The Uromigos Episode 140: Ipilimumab in Renal Cancer

Dr. Mike Atkins discusses the role of CTLA4 inhibition in RCC.

Episode Transcript

Brian:
Hey, welcome everyone to another Uromigos podcast. We’re pleased to be joined by Mike Atkins today. We’re going to talk about a topic that came up during the last ESMO, which was the exact role of ipilimumab in kidney cancer, which Mike is, of course, an expert on, and also has some relevant melanoma experience. We thought it’d be interesting as we, the kidney cancer community, starts down this path to get his input. So, Mike, welcome. If you just want to briefly introduce yourself, and then maybe you can start by maybe just giving us an overview of what you think the role of IPI is in kidney cancer in terms of dose and schedule, we can start there.

Michael Atkins:
Sure. So, I’m Michael Atkins. I’m the deputy director of the Georgetown Lombardi Comprehensive Cancer Center in Washington DC and have several years’ worth of experience in immunotherapy as applied to both melanoma and kidney cancer. And therefore, because of my interest in melanoma experience, that goes to CTLA-4 blockade that extends back nearly 2 decades now.

And so, what is the role of ipi or CTLA-4 blockade in kidney cancer? Well, I think it’s similar to what it would be for most cancers, but particularly from melanoma, I view it as bringing more cells to the battle while the anti-PD1 is role is to keep those cells active while they’re killing the tumor, and so I think many patient’s tumors already have enough T cells in them and all they need is to have the brake taken off with a anti-PD1 antibody or anti-PDL1 antibody in order to do the job, but their other tumors where they need more T cells there, or they need to have their T cells more active where CTLA-4, which acts at the priming phase of the immune response can actually serve that function.

And we see in melanoma that the combination of anti-CTLA4 plus anti-PD1 results in about a 10% increase in the tail of the overall survival curve, and higher response rates, and more activity in patients who have PDL1 negative tumors or have tumors that are driven by oncogenes, which may not have as many neoantigens in there. And- [crosstalk 00:03:16].

Tom:
Mike, we think there are biomarkers that …

Michael Atkins:
… we know that you don’t need to give of a lot of CTLA-4 to prime the pump. In many patients, 1 or 2 doses at the 3 milligram per kilogram level is enough. And even at the 1 milligram per kilogram level, in our studies in melanoma with neoadjuvant treatment where we then resect lymph nodes after giving the same schedule of nivo-ipi that we use in kidney cancer, we see that about 60% of patients have pathologic complete response or near-complete response. And that’s enough to actually take care of any metastatic disease that was there, and they don’t need additional therapy.

And nearly 100% of those patients will be relapse-free in 2 years. So, extrapolating from the melanoma experience to the kidney cancer experience, I think there are patients who need priming with CTLA-4 to increase their chance of response. I think it raises the response rate, particularly in patients with intermediate and poor-risk disease. And I think that it can induce a response quicker.

Brian:
So, Mike, you talked about mechanistically what it’s doing. In melanoma, how’s that translated clinically? Meaning that are people modifying ipi based on certain things, are they spacing it out, are they modifying dose, are there biomarkers, or is it kind of as the original checkmate, is it still being used? Because I think that’s going to be the question for RCC moving forward.

Michael Atkins:
So, there are a few different modifications that are being explored. One in melanoma, one is to look at the kidney cancer schedule in comparison to the approved melanoma schedule. So, comparing ipi 3, nivo 1 induction, to ipi 1, nivo 3 induction, and that was done in the CheckMate 511 study, and there’s reduced toxicity from that lower schedule of ipi, lower dose of ipi, about a 30% reduction in toxicity, but similar efficacy. In addition, there was a study in Australia that looked at pembro plus that lower dose of ipi in patients with melanoma. And the response rate looks very similar, and the tail on the survival curve, very similar, if not even superior, to what we’ve seen with the approved dose of nivo-ipi, with 73% of patients being alive at 3 years. And third of those [crosstalk 00:06:14]

Tom:
Mike, in what proportion of patients …

Michael Atkins:
… is to look at just 1 or 2 doses of nivo-ipi priming.

Tom:
Mike?

Michael Atkins:
And this is a study done by Mike Pascal in Memorial Sloan Kettering, where they gave 2 doses of ipi-nivo and then did a CT scan, and if the patient was exhibiting a response at that CT scan 6 weeks in, they went directly to nivo maintenance, and if they didn’t have a response or they had some evidence of tumor growth more than 4%, then they continued to 4 doses of nivo-ipi. And the response rate was the same if you were got the 2 doses of nivo-ipi, and the correlative data suggested that 1 dose was probably enough to fully activate the immune system.

Brian:
So, Mike, I think [crosstalk 00:07:04]

Michael Atkins:
So, I think we’re giving much more nivo-ipi in melanoma than we need to, and it may be the case in kidney cancer as well.

