Episode 138: Renal Cancer ESMO Update

Dr. Laurence Albiges gives an overview.

 

Brian:
Hey everyone. Welcome to another Uromigos podcast. Tom and I are with our good friend and fellow Uromigo, Laurence Albiges from Institute Gustave Roussy. And we’re going to talk today about the updated ESMO guidelines in kidney cancer. Laurence, if you could just briefly introduce yourself and then I think Tom’s going to give us a little background on sort of the structure of the committee.

Laurence Albiges:
Sure. Thank you, Brian. Thank you, Tom. So my name is Laurence Albiges. I’m a medical oncologist working at Gustavo Roussy Institute. My main focus is RCC. And so for the purpose of today will be discuss the ESMO updated guidelines. So Tom, maybe you can present the group. This is a group that you, you built and you animated very well.

Tom:
Well, that’s sweet of you. Lots of people are involved in the group. So essentially it’s a pan-European group, as you would imagine, on the ESMO guidelines. It’s I think it’s really well balanced. It’s got… I took over from Bernard Escudier, which was an easy job because he’d done such a fantastic piece of work in bringing people together. And essentially, we’ve now recently updated the… [inaudible 00:01:30] They actually only came out two years ago, but we felt the need to update them because so much has changed so quickly. One of the challenges we have is how often… Does the guidelines keep up with the rapid changes?

Tom:
The guidelines that we produce, and the group agreed on this, we produce guidelines that we think would be useful for patients and doctors. Therefore, we don’t just produce a long list of drugs, number one. We tend also, we don’t always go with exclusively with EMA advice. So we’ll approve things if we think they’re beneficial [inaudible 00:02:07] the EMA has not yet approved them. And occasionally, it’s possible we might even disagree with different bodies and different groups. We try and be as balanced as possible. We very rarely get a hundred percent consensus, but essentially when it’s over 75 percent, we tend to go with that. So…

Brian:
How many people are on the committee?

Tom:
About 10 people.

Laurence Albiges:
Yeah, I would add on that we do have the chance to meet for the moment quite virtually on a regular basis. And I think really Tom mentioned two key things. And the e-update allows us to be a very proactive and reactive to what are the data being presented. We want the guidelines to be clinically meaningful, and we try to capture the discussion that we have within Europe when looking at the new data coming up so that we can measure them for a very applicable resource.

Brian:
And so you update basically whenever the group feels like there’s enough accumulating data that it would impact practice. Is that fair?

Tom:
Well, it’s a bit more structured. [inaudible 00:03:11] No, you’re absolutely right.

Brian:
Educate me.

Tom:
Where we feel it’s good to go, it’s good to go. And I think, and it’s happened twice in the last year, we did an update about a year ago where we put the frontline VEGF TKI IOs with equal support with level 1A evidence. And why don’t you ask us what’s in our most recent update, Brian?

Brian:
Tom, what’s in your most recent update?

Tom:
Perfect.

Brian:
Thank you for feeding me the questions. I’ve not done many of these, so I appreciate it.

Tom:
So listen, we put len pem, lenvatinib pembrolizumab, in. And so that has now level 1A evidence in good, intermediate, and poor. That joins the other VEGF/TKI IOs. We also looked at papillary renal cancer. We did a review of papillary renal cancer.

Brian:
Can we stop on the VEGF IOs first?

Tom:
Okay. Go then.

Brian:
See, it’s not fun being interrupted, is it?

Tom:
It’s not, is it?

Brian:
A little lesson for you.

Tom:
Go, go, go.

Brian:
I’m going to ask Laurence, since she’s our guest. So among the VEGF IO combinations, do you differentiate? Do you give one a different recommendation than the other? Or do you sort of lump them in a bucket?

Laurence Albiges:
No, I think the key message here is that if you look at the new figure of these guidelines, that we go for a combination in all our patient with advanced clear cell RCC. That’s the key message. So in good risk patient, we will be offering TKI/IO strategy, for those regiments that have demonstrated overall survival benefits. And you mentioned that we rank them all the same in a way that these are randomized data against active agent with a significant and meaningful clinical endpoint, that is overall survival. That’s what…

Brian:
Let me push you on that a little bit. Since the guidelines are meant to be very practical and clinically useful. In essence, you’re giving a list of drugs, right? You’re giving a list of regimens. So how does a practitioner interpret that? Like, “Which one do I pick?” is what they want to know.

