Mark A. Socinski, MD: I want to transition into another abstract that we’ve gotten data from. In all of these trials that we have all participated in, we have excluded patients who were on higher-dose steroids for whatever reason. That’s a question that comes up in clinical practice—how much steroid is too much steroid, and does it really make a difference? We saw, Ben, a retrospective analysis here. Could you talk us through that?

Benjamin P. Levy, MD: This is a retrospective analysis from 2 very large academic institutions. It looked at 640 patients who received single-agent checkpoint inhibitors. They reviewed the pharmacy records and the clinical records and really identified 2 groups—those who had received steroids of greater than 10 mg of prednisone versus those who had used less than 10 mg or no steroids. Interestingly, roughly 15% of the patients had received greater than 10 mg of steroids. Common reasons for having received steroids included fatigue, dyspnea, and brain metastasis. I think we’ll learn more, but when you adjusted for smoking history, performance status, and brain metastasis, those patients who had received steroids did far worse than those who did not.

I think the devil’s in the details with this trial. We need to learn how long they were on them for. But clearly, this is important. It’s trying, albeit it in a retrospective way, to answer a very clinically meaningful question—do steroids negate the effects of immunotherapy? Based on this data, I think we can say that they perhaps do, but I think we need to learn a lot more about this data and, again, how long they were on it for.

Mark A. Socinski, MD: I think the central question from this abstract is, is it the prednisone or is it the protoplasm?

Benjamin P. Levy, MD: That’s right.

Corey J. Langer, MD, FACP: It’s a retrospective study with all sorts of confounding variables. There is a world of difference between a patient with asymptomatic brain metastasis, minimal vasogenic edema, who gets SRS (stereotactic radiosurgery) and does not require steroids, and one who is still on steroids off trial because of vasogenic edema. I can assure you that the prognosis for that patient with symptomatic brain metastasis is far worse than for the asymptomatic patient.

Benjamin P. Levy, MD: But they adjusted for performance status in brain metastasis.

Mark A. Socinski, MD: Retrospectively.

Corey J. Langer, MD, FACP: I congratulate them on that, but we really need to burrow into the details of these patients, as you’ve indicated, to have some sense of, are the steroids really hurting? Remember that in KEYNOTE-189 or, frankly, in any study that doesn’t use on nab-paclitaxel, steroids are at least a part of the conditioning regimen to prevent toxicity.

Karen Kelly, MD: Hypothetically, it does make sense, right?

Corey J. Langer, MD, FACP: Of course.

Karen Kelly, MD: We know they’re immunosuppressive. I think it’s probably the chronic use of the steroids versus when you give them for a short course with chemotherapy for a couple of days. That’s very different from somebody who has chronic use of steroids, like a COPD patient who has been on 20 mg of prednisone for 3 months or something. That goes on to that protoplasm question that I think you brought up. And I’d like to go on a tangent about antibiotic use, as well.

Remember that you could not go on these trials if you had used an antibiotic within a couple of weeks. And now we’re learning that perhaps that is changing the microbiome, which, again, goes to the host immune system, as well. Another interesting retrospective analysis could be to look at that, as well. I know I’m kind of going on a tangent, but there have been many scientific articles, now in Nature and Science, really promoting the microbiome and its importance in immunotherapy. And trials are going to be looking at fecal transplants to, again, do that. So I know I’m off on a tangent here, but I think it goes to the point that we still have a lot to learn about the host microbiome, what influences the host microbiome, and then how that influences these immunotherapy agents.

Corey J. Langer, MD, FACP: I think the other important point is that this should not be a deterrent to the initiation of steroids in patients who are having toxicity.

Mark A. Socinski, MD: Toxicities, yes.

Karen Kelly, MD: Yes, that’s a different issue.

Corey J. Langer, MD, FACP: All of us have treated folks. They’ve had toxicity. We’ve been able to wean the steroid down to some acceptable level. I usually try to get it to 10 mg, and then I resume the immunotherapy, particularly in individuals who are benefiting. I don’t think we need to stake ourselves to completely eliminate steroids in those individuals.

Mark A. Socinski, MD: Right. This abstract reminds me how I was trained to apply the results of a phase III trial to the population that was studied. These patients were excluded. Your point about the antibiotics is a good one, and I think that’s the take-home message here. Whether it’s the prednisone or protoplasm, I think we can debate. To me, I think it’s more of the protoplasm here. But it reminds us that we can’t broadly apply the results of the trial to populations that were excluded from the initial trials.

Transcript Edited for Clarity