PD-1 Inhibitor Use in Advanced Lung Cancer

Transcript:Mark A. Socinski, MD: Let’s turn to some of the clinical data, and I want to start with my good friend, Jared, over here at the other end and just make an initial comment. And that is, the year 2015 in lung cancer brought us to PD-1 inhibitors, both nivolumab and pembrolizumab. I would like for you to give us your perspective on those agents and where they fit in lung cancer at this point.

Jared M. Weiss, MD: It was a very good year for thoracic oncology in 2015. So, to bring this to a clinical level, in the clinic, our most common standard second-line therapy—prior to these advances that you described—was docetaxel, perhaps with the addition of ramucirumab as well, but it was a cytotoxic agent. And I think clinical docetaxel was a challenging agent. It’s one of the most toxic agents that we use in thoracic oncology, and efficacy is somewhat limited. And so, while it was king of the hill, perhaps it was a king of the hill that we were all looking to dethrone at some point.

We now have two agents that have randomized clinical trials going head-to-head against docetaxel in the second-line setting. So, the CheckMate-017 study was a randomized study of docetaxel, the standard of care, versus the PD-1 antibody nivolumab. And here we saw an improvement in median overall survival from 6 to 9.2 months…

Mark A. Socinski, MD: In squamous.

Jared M. Weiss, MD: In squamous histology, an area where we really have had a dearth of progress in lung cancer. Most of the targeted therapy advances that have really revolutionized the care for some of our patients have been pretty restricted to the adenocarcinoma histology. And so, yes, there was a huge thirst for advances, a huge unmet need for these patients. And that’s a very large survival advantage. Now, a natural clinical question, from a historic perspective in lung cancer would be to ask, “Okay, but what was the extra toxicity?”, and historically over the decades the experience has been that as we layer on to more and more intense approaches, the toxicity goes up.

There were two things remarkable here. One is that, in terms of the toxicity profile, the newer, better drug was actually quite a bit less toxic. We saw a lot of zeroes and ones (0s and 1s) in the grade 3, 4 toxicity column in a pattern that we’re not used to seeing. Now, they’re not risk-free drugs; there are still risks of toxicity. Of course, they’re mostly autoimmune since these are immune-stimulating drugs. But when compared on a simple level to the alternative, I think it’s fair to say that they’re a whole lot less scary.

The other thing that’s remarkable, from a human perspective, was the tail to the curve. So, patients don’t come to us asking for an extra month of survival. Of course, they’ll take it over not having it, who wouldn’t? But what patients really want is a chance at being alive, counting many months or years down the road. And in that trial, the 18-month survival was improved from 13% with docetaxel to 28% with nivolumab. And so from a human perspective, this is what our patients are asking for, this is what our patients care about, so we better care, too. This was a very, very meaningful improvement.

Now, the second trial that’s important in this realm was the non-squamous trial of the same agent. This was nivolumab versus docetaxel in the second line—but this time, for non-squamous patients. This is the CheckMate-057 trial. Their median survival was improved from 9.4 to 12.2 months, and again, the tail to the curve was improved. One-year survival went from 39% to 51%.

A very similar trial has been done with pembrolizumab. In this case, patients were not selected by histology; they were instead selected by PD-L1 status. They had to be PD-L1-positive, but both squamous and non-squamous patients were allowed on this trial. And, again, survival was improved from 8.2 to 14.9 months. So, we have less toxic drugs, or a less toxic class of drugs, with better median survivals and better tails to the curve. This is a major advance for our patients.

Mark A. Socinski, MD: So, it tastes great and is less filling, right? I know, I agree. I think the impressive aspect of those is obviously the efficacy. But I was actually impressed in the phase III setting: a large number of patients, how well tolerated and how infrequent the immune-related adverse events were with these agents once you got to the phase III. I think it’s obviously a win-win.

Jared M. Weiss, MD: One more point might be worth making about the nivolumab trials. These trials were unselected by PD-L1 status, but they did give us the subgroup data later of a level of staining that was present in tumor cells. And it was remarkable that there was no class of PD-L1 status that did not benefit. So, across the staining levels, patients did better with nivolumab.

Mark A. Socinski, MD: But now, pembrolizumab is associated with a companion diagnostic, whereas nivolumab has a complementary diagnostic.

Jared M. Weiss, MD: That’s right. What this looks like, in practice in the office, is that in the second line, you can give nivolumab without knowing the PD-L1 status. And I think that’s pretty fair based on the phase III data that existed. The negative predictive value wasn’t there to deny anyone the PD-1 therapy. In contrast, pembrolizumab requires at least 1% staining with the 22C3 antibody. And there you might ask, well, why would you ever bother? And the practical answer is that pembrolizumab is a 3-week drug where nivolumab is an every-2-week drug. So, you used to be my boss down South, and you know that our average travel distance was hours. For that kind of population, an every-3-week drug can be more convenient.

Transcript Edited for Clarity