Tom:
Laurence, do you want to take that?

Laurence Albiges:
Yeah, no, no, it’s a good point. I think our job as a guideline committee is to make sure that people get where do we stand in terms of knowledge of evidence. So we have randomized data over single agent TKI, and now we know that TKI/IO in good risk patient, for instance, or at least across all risk group, is our recommendation. We do not differentiate one combination over the other because we don’t have the level of evidence to provide recommendation on that. This is no head-to-head comparison, Brian. So I’m sorry that you do have a list there, but you’re not supposed to provide one recommendation over the other.

Brian:
Okay.

Laurence Albiges:
If you don’t have…

Tom:
Yeah. Brian, just to push back a bit more. So the EMA, for example, has approved Axi Avelumab, and we don’t support that because it doesn’t have an overall survival signal. So it’s not just a long list. And that’s for Bavatezo… The area of controversy and the good [inaudible 00:06:10]. And I know that some people feel that Pazopanib should be on the same level as the VEGF/TKI PD-1 combinations, but we haven’t gone for that.

Brian:
So you… So single agent TKI is not recommended at all in favorable risk? Or it’s a lower recommendation?

Tom:
There’s a caveat at the bottom, so it’s not in the table. In the table, level 1A evidence is for the PD-1 IOs. In the text, it says that as of yet, an advantage has not been proven in this population, response rates and PFS is longer, and sunitinib remains an option for these patients, particularly obviously where it, where the IO combinations are not available.

Brian:
Do recommend Ipi/Nivo at all in favorable risk or no?

Laurence Albiges:
No, we don’t.

Brian:
No, not even as a footnote or a considering select circumstances or anything?

Tom:
Nope.

Brian:
Okay. And then Tom, you were talking about papillary. So tell us about recommendations.

Tom:
Well, can we go…

Brian:
Have you guys done papillary before?

Tom:
Well, I mean, Bernard did them when he made guidelines. Yes, of course he did. Yeah. I mean…

Laurence Albiges:
Yeah, I think we moved away from like having non-clear cell being enrolled in clinical trial to… And that was at the time that Bernard was doing this, splitting into papillary versus other subtypes.

Tom:
That is a good point.

Laurence Albiges:
Including non-clear cell. And now we are actually getting to the next step, which is within papillary. Do we have a preferred option? And so on the ground of data that were presented this year, Tivozanib versus Sunitinib, in the PAPMET study, we now have a preferred option. But because this is a phase two PFS endpoint, our level of accommodation is 2B. We do stand in our figure, in our guidelines that there are alternative option, that single-agent TKI is still something that you may want to consider in those patient. And maybe Tom, you want to add on those two that we have added as alternative option, both pembrolizumab based on single agent, non randomized data, and Sunitinib as well?

Tom:
Yeah. So I that’s… I think it’s nice that we’ve really focused on papillary cohort. I think maybe in kidney cancer, lumping the non-clear cells together with hindsight is probably not sustainable. I think papillary does need to be considered in its own entity. I think some people say, “Well, hold on a second. You haven’t got a survival advantage for Cabozantinib. You do. And so without your… You have one set of rules for clear cell and another set of rules for papillary. One, you have to have OS the other DFS is all right.” I think that’s fair enough, but I can see the inconsistency around that. I think the PAPMET study suggest Cabozantinib is a bit better than Sunitinib. It’s obviously not a randomized phase three and therefore survival’s more challenging to achieve. And I think that for that reason, that’s reasonable.

Tom:
I also quite like Dave McDermott’s Pembrolizumab study, the 427 study. I realized PFS was short, but the OS looked pretty good, I thought. And remember that one had over a hundred patients in it. Although it wasn’t randomized, it actually is probably the most robust data set of any of the arms we have.

Brian:
Yeah.

Tom:
So I think that’s fair enough. I think it’s okay for Sunitinib to appear there. But the Savolitinib one is obviously controversial because you can’t buy Savolitinib don’t think. And, and so it’s sort of one of those, it’s a bit of an extreme example where we’ve gone for something which actually isn’t necessarily that helpful for patients. But it does underpin the point that we do feel that targeted therapy is important. And my fear is we do these trials and we just let these good opportunities go. There’s obviously the Savoir trial, which is enrolling, which is a randomized trial of Savolitinib plus Durvalumab versus Sunitinib in patients with MET alterations.

