Jeffrey S. Weber, MD, PhD: At this meeting, we’re now beginning to see new combinations, meaning the combinations of a checkpoint inhibitor plus drug X have matured to the point where we actually can see some data. And I think Paolo Ascierto from Naples had a pretty nice poster discussion on LAG-3 inhibitors plus nivolumab. So, Georgina, is there anything encouraging there? LAG-3 inhibition, preclinically, seems very active, but again, that’s in a mouse. How does it do in a patient?

Georgina Long, MD, PhD: First of all, LAG-3 is a checkpoint for T cells. It’s usually upregulated quite late in an exhausted T cell. That’s the biology of it, and so in that study, they included patients who progressed on anti—PD-1. They had 2 types: patients who had primary resistance and a few patients who had acquired resistance. I do want to point out that primary resistance is clinically different than acquired resistance—possibly not biologically different in terms of mechanisms, but clinically, it is different. And what they saw were some responses in patients who had progressed on anti–PD-1 when they added LAG-3 inhibition. What’s more, what was really nice about the work was that they had a biomarker, which was LAG-3 immunohistochemistry, and they saw that if there was higher expression of LAG-3, they got more responses in those patients.

The interesting thing to me is, regarding the acquired-resistance patients, how did they measure response? Because often, patients with acquired resistance will have just 1 or 2 lesions responding to anti—PD-1. So, what were the target lesions? What were they measuring? Were they measuring the growing resistance of target lesions to report those responses? Nevertheless, by adding in LAG-3 inhibition, they got responses. I think they’re early data and small numbers of patients, but exciting in terms of that there may be other options. We do have to remember, though, that it’s still focusing on the checkpoints of the T cells, and if you don’t have T cells in the tumor in the first place, we may have to come up with other ways of getting an immune response against a tumor.

Jeffrey S. Weber, MD, PhD: That’s true, and they did find that with LAG-3 expression, peripheral blood did seem to be associated with the modest number of patients who benefited. I thought those were very interesting data. As we’re getting toward the end of our discussion, let’s think about the future. I think at 1 count there were over 100 combination melanoma immunotherapy trials, and there are 800 combination immunotherapy trials you can find on cancer.gov that patients can look up and try to get on to. So, it’s a huge, burgeoning industry. Let’s start with you, Robert. You tell me, what do you think look like the most promising combinations? Usually, it’s a PD-1 backbone. I don’t think ipilimumab is as popular a drug. It’s always the bad guy. But most of these are PD-1, PD-L1 in combination, or sometimes ipilimumab/nivolumab/drug X. What do you think are the most promising combinations?

Robert H.I. Andtbacka, MD, CM: What I’m going to do here, actually, Jeff—I know we’ll have comments on non-intralesional therapy—so, let me actually take that from the intralesional therapy perspective and give my perspective on it, if you don’t mind. I think that, if we were to look at intralesional therapies in the preclinical models that we have in mice, we often find that it’s the anti-CTLA4 therapy that works better with the oncolytic immunotherapies, not necessarily the anti—PD-1.

Now, whether or not that’s because it’s a mouse model and it may not be reflected in humans, we don’t know. Having said that, we also have presented data here with the coxsackievirus A21 for patients in combination with ipilimumab. And there we’ve seen some dramatic responses in patients with a large amount of disease burden—responses over 60%. So, I think that we still should not discredit ipilimumab in this setting. There are certain patients who may benefit from getting these oncolytic immunotherapies in combination with the ipilimumab.

Having said that, we also presented data with the PD-1 inhibitor combination, where the response rate seems to be over 50%, as well. Additionally, I think where we’re moving with these oncolytic immunotherapies—and some of them, we’re actually administering intravenously, especially the viruses that have a receptor on the cell surface that allows them to gain access into the cell…so, the coxsackievirus, for instance—we are seeing some good responses with that, both as monotherapy but also in combination with the PD-1 inhibitors, specifically in lung cancer and bladder cancer, but also, more recently, we’re doing it in melanoma.

I think that these oncolytic immunotherapies, moving forward, will really have several roles. First of all, they’re monotherapy in patients with earlier disease. They’re in combination therapy for patients with more advanced disease, where we try to get a better response. And I think thirdly, they’re important in patients who are not responding to some of the current available checkpoint inhibitors, to try to change the tumor microenvironment so that we can then get a better response, or re-response, to those checkpoint inhibitors.

Transcript Edited for Clarity