Novel Strategies for Intermediate-/Poor-Risk mRCC: Optimizing Therapy

Sumanta Kumar Pal, MD: Brad, I’m going to jump to you next. What are you using for intermediate- and poor-risk disease?

Bradley A. McGregor, MD: We have probably favored nivolumab-ipilimumab for most patients with intermediate- and poor-risk disease. Similar to what has been said, if I look at a patient and they’re extremely symptomatic or I’m worried that if they don’t respond to my frontline therapy, they’re not going to get to second-line therapy, I’m going to choose the regimen that has the higher response rate. That may be pembrolizumab-axitinib, or more recently, I’m very impressed by cabozantinib-nivolumab and lenvatinib-pembrolizumab. It comes down to the situation. If I can afford to be wrong and the patient can have PD [progressive disease] as best response and get to that second-line therapy, I’m going to go with nivolumab-ipilimumab. But if I have a patient and I worry that in 6 to 12 weeks, if their disease progresses, I’m going to have to talk about hospice care because they’re not responding, I’m going to look at a TKI [tyrosine kinase inhibitor]–I/O [immuno-oncology] combination because, as Tom pointed out, the clinical benefit rates are over 90% with both cabozantinib-nivolumab and lenvatinib-pembrolizumab. It’s quite remarkable activity. If I need rapid control of the disease, that’s really appealing.

Sumanta Kumar Pal, MD: Brad, can I ask you more about that? How do you know if that patient is going to do OK or not? How do you pick the patient who you think is going to be all right if you lose out on that 3-month scan? Unfortunately, we’re sometimes a little baffled by that. We have patients we think are going to do OK, and then their disease explodes after you put them on 3 months of I/O–I/O and they don’t have a response. You just think, “Gosh, what if I had started with a TKI?”

Bradley A. McGregor, MD: It’s really challenging. To that end, we’re not perfect, so if I’m the least bit worried, I will often image patients at 6 weeks. If they’re clinically worse or not feeling well at 6 weeks, I’m not going to say, “We need 3 months of nivolumab-ipilimumab to get that benefit.” I will switch quickly. Something we haven’t talked about is the presence or absence of sarcomatoid features. Sarcomatoid and rhabdoid features drive me toward nivolumab-ipilimumab. The CR [complete response] rate is approaching 20% in patients with nivolumab-ipilimumab. I have had patients with large liver metastases that just melt away within 3 cycles. For patients with sarcomatoid features, even those who may be more worried that I can’t afford to be wrong, I prefer nivolumab-ipilimumab.

Sumanta Kumar Pal, MD: That’s very reasonable. Dave, do you have any other thoughts on this topic?

David F. McDermott, MD: The last few comments have been very well said. Hopefully some of the early end points that we’ve looked at in the VEGF TKI data will translate into later and durable end points, so that the deep responses you’re seeing with VEGF PD-1 will allow you to live not only longer, which we would expect, but also free of progression and, ideally, off treatment. If the VEGF in those combinations is enhancing the immune response to the tumor, then you’ll see that. You’ll be able to stop treatment, and the benefit will last. If it’s just additive, you won’t see that, and when you stop 1 or both of the drugs, patients will progress. The reason that is going to be harder to show in some of the VEGF PD-1 trials is that, while the PD-1 may be limited by time—they stopped it at 2 years in the pembrolizumab trials—they didn’t require you to stop the VEGF. For example, on the axitinib-pembrolizumab data, there are a good chunk of patients out at 2 years still benefiting, but many of them are still on treatment. I’d like to see what happens to them when they stop. Hopefully, we’ll be able to see that as well because that will tell us that blockade of VEGF signaling enhanced the immune response to the tumor and added to PD-1, which is what we’ve been trying to show from the get-go with these combinations. We’ll see.

Sumanta Kumar Pal, MD: Can you give us any hints on that, Dave? I’ve treated more than a handful of patients on these VEGF–I/O combination trials and, unfortunately, if I get durable responses, if I need to stop the VEGF because of adverse effects like diarrhea or proteinuria, I’m usually losing response at that point. Are you seeing lots of sustained response?

David F. McDermott, MD: No, we’re not. But my experience is less than many others. We certainly haven’t done the right trial. The right trial would have required, or at least explored, stopping in deep responders. The 1 piece of both the axitinib-pembrolizumab data set and the lenvatinib-pembrolizumab data set that makes me a little nervous about what we’re going to see when we get the 3- and 4-year updates, which Tom was alluding to, is where the OS [overall survival] hazard ratios are going in both of those trials. The clear data, as published in the New England Journal of Medicine, are great. But if you look at the OS hazard ratio at around 1 year, it was 0.47. If you look at it at 27 months, it’s around 0.66. I would have hoped that that deep PFS [progression-free survival] benefit that you’re seeing with clear would have translated into a stable OS. You may see that, but I worry because, ideally, in a true I/O therapy, its OS should be better than PFS and should improve or remain stable with time. I hope we see that, but those PFS curves keep drifting downward with axitinib-pembrolizumab. I’m a little anxious, but I’m happy to be proven wrong. If Dr Pal comes up in 3 years, shows the data, and it looks awesome, I will bow down.

Sumanta Kumar Pal, MD: Dave, thank you for saying that, because I put a graphic on Twitter, and I share your thoughts in terms of what we might see down the line with that data. Tom, do you want to forecast for us? I’m getting a little ahead of myself in our agenda, but you have had a lot of experience with lenvatinib-pembrolizumab. Do you think we’re going to see the OS curves pulled? Is the hazard ratio going to remain as robust over time?

Thomas E. Hutson, DO, PharmD: I wish I knew. If I did, I wouldn’t be an oncologist anymore. I would always cherish the ability to have someone like Dave McDermott to prove him wrong, which doesn’t happen most of the time. Lenvatinib-pembrolizumab is best positioned of the therapies we see so far to reach that if it’s going to. Cabozantinib-nivolumab would probably share the same fate as lenvatinib-pembrolizumab when it comes to that. It would be hard for me to see that there’d be that much difference between them when it comes to the tail of the curve. We’ll see. No one knows.

TRANSCRIPT EDITED FOR CLARITY