Laurence Albiges:
So that’s Samita. Okay.

Tom:
Oh Samita! Sorry. Yeah, Savoir was the one before. So sorry, Laurence!

Laurence Albiges:
No problem.

Brian:
So it just to clarify.

Tom:
I can’t recall half of it, rogue, apparently. I’ll have to go back and look at the…

Brian:
For papillary, what is the… So Cabo single-agent is the highest recommendation? Or is it equal to Pembro single-agent?

Tom:
It’s got 2B recommendation.

Brian:
What does?

Tom:
And it has, and it is described as the preferred option.

Laurence Albiges:
Cabo does.

Brian:
Cabo monotherapy preferred option?

Laurence Albiges:
Yes.

Tom:
Yep.

Brian:
Pembro monotherapy, excuse me. Is another option, however, as alternative and Savolitinib is also an alternative.

Tom:
Yes.

Laurence Albiges:
Yep.

Brian:
Yeah. At the same level as pembrolizumab.

Tom:
No, Savolitinib is 3C because it’s much weaker data.

Brian:
I see. And what about IO/TK? I mean, how do you actually treat these people when they come into clinic? I mean, why shouldn’t they get an IO/TKI?

Tom:
Well papillary renal cancer patients.

Brian:
Yeah. Well you just said IO is active and TKI is active. We know the combination is more active in clear clear-cell.

Tom:
Laurence, do want to take that?

Brian:
There’s combination of data.

Laurence Albiges:
Yeah. Yeah. So I think…

Tom:
Humor him.

Laurence Albiges:
We started this podcast by the fact that we wanted evidence-based data. And so here we are clearly lacking combination regimen. Tom just mentioned some of them that are being currently up and running, trials up and running, but we don’t have generated data yet at a sufficient level of evidence that we would recommend TKI/IO for an unclear cell, especially papillary.

Brian:
You think there’s more data for Savolitinib as there is for IO/TKI? I mean, they’re single…

Laurence Albiges:
We do have a randomized… We do have a randomized trial.

Tom:
That’s published in Lancet Oncology.

Brian:
Yeah. With a very limited… What number of patients were in that trial?

Tom:
I think it’s 50 in both arms.

Brian:
Yeah.

Tom:
It might be smaller, but it’s… Yeah.

Laurence Albiges:
Yeah. But it’s over 200 patient that have been screened. So at least it’s a biomarker-based trial.

Brian:
No, I’m not saying it’s not a good trial. Just think IO/TKI has as much evidence. And if you want to be helpful to patients and clinicians, then I think IO/TKI is perfectly viable for papillary and it’s my preferred regimen. I think it’s many people’s preferred regimen.

Laurence Albiges:
Well, we would love to to have more data on this and actually get approval on this.

Tom:
It’s very implicitly [inaudible 00:12:27] Love it.

Laurence Albiges:
We’re getting back to approval and access, and I must say that in Europe, we do have to face this issue of how do we access these drugs or this combination. What’s approved, what’s reimbursed?

Tom:
Brian, we’ve got a last thing to talk about.

Brian:
Adjuvant therapy. What about adjuvant therapy? Is that the last thing?

Tom:
Well, that’s the last big area.

Brian:
Since apparently you’re asking the questions, Tom. What’s what’s the last thing?

Tom:
Well, no…

Laurence Albiges:
I think that’s the one where we had the most debate, Tom. I think it’s fair to say that we had to have several meetings so that we can match rate the data analysis. And that was actually a lot of discrepancy in the way we were considering these. And it was quite a hard job on Tom’s shoulders to make sure that we could come to a compromise or at least a recommendation that would fit with everyone.

Brian:
So where did you land on adjuvant pembro?

Tom:
Laurenence, do you want to summarize that?

Laurence Albiges:
Yeah, I think… So, the big picture is that we found the consensus on the fact that we had randomized data. So that’s a level one evidence. However, there was a lot of debate on how substantial the clinical benefit was considered. Meaning, would we consider that DFS is sufficient in adjuvant setting? A lot of people in the group of course, were looking forward to have oral survival benefit before making a strong recommendation. So that took us to a consensus or at least some kind of trying to find a consensus towards finding that pembrolizumab is an option that can considered in the patient population that was defined by the keynote 56 core study. For pembrolizumab adjuvant, as it is within the trial for intermediate high and high risk patients. There has been a lot of debate on the M1. Maybe Tom, you want to comment on that as well?

