The Uromigos Episode 149: Treatment-Free Survival for Ipilimumab and Nivolumab

Dr. David McDermott describes his clinical cancer research paper on treatment-free survival in kidney cancer.

Episode Transcript

Tom:
Welcome to this special podcast, we’ve elected to have 2 papers of the month this month for a special reason. That special reason is we’ve got 99,700 listens, and we only need 300 people to listen to Dave McDermott’s paper of the month to get over the 100,000 line. Surely, David, we’re capable of that.

Brian:
We’d like them to listen in the next week, before the New Year.

Tom:
Before the New Year is the goal.

Brian:
Please.

Tom:
David, can you attract those 300 people? If so, please sell your paper.

David McDermott:
God. All right. And so, there is pressure there, I didn’t realize there was pressure. But …

Tom:
That is pressure.

David McDermott:
I’m glad to be included as a friend of the show, a friend of the pod as they say. So, I’m here to talk about our treatment-free survival paper, which was actually just published in Clinical Cancer Research.

Tom:
Do you want to introduce yourself, David, really briefly?

David McDermott:
Oh, so I’m Dave McDermott, medical oncologist in Boston at Beth Israel Deaconess.

Tom:
One of our fellow Uromigos.

David McDermott:
Correct. Anyway …

Brian:
You might have been on the show more than anybody else. I don’t know, it’s pretty close.

David McDermott:
Okay.

Brian:
There’s no prize or money though, [crosstalk 00:01:29] …

Tom:
There are no prizes for anything. Even at 100,000, there’s no prize. But, David …

Brian:
Just our friendship.

Tom:
Keep going. Just ignore Brian. Keep going, David. Keep going.

David McDermott:
Okay. So, as I was saying, treatment-free survival in kidney cancer. Paper, Clinical Cancer Research, this month. First author is Meredith Regan, who’s led most of this work over the last 5 years. And it’s actually a work in progress, meaning our treatment-free survival model continues to improve over time as we get more data, as we analyze more trials. But it’s part of a larger effort to try to develop novel endpoints, which we’ll actually use in future trials, particularly future trials with immuno-oncology agents.

And the original goal of measuring treatment-free survival was to measure the durable good or the remissions that happened in some patients after they get immune therapy, as well as the durable bad, the ongoing side effects that patients can get. So that we would have a measure as we go forward, to measure both, to make sure we’re on the right track as we start stacking agents on top of one another in hopes of improving long-term outcomes.

I think this started probably 6 or 7 years ago. I remember when I was giving a talk at ASCO, I think it was a few years ago. I don’t know if you guys remember this, but we were sort of summarizing the 003 and 010 data from the BMS kidney cancer Phase I and Phase II trials. And I showed what a lot of us have showed, which is a swimmer’s plot. And at the end of the talk, this guy got up and asked me a very angry question about my swimmer’s plot. In a thick German …

Tom:
I think I was in O’Callaghan’s by this point, by the way.

David McDermott:
You might have been.

Brian:
Likely.

David McDermott:
And I mean a thick German accent, it scared the crap out of me. It’d be any … but he was essentially upset with 2 things. One, was that the swimmer’s plot is only focusing on the winners. Meaning it’s only focusing on the patients who are benefiting from treatment, so good point by him. And the other was that the …

Tom:
Well, he was right to be angry?

David McDermott:
No, right. Right. No, I mean it … right. So, the other piece was we weren’t being clear about how patients did off-study, when they died, we had some asterisks on the slide that were in the wrong spot. And he was very riled up about this.

Brian:
Wow.

David McDermott:
And what sort of was the takeaway for this, for me anyway was we need to be talking about all patients. As the field matures, we need to not just focus as we had in decades prior with just the few patients who were benefiting, but we need to create a way of measuring how all the patients are doing from the start of treatment. So that that would allow us to create a decision tool.

So that when you were sitting with a patient, trying to decide what treatment to choose, you would have some sense of how that entire cohort did. That you could talk to your patient about what they might expect, if they choose treatment A or treatment B, and that was sort of the beginning of treatment-free survival.

And we’ve gotten data from trials over the years that helped build this model, much of that data up until this point has come from Bristol Myers Squibb and their studies. I just want to make it clear this has been a partnership, and we could not have done it without them.

