Safety of PD-1 Inhibitors in Squamous NSCLC

Transcript:Benjamin Levy, MD: Just a few more questions related to checkpoint inhibitors. One is, is there a patient which you would not offer this drug to, and who is he or she? Is it a patient with an autoimmune disorder? Is it a patient who’s on steroids for other reasons? Is there a clinical scenario where this does not become the de facto second-line regimen for a squamous cell patient? Ed?

Edward S. Kim, MD, FACP: I think that’s who you mentioned. Generally, we’re guided by the clinical trial and what exclusions there were. We will learn over time how exclusive they were, if they were overexclusive or not. Even the steroid aspect, we talk about physiologic doses versus non-[physiologic doses]. You want to avoid people who are on chronic steroids, as well as those who have autoimmune disorders, and the never-smokers I think should be equally in there. That was not an exclusion criteria, but I think it should be in there as well.

Paul K. Paik, MD: I think that’s absolutely right. I think the other thing to call attention to is that while overall it was much better tolerated, and they are much better tolerated than docetaxel, there is that small single-digit frequency of grade greater than or equal to 3 autoimmune events, which can be quite severe and can come out of nowhere if you’re not observant of these things early on. The other thing is that we need to be diligent providers, as well as the patients need to help us with this to call anything that might be brewing so we really can keep an eye on it.

Benjamin Levy, MD: Yeah, the AE profile of these drugs, I think, has forced us to work with our other colleagues from other disciplines. Even though the pneumonitis, and the colitis, the hepatitis, the endocrinopathies, those AEs are on the low end but they’re there. Have you guys seen these? And if so, how do you manage those? Are you employing the expertise of a pulmonologist or a gastroenterologist to help you manage them on your own? How are you doing this?

Paul K. Paik, MD: We have. We were, I’m sure as all of you were, involved in the early phase trials for this. And we’ve seen not a high frequency, but still a number of cases of pneumonitis, colitis, hepatitis, encephalitis, any -itis that you can imagine. And it can be difficult because the reflex is to start high-dose steroids and stop the drug. But what do you do if that doesn’t work and then you’re relying on your colleagues in hepatology, also in rheumatology, for adding anti-TNF—alpha agents on top of that for T-cell suppressors, like cyclosporine? Then you are in the weeds in terms of what you’re doing and in terms of the long-term consequences of that degree of immunosuppression. So it’s one of those things where most people will not get to that point. But if there’s a sense that your patient is getting to that point, get all that help through specialists and consultants to help you try to manage them to get the patients over it.

Edward S. Kim, MD, FACP: I think one of the biggest struggles is actually more on a frontline. If your patient is seeing you in clinic, then you can refer them to the specialist. They’re going to go the emergency room or the urgent care clinic after hours, and they’re going to get Imodium and that is a problem. What we’ve done is we have posted immunotherapy side effect—management pathways up, and we’ve also educated the patient strictly to say, “Hey, before you touch me, I am on an immunotherapy agent.” That triggers that provider—whether it’s an ACP that’s seeing them, whether it’s a clinician—to look at that pathway and then follow that pathway; the colitis pathway, the pneumonitis pathway, the LFT pathway, any hepatitis, uveitis, all of those things we have listed on an electronic format.

Benjamin Levy, MD: We do the same thing; we educate. We have a packet that we give to patients, and a card, so that they’re aware and their other care providers are aware that they’re on an immunotherapy drug. I agree. A patient walks into the ER and says, “I’m on some chemotherapy, and I have diarrhea.” They may be giving Imodium not realizing this is a colitis. Although these AEs are a little bit on the lower end, they do happen and we have seen them. So I think that’s an important point. The other thing that I want to get to with the immunotherapy drugs is this concept of pseudoprogression and how this is now factored into immune-related response criteria.

I won’t get too granular, but immune-related response criteria was a criteria put forth by Jedd Wolchok in melanoma a few years back, and we tried to carry it over. It does factor in the potential for pseudoprogression and not calling these patients’ potentially progressive disease. Have you guys seen pseudoprogression? How do you manage this? Do you confirm it with a scan 4 to 6 weeks later? Is this something relevant? If you look at the melanoma data, their rate of pseudoprogression is on the low end, 5%. Experiences?

Edward S. Kim, MD, FACP: It goes back to what I tell patients no matter what therapy I’m giving them. You’ll commonly see scans where tumors look a little bigger than from 8 weeks, 6 weeks ago, and the radiologist has read it as definite growth because it is. They’re black-and-white, and they see it as bigger or smaller. It’s important. And what I tell them, if the scans look bad and you’re looking bad, we’ve got to do something different here. If the scans are bad and you’re looking good, there’s a disconnection here. And the exact opposite is also true. If scans look good and they look bad, something’s going on, and if the scans look good and they look good, that’s the best-case scenario. It’s that sort of gradient between the best-case scenario and the worst-case scenario. I think we have to not take patients off quickly due to a scan report when they look great in front of you. They might say, “I’m tolerating the drug, I didn’t have any side effects, I went skiing last weekend,” or you take them and continue them on a therapy just because of the scan. So modified immune RECIST is just clinical judgment characterized in writing.

Paul K. Paik, MD: That’s right.

Benjamin Levy, MD: And it speaks again to what you’ve spoken about before, what you alluded to before, which is personalized and individualized and getting as much information you can from the patient. I agree. We have had plenty of patients in which we’ve seen some growth, and maybe it’s not even pseudoprogression and they’re doing great and there’s weight gain and there’s pain relief. I’ve kept them on. And I think we have to have that sort of individualized approach. Paul, your experience with pseudoprogression?

Paul K. Paik, MD: I agree with all of this. Part of the problem with pseudoprogression is we invoke it, patients know about it. The question is, how far do you follow this through? Do you wait a little bit longer, wait a little bit longer, wait a little bit longer? One of the things that I’ve seen, at least for my patients who have pseudoprogression that could help, is that the kinetics sometimes are different. You have the natural history before you start, and for some of the patients who had pseudoprogression, the kinetics just changed. The growth rate seems to be faster in a relatively short period of time than what we saw in the natural history. This can help tease apart whether or not that might be pseudoprogression, because it would be unusual for the natural history just to change from starting a new therapeutic in the bad direction. That can be challenging, though, because these patients can also be sicker, so it’s not like you can continue the drug because they’re getting sicker. It just means you have to wait it out and see what happens on a subsequent scan.

And there are some patients again who have had this change in kinetics who, on a subsequent scan, we get them through it. It’s a nice response they have seen. But it’s exactly as Ed had said; because of that, it really needs to be personalized for any given patient. It’s going to be a low frequency, but you can’t take it away prematurely and you can’t keep them on it for too long and risk not giving them something that’s going to be effective.

Edward S. Kim, MD, FACP: One of the things I do for my patients—and I’m bringing the cup into the photo, by the way.

Benjamin Levy, MD: We love visuals.

Edward S. Kim, MD, FACP: is that I’ll hold something like this up to them and I’ll say, “Well, if this is where they measured your tumor and then the next time the cut was through here [points lower on cup], that’s a response. If they cut it here [higher on cup] the next time, that’s progression. We don’t weigh tumors, and that’s the way we would know if something is actually growing or not. Just two-dimensional measurements is not really good.

Benjamin Levy, MD: Yeah, RECIST is maybe not the best way to understand whether a patient is truly garnering a benefit or not.

Transcript Edited for Clarity