Tom:
Yeah, I think it was a very difficult discussion. And in our dialogue, we actually write quite a lot with caveats. And the recommendation essentially is a 1C, which is after careful consideration and patient counseling, it should be considered. There were people in the group who felt much strongly about it being a 1A or a 1B. And there were people who felt it was premature at 1D, so I don’t think… And actually when you go around, that’s kind of where I think we landed and people were comfortable with that. But I have to say the recommendation changed twice during our debate.

Brian:
So one… Just to clarify, so I was going to ask you if 1C was the lowest level one, but it sounds like there’s a 1D.

Tom:
Well D is recommended against.

Brian:
Oh, okay. So it’s the lowest positive recommendation, so to speak, that you can get for level one evidence. Is that fair?

Tom:
Yes.

Brian:
Okay. And it changed, and if you’re willing, you’re able to say, what did it… Was it fluctuating between 1B and 1C or 1C and 1D or…?

Tom:
It was changing on a daily basis at one point, Brian.

Laurence Albiges:
That’s fair to say.

Brian:
But you did achieve consensus, so you said. Seven or eight out of 10 of the members agreed that it was, quote unquote, “an option,” which is sort of, to me, “an option” is a very weak recommendation. It’s almost damning with faint praise. It says, “Yeah, sure. Use it if you want, but we’re not that excited about it.” Is that fair or unfair?

Tom:
I think we are excited about it. I think we’re very excited about it. I think there’s a difference between being excited about it and giving very strong recommendations for using it. The people in the group who were reluctant for its use made the point that we don’t yet have overall survival. We are inevitably overtreating a significant subgroup of patients and the M1 NED population, that population of metastatic disease, may be skewing the results one way or another. And so…

Brian:
There was only 6% of the population, right? That was M1 NED.

Tom:
Yes. And then the other group, the other people who wanted level 1A, and there were people who wanted level 1A evidence, and the people who wanted level 1B and I have to say there were more people who wanted 1B than 1D, the people who wanted level 1B were saying, “Well, at the moment, it’s a very strong PFS signal. It’s at a high area of risk for patients that the survival is trending in the right direction. And why wouldn’t patients want that as it currently stands?” I think with a considered recommendation, I think you can then put it to patients and then they will have the option about whether they take it or not. And some of the people who argued against level 1A and 1B said, “Doctors probably shouldn’t be going in saying, you need this therapy now until we hear overall survival.” And I think as the conversation matured, those people with stronger feelings that we should give it were happy that it could be more of a patient-focused discussion rather than doctors telling patients what to do.

Brian:
Which is true for everything, right? I mean, even for higher levels of evidence, right? I mean, that’s kind of true for everything that it’s a benefit risk discussion with the patient. But I hear what you’re saying. So if survival doesn’t come through… Let’s say, the early signals based on, 50 events or less than 50 events, a lot of events to go, so to speak, and obviously driven by the early recurrence in death. If it doesn’t come through on the final analysis, that’s probably years away. Does it move to a 1D? Does it move to some different category?

Tom:
Laurence, do you want to take that?

Laurence Albiges:
Yeah, well, I’m not sure I have a straight answer to that one. I think I will take into consideration of course, the updates on this trial, but it’s very likely that at the time, we’ll have more major data with further immune checkpoint inhibitor in the adjuvant space. And that is clearly something that will be taken into consideration.

Brian:
Meaning if all the other trials are negative as what happened with the TKIs, where there was a lone positive trial. And I don’t know that anybody expects that, but it’s certainly possible, that would dampen your enthusiasm for pembro in this setting.

Laurence Albiges:
I think it will at, at the community level. With regard to guidelines, what we try to do, once again, is to stick to the level of evidence. What we have is randomized data. And we build on the clinical outcome out of this study. So DFS, OS, and that is what we’re looking at. And of course, paying attention of the risk ratio benefits.

Tom:
The EMA previously has asked for updates on data cuts. And it’s likely that we will… I hope we will get more survival signals as time goes by with this. And that I think would be helpful. There was an issue as well around this M1 NED population, which was equally controversial. The reason that was controversial was many people on the committee felt that if you relapse within a year… Not you. If a patient’s cancer within a year of nephrectomy, those patients receive PD-1 VEGF/TKI combinations, and that’s the standard treatment, not metastasectomy.

