Naiyer Rizvi, MD: More recent data came out for CheckMate-227, which combined nivolumab at a standard dose every 2 weeks with ipilimumab 1 mg/kg every 6 weeks versus chemotherapy. And the I/O [immunotherapy] doublet improved overall survival in the group with PD-L1 [programmed death-ligand 1] expression of more than 1%. We are potentially going to see approval in that setting. That's what the statistics were powered for. Interestingly, there was a benefit seen in the 0% expression group, as well as the all-comer group.

Leora, what do you think? Are we going to do this?

Leora Horn, MD, MSc: It's hard to imagine that this is going to change anyone's current practice. The less than 1% group is kind of interesting. It’s a small subset analysis from the study, but in the greater than 1% group, pembrolizumab despite its toxicities is still easier in my mind than nivo/ipi [nivolumab/ipilimumab]. It’s significantly cheaper, and it’s given less frequently, every 3 weeks versus the every 2 weeks. And so I cannot imagine this suddenly making a big splash in terms of practice-changing management.

Naiyer Rizvi, MD: Jacob, I would like you to explain to me why the hazard ratio was really good in the 0% expression group and really good in the more than 50% group, but it was really not very good in the 1% to 49% group.

Jacob Sands, MD: Well, it's a great question, and I don't have a great answer. I can just say that we do see this consistently across other data sets as well, and I think it goes to the point that there's still a lot to know about PD-L1 expression. And we need a better biomarker because this is something that crosses other data sets as well.

Naiyer Rizvi, MD: Will you use ipilimumab/nivolumab?

Jacob Sands, MD: Not really.

Naiyer Rizvi, MD: Or do you use ipilimumab/nivolumab?

Jacob Sands, MD: I don’t. This isn't any more compelling to me than carbo/pem/pembro [carboplatin/pemetrexed/pembrolizumab], which is generally well tolerated. I'll say that ipilimumab, 1 mg/kg, the data on that are it's far better tolerated than the groups getting the 3, which in the small cell lung cancer groups is the dosing. But even so, the adverse effect profile can be really be limiting.

Naiyer Rizvi, MD: Josh, ipilimumab/nivolumab?

Joshua Bauml, MD: I don't use it right now, and I don't plan to use it based on the data that we have available. I've heard a lot of people who are proponents of it say, “Oh, this is a chemotherapy-free regimen,” but it's important to remember that, again, this is an education problem. If you look at the grade 3 to 5 adverse events with nivolumab and ipilimumab, it’s pretty darn close to what you get with chemotherapy. It's pretty significant toxicity. That being said, I think that if you look at the CheckMate-227 data, there are patients who are benefitting, and so to the point that a lot of us have made today, we need to figure out who those people are. We are really in our infancy of using immunotherapy, and I think that there is likely a group of patients who benefit from the addition of CTLA-4 [cytotoxic T-lymphocyte—associated protein 4] blockade with ipilimumab, and we need to do the science to understand who those people are, so we can give them the appropriate therapy.

I am hopeful that at some point in the future we'll be able to do that, but at this time, based upon the biomarkers that we have available, I do not plan on doing it.

Tim Kruser, MD: And the problem is it seems like such a small subset. When you talk about nivolumab alone versus adding ipilimumab, how much does that really add? It seems like maybe there is a small group that really does get durable benefit when adding that, but it just seems so small and you're exposing everyone to the adverse effect profile. So being able to identify that very small population would be meaningful, but we're just not able to do that right now.

Transcript Edited for Clarity