PIVOT-09 Trial: An In-Depth Analysis of Revisiting IL-2 Therapies in Advanced RCC - Nizar Tannir
January 25, 2023
Nizar Tannir, MD, FACP, Professor, Department of Genitourinary Medical Oncology, Division of Cancer Medicine, The University of Texas, MD Anderson Cancer Center, Houston TX
Pedro C. Barata, MD, MSc, Leader of the Clinical GU Medical Oncology Research Program, University Hospitals Seidman Cancer Center, Associate Professor of Medicine, Case Western University, Cleveland, OH
ESMO 2022: Bempegaldesleukin plus Nivolumab Compared to the Investigator’s Choice of Sunitinib or Cabozantinib in Previously Untreated Advanced Renal Cell Carcinoma Results from a Phase 3 Randomized Study
ESMO 2022: Invited Discussant: The Results of the PIVOT-09 Trial & the Role of Enfortumab Vedotin
Pedro Barata: Hi and welcome. My name is Pedro Barata. I'm a GU Medical Oncologist. Today, I have the pleasure to be joined by Dr. Nizar Tannir. He's a professor at the GU Group out of MD Anderson, Houston. Dr. Tannir is really well known in the field as one of the senior KOL's and lead investigator for many, many trials we've used, and we use that data to help treat patients on a daily basis. He happens to be also a mentor and a friend. So, thank you, Dr. Tannir, for taking the time to chat with us today.
Nizar Tannir: Thank you, Pedro. It's a pleasure and an honor to be with you here. You are a superstar, rising star in the field as well. I'm proud to see your career blossom and I will be following your trajectory, your career here and the years to come.
Pedro Barata: Well, thank you. Actually, we're here, I should congratulate you first because we're going to be talking a little bit about PIVOT-09, like behind the scenes. You delivered a fantastic presentation on the PIVOT-09 trial, right? We were excited, and this is a large Phase III trial. I'll ask you to summarize the design really quick for us but before I do that, just to say out there, as many of the audience know, this is actually a very important study. We're all waiting anxiously for those results because it was revisiting the concept of IL-2 again in the PEGylated formula. So, maybe we should start there. Of course, we used to do it back in the days, we know the data. I guess, what was the rationale, what did you see when you look at the preclinical data about the PEGylated formulation for IL-2? Why did you take it and try to develop in a Phase III trial in the context of advanced RCC?
Nizar Tannir: Thanks, Pedro, for the question. Very important question. If we look at the landscape of RCC therapies, there are several classes that have been validated and IL-2 pathway therapy, with high-dose IL-2, is one of those. That was the first one that was FDA approved in 1992 in the US. It delivered a consistent 6-8% complete response rate, but 85, 90% of these CRs became durable over 10, 15, 20 years. I think that was the first pathway that has been validated in RCC as well as melanoma, I would say. Then, of course, we had the VEGF directed therapies. That is a class by itself. That's been also validated with multiple VGFR-TKIs approved in the space and then of course the mTOR inhibitor pathway, another one, and the most recent one is the immune checkpoint inhibitor.
I think we all appreciated the benefit of high-dose IL-2, but we also recognize the difficulty and the complexity and the toxicity of giving high-dose IL-2. It required inpatient administration in a monitored setting, ICU setting. It was delivered or given at only few centers that have been, their staff trained to do it and it was difficult to combine it.
There was one study that tried to combine it and combining it with new immune checkpoint inhibitors like an anti-PD-1 which was really not deliverable. Then there have been many compounds looking at novel IL-2 compounds. Instead of the native high dose that had short half life of 20 minutes and you had to give it every eight hours in the ICU setting as I said. One of those compounds that attempted to reproduce the efficacy of high dose IL-2 but in the outpatient setting and importantly to try to combine it with the immune checkpoint inhibitors, particularly anti-PD-1.
