Suresh S. Ramalingam, MD, FASCO: With the maturation of data from some of the phase III trials that we’ve seen over the years in the immune checkpoint inhibition landscape, we’re beginning to see long-term survival data. Pasi, we saw the data from the CheckMate 017 and CheckMate 057 trials with NIVO [nivolumab]. Talk us through those data and what they mean for patients.

Pasi A. Jänne, MD, PhD: I think all of us have now seen patients who are living longer than what we would potentially have expected a priori when they’re getting single-agent or combination checkpoint inhibition. The trials there gave us some real numbers as to what they are in a trial setting. In the NIVO-versus-docetaxel trial, 5-year survival was looked at, so 13% versus 3% 5-year survival. That’s a 4-fold improvement in 5-year survival, which is really a significant number. In the single-agent PEMBRO [pembrolizumab] versus chemotherapy trial, at 3 years, it was 44% versus 24%, so that, too, is a fairly significant number. These are really meaningful numbers. These are numbers that we can also tell our patients about. If people ask, “Well, what is the likelihood of being alive at different landmark points?” I think we’re happy to be able to say they’re better than what they were, and immunotherapy is clearly contributing to that.

Suresh S. Ramalingam, MD, FASCO: How long do we treat these patients with immune checkpoint inhibition? What is your practice?

Pasi A. Jänne, MD, PhD: Yeah, I think that is another area where we don’t have clarity. How long do you need to treat someone who’s really had a response and has had a great response for multiple months? Do you need to treat them infinitum, or do you stop at some point? Oftentimes, 2 years has been kind of the point at which stopping happens. Sometimes you run into toxicities that start to build up between the first and second years, and you have to weigh that into consideration as well when you’re treating individuals. How good has the response been? What are the adverse effects, etc?

I don’t think we have clear guidance on that about what the minimum amount is. Again, if you go back to some of those early-phase clinical trials of single-agent I/O [immuno-oncology], there are patients who had only very little in terms of treatment, had to stop because of toxicity, and had tremendous clinical benefit. It’s certainly possible that limited treatment can still lead to benefit, but I just don’t think we have a clear idea of that.

Byoung Chul Cho, MD, PhD: Adding to that, a recent CheckMate study suggested that a 2-year treatment is better than a 1-year treatment without a significant increase of immune-mediated adverse events. If I had to choose between a 1-year treatment or a 2-year treatment among my patients, I would choose a 2-year treatment.

Johan F. Vansteenkiste, MD, PhD: Sure. The best evidence is exploratory. It comes from the pembrolizumab trials because they were designed to give the immunotherapy for 2 years, while the other trials gave it until progression or even beyond progression, so you can’t say much. But pembrolizumab was stopped after 2 years, and in the second-line study in patients who had pembrolizumab for 2 years and then they stopped, about one-third of them relapsed from year 2 to year 3. So that’s quite more than it is in metastatic melanoma.

Even for lung cancer, stopping after 2 years is debatable. On the other hand, in the recent update of the KEYNOTE-024, where pembrolizumab was also given single agent for 24 months, it was reported recently by Dr Martin Reck that in year 2 to year 3, at 3 years 97% of the patients were still doing fine. So it’s not the same as in the second-line setting. There’s no certainty at all. You can use the 2-year barrier, and actually we don’t do it at the present time. We try to prolong the intervals to 6 weeks, but we continue.

Pasi A. Jänne, MD, PhD: I think it’s certainly an area where we need more research.

Suresh S. Ramalingam, MD, FASCO: Right. In non—small cell lung cancer, to wrap up this part of the conversation, we have highly effective options for frontline therapy for patients without a targetable mutation and immune checkpoint inhibition alone or with chemo. Now, more recently with the PD-1 [programmed cell death protein 1] and CTLA4 combination, we’re able to see durable benefits. Still, we’re benefiting only a subset of patients. There’s still more work to be done in developing novel approaches to broaden the subset of patients who benefit from immune checkpoint inhibition.

Transcript Edited for Clarity