But the goal of the treatment was not, the goal of the project at least initially was not to try to make their agents look better. It was just to measure both the good and the bad, of particularly PD-1 and PD-1/CTLA-4 and CTLA-4 alone. First in melanoma, and now we’re moving it to kidney cancer with the paper that you guys are on with CheckMate 214.

And ultimately, we are hoping that this endpoint, treatment-free survival can become a validated endpoint that we can use for future studies. Much like we use overall survival and progression-free survival, but it’s one of other endpoints that are also probably useful in future trials. Like …

Tom:
But …

David McDermott:
Landmark progression-free survival … just a second. Just a second, Dr. Powles. And long-term survival, if you want. And so, we’re trying to develop a number of these endpoints, it’s a group effort. We are looking for other people to join this effort. Ultimately, I think it’s essential not just for patients to focus on these endpoints, but it’s important for the field. Because without endpoints that focus on the true benefits of immuno-oncology, the field will have trouble moving forward. Now, go.

Brian:
David, let me ask first, Tom. So, let’s back up and talk about methodology. So, in essence, you’re defining disease states. Or as you’ve put it, which I like, how are patients spending their time from day one of starting treatment? Basically, until death, right?

David McDermott:
Correct.

Brian:
And so maybe just explain for those not in the treatment-free survival world or as familiar, what’s the methodology and what are the different disease states in which patients can find themselves in?

David McDermott:
Right. So, we take the entire cohort. Of both arms, of in this case, the CheckMate 214 study and measure their different survival states over time. So, the first survival state is the time that they’re alive on protocol therapy. And the second is the time they’re alive, but off all therapy, so-called treatment-free survival. The third is where I mean patients spend a lot of time, which is alive, but on subsequent treatment. And obviously, the fourth is the patients who have died. Treatment-free survival represents an area under the survival curve, for the time you spend either between protocol therapy and your next treatment essentially.

And the goal is to try to not only define that time, but also define that time when you’re spending it with and without side effects. So, some of that time you’re alive with toxicity, and some without. And by establishing this measure, what we hope to do is show in the future that we’re amplifying the time patients are alive and off-treatment. In remission, but not amplifying the time they’re alive with side effects, so measuring them both at the same time.

Tom:
So, David, this is a guess, a Q-TWiST analysis back in day.

David McDermott:
Correct.

Tom:
We did that. But this is a next leap on top of that, because you interested in those patients who’ve stopped therapy, who then go on without progression and further therapy. Why is that important?

David McDermott:
Well, ultimately, I think remission or time alive off-drug is the patient’s goal. So, it’s a very patient-centered endpoint. Patients not just want to live longer. But if they can, they’d like to live of all treatment. There are other benefits to having your patients off-treatment besides just quality of life, there’s also a cost-benefit potentially there. So, there are some benefits to trying to expand this outcome, but it’s not an outcome that many patients get.

By focusing on it as our main goal, maybe future regimens can be developed to increase treatment-free survival. That means better drugs, drugs that produce deeper responses, which might lead to remissions, but also things like stopping treatment sooner. For example, one thing that could clearly improve treatment-free survival for some patients potentially is stopping treatment earlier. Maybe not giving indefinite PD-1, for example, to all patients.

Tom:
David, does this data support stopping therapy earlier?

David McDermott:
It doesn’t. Not yet, because that wasn’t the design of the trial. To confirm that, we’d need to design a prospective trial with treatment-free survival as one of the endpoints. So, for example, you could imagine a trial in kidney cancer where you treated patients on immunotherapy combo X up to a year and then took the patients who were in deep response at a year and randomize them to staying on treatment versus coming off treatment. That would help confirm that in some responders, you can stop treatment while they’re in response. A lot of the … some of these patients stopped while in response, but some of them stopped from side effects.

So, one of the criticisms we get of this work is that people are coming off because the regimens are too toxic. And we certainly got that criticism in our melanoma paper, which was JCO a couple years ago on a similar methodology. That’s a valid criticism. I say bring it on, develop regimens free of the …

Tom:
My …

Brian:
Do you …

David McDermott:
Well, let me finish my sentence. The point of that being I know some of these regimens could be toxic, let’s improve toxicity. So how … and if we do that, treatment-free survival without toxicity will increase.

Brian:
David, is it reported in the paper, what percent of patients who stopped treatment did so for toxicity? Or what are the different reasons people stopped? Where you had a cohort of patients focusing on the ipi-nivo group, who stopped therapy quite obviously, proportionally who stopped for toxicity? I mean, the protocol didn’t allow you to stop for response.