That brings us to the agent or compound that we tested in Phase I in a basket trial in many solid tumors including RCC, melanoma, bladder cancer, et cetera. That's NKTR-214 or bempegaldesleukin. I'll call it now for short BEMPEG. It was interesting that we have done, and this is many years ago now, preclinical studies showing that this agent recruited T-cells to the tumor micro environment and in fact in human tissue, in biopsies obtained from patients, RCC and melanoma, we were able to show that when you give BEMPEG and anti-PD-1 like nivolumab in that situation, you basically increased exponentially CD8-positive T cells. These were able to, they were able to proliferate, activate, as well as the NK cells.
Also, in many tumors with PD-L1 negative tumors, we were able to convert these to PD-L1 positive tumors. The preliminary results of efficacy in a Phase I trial that was presented a few years ago in RCC, small number of patients, 14 patients, we were able to show a high response rate, about 70%. That led to an enthusiasm to try to take it to a Phase III trial in melanoma and in RCC. In melanoma versus nivolumab monotherapy. In RCC, we can discuss that if you will, in first-line as well the combination of BEMPEG plus nivolumab and then here we can discuss the choice or the selection of the competitor but in PIVOT-09 we selected at the time, this was when we designed the trial in late 2017 and the trial was launched in December 2018, we set on using a TKI.
Pedro Barata: Let me pick on that Dr. Tannir because, as you said, we know a lot more about combination regimens. We didn't know as much, actually I don't think we actually had any data from Phase III data, at least with IO-TKIs, right? The IO based combos were emerging when that was designed. I think it was actually quite interesting that, although you did use a TKI as a control group, you did allow for Cabozantinib in addition to Sunitinib, which might reflect what you learned in the context of frontline comparing Cabozantinib, which I think was good news to you guys because you had that. I'll let you expand on the design of PIVOT-09 for patients with advanced RCC. Just walk us through the main points. How do you design it, the way you design it and what you're looking for as far as endpoints?
Nizar Tannir: Sure. Ideally, in the ideal world, we want to use, when we are testing a novel combination. We'll call it an experimental combination. In this situation it's BEMPEG plus Nivolumab. Initially, we were aiming to have a trial focusing on intermediate risk, poor risk. When we launched the trial in 2018, as I said around December, when was the first patient was enrolled but of course the trial had to be sent to for IRB approval, et cetera, earlier than that. So really, the only data that was available back then, as you mentioned, was the CheckMate-214 data in first-line and we all know that in intermediate-risk, Nivolumab/ipilimumab produced benefit in terms of OS and ORR versus sunitinib, the comparator arm.
So, we knew that ideally one would, in an intermediary high risk category, would choose nivolumab/ipilimumab for comparison but one thing, what we wish, our wish list and another thing is what you end up negotiating and working with a pharmaceutical company to bring a novel therapy. We all know the limitations or the challenges of giving nivo/ipi. I mean, I am a nivo/ipi advocate for intermediate high risk, not because I work with BMS as well as colleagues at other academic centers to bring this therapy to FDA approval through the results of CheckMate-214, but because I recognize at the end of the day patients want to have a durable CR like with high-dose will lead to cure.
At the same time we recognize that the difficulty of giving nivo/ipi in some settings, the high discontinuation rate of 22-23%, as well as the, not low, progressive disease rate of about 18 to 20%. We really wanted to bring in a novel therapy that could deliver on the promise of nivo/ipi in intermediate/poor risk patients and maybe also in favorable risk because high dose L2 work best in patients with favorable risk, those that had as good performance status.
Based on the preliminary, very preliminary data from the Phase I, the earliest Phase I with BEMPEG/nivo, when we saw 71% response rate initially in a small cohort of patients with RCC, we wanted to bring in a novel therapy that will produce durable responses as well as less toxicity. We saw that when we give bempegaldesleukin and nivolumab, the toxicity was by far less. Definitely less immune related adverse events compared to nivo/ipi. That's the rationale for doing that. Ideally we would have chosen nivo/ipi as the comparator, but then you would have to really have maybe a non-inferiority trial and then maybe you will win, maybe you won't win. As a compromise, we ended up choosing a TKI.
