Emerging Landscape of Immunotherapy in NSCLC

Transcript:Mark A. Socinski, MD: The field of immune-oncology is an exciting one, with multiple breakthrough therapies advancing for lung cancer. So far we have one PD-1 inhibitor, nivolumab, approved, and pembrolizumab is not far behind [editor's note: since filming, pembrolizumab has gained approval]. But there are still a lot of questions remaining about the optimal way to use these agents. Let’s discuss what we know so far. Naiyer, I’m going to have you take the lead on this and give us a summary of the immunotherapies.

Naiyer Rizvi, MD: We first started to use nivolumab a number of years ago, back in the phase I trial, and we certainly had little expectation that there would be this magnitude of benefit. However, when the first patients with lung cancer were treated in 2008 and activity was seen, that’s when the revolution began in terms of immunotherapy for lung cancer.

In terms of my view of the landscape of immunotherapy for lung cancer, in terms of science, it makes sense. These tumors, in particular those related to tobacco carcinogen exposure, are like melanoma in that they are highly mutated tumors, which leads to T-cell recognition of the tumors and, ultimately, T-cell exhaustion. These drugs do what they’re meant to do, which is to turn that recognition back on so the T-cells can do their job.

From the phase I to the phase II data there has been very consistent activity in about 20% of patients. We finally have the first positive trial in lung cancer, CheckMate-017, that met all of its endpoints and showed a survival benefit in pretreated squamous cell lung cancer patients being randomized to nivolumab every two weeks versus docetaxel every three weeks. There was a survival advantage for nivolumab, and the survival advantage extended to patients whether they were PD-L1—positive or –negative, which is really a whole other discussion we’ll get into, I’m sure.

Mark A. Socinski, MD: For many of us, and I’m sure it’s true for many community oncologists, our understanding of immunology is relatively rudimentary. Roy, I’m going to ask you, as a professor of medicine at Yale, if you were to get invited to a first grade class somewhere in New Haven to talk to youngsters about immunotherapy, how would you explain the mechanism of action?

Roy S. Herbst, MD, PhD: For the first grade, I’m not going to talk about signal 1, signal 2, or MHC. Maybe I’ll leave that alone. What I would do, and what I actually tell my patients, is that you have a tumor cell and you have the immune system, and you want the immune system to kill the tumor cell—but that doesn’t happen. And why doesn’t that happen? Because the tumor cell is making a blocker, and when your immune system tries to kill the tumor cell, it gets repelled.

What we’re doing with our therapy is that we have drugs (and we hope that someday you’ll go to medical school and work with us), and these drugs can take away that block and then we can actually have the immune cells kill the tumor. That occurs in some, but not all patients. It’s a therapy that really is helping people who weren’t helped before.

I can give them hope: many patients are benefitting from this. Also, I wouldn’t tell a patient this, but I’ll tell you: many people are not benefitting. You and I have been doing this long enough to see that this is clearly a breakthrough, that 20% of people who wouldn’t be alive in six months are alive at three or four years. Now we have to figure out what to do for the other 80%; the good news is we have these great new therapies that can relieve that immune block.

Mark A. Socinski, MD: Is there only one blocker?

Roy S. Herbst, MD, PhD: Oh, no, that’s a gross oversimplification. Actually what we’re finding in some of the research that we’re doing is that there are other inhibitors, as well. In those patients who are not susceptible to the PD-L1 or PD-1 blockers, there are probably other PD-L3 or PD-L4 blockers, sometimes they’re known as B7H3 and B7H4. It’s also possible that we can block these agents as much as you want, but if the tumor doesn’t have T-cells, immune cells in it, it’s not going to make any difference. You have to inflame the tumor.

We’re beginning to look at different classes of tumors, and the nice thing is that through our patients, we can do reverse translation. We can learn from the patients and their biopsies, who responds and who doesn’t. We can take some of that information back to the laboratory where we can develop animal models. And this is a lot of what we’re doing in some of the research, actually, where many of us are working together on projects to figure this out.

Mark A. Socinski, MD: I must say, having been in lung cancer research for a very long time, we’ve never seen a hazard ratio for overall survival like we saw in CheckMate-017 in the squamous population, of 0.59.

Naiyer Rizvi, MD: And it wasn’t chemo plus something. It was something.

Mark A. Socinski, MD: No, it was something new.

Naiyer Rizvi, MD: Something else.

Mark A. Socinski, MD: Compared to our old friend, docetaxel, that we’ve used for more than a decade.

Roy S. Herbst, MD, PhD: Right. I think that was great! Hazard ratio might not be the best way to look at this. What I would say is look at that tail of the curve in those patients who are alive at one year. We have to figure out how to identify those patients early on because it does look like there are many who are not benefitting from nivolumab, as well. If we could figure out who those are, they might need to receive combinations of therapies and other sorts of cocktails.

Mark A. Socinski, MD: Right! The response rate with nivolumab was doubled; it went from 9% to 20%. But if you looked at the median duration of response, I don’t think it had been reached in the squamous population. That’s the tail on the curve.

Roy S. Herbst, MD, PhD: Right. In patients who respond, they respond for a really long time. Even for those who don’t respond, don’t you think, Naiyer, that they benefit? So you don’t have to have a documented response to benefit. Many of those with stable disease benefit. Of course, it’s hard to make a case for those who are progressing through therapy, but even some of those patients might have some benefit.

Naiyer Rizvi, MD: I didn’t like his analogy by the way.

Mark A. Socinski, MD: Well, let’s have yours.

Naiyer Rizvi, MD: Remember the movie, The Matrix, and you’ve got those things that are going around scanning for the spaceship and you have to have the shield up? I see what’s happening is that the immune system is always scanning because it’s scanning for viruses, bacteria, or anything that’s foreign. Any DNA sequence that’s different, it’s scanning. This is like The Matrix where the tumor has this protective shield up, and so what these drugs do is power down that shield so the immune system can recognize the cancers.

Mark A. Socinski, MD: It’s the same thing on the old Star Trek.

Geoffrey R. Oxnard, MD: I use a Star Wars analogy. I use that the tumor is saying to the immune system, “These are not the drugs you’re looking for, move on by.”

Roy S. Herbst, MD, PhD: They’re all good ways. I think you might have scared some of the kids with yours.

Naiyer Rizvi, MD: Well, I have boys, so you know.

Mark A. Socinski, MD: I’m going to come back to nivolumab in a moment, but there are other agents; pembrolizumab, atezolizumab, the AstraZeneca/MedImmune compound.

Geoffrey R. Oxnard, MD: Durvalumab.

Mark A. Socinski, MD: Yes, yes. Tell us a little bit about the other agents.

Naiyer Rizvi, MD: Well, I think what makes it even more exciting is that there are other compounds that block the PD-1 pathway. PD-1 is the ligand on T cells. PD-L1 is what’s expressed by tumor cells. So the PD-1 agents such as pembrolizumab and nivolumab bind to the T-cell side, whereas durvalumab and atezolizumab bind to the PD-L1 side.

The bottom line is that they also show very comparable levels of activity. For atezolizumab, there’s very nice data in the POPLAR trial showing a very powerful hazard ratio in those patients that had a high level of PD-L1 expression within the tumor.

Transcript Edited for Clarity

I don’t know that we can necessarily distinguish these compounds in isolation, but as we think about the strategies that are being used to combine therapies, based on a company’s portfolio, about what their individual science is, collaborations with us are differentiating the companies a little bit. There will be a lot of data coming out in the very near future. Every ASCO is really jam-packed with data. I think that how we use these drugs in combination, the line of therapy, and dependence on PD-L1 expression will differentiate them.