Read the Full Video Transcript

Pedro Barata: Hi there, and welcome. My name is Pedro Barata. I'm a GU medical oncologist at Seidman Cancer Center, Case Western University. Today I'm very, very happy to be joined by Dr. Martin Voss. Dr. Voss is the Clinical Director of GU Service at Memorial Sloan Kettering. He's a guru in kidney cancer. First of all, I should welcome him today. So welcome, Dr. Voss. Thanks for joining us.

Martin Voss: Thank you, Pedro. Dr. Barata, thank you for having me, and it's my pleasure to be discussing our findings here today on behalf of my co-investigators.

Pedro Barata: Absolutely. First of all, congrats. Very good presentation during KCA. We've seen very important data that you presented on subsequent therapies for patients with advanced RCC enrolled in the CLEAR study investigating lenvatinib-pembro versus sunitinib. So I guess this is very important data, and I think we've been talking recently in this endpoint which has to do with this progression-free survival 2, PFS 2. Maybe for those of us who are less familiar with this concept, can you explain a little bit what that endpoint is, and what's the significance of that endpoint in advanced RCC?

Martin Voss: Yes, of course. This is an abstract that we recently presented, and we have discussed this data also at ASCO before that gets to progression-free survival 2, PFS 2, which is a somewhat novel endpoint applied to clinical trial data sets to try to look beyond the scope of the primary analysis itself. The definition of PFS 2 is the time from randomization to disease progression on the subsequent line of therapy. So a patient goes on therapy, is randomized on study, in this case the CLEAR trial, and now information is not only collected on how the patient's fare on the study treatment itself, but after progression on study treatment data is collected on subsequent therapies.

Here now we look at the time from randomization to progression on that subsequent therapy or death, whichever occurs first and can analyze that data in a fashion very similar to what we do for PFS, OS analyses. We can do Kaplan-Meier calculations, log-rank regressions to compare the treatment arms, and it provides somewhat of an impression whether the benefit one were to see on the investigational regimen carries forward beyond treatment exposure itself.

I do think it's relevant, to your question, in kidney cancer specifically because we have many data sets from large randomized trials that apply immuno-oncology strategies. Of course, the question is whether or not the effect of such therapy carries forward beyond time on treatment, and PFS 2 is one of the ways to look at that.

Pedro Barata: Got it. No, that's very, very helpful. Thank you. Maybe I should ask you ... I mean, I've looked at the data ... It looks very interesting favoring the combo of lenv-pem. Can you summarize perhaps for those less familiar with that data what you guys found?

Martin Voss: Of course. I mean, first of all, we applied this strategy now on the CLEAR trial, and to all who are not instantly familiar with the dataset, this was an open-label, randomized Phase III study. It was a three-arm trial that compared in the first line, three different treatment strategies for metastatic clear cell kidney cancer patients.

The standard arm on the trial was sunitinib, and the two comparator arms were lenvatinib plus pembrolizumab and then lenvatinib plus everolimus. Lenvatinib plus pembrolizumab, as you mentioned, based on this pivotal trial has received FDA approval. Has been a standard regimen now for some time, and the aim of this analysis here was to do a PFS 2 comparison between that approved regimen, lenvatinib- pembrolizumab and the then standard sunitinib and to see what would happen beyond the primary efficacy analysis.

One of the thoughts that immediately comes to mind of course is that we apply immuno-oncology combination therapy successfully here in the first line. As a reminder, this study had a PFS primary endpoint and strongly favored the investigational regimen with a hazard ratio that now with extended follow up is actually south of 0.4. So very strong signal. But to play devil's advocate, you could say, "Well maybe the patients randomized to sunitinib would later 'catch up' if they are given opportunity to receive checkpoint inhibitor therapy in the second line." So a very simple question that goes into this PFS 2 analysis is does it matter to receive the combination of IO- TKI upfront or is that benefit blurred once you give PD-1 inhibitor subsequently to the TKI treated patients? In other words, does upfront combination equate to sequential therapy of a TKI plus IO or is there truly something to it being a synergism additive effect when patients will get exposed early on?

To our findings, we compared ... The lenvatinib plus everolimus regimen was out of this analysis, so we compared lenvatinib-pembrolizumab to sunitinib, and on this trial about 350 patients went onto each of these arms. So 350:350 were technically eligible but we had to account for only those who were recorded to have received subsequent therapy. That was in this case 117 under lenvatinib-pembrolizumab arm and 206 on the sunitinib arm. So 33% of investigational arm and 57, 58% on the comparator arm had documented second line therapy with information on such treatment in the dataset.

Then we could look at things like the time to receipt of second line therapy. We could look at duration on second line therapy and so forth. But ultimately the primary aim of this analysis was to compare PFS 2, which strongly favored the lenvatinib-pembrolizumab combination. In fact, median PFS 2 with median follow up of about 27 months in this dataset, the median PFS 2 on the investigational arm was not reached, and the median PFS 2 on the sunitinib arm was about 29 months.

