Jack West, MD: One of the issues with pembrolizumab monotherapy, or in maybe even nivolumab/ipilimumab, is that you still have chemotherapy available. When you give chemoimmunotherapy, we’re right now consolidating what has been 2 lines of therapy into 1. What are we left with after chemoimmunotherapy, which is going to be the prevailing standard? I would say it really seems to be docetaxel with or without ramucirumab by default. Any other thoughts on this?

Gregory J. Riely, MD, PhD: Yes. You know, I think it was really interesting how we went through a time where every second-line critical trial was a trial comparing with docetaxel. And then quickly, that was supplanted with every second-line trial compared with immunotherapy. But now we’re back to docetaxel as the standard of care.

Jack West, MD: Can’t get rid of it.

Gregory J. Riely, MD, PhD: With or without ramucirumab is the only question that we’re asking. And I think that’s a question that we can debate for hours at a time. But I think that docetaxel is probably the new standard in the second line.

Jack West, MD: Did that factor in for any of your thoughts about, you know, how much does it matter that you’ve saved chemotherapy if you start with nivolumab/ipilimumab or pembrolizumab?

Gregory J. Riely, MD, PhD: I think that shouldn’t really play into thinking. I think it’s a legitimate question, but the implied, and therefore, preserve your drug doesn’t hold. I think we’ve all accepted over time, and I’ve heard no major challenges to this, the idea of use your best treatment first—whatever you believe that to be—because if you don’t, you may not get a second chance of coming back…

Jack West, MD: Well, again, I think that when you raised that point earlier about that, there’s a difference, if it’s likely, in a negative PD-L1 that chemotherapy is your leading light, they’re still going to almost always be fit enough for doing immunotherapy. I’m less confident about saying that about platinum-doublet chemotherapy if they decline. And so, I think there may be a real benefit. I agree with you, there’s no obvious synergy here, but we have seen some suggestions that there may be a benefit to hitting from 2 angles and getting 2 different populations inhibited in the first line.

Let’s turn to another presentation at ASCO that was from your institution, by Dr. Arbour and colleagues, that looked at retrospective work from Memorial Sloan Kettering Cancer Center and from Institut Gustave Roussy in France: That showed a really consistent theme between the 2 centers that patients who were on daily chronic steroids at a dose of prednisone 10 mg or its equivalent had lower response rates, progression-free survival, and overall survival. The question is, first of all, what does this mean and is this a direct effect of the steroids inhibiting immunotherapy? These were all patients who had received checkpoint inhibitors. So, is it the steroids? Are they doing worse because they needed steroids?

Gregory J. Riely, MD, PhD: I think clearly this is a retrospective effort, and I can highlight the flaws probably as well as anybody. One of the key flaws is that people who you’re giving steroids to are those people with poor performance status, those people with brain metastases. It’s a long list of things that are negative prognostic factors. And so, you’ve identified them through giving them steroids. You take people with poor prognosis and we see that they actually do worse. But I think it is a little bit more clear than that because we did a lot multivariate work. We normalized for performance status, we normalized for a lot of other things, and it still shows up. And it’s important to think, this isn’t just some sort of funny finding, this is mechanism based. We know that steroids do affect the efficacy of immunotherapy, so it makes sense. But I think the way I use these data is I’m a big user of steroids. I give steroids to a lot of patients.

Jack West, MD: Just to look at a patient.

Gregory J. Riely, MD, PhD: That’s right. So, my bar has always been very low. And at a new visit, I would oftentimes hand out the prescription for prednisone. And I’ve stepped back from that for people who don’t have an absolute need. So, patients with brain metastases who need steroids, they get steroids. But patients who are just kind of a little bit more tired and their appetite’s not so good, I hold off on that prednisone prescription now.

Charu Aggarwal, MD, MPH: I will just point out one thing. It’s that these data should not be interpreted on how to manage patients when they develop immune-related toxicities. And we should not be shy about giving steroids to somebody who may have colitis or who may have pneumonitis. And I worry that the interpretation of this data may be that steroids with immunotherapy are back, preexisting steroids for a patient coming into immunotherapy may portend to poorer response rate. It doesn’t say anything about management of toxicities as they appear when a patient is on immunotherapy. So, I think that’s a very clear distinction, and nobody should take away the point saying you should not be giving steroids to patients on immunotherapy.