Brian:
So, it sounds like, and I think, I don’t know that you can hear Tom. I can hear him, but you can’t. Is that right? Mike, can you hear Tom?

Michael Atkins:
I can’t hear Tom.

Brian:
Okay, perfect. This has happened before.

Michael Atkins:
It’s perfect for you.

Tom:
Can you hear me now, Mike?

Michael Atkins:
It’s your dream come true.

Brian:
No, Tom, I can hear you, but Mike can’t.

Tom:
Okay.

Brian:
So, Tom can ask me questions, but before he does, so Mike, you described a number of studies in melanoma that are looking at different dose and schedule, maybe using CTs as a biomarker. I guess really, the fundamental question is, should we repeat some of those studies in kidney cancer, i.e., is practice actually changing in melanoma or are these small studies that are interesting, but they aren’t really changing practice?

Michael Atkins:
These small studies are not changing FDA approvals, they’re not registration studies, but people who look at this data can get some comfort in using a reverse ratio of nivo-ipi in patients where they’re worried about ipilimumab toxicity and some comfort in stopping ipi-nivo induction early in patients who don’t seem to be tolerating it well.

Brian:
So maybe [crosstalk 00:08:32]

Tom:
Brian, ask Mike about booster doses of ipi. How frequently should it be?

Brian:
So, Tom’s asking about booster doses of ipi. So, I assume he’s talking about patients who lose their response, so they get whatever induction they get. They have some sort of response or stable disease and then start to progress. Is there any experience with adding ipi in later with sort of, I guess, reinduction would be the other way to say that?

Michael Atkins:
Yeah. Some of the initial studies with ipilimumab in melanoma had a reinduction in patients who initially responded then progressed, and there was a response rate in that patient population, and so I’ve extrapolated from that to the kidney cancer world where I’ve seen some patients who had a response to ipi-nivo at their 12-week scan, but after 6 months or a year on just nivo monotherapy have started to have progression and wondered whether we could restore activity by giving them nivo-ipi again. And I think we only have anecdotal information on that approach, but it’s something that I [crosstalk 00:09:40]

Brian:
But you’ve done it?

Michael Atkins:
… looked at before wanting to switch to the TKIs.

Brian:
Right. So, I’ve not done it, but I mean, it makes sense, but is there published even anecdotal experience? I’m not aware if there is at RCC.

Michael Atkins:
I don’t think so.

Brian:
Okay. Just sort of … [crosstalk 00:09:56]

Tom:
Brian, what about 3 monthly? This is 3 monthly … [crosstalk 00:09:58].

Michael Atkins:
What we do see published is nivo-ipi after anti-PD1 therapy, and that’s, I guess, could be single-agent anti-PD1, which likely was given either on a protocol or in the second-line setting, or anti-PD1 plus a TKI and then giving nivo-ipi after that. And we have direct data for that from our Hoosier study of nivo monotherapy, where we tried to give nivo-ipi to every patient to either had progressive disease or stable disease at the year. And we were only able to treat about half the patients who fit that category because of some of the exclusion criteria in our protocol.

We wouldn’t give ipi-nivo to patients who had grade 3 or higher immune-related adverse events on nivo monotherapy. And we required a biopsy of a metastatic site for patients who were progressing before they went on to nivo-ipi because the protocol was designed as a biomarker study.

Brian:
So, let’s talk about … [crosstalk 00:11:12]

Michael Atkins:
And interestingly, I think we found that some patients when we did those biopsies, we found no tumor there, but when we gave them nivo, the ones we did give nivo-ipi to, their response rate was not that great. It was about 11%. I think there was 1 complete response. And so, the frontline response to nivo monotherapy in our updated data was about 34%. So, when you combined that you will probably total, we were seeing responses in maybe 40% of patients, similar to the 214 data without everybody needing to get nivo monotherapy.

Brian:
Let’s talk about that a little bit. I’m going to ask you a question about schedule in a second for Tom, but let’s talk about the sort of the take-home from your HCRN study. There’s a study called TITAN done in Europe, reported at ESMO, I think a couple years ago now, and then OMNIVORE, which were broadly similar in that they were nivo monotherapy upfront with ipi at either progression or lack of response at a given time point. And if I would just sort of summarize them as is the response rate to nivo upfront, it kind of varied across studies, but at least for 2 of them, it was in, I think, the 30% range. Adding ipi added about a 10% response, but not many complete responses added with the addition of ipi. I think maybe just one in each study if even that.

So, I think my take-home from that, and I’d be interested in what you think, is that you kind of need to give both upfront that yeah, there probably are patients who can get away with nivo monotherapy, but we don’t really have a good way to identify them, and we don’t want to miss the opportunity to give it be upfront. Let’s put aside the IOTKI combos, but if you’re giving ipi-nivo, you probably want to give both upfront. Do you agree with that or not?