David McDermott:
Correct. Well …

Brian:
You wouldn’t …

David McDermott:
Correct.

Brian:
You’ve really stopped for that, right? I mean or you could have stopped, I suppose, for progression?

David McDermott:
Yes.

Brian:
But …

David McDermott:
You could stop if you were … some patients who stopped were in a complete response, stopped early. I think you could stop for that.

Brian:
Okay, fair.

David McDermott:
I’m just looking, I don’t know that we know the answer to that question. Because it’s very …

Brian:
Because I think that’s … I mean getting to the criticism. If all of your … if in order to get into a treatment-free survival state, you have to stop for toxicity. And I realize it’s an artifact of how the protocol was written. That’s very different than what you’re saying, is the next step is stopping for deep response. Stopping at a fixed time point, da, da, da?

Tom:
It’s like diving into a car with David, he’s driving along the street. You say, “Oh, we’re going very nicely, David. Isn’t it great, how long is it until we get there?” He says, “Oh, about 15 minutes.” And about 10 minutes later, you say, “Okay, are we going to stop now, David?” You say, “Oh, well, the problem is that, because we haven’t got any breaks on the car.” And there’s, “Oh, what do you mean?”

Brian:
I have no idea …

Tom:
He says, “Yeah.”

Brian:
What you’re talking about. Seriously, it’s …

Tom:
“Well, how do we stop the car?” And he says, “Well, I normally sort of plow it into a small wall, or we go into a water feature.” Or, “You have to crash the car to get out of it, you have to come into harm’s way.”

Brian:
Well, I don’t know if it’s a crash. I guess what I was going to ask, David, is …

Tom:
Well, but let’s talk about a grade 3 pneumonitis, let’s talk 3 scenarios.

Brian:
Well … sure.

Tom:
A patient gets a grade 3 pneumonitis after 4 cycles of ipi-nivolumab, comes off therapy, and then remains off therapy progression-free for 6 months. That would be just fine, there’s some degree of a success?

Brian:
It depends how bad the pneumonitis is, right? How disabling it is, right?

Tom:
But the principle is that there is inherent, there’s quite a powerful bias in that process saying, “These patients have all got to had grade 3 toxicity to get that subsequent benefit.” And of course, we already know from previous retrospective data, there is a link between adverse events and better outcomes. Now I know that’s quite complicated because there is a lead time bias and landmark analyses are pretty difficult when you try and compare, at the occurrence of toxicity and outcome. But one of the potential problems with this is this isn’t just a case of benignly stopping the drugs, wait and looking for what happens. We know those patients who come into harm’s way, we have to stop the drugs. But we also know that those patients actually tend to do quite well anyway.

David McDermott:
Oh. Oh, right. I’m not … there are issues related to therapies. That by measuring this, you’ll capture, okay? The idea is not to prove that one regimen is better than the other, okay? I know people find struggle to believe that. But that’s the idea, okay? Number one. Number two, let’s talk a little bit about the results of the trial, before we get into [crosstalk 00:13:03] …

Tom:
Let’s do that, David. Fire away, fire away.

David McDermott:
Okay.

Brian:
Great idea.

David McDermott:
This was, look … right. Listen, so CheckMate 214 is ipi-nivo versus sunitinib. So, we had the data at 42 months since randomization. And at that point, more patients with nivo-ipi were alive than sunitinib. So, OS favored the combo of IO. And more, 3 times more were alive treatment-free on the combo of IO. That was in favorable risk patients the net percent alive was very similar, but the percent alive treatment-free was twice in the absolute patient …

Brian:
What were the absolute numbers, remind us?

David McDermott:
20% and 9% in good risk, 18% versus 5% in all-comers.

Brian:
Overall.

David McDermott:
So, treatment-free survival …

Brian:
About …

David McDermott:
Which is the area under the curve, not just where they are at the endpoint, at 42 months was twice as long for the patients with intermediate and poor risk and 3 times as long for favorable risk. So, it gives you a balance. Which is overall survival in, favors ipi-nivo for intermediate and poor risk. And you’re twice as long likely to be treatment-free. So, you can sit with a patient and say, “This is what happened in this entire population. Are you willing to expose yourself to, admittedly more, greater risk? Not of side effects, because the side effects actually were greater in the sunitinib arm, but of potential serious side effects for that benefit of being twice as likely to be alive and treatment-free?” And you can pose that issue to the …

Brian:
Well, and risk of earlier progression, right?