Tom:
And so many people felt that we’re starting in the wrong pace. Those patients shouldn’t have had surgery, or most of those patients shouldn’t have had. Now it’s an impossible conversation, but…

Brian:
Right.

Tom:
I guess the question… And it’s happened and you can’t go back and say, “Well, I wish we didn’t start here.” I think there’s an Irish proverb about that which I can never remember the detail of. But the key to it is that if you are in that position, it does seem like a reasonable treatment, but the committee was very key, not to encourage patients to get to that position if you know what I mean.

Brian:
So, just to clarify, keynote required that the metastasectomy population be within a year of that the metastasectomy was within a year of nephrectomy. Is that correct?

Tom:
Yes.

Laurence Albiges:
Yep.

Tom:
And what the recommendation says metastasectomy as an alternative to systemic therapy in patients with synchronous or early oligometastatic disease is not usually recommended and requires a multidisciplinary decision. Adjuvant pembro can be offered to these patients after complete resection with level 2B evidence.

Brian:
But what if a patient walks in the door, and you may or may not know this, has a slow growing primary and a slow growing solitary met. You would absolutely resect or irradiate that met. I think that would be standard. But you’re saying you shouldn’t do that. Is that right?

Laurence Albiges:
The example you’re giving us, Brian, is a patient who has slowly growing disease. It’s unlikely that this is within one year. And so it’s actually not this patient population that is…

Brian:
Yeah, but synchronous disease. So they present with a primary and a solitary met, and they have some old imaging and because of COVID, they couldn’t get their surgery or whatever. So they’ve been sort of lingering for 8, 10, 12 months, whatever.

Tom:
Yeah. So Brian, if you get there, but you agree, coming back to the Irish proverb, if you get somewhere you don’t want to be, where do we go next? Yeah, that’s fine.

Brian:
I don’t think that’s the proverb, by the way, but okay. We’ll investigate this, but okay. But you’re basically saying it’s not really about getting the therapy or not. You’re just saying those patients really shouldn’t exist. Those patients should have gotten systemic therapy per our other guidelines.

Tom:
No, I think we acknowledge those patients exist, but I think they were pity and it’s not necessarily Laurence and I who felt this way. It was the committee as a whole.

Brian:
Hiding behind that committee. Okay, go ahead.

Tom:
It was Camillo. It was Camillo. That as a whole, we shouldn’t be encouraging metastasectomy within [inaudible 00:22:27].

Brian:
Okay. That’s a different issue. That’s a different issue,

Tom:
But it’s okay bringing other issues in. Because if you’re not entirely happy with some of the criteria, you have to bring those issues in.

Brian:
Fine. Fine, fine, fine. But I’m just saying a patient walks into your clinic. Somebody else made the decision to do metastasectomy within a year. You weren’t involved with that. Now the patient’s before you.

Laurence Albiges:
Yeah, then we consider Pembrolizumab.

Brian:
And is it the same level of recommendation as a straight adjuvant?

Tom:
No, it’s 2B. It’s 2B.

Brian:
But why?

Tom:
Well, because it’s a smaller subgroup, Brian. It’s not a level… It’s a smaller subgroup, isn’t it? It was only as you say it was only 6% of the population. There is some uncertainty in that group.

Brian:
Well, there’s uncertainty in a lot of small subgroups. So are you making these recommendations for all the small subgroups in the trials?

Tom:
It sounds like you’re not super happy with this recommendation.

Brian:
Well, I’m just trying to understand the logic of your… You’re sort of picking on this one, small subgroup whose hazard ratio by the way, was the best of everybody, right?

Tom:
I think it was 0.29.

Brian:
And then you have the highest risk of recurrence.

Tom:
But it was only 50 patients as opposed to 1000 patients.

Brian:
I understand it’s a small subgroup.

Tom:
Yeah.

Brian:
I’m just saying that there’s small subgroups in all these trials. Right? [crosstalk 00:23:32] If we go cherry… If we go… I’d love to be. If we go cherry picking the small subgroups and saying, “Well, we’re going to make a lower level of recommendation for this, that, and the other small subgroup,” you’re going to get into trouble. Because where do you stop? Where do you start and stop with small subgroups?

Tom:
So I think this comes back to the original Ipi/Nivo discussion with a good, intermediate, and poorest group. Remember in the original trial, it was an ITT trial. And then there was the intermediate and poor subgroup and the ITT trial was positive. And so we’ve been down this route, the global community’s been down this route.

