Suresh S. Ramalingam, MD, FASCO: Byoung Cho, obviously, as we think about combinations of immunotherapy, the safety profile tolerability is an important issue. In the lung cancer experience at least, the IPI [ipilimumab]—NIVO [nivolumab] schedule is very different from the melanoma regimen where they give higher doses of IPI. Can you share your thoughts on tolerability profile with the IPI –NIVO regimen in lung cancer as you’ve seen it?

Byoung Chul Cho, MD, PhD: In general, the combination of a PD-1 [programmed cell death protein 1] and CTLA-4 inhibitor increase the immune-mediated-data adverse event, and it expands the diversity of immune-mediated adverse events in our patient. Toxicity management in terms of receiving an I/O [immune-oncology]—I/O combination is really important. But I think that for patients who do not want cytotoxic chemotherapy and are intolerant to chemotherapy, this combination, the I/O–I/O regimen, may be alternative in the first-line setting.

Pasi A. Jänne, MD, PhD: I agree with you, but I do think that what Ram was saying about the lower dose of IPI that is now used compared with when it was first tested, does make it better tolerated. I remember treating some patients early on when the higher doses were being tested in lung cancer, and there definitely was more toxicity. I think this is a more friendly combination regimen.

Suresh S. Ramalingam, MD, FASCO: Johan, where do you see this fitting into the treatment landscape? Let’s assume that the drug combination will get approved for lung cancer frontline.

Johan F. Vansteenkiste, MD, PhD: Yes. Well, it’s remarkable because it turns out now, at least with this presentation—and there are other things to follow—that again in CheckMate 227 it’s more PD-L1 [programmed death-ligand 1] as a biomarker that comes to the forefront. While in last year in our mind, most of the people had TMB [tumor mutational burden] in the forefront of their thinking. So now, again, PD-L1 expression becomes very important because that is where it is reported now. I think these data are quite good, quite solid, and it means in patients that do have some degree of expression of PD-L1, it is a chemotherapy-free option that you can give to a patient—not necessarily to every patient. You need to tailor treatment according to patients, but certainly I think it becomes a valid option. If you want to give a chemotherapy-free regimen, and you spare the chemotherapy for later lines, the patient has therapy without chemotherapy in the beginning. What was most exciting for me were the plateaus at 1 and 2 years, which are rather unprecedented with chemotherapy.

Suresh S. Ramalingam, MD, FASCO: Adding on to that, I was also impressed with the CheckMate 227 results for the PD-L1—negative patients where we saw superiority in survival for the IPI–NIVO combination versus chemotherapy. I would take a slightly different take on that from what Johan just mentioned. Because to me, if you see survival benefit in PD-L1-high and in PD-L1-low patients, then do we really need PD-L1 at least when it comes to the IPI plus NIVO combination to make treatment decisions? We will have chemotherapy plus immune checkpoint inhibition as 1 strategy, immune checkpoint inhibition alone as monotherapy as 1, and now IPI plus NIVO as another strategy. In the upcoming months we’ll be talking about relative merits of 1 approach versus the other for patients with newly diagnosed non–small cell lung cancer.

As we think about these studies, these are all done in patients with good performance status—fit patients. What do we know about the use of checkpoint inhibitors in patients with poor performance status? Johan, we’ve seen results from the CheckMate 817 study that was reported at the International Association for the Study of Lung Cancer World Conference on Lung Cancer. It was a small cohort, about 100 patients, PS2 [performance status 2] patients.

Johan F. Vansteenkiste, MD, PhD: It was a report of a subgroup, of course, and the subgroup was patients with a PS2, and that was the majority of patients who were special because of certain comorbidities. That was one-third of that subanalysis. What was noted is that the tolerability in that cohort, in terms of needing to stop needing to discontinue the therapy because of adverse events, was rather similar in the subcohort reported there compared with the overall trial. What was not similar was the outcome. The outcome in PS2 patients was lower in terms of PFS [progression-free survival], in terms of response rate, but this is what you can expect with most of the therapies. The exception, perhaps, are the targeted therapies, which are far more PS independent. The overall outcome—and it was mostly a prognostic effect—was lower, but it was tolerable. It can be an option because chemotherapy is an issue for these patients as well.

Suresh S. Ramalingam, MD, FASCO: As we look at what are the next steps when the immune checkpoint inhibitors are used in lung cancer patients, understanding why patients develop progressive disease while on immune checkpoint inhibitors—and what molecular or immunologic mechanisms are in play—are all important for us to understand how best to develop new therapies to overcome the resistance. Pasi, any insights on what we know about resistance to immune checkpoint inhibition?

Pasi A. Jänne, MD, PhD: I think it’s still a little bit of an evolving area. I would say that the 1 area, in terms of a single molecular marker, where there’s been a lot of interest is cancers that have the STK11 or also known as LKB1 loss. This most often associates with KRAS-mutant cancers, so about a third of KRAS-mutant cancers have this concomitantly mutated, although it’s not exclusive to KRAS. But it tends to create an immune cold microenvironment. Data exist, both on single-agent immune checkpoint inhibitors as well as combination with chemotherapy, showing that this is a population that is unlikely to derive benefit from the immune checkpoint inhibitors. That opens up the possibilities of why is that the case, can we learn something from that, and can we make this population more responsive? There is more to come there.

That’s really in the de novo resistance. In the acquired resistance, we have some understanding of how immune escape can happen. Again, there it’s not 1 mechanism, there are multiple mechanisms. It’s not clear that there’s a strategy yet or a clear demonstration of reversing that resistance to once again regain an immunologic response analogous to a second-generation TKI [tyrosine kinase inhibitor] that overcomes a resistance to first generation. I think we’re still missing that piece there.

Transcript Edited for Clarity