Jack West, MD: Nor that this applies to the patient getting premedication for a couple of doses before paclitaxel or around pemetrexed. Because those trials were very, very positive and there’s no hint that that’s really a relevant issue. This is a different situation.

One other question that comes up is that we have a significant minority of patients who have autoimmune disorders. I think we would want to make the distinction between somebody who has very significant ulcerative colitis and somebody who has more minor but still arguably relevant autoimmune diseases. Some may or may not require treatment.

There was a publication just in the Journal of Clinical Oncology, by Leonardi and colleagues, that’s actually looking at several institutions that suggest that, yes, you do see some worsening potentially of the autoimmune disease, you can have side effects, and some will need steroids, maybe one-quarter of the patients. But the response rate is still over 20%. Jared, what’s your view about the role of immunotherapies in patients with an autoimmune disorder?

Jared Weiss, MD: I think that it depends what the autoimmune contraindication is. For example, a patient with an allograph, the PD-1 pathway is the rejection pathway. You have near a 100% chance of graph rejection if you give that patient a PD-1 or PD-L1 agent, so that’s a “Thou shalt not.” But I think these data were practice changing. Of these patients with autoimmune conditions who got drugs, I think it was 23% who had an exacerbation of their existing autoimmune condition. It was a higher number than typical. I want to say it was in the 40% range—help me out if you remember the number—who had a new immune-related adverse event, so a little higher than what we otherwise see. But the take-home point to me is that relative to the risk of the lung cancer and the extent to which the PD-1 inhibitor was helping there, including in this population, it was worth it.

And so, I think that this provides us some cover and some comfort in treating a broader range of patients. Based on the rate of exacerbation of the existing autoimmune condition and new events, I think we have to monitor these patients more carefully, be prepared to intervene at a low threshold if they get into trouble. But it opens up the potential for benefit to a broader population.

Gregory J. Riely, MD, PhD: I’ll say one thing that I have to admit is a little bit of a frustration, that this retrospective study had to be done. These drugs have been under study for almost 10 years now. And while I can totally understand the exclusion of patients with rheumatologic disorders from the initial phase I, maybe even the initial phase II, to not enroll all these into trials is really an example of how we too often narrow the inclusion/exclusion criteria for our trials and make it so it doesn’t apply to a significant number of our patients.

Charu Aggarwal, MD, MPH: But it also speaks to the general practice. Because if you look at that study, there were only about 50 or 53 patients across several institutions who are being treated. So, it has seeped into the clinical practice. Automatically you’re checking off patients with these diseases getting these drugs.

Jack West, MD: Let’s go back to the group of patients with the driver mutation EGFR ALK. I think we all agree that nothing we’ve been talking about suggests that we should not be doing up-front testing. If anything, we’re every bit as committed to doing that and doing it up front. What would you say is the importance of IMpower150? Is that the regimen of choice for these people? Are you not convinced that immunotherapy should be integrated with chemotherapy in this setting? So, where would you say we are now for these folks?

Gregory J. Riely, MD, PhD: I think we’re very much in need of data. I think that it’s nice that IMpower150 included these patients. And certainly, I think that takes me a little bit further to feeling that I’m comfortable doing it. It doesn’t take me further to believing that there’s a clear benefit to giving immunotherapy plus chemotherapy for these patients. I think that…

Jack West, MD: Or that that’s the regimen.

Gregory J. Riely, MD, PhD: Right, or that that’s the particular regimen. I think we need trials specifically in this patient population. I think we can do trials specifically in this patient population. I look forward to those results.

Charu Aggarwal, MD, MPH: It highlights the importance of using antiangiogenic agents. I think in patients with EGFR mutation, it’s known from different trials that adding bevacizumab in the second-line setting can be associated with an overall survival advantage. And I think it needs to be studied a little bit more. But in no way can we say that this quadruplet is the second-line therapy for these patients.

Transcript Edited for Clarity