David McDermott:
Correct. Absolutely. You can have that kind of … that this allows that conversation. And it also supports that conversation in favorable risk, where overall survival is the same. You can talk to a patient about the risks of treatment and say, “But you’re 3 times more likely on this trial to be alive off-treatment.” Having that conversation, based on all the data, not just the winners … and not just the positive, not just the swimmers. We’re focusing on everyone here. It to me is a more balanced way to talk about it, and it’s a useful tool when you’re trying to decide what to do first with a patient.

And some patients will choose less risk, which may … is fine with me. Others will choose greater risk, but a chance to live off-treatment. But the goal was not to prove one better than the other. The goal was to create a methodology that could be used for future studies that could add to our way of judging them. Because future regimens as we move from singlets to doublets to now triplets, it’s going to, we’re going to have to be able to measure everything that happens to that patient across the course of their survival. And we don’t do a great job of measuring what happens after they come off protocol treatment. So, this is another way of emphasizing to investigators, “You got to measure the whole thing going forward.”

Brian:
I think it’s a good way after the trial’s matured to look back and say, “All right, let’s look at the entirety of time.” And it’s like looking at landmarks that are distant, et cetera. Is it a good endpoint? So, if you’re developing a novel triplet, right? Compared to a doublet or just single-arm trial, can you use it as an endpoint in a trial to generate a signal that you can use to develop the drug? So, an early enough signal, if you will, that it’s actually useful in drug development? Or is it just something you’re going to have to look back, look at later on?

David McDermott:
Well, I think it remains to be seen. And it would be different for different regimens. Ideally, we were not developing this to speed drug development. I think we should work on that too. And as you know, we are. Obviously, Brian, you’re part of this, with trying to come up with surrogate endpoints that might predict for long-term benefit. Obviously, long-term benefit is not just treatment-free survival, but things like landmark progression-free survival and long-term overall survival. As you know, are important. We, you, and I wrote a review article in Lancet Oncology about this debate, essentially. Which I think is important, if people are interested to read.

Brian:
I’m pretty sure I will not. Yep.

David McDermott:
Yeah.

Brian:
Go ahead.

David McDermott:
Well, you probably did. But the point of the story being, we need endpoints that are shorter, and we need … and this is not that. So, things like depth of response, as you know, may be a surrogate for some of these long-term endpoints. And trying to connect early endpoints with these later ones, I think is an important thing. Also create, connecting a biomarker analysis with some of these IO endpoints is also important. And one trial, that this is off the subject, this is a melanoma study. But I think it’s really important because it gets right to this question of IO endpoints. Which is the DREAMseq trial, which is a trial that Mike Atkins led in the US cooperative group that compared in BRAF-mutated melanoma patients, BRAF combo therapy. BRAF/MEK versus CTLA-4/PD-1 and used 2 -year overall survival as an endpoint for that study, used an IO endpoint for that comparison and showed a clear improvement.

The study was stopped early, favoring the IO endpoint when you used an IO outcome. That’s not just important to the A regimen, but it’s important to patients. And that we’re all … in that context, we’re also going to look at treatment-free survival in that group of patients. And it would be good to do a similar thing with the VEGF/PD-1 regimens in kidney cancer, I think.

Tom:
Well …

Brian:
David, was … go ahead, Tom.

Tom:
David, pulling it back to the results …

David McDermott:
Sorry.

Tom:
Let’s talk about an average patient, who has to stop therapy for whatever reason, were how long into their journey, are they stopping therapy? And then, how long can they expect to be off-therapy for subsequently? Because if we’re going to explain to patients, “You’ve got a 3 times greater chance to be able to stop therapy and not progress,” how long is that period of non-progression?

David McDermott:
Well, I think it’s, it varies per person.

Tom:
Sure. Well, during the study, what’s the median? Was sort of my question.

David McDermott:
What’s the median of what?

Brian:
Was the median treatment-free survival, in the ipi-nivo arm?

Tom:
What was … yeah.

David McDermott:
The point is we’re not measuring medians.

Tom:
Okay.

David McDermott:
We’re measuring mean times, area under the curve.

Brian:
Oh, boy. But there was a median?

Tom:
Okay, but there must have been a period of time? So, let’s say there were 100 patients who stopped therapy early. And of those 100 patients, they would’ve stopped from between, I guess north 0.1 months until an undetermined age. You could generate a median for that. I mean, do you know the amount of time? Because where I mean, when I read the paper, there were a couple of numbers pricking around the region of 6 to 8 months? Is …

David McDermott:
Right.

