The Association Between Depth of Response and Clinical Outcomes in the CheckMate 9ER Trial in Advanced Renal Cell Carcinoma - Andrea Apolo

January 19, 2023

In this conversation, Pedro Barata and Andrea Apolo highlight an exploratory analysis of the CheckMate 9ER trial in patients with previously untreated advanced renal cell carcinoma (aRCC), looking at the association between depth of response and clinical outcomes with extended follow-up (minimum 25.4 months; median 32.9 months). In January 2021, the Food and Drug Administration (FDA) approved the combination of nivolumab and cabozantinib as first-line treatment for patients with advanced renal cell carcinoma based on the efficacy evaluated in the CheckMate 9ER trial. In the analysis being discussed in this conversation, patients were subdivided into depth of response subgroups based on best overall response (blinded independent central review [BICR], RECIST v1.1) and best tumor reduction thresholds (CR; partial response subdivided by tumor reduction of ≥80% [PR1]; ≥60%-< 80% [PR2]; or < 60% [PR3]; stable disease; and progressive disease). The authors examined progression-free survival (per BICR) and overall survival stratified by depth of response subgroups, after a 6-month post-randomization landmark. Andrea Apolo details the findings of this analysis in this conversation.


Andrea Apolo, MD, Medical Oncologist, Senior Investigator, Center for Cancer Research, National Cancer Institute, Head of Bladder Cancer Section, Genitourinary Malignancies Branch, Director of the Bladder Cancer and Genitourinary Tumors Multidisciplinary Clinic, Bethesda, MD.

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

Read the Full Video Transcript

Pedro Barata: Hi, and welcome. My name is Pedro Barata. I'm a GU Medical Oncologist at Seidman Cancer Center, Case Western University in Cleveland, Ohio. And today, I'm joined by the great Dr. Andrea Apolo. Dr. Apolo, probably everybody already knows her, she's a Senior Investigator at the GU branch at the NCI. She also is the Chief of the bladder cancer section at the NCI. And she was, in fact, a PI in the CheckMate 9ER. And, today we'll be talking a little bit about 9ER and the subgroup analysis or exploratory analysis that she helped conducting. So with that, thank you, Dr. Apolo for joining us today.

Andrea Apolo: Thank you for having me. It's great to chat about the details of this trial and this sub-analysis. I think it's really important.

Pedro Barata: Absolutely. So first, great presentation out of 9ER. I mean, this is very important data, right? Established cabo/nivo as a frontline regimen for patients with vascular cell, renal cell carcinoma, and beyond what was disclosed to all of us when the study read out about endpoints, primary, secondary endpoints, PFS, OS, et cetera, I think is really cool to go beyond that and show other data of interest to us. And one of that, actually, we're going to be talking today, which is that depth of response. Maybe I'll just start there and ask you in general, right, we've talked a lot about endpoints in the context of TKIs and then in the cons of immunotherapy endpoints are relevant in RCC. What are your insights? How important do you think the depth of response really is? And what MI mean for investigators on one end treating physicians and patients?

Andrea Apolo: Yeah, it's a great question. And I think that we're still learning, and there's been data in other tumors including kidney cancer that have shown that the deeper the response, so the more the tumor shrinks, the better the patient does overall in terms of progression-free survival and overall survival. And to have this large data set of the CheckMate 9ER and to be able to go back and look at this and assess and analyze the PFS and the OS data, I think is really important. So this subset analysis was presented already as an oral presentation at ASCO, and we did an encore presentation of this because I think the data is really important and really understanding who are those patients that are going to do the best and what the percentage of the shrinkage really means in terms of the patient outcomes. So playing around with that data is really important for us. And so we can learn more as to how to advise our patients and how for us ourselves to understand how those patients are going to do overall with these treatments.

