mean, my important point which I'm making at the moment is, the most important thing is pick one of the regimes, they look quite similar, learn to do it really well, because it looks like actually duration of therapy in the real world is much shorter than the trials. And that's probably because we're not as good or education training is really important in this to try and maximize the outcomes of the regimes we use. So I'm not keen on getting into a debate about which of the three is best. I'm keen to get into a debate about the best way of giving the drugs.

Andrea Apolo:
I completely agree, I think we have three active combinations. It's hard to say, without doing head to head comparison, which one is better. The arms were a little bit different. If you look at the favorable group versus the intermediate group, and just a little bit, but possibly enough to kind of sway the tail of the curve one way or the other. So I completely agree. I think managing toxicities is really important. Not stopping therapy too quickly, which is what I see people doing just really supporting patients with frequent visits and TKIs are difficult. And just kind of they can be for some patients, just managing the adverse events seen with TKIs I think is really important to keep patients on study for a long time, holding if you need to, and then resuming or dose reducing if you need to. I think all those things are very important. And it's an art just like we talked about earlier, we're going to learn a little bit more about how to manage the adverse events of enfortumab vedotin and I think it's a learning curve to manage the toxicity from these combinations.

Arlene Siefker-Radtke:
So with so many great standards being compared to Sutent (sunitinib), but not against each other. We see TKIs plus IO playing a huge role in renal cell. But are there patients that you would still choose nivo-ipi?

Thomas Powles:
Yeah, it's really important to remember that in intermediate and poor risk patients, it's a really attractive combination for patients. It again, has attractive data points and less attractive data points. It's attractive data points: its long term durable PFS at about 30% at 48 months is a landmark figure which David McDermott will remind me day after day. And on the other hand, the initial response rate is lower at only 40%. So it's probably less good at initial control of disease. But that CTLA-4 may be really important in long term durable remission, which why Andrea's work with cabo-nivo plus ipi to work how you can keep that curve, maintain that curve, is for me, the most important next study along with the adjuvant studies in renal cancer.

Arlene Siefker-Radtke:
So what about you, Andrea? Would you treat patients or any patients with nivo-ipi, with the lenvatinib and cabo data coming out? I mean, a durable response could be quite attractive, even if the immediate response rate is lower.

Andrea Apolo:
Yeah, I think I would still favor a TKI plus immunotherapy just because I don't know if I can call it a synergy. But I just have seen such great durable responses and in the Phase I trial that I had with the kidney cancer patients, there was not one patient that progressed. I mean, there was like the worst case scenario was stable disease. So it's just the combination of the TKI and immunotherapy are just really, really active. And I would want to start off with the best therapy first.

Thomas Powles:
I kind of agree with what was said and I think the pendulum has swung a little bit away from ipi-nivo towards VEGF TKI plus PD-1 therapy, because although at the moment, the follow up of the VEGF TKI immune therapy studies are not as long, they never will be as long as they were done afterwards. As those landmarks go by, if you look at the axi-pembro arm which has the longest, the OS just continues to look terrific in that group. So that's my personal preference. But again, I can understand why so many people want to give ipi-nivo. It's not the easiest regime to give the first 12 weeks are tough because the toxicity is quite difficult to manage.

Clearly, the nivo by itself is easier to give. And again, the most important thing for me if you're going to make that choice, there's not a difference between len and pem and ipi-nivo. Are you experienced to give a ipi-nivo? Does your team have a 24 hour care? Do you know how to look after grade three toxicity? Are you measuring transaminitis on a regular basis? The neuro toxicity, have you got of experience that? Do you have [inaudible]? So for me, those are the important issues not so much to from choice.

Arlene Siefker-Radtke:
So with all of these choices and the difficulties choosing between them, are there any biomarkers that you find of interest in the setting of kidney cancer that are helping guide what's done?

Andrea Apolo:
We did a lot of biomarker work in the Phase I trial and from immunohistochemistry staining of MET to looking at gene expressions to, looking at immune subsets. And we haven't found anything definitive yet and that I would say should be incorporated into Phase III trials. I know the PD-L1 marker is always something that comes up. It's not something that I have seen to be as robust in these combinations. I don't know. Tom, do you have any thoughts on that?

Thomas Powles:
Yeah, I did a debate with Bill Kaelin recently who I think I thought I wanted it everyone tells me I didn't. But so Bill said, "We need more drugs." And I said, "We need to know how to pick patients better." And of course, he won the debate. I think that's because he won a Nobel Prize to be honest, and mine hasn't arrived in the post yet. But I'm sure it will very soon, but, there's been some mix up over COVID. But the reality I think of this is that we do need to try and select these patients, so we probably are going to reach a plateau fairly soon.

And the PD-L1 biomarker in ipi-nivo look pretty good with hazard ratios down to 0.4 for survival. And there's been some work with gene expression signatures, many call them the GEP or T-effector signatures that are active signatures in kidney cancer, which were shown in the JAVELIN trial and the bev-atezo studies, IMmotion150 and IMmotion151, to be what appeared to be predictive.

So I think there was some beautiful work on Tony Choueri, [Robert] Motzer in the [JAVELIN Renal] 101 study too. And there's been this recent classification, which Brian Rini led with the six signatures based on gene signature analysis. So we are beginning to make progress at last in kidney cancer. I think we've got a lot of catching up to do, but I genuinely think we can be combining biomarkers and of course, I'm excited by circulating biomarkers as well.

Arlene Siefker-Radtke:
So I think this has been a wonderful discussion and clearly all of my colleagues are prize winners whether their prizes have arrived or not due to the excellent work they're doing in the field of kidney and urothelial cancer. We've seen several practice changing abstracts presented at one GU ASCO meeting that have resulted in New England Journal papers for the lenvatinib-pembro data in renal cell and enfortumab vedotin in the setting of bladder cancer.

So clearly the standards are changing, the field is evolving. And hopefully in the future, we'll be heading toward that personalized strategy where we can start to pick the right treatment for the right group of patients who will have the most benefit from that particular therapy. And I'd like to thank my colleagues for being so generous with their time and thoughts and not holding back at all on any of the discussion, as we ask these difficult questions on how to best treat our genitourinary cancer patients. Thank you all for you time.