We knew that in intermediate risk, poor risk, an agent such as sunitinib would not produce the best results. Every pharmaceutical company, at the end of the day, wants their therapy, their compound, their agent to succeed. They want their trials to be positive and succeed in bringing a new agent. It is, you see this in many, and I can give you examples of many Phase III trials where the comparator was really the low bar for the experimental arm to really win. To come back to your question, why did we choose cabo? Okay, at the time back in 2018, even, except for the US where nivo/ipi was FDA approved for patients with intermediate risk, there weren't many countries that approved nivo/-ipi. Even for intermediate risk and, in fact, in Europe, while the EMA approved, saw that this was a reasonable choice for first-line showing OS for the first time versus a TKIs end of care, there were demanded to have a trial showing nivo/ipi versus nivo and so it wasn't reimbursed in all the European countries, and certainly in the majority of the countries in the rest of the world, nivo/ipi wasn't approved.
As you pointed out Pedro, we, the 2018, that was a year before the IO/TKI like pembro/axi from Keynote-426 and then later on cabo/nivo and then later the most recent one, pembro/lenbatinib. Those didn't come till later. So really, we didn't have any data with IO/TKI. The only data we had was with nivo/ipi, which wasn't approved in many countries, as I said, so it was appropriate in my mind to have a TKI as a comparator.
Now, cabozantinib based on the results of cabo/sun, a randomized Phase II trial of 157 patients, as you know between cabozantinib and sunitinib in intermediate risk / poor risk showed benefit for PFS with cabozantinib over sunitinib. But again, while cabo was approved in the US for intermediate risk / poor risk and in fact for any risk as firstline therapy based on cabo/sun, cabo was not approved in the rest of the world. In few countries it was, but not in many other countries.
We had to give investigators a choice, we had to give investigators the flexibility to choose the TKI and that's why we allowed either sunitinib or cabozantinib, depending on approval of one of these two TKIs in their countries. You couldn't provide a drug to, say Brazil, to patients in Brazil if the drug is not approved there by their health authorities. You had to have that. For that reason, which shows TKI selection or TKI choice as one of the strata. Typically, when you design a Phase III trial, a randomized trial, you like to stratify by prognostic groups. That's why you choose, so IMDC risk group as one of the stratification factors.
Other factors that you could use for stratification would be geographic location. The stratification by TKI chose was a reflection of geographic location. And I have to say that all for Phase III trials in first line RCC therapy, the nivo/ipi in CheckMate-214 and the three other IO/TKIs used geographic location as one of the stratification factors. That's the rationale for using sunitinib or cabozantinib as a comparator.
Pedro Barata: Let me thank you for that. That was very thorough and really appreciate the, again, behind the scenes discussion when the trial design was made. Priceless information for us and for the audience of course. So moving forward, we saw the results and we'll have access accompanying this video about the results. I have to say the word disappointing. I was expecting much better when you look at response rate, which was one of the primary endpoints. Lower with the cabo compared with TKI. The same for median OS, right? Actually not reached for TKI but it was relatively low, if you will, for the TKI.
I guess for the, excuse me, for the combo, I guess the question that I have for you is, is that it for IL-2 or are you thinking of other ways? For instance, you did mention, which I was going to bring it up, that IL-2 back in the days used to be used mainly in patients with good risk, as you said. Low levels, low number of mets, good performance status, low volume of the disease. That kind of describes what today we see a lot of those patients with good risk disease, right? Nonetheless, now even with CheckMate-214 and of course in these PIVOT-09, we are looking for patient intermediate or poor.
Then we can even make it more complicated by bringing molecular signatures into play, right? Because as we know, we see the breakdown and we can see antigenic signature there in the good risk, actually in all risk but also in the good risk, but we also see an immunogenic signature across risk groups. My long question would be kind of a teasing you a little bit. Do you think there's a role for us to find the ideal population that can definitely benefit from a combination of PD-1 with an IL-2 in this formulation for instance? What are your thoughts of that as you think of further research?
Nizar Tannir: It's a very valid question, Pedro. I think first let me answer the first part of your question or first question about is that it for IL-2. I really think that there is still room to find a novel IL-2. I think high dose IL-2, again after 30, 40 years of research and treating patient with high dose IL-2, I don't think high dose IL-2 is going to really, and in fact we saw that fewer and fewer patients have been treated with high dose IL-2 since the advent of immune checkpoint inhibitors. But I don't think it is over, game over, in my opinion, for the interleukin 2 pathway. The BEMPEG program has been terminated based on the three registration trials that were negative.
