Jeffrey S. Weber, MD, PhD: Let’s say you’re a European who can travel or has access to an outstanding center like the ones you work with, what do you like best? Reinhard, what would you put a patient who had failed ipilimumab and nivolumab, who was BRAF wild-type, who absolutely doesn’t want chemotherapy, and for whom you can’t use IL-2 on? What trials do you think look good?

Reinhard G. Dummer, MD: I think the new trial combining anti—LAG-3 antibodies and nivolumab is a very promising one. Actually, we are participating in this trial, and it’s really a surprising result that you can include patients who have failed nivolumab, our most potent immunotherapeutic drug. And if you add it on another antibody, you see disease stabilizations and quite a significant number of responses. This is one observation, but the more interesting observation is that you can even enrich it by applying a biomarker, by staining for LAG-3. This is a step in the direction of what I would call precision immunotherapy. We will have so many different drugs in the future, and we need a panel of biomarkers that helps us to use them.

The other good news from this trial is that the toxicity is mild. There is actually not more toxicity than with anti—PD-1 monotherapy, so it is a very promising approach. This goes also in line with the data that we have with IDO inhibitors. Sometimes, you think if a drug has no adverse events, it will not really have a benefit, and IDO inhibitors are extremely well tolerated. Even if you see data from 1 company, you might be critical. But now, with IDO inhibitors, we have 2 different molecules, both of them in combination with an anti–PD-1, both of them with excellent tolerability, and both of them showing, consistently, response rates that I would assume are higher than those expected with PD-1 monotherapy. It’s very possible that this combination will substitute the ipilimumab/nivolumab combination that has much more toxicity in many, if not all, patients with advanced disease. So, these are really promising developments.

Jeffrey S. Weber, MD, PhD: It gets complicated though, because the response rates for indoximod, pembrolizumab, or any of the other inhibitors—like epacadostat plus pembrolizumab or epacadostat plus nivolumab—appear to be in the 55% range, the same as ipilimumab/nivolumab. But do we know anything about the duration of response or progression-free survival? Or, are we going to have to wait for the phase III studies?

Reinhard G. Dummer, MD: What we have seen on the response rate and response duration is encouraging, but, in the end, we have to wait for the randomized trials. Some of them have already recruited, so it will be a few months or years until we see the results, and other ones will start very soon. We will have a lot of data on these combinations.

Jeffrey S. Weber, MD, PhD: I guess in the pembrolizumab/epacadostat trial, the endpoint is PFS, or the primary endpoint is PFS. That should read out, I would hope, soon, because I think everyone likes that combination. Everybody likes that trial, and I would agree that people might avoid the toxic ipilimumab/nivolumab combination, although we’ll see what happens with yet another trial. Isn’t there a trial, Michael, of low-dose ipilimumab/nivolumab versus high-dose ipilimumab/nivolumab?

Michael A. Postow, MD: Yes. We’re trying to improve the ipilimumab and nivolumab combination. We all acknowledge that the current dosing, 3 mg/kg of ipilimumab plus 1 mg/kg of nivolumab, has a high rate of grade 3 or 4 toxicity. And so, ongoing studies are lowering the dose of ipilimumab, and there’s a randomized study that’s being done, 3 mg/kg of nivolumab plus 1 mg/kg of ipilimumab versus 3 mg/kg of ipilimumab plus 1 mg/kg of nivolumab. So, that will give us an idea, if we lower the ipilimumab dose, that maybe it will be more tolerable. The other study that we’re doing at Memorial Sloan Kettering Cancer Center is where we’re giving patients 2 doses of standard ipilimumab and nivolumab, and, if they’re already starting to have a good antitumor effect, dropping the ipilimumab and continuing with nivolumab maintenance. This is with the idea that hopefully, you don’t need to continue with the combination dosing to an arbitrary 4 doses, that maybe we should be dosing patients dependent upon their antitumor responses.

I think as we’re trying to improve the ipilimumab and nivolumab combination, that’s a prototype of how we build combinations in PD-1 backbone immune therapy approaches. I think we’re all excited about IDO, we’re all excited about LAG-3, and hopefully we’ll have better combinations that we don’t have to necessarily tweak to get better tolerability. They’re already going to be pretty tolerable as they are. I think the question is—we’ve been very impressed with the response rates—what the long-term survival is going to be, particularly in a randomized context.

Jeffrey S. Weber, MD, PhD: Nobody wants to use ipilimumab anymore. It would seem as if there are now new frontline treatments, say pembrolizumab/epacadostat, that look promising. When patients, as they invariably will, fail that regimen, then what do you do? Are there any other trials being developed?

Caroline Robert, MD, PhD: I think there is also room for evaluating epacadostat plus ipilimumab in anti-PD-1 further, because we had some previous results and then we didn’t hear about it. But it was not that bad, the ipilimumab plus epacadostat.

Jeffrey S. Weber, MD, PhD: Actually, I presented those data at ESMO in 2014 or 2015. It’s interesting.

Caroline Robert, MD, PhD: We would like to know the follow-up, yes.

Jeffrey S. Weber, MD, PhD: It will come out soon, and it did—like the Chesney abstract with T-VEC/ipilimumab—pretty much double the response rate.

Axel Hauschild, MD, PhD: I’m not convinced that ipilimumab is dead, and I’m not convinced that ipilimumab plus nivolumab—I call it high-dose nivolumab—is dead, because it could well be that the 3 mg/kg ipilimumab is needed for long duration of responses. We don’t know it yet, because all the other data have an observation period that is simply too short to convince us that the quality of the individual response is the same. You might come up with lower response rates, but once the responses are better, it’s a very individual decision if patients want to take more toxicity or less. But they have less opportunity to have a long-lasting response, and maybe we are in a fortunate situation where we should focus on patients with complete responses—count the number of complete responses. In the end, if this is in the range of 20%, we are happy. If it’s 30%, we are even happier. So, I’m not convinced.

Caroline Robert, MD, PhD: I would like to mention a trial that we had initiated some months ago, with Aurélien Marabelle at Gustave Roussy. It’s a multicenter trial in France where we compare intra-tumor injection of ipilimumab plus systemic nivolumab versus the 2 drugs given like as usual. So, we don’t know yet. The trial is running, but right now it’s quite encouraging, and the rationale is that you deplete your regulatory T cells locally and you might do something, even priming locally, which is a little bit contraintuitive but quite interesting.

Jeffrey S. Weber, MD, PhD: There are drugs that are being developed to make existing drugs better, such as a masked ipilimumab that would only be active in the tumor, which would reduce the systemic toxicity. There’s a genetic variant of IL-2. There’s a pegylated IL-2, and I’ve heard some very nice responses were seen with it in combination with nivolumab in kidney and melanoma cancer. So, it sounds like there are a lot of really interesting trials.

Transcript Edited for Clarity