Jeffrey S. Weber, MD, PhD: Jason, what’s your opinion on the use of IDO inhibition? There have been at least 2 or 3 trials, small phase I and II trials, of either indoximod or epacadostat, both IDO inhibitors. There are data suggesting that IDO inhibition is an immunosuppressive or that it’s potentially a way to overcome immunosuppression, independent of checkpoint protein inhibition. What do we know about those phase II trials? Now I guess they’re in phase III?

Jason J. Luke, MD: Yes. To build on things that other people have said, the model that we’re building about how to develop these combinations in a more rational fashion is really taking the idea of whether or not the immune cells are present or not and trying to harness that knowledge into deciding which combinations to give. Because if there is a robust microenvironment that has immune cells in it, several of these approaches of hitting multiple checkpoints in the tumor microenvironment are likely to be affected. And so, you mentioned indoleamine dioxygenase, or IDO, which is another putative checkpoint. It’s a metabolomic checkpoint in combination with PD-1 where, at previous meetings, we’ve seen that the response rate has been elevated over 55% with minimal, if any, increase in toxicity, relative to the monotherapy PD-1. So, I, in fact, think that is actually the most exciting combination that’s coming forward.

You referenced how it has gone into subsequent trials. We participated in that phase I trial, and it went from a phase I trial to a phase III trial—accrued—very, very rapidly. And we’re now basically waiting for the results. I think that if the results look anything like what we saw in the preliminary data set, it will be a very attractive combination, with an increase in response rate. If those responses are like what we’ve seen with pembrolizumab at this meeting, they will likely be durable, and if there’s no increase in toxicity, that would be quite attractive. I think that’s in the hot tumors. LAG-3 might also go with hot tumors because it’s on T cells, and there are others that go with that. But then the flip side was just discussed here: those patients do not have immune cells in their tumor. We’ve got to get them there somehow. The best way to get them there is innate immunity, and it could be a virus, it could be a toll-like receptor agonist, or it could be a STING agonist.

Jeffrey S. Weber, MD, PhD: And there are at least 3 or 4 trials, I gather, of either pembrolizumab, nivolumab, or atezolizumab with a TLR9 or a TLR4 agonist. So, do you see these as making a cold tumor hot?

Jason J. Luke, MD: Absolutely. But I think the broad concept is bringing in the innate immunity. Will oncolytic virus be better than TLRs, be better than STING agonism? We don’t really know those answers because they’re too preliminary. I think the data set for some oncolytic viruses is the furthest clinically developed, but the biology and the framework that I’m talking about, I think can really guide these clinical trials to be more rapidly successful as opposed to just treating all patients.

Jeffrey S. Weber, MD, PhD: Okay, but here’s the development conundrum. You have an active drug, pembrolizumab, with a 42% response rate and a 33% median survival. You’re going to add drug X to it. Drug X has no independent antitumor activity, so you’re forced to do a randomized trial. In general, I assume, in a randomized trial, it’s the number of events that has to occur to see a difference between arm A and arm B. But won’t we be waiting for years with these IDO trials, Georgina?

Georgina Long, MD, PhD: Well, it depends on the endpoint, which goes back to my point about the progression-free survival versus the overall survival. So, if you’re talking about events being progression events, yes, we may have to wait a little longer, but I don’t think years. If we’re talking about survival events, yes, we’ll be waiting a long time.

Jeffrey S. Weber, MD, PhD: Will the FDA, the Australian authorities, or the EMEA allow progression-free survival to be used as an endpoint for registration?

Georgina Long, MD, PhD: I don’t know, but I think we’re really going to have to change the paradigm as we get more sophisticated in terms of understanding the biology and then translating that directly and seeing exactly what we hypothesized. This last decade has seen that time and time again, even in lung cancer, in ALK inhibitors and the immune therapies. It’s amazing that we had a hypothesis, we do it, and wow, it worked. So, I think we have to change our attitude. But I would propose that we need to consider the landmark progression-free survival. We’re in it for cure. I think melanoma is different than solid tumors. and maybe the whole cancer field of solid tumors is shifting because it has been a desert for us, and all of a sudden, we have these active treatments.

As cancer physicians, we’re a bit different. We’re like hematologists. We’re going for cure, and so what we’re after is this tail and getting that tail higher. And I think the best way to measure that, the quickest way at this point in time, would be landmark progression-free survival. So, yes, we’ll have to wait a little bit longer, but not that long. I think overall survival is hard and we should get on with it and do these trials. The trick is choosing the right combinations and not wasting time, effort, and money in clinical trials.

Jeffrey S. Weber, MD, PhD: Because patient resources are limited, we now have so many trials, we don’t have enough patients to put on them.

Transcript Edited for Clarity