I mentioned PIVOT-09, that's the RCC Phase III trial. The melanoma or 01 study, the nivo/BEMPEG versus nivo was negative and there was a single-arm study of 110 patients or so in bladder and that was negative as well. It may be over for Bempegaldesleukin, but I don't think the negative results from these Phase III trials has dampened, in my opinion, in any way the enthusiasm of pursuing still some other molecules, some other drugs, novel agents with high, that are IL-2.
So that basically signal through the interleukin into receptor beta gamma away from the alpha because we know the alpha unit subunit of the receptor is the one that crosses the toxicity of high dose IL-2. So there are still many companies pursuing IL-2 compounds that are signal away from the alpha subunit and signaling through the gamma and beta and gamma. Without naming companies or without naming compounds, some of the options is to combine in that one agent an IL-2 signaling agent as with say IL-10 or IL-15. And therefore I think we will see and we will hear about trials launched from different companies with different compounds still aiming at the IL-2 pathway.
Pedro Barata: That's awesome.
Nizar Tannir: The other part of the question you asked is, can we look at molecular signatures? This is the holy grail of oncology. We always talk about biomarkers, we always talk about if we just knew what is the achilles heal of the tumor and if we can enrich our trials, our patients population that we recruit into trials for that one particular, for what in the lab we call synthetic lethality. If we can just say, "Okay, I've got the right patients where I know this drug is going to hit tumor and then boom." I think it is, really what we want to aim for.
That's really how we're hopefully, I think, win the war against cancer is by personalizing, I mean because we know the heterogeneity of tumors, solid tumors and [inaudible 00:20:57] and everything. I still think we are not there in kidney to really know. I wish we can say, "Okay, what were the patients who really benefited the most?" I can tell you from PIVOT-09, the patients who do the worst in terms of outcome with a combination of BEMPEG/nivo, were the patient with poor risk. I'm sure you watched, like I did with disappointment, the results of COSMIC-313, where the patients with the triplet of cabo plus nivo plus ipi, did not do well in the poor risk but did better in the intermediate risk.
My opinion about this is for novel agent to really make it, I think you really have to have efficacy as monotherapy. I think that's a cardinal rule. That was never really tested as monotherapy because in combination we thought we really were recruiting T cells, proliferating and activating CD-8 positive T cells and NK cells and not T-REGS. Then we thought it's valid. Let's go ahead and do quickly, jump quickly.
I think, in retrospect we need to demonstrate efficacy with monotherapy, okay? That's really important. And then try to, and that's not what, unfortunately wasn't done, is get biopsies pre and post monotherapy and in combination. I think this is in retrospect, if we had the choice of going back in time, revisit, I think that would be those things. We would do biopsies, pre baseline, pretreatment, monotherapy and then post and then also combination and post to really see what's going on. So I think, hopefully other companies that have different compounds will hear this message and will be heading this advice and then following through.
Pedro Barata: This was wonderful. I really thank you for the time that you gave us to give us this wonderful discussion around PIVOT-09. Again, congratulations for your fantastic presentation and actually that's younger folks who look at to be lead PIs in their future, should look at your presentations and see how it's done. Yeah, absolutely.
Nizar Tannir: Thank you for the compliment. Again, I wish for the sake of our patients and the field, the study was positive because we really need... Unfortunately we have not broken the barrier of cure for every patient. Our goal is to make RCC history as I'm sure you have that passion in you at your young age to make all GU cancers and all cancers history. That's our mission at MD Anderson. We fell short in this trial, but I don't think we should be discouraged. We should continue. As Einstein said, "There is no success unless you have failures," so I think we should learn from our failures, learn what we did wrong, what we did right, and then hopefully we'll continue our march, we'll continue our path towards making cancer history.
Pedro Barata: I really appreciate. Thank you so much. Congratulations. Always a pleasure to chat with you and I'll see you soon. Thank you.
Nizar Tannir: Thank you, Pedro. Pleasure to be with you.