The hazard ratio for comparing PFS 2 between the two treatments arm was 0.5. So in other words there was a 50% decrease in the risk of progression on second-line therapy or death when you compare lenvatinib-pembrolizumab to sunitinib. It sort of gives further support to the notion that upfront combination therapy has a lasting effect for patients to the extent that we can track this prospectively in a trial like this.

Pedro Barata: Got it. No, that's awesome. Yeah, indeed, it does support this combination and establishes that or this data goes in line with the data we saw previously supporting the good effect of lenv-pem lasting even beyond progression. So thank you. That's really, really helpful.

One of the things I also find it very informative if you will, is actually the breakdown you offer regarding subsequent therapies. We always want to know what did patients received upon progression. I think that's super, super important, and I really like the fact that you actually present that information because we don't quite often see that. Its always welcome and now you see VEGF therapy, checkpoint inhibition, mTOR inhibition, even CTLA-4. I find it interesting the different rates of exposure that the patients who progress on lenv-pem compared to patients who progress on sunitinib got exposed to. Maybe I'll ask you what are your thoughts and insights about what actually patients end up getting upon progression and whether or not the fact that a lot of patients were exposed to the life-prolonging therapies as we know it would reinforce the message that actually we should not save the best therapies for later, we should do it upfront. I'll let you comment on that.

Martin Voss: Yeah, thank you. So it's a super important question, and thanks for pointing to those details. It gets to be very difficult to draw comparisons when it comes to the less commonly applied drugs, the smaller categories. Yes, we report on mTOR inhibitors and CTLA-4 inhibitors, but we can't really easily compare or interpret what effect they may have had big picture.

But the more relevant consideration is looking at the two most commonly applied strategies in its subsequent lines and that of course is TKI therapy and PD-1 based checkpoint inhibition. It was very important when we undertook this analysis to ascertain that patients on the control arm that were treated with sunitinib on CLEAR actually had apparent access to checkpoint inhibition in the second and onward line. Indeed, that is exactly what we saw to be the case. In that breakdown that you mentioned, we showed that the vast majority of patients who first received sunitinib on the trial were subsequently treated with a PD-1, PD-L1 inhibitor.

We see that most of the patients treated with the combination of lenvatinib-pembrolizumab on the study subsequently received therapy that was TKI based. There were some patients with checkpoint inhibition also in later lines, but the most commonly applied strategy in those patients was TKI therapy. So you are right. We are able to at least get a sense of what is the comparison of sequential treatment with TKI-IO combination on trial, mostly TKI-based therapy after versus TKI monotherapy on trial followed by heavy exposure to PD-1 based checkpoint inhibition. I do agree with you that that allows us to ask the question, should we be treating a patient with combination upfront or is there the notion that patients who as long as they receive checkpoint inhibitor therapy at some point or later do more or less evenly well?

We saw here very strongly that there seemed to be a benefit in this modern regimen when given upfront despite broad exposure to PD-1 inhibitor on the comparator. Once they have progressed on sunitinib, the PFS 2 which now integrates first and second line, strongly favored starting with lenvatinib-pembrolizumab for those patients treated on CLEAR. There's lots of limitations to this analysis, of course. We do not have information for all the patients, but the signal and the hypothesis that comes from it is very strong.

Pedro Barata: Got it. That's fantastic. Before I let you go, one final question. Since you are heavily involved in this trial and very important data set, should we expect additional analysis? Is there something you can share with us? If you can't, we understand.

Martin Voss: No, I do think ... For one, we want to write this analysis up as a manuscript. I do think it's important to make these numbers available to the broader audience beyond presentations at abstracts. Then I would make the point that PFS 2 as an endpoint, and we started our conversation with that, has somewhat newly emerged. Five years ago no one was really talking much about it. I think there is merit to it, there's limitations to it, but it definitely adds to understanding of these novel agents. Considering how many combination therapies are now approved in the first line, I find it very helpful to think beyond just the data set itself, and kudos to companies for starting to collect more data like this and for helping us look into these things.

This is not the only dataset that has been analyzed and reported in this fashion. When we first presented this data at ASCO earlier this year, it was presented side-to-side with KEYNOTE-426 PFS 2 analysis. Tom Powles, in the same session presented data from that randomized trial doing a very similar comparison. It is interesting for us then to look through all the different regimens and see what efficacy signal they project beyond the lifetime of treatment on trial.

Pedro Barata: Yeah, no, that's fantastic points. Dr. Voss, this was great. Thank you. Congratulations again. Fantastic work. CLEAR is definitely important data set out there and is driving what we're doing for patients. So with that, thank you so much for taking the time and being able to sit down with us, and congratulations again.

Martin Voss: Thanks Dr. Barata, I really appreciate it. My pleasure.