Tom:
Have I interpreted those correctly?

David McDermott:
Those were means, not medians.

Tom:
Oh, I apologize. I apologize.

David McDermott:
No, it’s fine. And it was 6.9 months for a nivo-ipi, versus 3.1 months for sunitinib, in that intermediate and poor-risk group. And it was even a bigger difference in the good risk group, I think it was something like 11 months versus 3.7 months. Those were means.

Tom:
And so, let’s go with the sunitinib for a second, because we’re quite comfortable sequencing therapy. So, let’s say you get, you stop sunitinib for whatever reason. You then can’t go back on, to something else. And then there’s a 3 -month gap between starting new therapy. I guess that’s fair, because often we would restart therapy at that point. Of this group of patients, how many of them are subsequently progressing and how many were the oncologists comfortable with retreating with immune therapy? Because you get a great response to frontline immune therapy, and then you come into harm’s way, how many are being rechallenged with the same drugs?

David McDermott:
Okay. You’re asking a very different question, which was not addressed by this work. So, we don’t know, and we don’t know exactly what every patient got as salvage therapy on this study. I haven’t seen that data published anywhere or looked at anywhere. And that argues to what I was talking about before, which is if you really wanted to do this well, you would know specifically what patients got as salvage. Each patient, and for how long they were on it as salvage. But in general, what the data is, is it looks like it looks at the time patients are alive on any subsequent therapy, not a specific subsequent therapy. Does that make sense?

Tom:
Yes.

David McDermott:
So, it argues, the larger argument is let’s follow those patients more closely. So, for example, quality of life, quality of life is followed for maybe only 2 months after coming off therapy. I would argue that if you followed it for longer, you would show clear improvements in quality of life for patients living in treatment-free survival. That may be hard for us to show, though, because those patients are not followed long enough. I would argue in future trials, let’s follow it longer. Let’s measure those patients over years. My guess is for those patients, and we all have some of them, that are alive and off-treatment, their quality of life is pretty good.

Tom:
Pretty good.

David McDermott:
But right now, we’re not measuring it. The other thing I wanted to say about stopping treatment, the way this protocol was written … as you guys know. Was if you couldn’t get 4 doses of ipi-nivo, you didn’t get anything else. You were off treatment. You didn’t get maintenance. That’s probably not how most of us manage ipi-nivo in the clinic. But it talks about, that it gets to the question of how you might design a regimen to amplify treatment-free survival. And in some ways that idiosyncrasy of the protocol, which is probably not the current standard of most people’s care, helped create more treatment-free survival. Because people stopped treatment, and they were able stay off it. Does that make sense?

Brian:
And those patients got followed? I didn’t remember.

David McDermott:
Correct.

Brian:
I mean they still got followed obviously for survival …

David McDermott:
Yes.

Brian:
And what not and …

David McDermott:
Correct. Correct.

Brian:
I have 2 questions. One is, were there any baseline or post baseline factors associated with treatment-free survival? I.e., patient characteristics. Obviously IMDC to some effect, which is disease biology, response, depth of response, grade of toxicity. I don’t remember that analysis being done. But was there, is there any way to know as the patient’s entering their treatment-free survival … or at baseline even, at getting their first dose. That they might be the kind of patient who’s going to exist in that state longer?

David McDermott:
Right. So, that is part of an ongoing analysis.

Brian:
Right.

David McDermott:
So, we did it in melanoma. We did, and we looked at baseline characteristics as it predicted for treatment-free survival and didn’t see any clear connections. Which in some ways is disappointing, but also some ways good. Because it suggests that all patients can get a benefit. We’re going to look for the same thing in kidney cancer as well. As we develop novel biomarkers, we hope to do that as well as part of future treatment. But as of now, we haven’t completed that analysis. So, I can’t tell you. But it’s part of several future directions that are interesting and I would say essential to the development of immuno-oncology agents across multiple solid tumors, not just kidney cancer. Because unless we start incorporating novel endpoints, many of these novel drugs are not going anywhere in randomized trials is, would be my opinion.

And unless we start talking to patients about how everyone does starting from the beginning of treatment, we’re not being honest about the pluses and minuses of these regimens as they certainly get more toxic. Because as, so a perfect example in kidney cancer is COSMIC-313. Which we’re all excited about seeing probably next year, which is ipi-nivo-cabo versus ipi-nivo, doing a TFS analysis on that regimen would be of interest to me.