Brian:
Well…

Tom:
The vast manner. I think this is only 6% of patients. It’s very hard to give level 1A recommendations on this. We had the discussion. Again, there wasn’t consensus on this issue. It was a difficult issue, but the group [inaudible 00:24:17] The first decision was made. Should we look at these patients in isolation? Is that fair? Yes or no. The group said, “No, they should be looked at in isolation, not as a whole.” And so we pulled that group out and that discussion took place. So the point you make is correct, is that it is an ITT trial and we pull that out. And then you could take that argument to an extreme and say, “the day of the week the patient was randomized.” Right? And so I accept that, but at the same time, people felt that it was such a different population from the original population in terms of metastasectomy versus the primary renal tumor. That that was the right thing to do.

Brian:
But if anything, those patients have the highest risk of recurrence, right? Metastasectomy patients, especially within a year, have, what, 70, 75% risk of recurrence? So if you’re going to offer a potentially toxic therapy, why would you not do it in the very high risk patients? It’s like pulling out the node-positive patients. How many node-positive patients are there?

Tom:
So we’re giving a 2B recommendation on that because it’s not super robust, but it is a B rather than a C. So it’s a stronger recommendation on less robust data.

Brian:
I just think… I don’t think it’s less robust. I think you’re picking on a small subset and I think…

Tom:
Well, it’s less robust when the numbers are smaller.

Brian:
If… Well, right. But I can go find some other small subsets that are less robust too, but you’re not going to pick on them. If it’s a small subset and the hazard ratio is… The confidence intervals are wide, which these are from zero to infinity. Why would you not default to the overall population? And say, “We can’t really say in this small subset that was 6% of patients.” So your default has to be the overall population that the effect is similar to the overall population.

Tom:
I don’t disagree with what you’ve said. I do think the approach that you’ve come up with with level 1A across the board will be reasonable. There were people within the group who felt strongly that that subgroup is an unusual subgroup. That metastasectomy within a year is not the standard of care. And therefore giving level 1A evidence for that approach might end up with more people wanting to do more surgery in that population.

Brian:
Oh, I don’t. I think that’s crazy. I think. To think that people are going to look at these guidelines and say, “Gee, we should do more surgery.” I think that’s crazy. And you’re giving yourself more credit than you deserve, frankly.

Tom:
Well, I’m looking forward. I’m looking forward to the letter…

Brian:
Yes. I’m writing an editorial. Okay. I’m sorry to pick on you guys over that. I just… Obviously we get into discussions about overinterpreting small subsets all the time. So it’s the larger point about overinterpretation of small subsets that I think we, the global we, does a lot and it really gets us into trouble, in my opinion. But we don’t have time to cover that.

Brian:
So, well, I’m satisfied with this discussion. I got to pick on you guys for 25 minutes. This has been a lot of fun for me.

Tom:
Thank you.

Brian:
Anything else come from the guidelines? Any other final thoughts on the guide…? Well, I have a quick question. Do you guys consider quality of life data in your recommendations?

Laurence Albiges:
So once again, the risk is to consider a quality of life across trial. And we clearly made a point that we didn’t want to compare those trial based on quality of life, given that it wasn’t assessed in the same way. And we don’t want that to be used as a surrogate to cause comparison.

Brian:
Okay. So I guess you generally don’t consider it.

Laurence Albiges:
We consider quality of life for our patients, Brian. What are you saying? Were you just saying…?

Brian:
No, no, no. I meant the data. Like you might consider it within the context of a single drug, but you’re not using it across regimens. Is that a fair summary?

Laurence Albiges:
It is.

Tom:
Brian, there’s a magnitude of clinical benefit score. The magnitude of clinical benefit score is the degree of benefit associated with the therapy. And it’s defined by a PFS advantage gives you a point, an OS advantage gives you a point, and it’s scored out of five. And actually having predefined positive quality of life gives you… So within the ESMO guidelines, there is clear role for quality of life in terms of how does the group decide where we are. Well, I think quality of life is pretty difficult. We’ve talked about it before. It’s a difficult endpoint, for example, in Cabo/Nivo in the text, we’ll probably say it has positive quality of life data, but we wouldn’t be recommending it over Len/Pem because of that.

Brian:
Yeah, I hear you. Yeah, yeah.