Transcript:Robert A. Figlin, MD: Let’s turn our attention to some novel approaches. David, I want to turn back to you. We’re now in the frontline setting, we’ve taken checkpoint inhibitors into the frontline setting. We are asking the question that we have asked over the last decade, Tom leading much of that effort, which was on combination strategies. We have lenvatinib/everolimus, the only combination strategy that’s really led to a remarkable demonstration of efficacy in a population of patients. And now we have ipilimumab/nivolumab, axitinib and a checkpoint inhibitor, and we have bevacizumab and a checkpoint inhibitor. So, talk a little bit about the science of why, and where, we are in that population of patients looking at moving checkpoint inhibitors closer to the frontline setting than in the second-line setting.

David F. McDermott, MD: We’ve talked a lot about the exciting parts of the checkpoint story. We talked about improvement in survival, we talked about tolerability. I think the tolerability piece allows us to imagine combinations. The fact that you can give someone a PD-1 or a PD-L1 drug for 1, 2, 3 years every 2 or 3 weeks suggests that this could be a platform to build upon, to hopefully improve the tail of the curve that you were talking about. I think it’s also potentially, from a biologic point of view, a platform, which is when a T cell recognizes a tumor. Most tumors defend themselves with PD-L1. So, blocking that interaction is one step in getting a response. But for many tumors, it’s not the only reason they avoid an immune response. There’s so much in many tumors, there may be three or four things that you need to block to get that tumor to die with immune therapy.

Probably the most exciting combination so far, the only combination for which we really have a large phase III trial experience, is the combination of CTLA4 blockade with ipilimumab and PD-1 blockade with nivolumab—which is now FDA-approved for melanoma. There you see two distinct mechanisms, two checkpoints, CTLA4 and PD-1, but they work in different parts of the immune response. CTLA4 earlier in the activating step probably globally activates far more T-cells. We’re still learning how it works; it may actually be able to deplete regulatory cells from the tumor microenvironment. So, that has one mechanism, and PD-1 we’ve talked a lot about already. Putting those two together, what do you see? Well, in all the cancers we’ve tested so far, we’ve seen an increased efficacy.

In many tumors, including kidney cancer, we’ve seen over a doubling of response rates. In the phase I trial where we looked at ipilimumab and nivolumab, we saw response rates north of 40%, which is great. We saw other benefits. We also see dramatic responses, deep responses with shrinkages, in some patients, of 50%, 60%, even 80% or 100%, which is exciting for us because they often translate into those tail-of-the-curve benefits. The but in that story is the toxicity. So, you see increased activity, but you could almost argue synergistic toxicity, with the combination, toxicity that is not different than single-agent. It’s just more severe and it can last longer, and it requires a much more rapid response to deal with it.

So, there’s a trade-off, but we’re now working at different doses and schedules that may be more amenable to use in the average patient. The dosing schedule that’s being explored in the phase III trial that you mentioned in kidney cancer is different than the FDA-approved regimen in melanoma. Hopefully, it will be less toxic, because I think our kidney cancer patients are less likely to tolerate severe toxicity. They’re older, they’re sicker. We need to come up with better strategies. But so far, that seems to have raised the bar in melanoma, although it’s in the early days.

The other approaches are building on PD-1 pathway blockade, mainly involved combining blocking VEGF and PD-1. There’s certainly a preclinical rationale for doing that. VEGF is thought to have many immunosuppressive effects on dendritic-cell function, and T-regulatory numbers, and a variety of things. Preclinical models suggest it’s a good idea, but so far, all we have right now is single-agent data with several different combinations. We have atezolizumab, which is a PD-L1 drug, and bevacizumab, a VEGF antibody—interesting single-agent, combined-agent, single-arm activity. It’s now in phase III testing in kidney cancer. We also have VEGF TKI blockade with….

Robert A. Figlin, MD: Axitinib.

David F. McDermott, MD: Axitinib, thank you very much. And pembrolizumab, and a separate study with axitinib and avelumab, both of which are moving into phase III trials. All the data there from the single-arm study is also interesting, exciting. You’re seeing high response rates, a large percentage of patients with both response and stable disease, but very small numbers, very early on. So, maybe in kidney cancer there’s a rationale. Because it’s such a VEGF-driven disease that these combinations could be particularly effective, moving both approaches that we’ve been talking so much about into the frontline setting. In the end though, these combinations, in my mind, have to improve the tail of the curve, have to improve that long-term survival that we’re all looking for and that our patients are all looking for. Because, if they don’t, then many of us will say, “Well, why don’t we just give them in some sequence as opposed to…because they’re all going to be an increase in side effects.” So, the phase III trials are enrolling. We’ll get the results over the next few years. It could be an exciting time. We had this long conversation about the options now. They could be completely different 2 years from now.

Robert A. Figlin, MD: So, Elizabeth, we’ve moved from a conversation about fourth- and third-line therapy to a conversation about using checkpoint inhibitors in the frontline setting. And, clearly, our colleagues in the community see many patients who are found to have resected disease with what is otherwise called high-risk resected disease in the absence of metastases. And I’m sure they’re wondering what the status of the world is in relationship to what we have to offer those patients. Currently, in 2016, it’s put a person on a trial, but maybe there’s something around the corner. Thoughts about that?

Elizabeth R. Plimack, MD, MS: I think it didn’t take us long to take those VEGF TKIs from the metastatic setting and move them into the adjuvant setting. The issue with the adjuvant, it takes many years for those trials to read out. So, we just saw the ASSURE data last year, which was unfortunately negative. We have many more trials that are going to read out, but we all think the results are likely to be similar. It, of course, makes sense now that we have active IO agents to think about using those in the neoadjuvant and perioperative setting.

I think one thing to keep in mind is the biology of the disease, which is that you need antigen and tumor for that interaction to happen. Whether that can still happen with micrometastatic disease in the resected setting or whether you need to give these IO agents a little bit before and a little bit after, we don’t know. Fortunately, there are two large phase III trials that are upcoming, hopefully, in the next few months. One is a pure adjuvant study of atezolizumab in resected high-risk renal cell carcinoma and another is more of a perioperative approach of nivolumab, giving a little bit before and then the rest after resection. So, I think we’ll look to those trials to see answers. Unfortunately, it will take a while.

Transcript Edited for Clarity