Pedro Barata: Right. No, that's really well said and I couldn't agree with you more. That's really important. I really like that endpoint. And we talk a lot about... as you know, we always say we do not compare responses across trials, but then we all get those tables together where we see ipi/nivo one column, and then we see cabo/nivo, axi/pembro, lenv/pembro, et cetera. And I guess one fair statement is in general, IO/TKI, whether it's cabo/nivo or different regimen, seems to be associated with a higher response rate in general and lower PD rates than actually IO-IO ipi/nivo in this case. So with that in mind, I think it's really neat that what you did is you took one step further. So we get it, we are in the 60, almost 70, 60% range or so for overall response rate. We actually take one step further and you did a deep dive into, okay, what about the responses? What does that mean? How much of that is really a response? How much was CR, how much was good PR? So with that said, can you briefly summarize for us what were the main findings in your exploratory analysis of 9ER?

Andrea Apolo: Yeah, so I just wanted to summarize again just the results of CheckMate 9ER for everybody. So CheckMate 9ER was a randomized phase III trial that asked the question in patients with advanced metastatic renal cell carcinoma, clear cell renal cell carcinoma in the first line setting the combination of cabozantinib and nivolumab versus sunitinib. So the study randomized 650 patients, and the results showed that there was a doubling in the progression-free survival with cabo/nivo, an improvement in the overall survival with a hazard ratio of 0.6. There was a doubling of the overall response rate from 27 to 56% with the combination of cabo/nivo. So those were the results. And based on that data, the FDA approved cabo/nivo as a treatment, first line treatment for patients with renal cell carcinoma. So I just wanted to give that background. So we wanted to look at the patient... and this analysis was done in longer follow up.

We have 33 month follow up now, and we looked at the patients that achieved a CR, a PR and then we broke down the PR by PR1, PR2, and PR3, and those are for the percentage of shrinkage. So a CR is a hundred percent response. PR1... And these were kind of arbitrary numbers. A lot of other studies have looked at different percentages and kind of what would constitute a deep response. We broke it down, we broke down the partial response, the PR into three, PR one, two, and three, and PR1 is anything, if it shrinks less than a hundred percent to 80%, PR2 is 60 to 80%, PR3 is greater than 30% to 60%. Stable disease of course is a really broad category because it constitutes anything less than 30% to PD. So that's a large category. And then of course there's PD.

So that's the way that we broke this analysis down. And when we looked at the patients that received cabo/nivo, we saw that only 5% of the patients had PD. And this was not surprising to me because I ran the phase I trials at the NCI. At the NCI usually, we do phase I, II trials. So it's really a treat to be able to be part of this phase III trial. It was a small subset, it was in all GU tumors. And when we did the kidney cancer expansion, we saw PR and CR in about 65% of the patients, and everybody else was stable disease. So we saw no progressive disease. So in this larger cohort of CheckMate 9ER to see only 5% PD is not surprising. And then for the CR, it was 14% for the cabo/nivo, and then PR1 was 11%, PR2 was 13%, and PR3 was 24%.

Of course, this is much higher than what we saw with the sunitinib. With the sunitinib, it was about half of the CRs, it was 7% CRs. And actually most of the patients had stable disease. There was 50% of the patients had stable disease. So that's kind of like the general breakdown of the responses and how we saw them. So we looked at PFS and OS, and we also looked at time to response, duration of response, and we also looked at adverse events because we were like, is there more adverse events in patients that have response? So we had all these different questions in this analysis. And looking at the depth of response in the patients in terms of PFS, the patients that had a complete response, of course, did the best. And that was true for both cabo/nivo and the sunitinib arm. And the patients that achieved a PR1 and PR2 did actually really, really well and had better PFS than those that achieved a PR3 and much better than those that achieved a stable disease.

So in terms of PFS, there was definitely an association in the cabo/nivo arm, and also a slight association but the numbers were really small in the sunitinib arm in terms of the PRs. But there was an association with the depth of response and the PFS. We looked at the 12 month PFS, and that was also numerically associated, and it was improved if you had a deeper response in the cabo/nivo arm and also in the sunitinib but again, the numbers were very small and not... the association was not as clear cut as it was for the cabo/nivo. And then we looked at the survival, and this is kind of the question I think that you were getting to. Is there a survival benefit in the patients that achieved deeper response? And in this subset analysis, we did see that those patients did better. The patients that had a complete response actually in both arms in the cabo/nivo and in the sunitinib arm, did really, really well. They did the best.