Michael Atkins:
I agree with that, particularly for the intermediate and poor-risk patients. In our study, the response rate was 25% to nivo monotherapy in the intermediate and poor-risk patients. And not many of them benefited from ipi-nivo salvage. I think it was only 2 of the 4 responses in the salvage setting for our intermediate or poor-risk patients. And so, I think there’s value to giving nivo-ipi upfront to those patients where I think there may be a role for nivo monotherapy is in the favorable risk patients where we saw a pretty remarkable 57% response rate, 20 responses in 35 patients, and only 1 patient had progressive disease at their 12-week scan, and that was just a new lesion, but overall, had no overall growth of their disease, and so we didn’t miss the chance to give nivo-ipi to those patients.

A couple more patients actually responded when we added ipi to the regimen in the group that didn’t respond. And I just think you have many chances in the favorable risk people to try different types of schedules. And if you can get a good response to nivo monotherapy, to someone who’s asymptomatic and may not want to be getting a TKI, which is the only approved treatment for the favorable risk patients. I think giving an anti-PD1 as a single agent is a treatment that is, I think, appropriate for the patient’s disease, and rate of disease progression, and asymptomatic status at baseline. And so … [crosstalk 00:14:53]

Brian:
So maybe in more [inaudible 00:14:54] patients?

Michael Atkins:
… I think we should need to identify who those patients are, who can respond to single-agent anti-PD1 in the favorable risk setting.

Brian:
And I know you have a lot of data from the HCRN study coming out, tissue pre and post biopsies, et cetera. So obviously we look forward to that. Because like you said, if we could identify these patients, then we’d be doing them a great service, assuming that those responses are durable, which you probably don’t know yet from your study but making that assumption.

Michael Atkins:
What we do know is that 17 out of 20 responses were ongoing at the time of our last data lock.

Brian:
So let me ask you about schedule. Tom mentioned about the study at ESMO that gave 3 monthly ipi instead of 3 weekly, roughly similar efficacy, as opposed compared to CheckMate-214, obviously different studies, and I think about a 20% reduction in grade 3 toxicity. What do you think the role of spaced out ipi is not based on toxicity, but just planning from the start to space it out, and do we need to study more, is it something that we should put our resources towards studying, or can we just do it in, can we take comfort, like you said, melanoma has done that. You can do it but don’t necessarily need to?

Michael Atkins:
Yeah, that’s a tough question. I think that, as I said, 1 dose of ipi might be enough to prime the immune system, although we’ve not tried 1 dose of ipi at 1 milligram per kilogram. So, I can’t say that we have support for that approach in melanoma. When we first were developing CTLA-4 antibodies, tremelimumab was tested at a every 12-week schedule. And there were patients who responded to that, but it didn’t do better than the control arm dacarbazine in the phase III trial, which is why tremelimumab was never approved. And many of us thought that part of the reason was that the CTLA-4 blockade was given every 12 weeks, and there wasn’t enough CTLA-4 antibody given early to fully activate the immune response, but that was at a single agent. It might not be as important in a combination.

Brian:
So, when you’re giving ipi-nivo upfront to kidney cancer, you’ll give it every 3 weeks, right? As per CheckMate-214, assuming you’re giving combination therapy upfront and you’ll continue to, you’re going to give 4 doses unless you run into limiting toxicity. Is that a fair statement?

Michael Atkins:
Right. Well, I have a low threshold though for stopping toxicity, stopping because of toxicity. So, if someone comes in to me and it’s time for their fourth dose, and they’re having a few loose bowel movements, that’s enough for me to say, “Okay, their risk is probably greater than the benefit at that point of giving the fourth dose. And I may just hold that fourth dose and see them 3 weeks later and then start maintenance. Or I may decide it’s not that worrisome, and I’ve already done a good job of activating the immune system and switch him right to nivo maintenance at that point.”

Brian:
So, 2 questions. One is, is it a specific CTC grade of toxicity, like number of stools per day in your example, or is it just your clinical gestalts like anything more than a little, and you might consider stopping?

Michael Atkins:
Well, I think it’s not… I don’t go based on grade. I’ve treated a lot of patients with nivo-ipi with melanoma, where I learned the hard way that patients were having a little bit of a problem, and I gave them that next dose, and then we dealt with that problem for a long time. And so, the fortunate thing about these immune therapies is they last a long time, they have a long half-life, and their biologic half-life is much longer than their actual pharmacologic half-life. And so, you don’t have to give the dose if you have a doubt of what’s causing a patient’s problem, and time will tell.