Brian:
Yeah.

David McDermott:
Anyway, are the trade-offs worth it? We all expect it to win on early endpoints, but will it win on OS? Will it win on TFS? I’d like to know, and I’d like to know what the trade-offs are. And this is a way of measuring those trade-offs, by measuring toxicity after you stop and before, on treatment. We measured both on and off-treatment, I think both of those are important.

Tom:
David, we’ve got 280 of the 300 required listeners. So, we can draw a line under this fairly soon.

David McDermott:
Is that right? So, you’re cutting me off here …

Tom:
I was just …

David McDermott:
Just as I’m getting rolling?

Tom:
I’ve got one thing I’d like to say. I’m not sure if, that Brian’s got something too. Many people have come to me and saying, “Listen, we’ve struggled with the TFS endpoint. We don’t really get it.” I talk about crashing the car and you not having the brakes, and they struggle with that.

David McDermott:
I can’t imagine why.

Tom:
If you were … had a lift conversation with someone now. And you’ve got 30 seconds to say how does this information that you have right now help doctors and patients, what would you say?

David McDermott:
So, treatment-free survival is a decision-making tool that helps you decide are the pluses and minuses of immunotherapy worth it for my patient, for the average patient? And it’s an endpoint that favors patients because it focuses on creating more remissions. Which is what we’re, we should be all about.

Tom:
Fantastic. Brian?

Brian:
I don’t know, maybe the last comment. And I think this, the work’s wonderful. Clearly in the last few years, through your effort and Meredith’s and many people, way more people are talking about this than they ever have. And they’re bringing it up as a viable endpoint, so congrats to you, but let me push back a little bit. If I’m a patient and I’m alive with controlled disease and I have minimal toxicity, do I really care if I’m getting a monthly infusion? I mean, I understand cost and convenience and things like that. But at the end of the day, if I’m prioritizing what I want, I want to be alive and I want to have disease control. I want to have an acceptable level of toxicity, right? So, whether I’m getting monthly nivo or not, to me, doesn’t matter that much?

Tom:
I guess, it depends on why you stopped?

Brian:
I wasn’t asking you, Tom, but go ahead.

David McDermott:
So, this is a great question.

Brian:
Great.

David McDermott:
At some level, I think it depends on who’s answering the question. So, I think if you ask the average patient, would they prefer more than disease control? If they could realistically achieve it, and we have to be honest, it’s not a common thing. I think they would say, “Yes, I’ll take a remission. I’ll take treatment-free survival.” Patients understand this concept a lot easier than oncologists is my experience. So, patient advocates love this idea. Other people who understand it pretty well are surgeons. Surgeons understand. We cut the tumor out, we walk away and that’s it. The same thing with BMT doctors, that’s what they do. They put patients through intensive therapies in desperate situations and create remissions. And they’re not into chronic therapy, they don’t need it.

So, part of the problem with this conversation is the audience. The audience of solid tumor oncologists doesn’t believe that most people trained in oncology don’t believe that the average solid tumor can be put into remission. They just don’t, because they haven’t seen it.

Brian:
Mm-hmm (affirmative).

David McDermott:
And they’re nervous about stopping treatment, because they’re afraid patients are going to progress. And sort of, they’re stuck in their ways. So, I think ultimately, we need to create forward looking data that shows that a subset of patients can be put into remission. This, make kidney cancer more like lymphoma, more like testicular cancer, less like pancreatic cancer. Because for some patients, it is. That, the way to measure that progress is through methods. Not just treatment-free survival, but some of these other things that you’re developing, so that we can convince ourselves that you can stop. I think that’s the biggest issue, is the audience. It’s not the data.

Brian:
Good point.

David McDermott:
It’s the audience.

Brian:
Well, and I think these analyses for IO TKIs are even more compelling, right? Because continuing in maintenance nivo, generally pretty easy.

David McDermott:
Right.

Brian:
The patients can get into trouble, certainly. Continuing TKI …

Tom:
Or …

Brian:
Is a much bigger deal.

Tom:
They’re on triplet therapies of the future, this may have a really … this may be really important in the future. I think it’s super cool, David. Congratulations, have a lovely Christmas.

Brian:
Happy New Year’s.

David McDermott:
You too, gentleman.