Patients that had a PR1 and PR2 and PR3 also did terrific in terms of overall survival. And we have Kaplan Meyer curves for these. And these patients are just doing really, really well. Very similar actually, PR1 and PR2 and PR3 very similar in terms of their responses, in terms of their overall survival compared to stable disease and compare of course to PD. So there was an association in this analysis in terms of the depth of the response and the overall survival. And we also looked at the 18 month overall survival probabilities. And these were similar between the treatments arms for the patients that achieved a CR, a PR1 and a PR2. And in both treatments arm the increasingly deeper responses led to better overall survival outcomes.

Pedro Barata: Well, that's a great... a fantastic summary of the findings. And when I saw that data, I think we've been talking a lot about CRs. I mean, I know you recall, I mean, we were looking really for the first... and fairly so right, because we saw double digits for the first time with ipi/nivo, and then we were waiting anxiously for the IO/TKIs, and we were talking a lot about CR and the meaning of CR. So I'm... first, I'm happy we highlight the low number of PDs because I actually, that means a lot to patients in clinic, right? Because it's more like, it's not what's the chance of cancer going away while you are on treatment? It's rather, what are the chances that treatment is not going to work, right? That's as best response. So I think that's really meaningful. So 5% is really, really low.

But also what I think these data suggests, it gives actually more hope because I think the conversation gets a little bit broader because it expands from the CR group that we know they're destined to do the best. And you say, wait a minute, even if you don't have a CR and you have a deep PR, you are destined to do perhaps as well or as CRs or close to CRs or closer to CR compared where patients who don't achieve a response. So I really think this data is important because I guess it's applicable to more patients that we have in clinical practice, and it's more informative from that perspective.

Andrea Apolo: And RECIST has limitations and what we call a PR1 maybe a CR because we're measuring scar, we're measuring something that's left over, but we don't know if that's viable tumors. So it's limitations of RECIST. And I think this shows that the PR patients, the deep responses, they do almost as well as a CR.

Pedro Barata: Right. No, that's wonderful insight on that. And maybe I'll put you on the spot now a little bit. I know there's not a clear answer, but I have to ask you that, right? Because as you said, I mean this is a updated follow up, longer follow up, all these IO/TKI trials including 9ER, were launched and read out later, ipi/nivo. We're always comparing for the follow-up data with ipi/nivo over five years now. And we love to look at the Kaplan Meyer curves and see how those subgroup of patients who remain without progression, how are they doing over time? Is that perhaps... curious... perhaps a very strong statement. But there's definitely durable responders out there with IO-IO, and I think all of us are anxious to see, can we identify those patients on an IO/TKI? So I guess my question applicable to cabo/nivo would be, can we actually have you identifying these longer follow up analysis? Do you start identifying patients where you say, here are my longer or durable responders? And do you believe some of them will achieve durable remissions even with we there to actually stop both therapies, cabo and nivo?

Andrea Apolo: Yeah. No, this is a important question. We did look in this analysis at the time to response and at the duration of response and with the association of the depth of response and the patients that achieved a CR, a PR1 and a PR2, these had a really short time to response. So that's important. So we saw the responses really quickly radiologically, and then in the patients that had a PR1 and a PR2 in the cabo/nivo arm specifically, they had a longer median duration of response compared to the patients that achieved the PR1 and 2 in the sunitinib arm. So the duration was longer for these responses and for the CR patients in both arms, they were actually very similar. So both of those, they hadn't been reached the duration of response. So these patients that are achieving these deep responses are doing really well.

Pedro Barata: So you think over time we'll start talking about plateauing of the curves in a couple of years?

Andrea Apolo: Yeah, that's the hope. Yes.

Pedro Barata: Great. So Dr. Apolo, this was great. Thank you so much for taking the time and breaking down the fantastic presentation that you did. And I'm looking forward to get more updates on 9ER from you and your colleagues. And again, congrats for the great job, and this is very helpful for treating physicians. So thank you.

Andrea Apolo: Thank you. Thank you so much for having me, and thanks for allowing me to chat more about this analysis. I appreciate it.

Pedro Barata: Absolutely. Take care.

Andrea Apolo: Bye.