But so, I always say to them if they’re having some issue, “I don’t really want to add any more fuel to the fire at this time. Let’s see you back in 2 or 3 weeks. And if nothing has become of that, we can either make up that dose. Or if it’s time for your scan and you’re responding, and we’ve done the job of priming the immune system, and we can just go to maintenance therapy.”

Brian:
So, my follow-on question is then in that patient where maybe you’re debating, let’s say you’re debating the fourth dose, right at the time or 3 weeks later, or whatever, would you get a scan early to help make that decision? And then is there a result on the scan that would say, “All right, we don’t need to do it.” Do they need tumor shrinkages, stable disease enough, et cetera?

Michael Atkins:
No, I would not get a scan early for that decision. The only time I would get a scan early would be if patient has new symptoms that make me concerned about disease progression and when I have other therapies to give them.

Brian:
So why wouldn’t, let’s say they’re due for dose 3 or 4. They have borderline toxicity, kind of like in the melanoma study. And I know they did it after 2 doses. Why wouldn’t, because I do this in practice, why wouldn’t you use if they had tumor shrinkage on a scan to say, “Okay, that’s enough. And maybe I don’t need to push that fourth dose?”

Michael Atkins:
Well, because I don’t feel like I have to give a dose during the time that the immune system’s going to stay active and the extra 3 more weeks gives me time to evaluate what’s happening.

Tom:
Brian, we’re running out of time.

Brian:
So, whether there’s … [crosstalk 00:20:54]

Tom:
I just wonder…

Brian:
… shrinking on a scan or not, doesn’t matter?

Michael Atkins:
No.

Tom:
I don’t think Mike can hear me. Could you ask … [crosstalk 00:21:00]

Michael Atkins:
I want to, yeah, you can see what’s happening at 12 weeks, and you might not see much shrinkage if you look too early and … [crosstalk 00:21:07]

Tom:
Could you, Brian?

Michael Atkins:
I think there’s also an overreliance on what’s happening on the scans to try to make decisions when it doesn’t really tell you what’s happening inside the lesions on the scans and not saying that we see a lot of pseudo progression because we don’t, but we also don’t look early because of some concern that we might see. So, the progression, so progression is a lot less common now that we use RECIST criteria than the WHO criteria. But I think that you can see a lot more shrinkage if you wait the extra 3 weeks.

Brian:
Yeah. So that’s interesting. I do tend to do it, but I totally take your points that it’s a little bit of a crutch, and maybe it’s not doing what I think it’s doing. All right, we just have time for 1 last question Tom’s been whispering in my ear this whole time. I kind of like better of me controlling the podcast, and I get to control what Tom asks.

Tom:
I find it very frustrating. Very frustrating.

Brian:
It was really about how to make … [crosstalk 00:22:05]

Michael Atkins:
But it’s not the same when you guys aren’t going back and forth.

Brian:
We are, but nobody can hear Tom. It’s great. So, it’s about COSMIC-313. And I think it’s really about, I don’t know, Tom, help me, but it’s really about what you think that data will show and how it might impact practice. And are we the global kidney cancer community going to be able to put up with the toxicity, so to speak. Do you think that’ll be limiting?

Michael Atkins:
Well, I would’ve thought that given that we couldn’t add sunitinib to anti-PD1s and that in melanoma, we couldn’t add BRAF inhibitor to ipi that the ipi-nivo-cabo combination would be difficult to tolerate. And so, I’m concerned about the toxicity. I have not used that combination. I wasn’t part of that study. So, I’d be concerned about the toxicity and the need to reduce the dose of Cabo or to hold a lot of the immune therapy in order to accommodate that toxicity and what impact that might have on the long-term efficacy of the regimen.

But I do think that it’s a given that response rate is going to be higher with the triplet and the medium PFS is going to be higher with the triplet, but I am not at all sure what we’re going to see on the tally, overall survival curve, or the PFS curve, because those are things that are driven by giving the immune therapy at a time when it’s most potent. And I’m not sure how much Cabo adds to that. And if it adds toxicity that causes you to not give as much immune therapy, it might take away some.

Brian:
Yeah. All good points. So, I think we’re all… I think it’s the next major piece of data in kidney cancer. And I agree with all your points. I think it’ll… I think it’s good we’re getting to the point where we have so many active drugs that we’re sort of facing these issues of intensification versus deintensification, which is sort of kind of what you were just alluding to. But we’ll see. I agree that the time of event or the tumor shrinkage endpoints I think will be in favorable, but I think some of those long-term endpoints we’ll see, and that’ll be balanced against toxicity. So, we will see.

So anyway, Mike, thanks for joining us, really insightful. I always learn a lot from you, and I think we, as a kidney cancer community, can learn a lot from melanoma since you guys are a few years ahead of us on this. So, thanks for joining. We hope to talk again soon.

Michael Atkins:
Yes, it’s my pleasure.

Brian:
Tom says thank you as well.