Updates in SCLC: Advances in Treatment - Episode 4

SCLC: Future Therapies and Advice to Community Oncologists

,

Drs Carl M. Gay and Jared Weiss highlight several novel strategies that show promise in optimizing treatment for patients with small cell lung cancer and respond to questions from community oncologists who manage treatment for patients with lung cancer.

Carl M. Gay, MD, PhD:There are a number of other regimens that are being tested. I'll plug a couple of these rather than go through them exhaustively. In the DNA damage arena, PARP inhibitors have not borne out quite as well as they looked preclinically, but there is some enthusiasm based around biomarkers there, especially with SLFN11. If any of you have Schwab 1929 trial open at your institution; you may be familiar with that. The targeting TIGIT [T cell immunoreceptor with immunoglobulin and ITIM domain] has gained a little steam because of how effective it's been in non-small cell lung cancer. The phase 2 CITYSCAPE trial data, which got an update last week, shows just this extraordinary benefit of anti-TIGIT plus atezolizumab in patients with PD-L1–high non–small cell lung cancer. I think it leaves people hoping to see some of that. Unfortunately, PD-L1–high small cell lung cancer [SCLC] is an extraordinarily rare entity, so hopefully, there's another biomarker for that population. DLL3 targeting left a bad taste in everybody's mouth after the failure of Rova-T [rovalpituzumab-tesirine], but now it has gotten a second life with Amgen's DLL3 strategies. I don't know if there are other drugs or strategies in development, Jared, that you're especially high on.

Jared Weiss, MD: I think there are some things. Small cell has a few preserved neoantigens, and I'm kind of excited about that as a therapeutic area. I spent a lot of time on personalized vaccination approaches, and I look to tumors that have preserved neoantigens with a little jealousy. In small cell, I agree, DLL3 is a really promising target. The updated data that Taofeek Owonikoko, MD, PhD, [Winship Cancer Institute of Emory University, Atlanta, Georgia] recently reported a response rate of 20%, but when I counted the patients at a higher dose, it looked like it was creeping toward 30%. I think a lot of people in the thoracic field sort of freaked out about the CRS [cytokine release syndrome] in that trial. But I kind of grilled them a little bit about that and it sounds like all events were treated in the infusion center, not in the MICU [medical intensive care unit] or even inpatient, so maybe it's a little less scary. I think that's going to be a winner of an agent. I reserve the right to be disappointed, as small cell research has disappointed all of us before. I don't have that 1 open, but I'm referring people for it. The other preserved neoantigen I'm excited about in small cell is GD2, and I think that that could be a very good cellular therapeutic target.

Carl M. Gay, MD, PhD: I think in general, a shift towards cellular therapeutic agents or antibody-drug conjugate agents, and doing something different other than just trying different cytotoxic therapies in small cell lung cancer is right now a very justifiable way to go. We'll see how those trials pan out over the next few years. We have a few questions and probably need to make sure we've covered those for everybody in our audience.

Jared Weiss, MD: We have a question about sequencing once a patient has progressed and can't receive platinum etoposide, preference of topotecan, lurbi [lurbinectedin], or other agents. I think you were very wise earlier when you said that when the standard therapies aren't very good, you look to something else. So if I have a patient that can't get a platinum doublet, I'm talking about hospice. I'm not pushing it, but I want to at least bring it up as an option and sort of assess a patient’s values. I'd favor 1 of the trials we've talked about, and if I don't have a spot for the patient, I encourage them to consider travel, which is possible for some patients and not possible because of values or resources and other reasons. Between topo [topotecan] or lurbi, again, we don't have comparative data yet, but in my practice, it's near 100% lurbi, with which I'm less scared of hurting people than I am with topotecan.

Carl M. Gay, MD, PhD: That's my practice as well, and obviously there's no right or wrong answer there. Some of it might be recency bias where lurbinectedin is favored and topotecan is something that has failed us for a number of years, whereas lurbinectedin hasn't had the opportunity to break our hearts yet in quite the same way. But it also seems like a more valuable agent to build off of because designing trials where you add things to topotecan—5-day regimens or something like that, it's just so difficult to enroll and design things. I think the enthusiasm about where lurbinectedin might take us in the future is also really valuable, whether it's the combination with immune checkpoint blockade or something that's not obvious already. I think that that's nice. I want to reiterate your plug for getting these patients onto clinical trials. I think there could be a tendency with patients with small cell lung cancer to look at the failed clinical trials that have plagued us for decades and the general patient population, which is not necessarily a robust one, and think perhaps that's an excuse not to push them towards clinical trials. But really, if there was ever a subgroup of thoracic patients that really has nothing to lose by going on a clinical trial—and I mean that in the best possible way. I find it to be a very easy sell to these patients when you give them the data, as it stands for the standards of care; they see it that way as well: Best-case scenario I'm going to live a year. Why not try something different? And I think that's the only way that we're going to move the needle on this.

Jared Weiss, MD: Yeah, I'm with you. My comfort level in recommending trials is higher than for other disease states. Stated much more simply, if tomorrow I got lung cancer, I would go on a trial because the standard of care just is not what we'd like it to be. But talking about that last question of toxicity, the next one is about performance status for systemic therapy.

Carl M. Gay, MD, PhD: I think the idea behind a poor PS [performance status] in a patient with small cell lung cancer and how to manage them is something that I often focus a lot of time on in my lecturing to fellows who are learning to treat small cell lung cancer. And it's a little bit almost paradoxical. These are patients that in general are not going to do well overall, so if it's a patient with poor performance status who is heavily pretreated, I think the answer is fairly obvious to not push them towards more therapy. But if it's a patient with poor performance status who is treatment-naive, to me, that goes all the way in some cases to include patients who are either in the ICU [intensive care unit] or on the brink of going to the ICU. I think as long as you have a frank discussion with them about what the long-term looks like, there's very clear data supporting that there is a symptomatic benefit to treating these patients, even when they have very poor PS. To me, I don't think a PS of 3 is a disqualifier for systemic therapy in a patient with small cell lung cancer. But maybe that's me being too much of a Pollyanna or too optimistic about where we stand.

Jared Weiss, MD: If you're wrong, I'm wrong with you. If the patient's precancer performance status was good and it's been trashed by the cancer, my experience is that if you blast with chemo—and I'm choosing my words intentionally—full-dose treatment, you very often improve that performance status. You get a Lazarus effect, including patients going from a ventilator to walking out of the hospital days after you're done. So my practice is very much the same. As long as it's not lots of comorbidities, as long as it's the cancer trashing the PS, I'll treat any PS with frontline. And my experience, we never really came back to trilaciclib too much, but my experience with retreatment has been a bit more friendly in the trila [trilaciclib] era. I recently had a patient that sort of hit me over the head a little bit, because it was the extreme of something I've observed in milder-case situations. It was a much older patient who in his front line went on one of the trilaciclib trials. Got his platinum etoposide and really had very little adverse effects, very little myelosuppression, very little fatigue. And then got second-line without it. It was in the interim before approval, and everything went downhill. And then I just more recently had the reverse of that. A patient who had a really tough time in front line with radiation then came back to get platinum-etoposide-atezo [atezolizumab] with trila, and the counts stayed really nicely preserved. So the superior supportive care [may vary], whether you're talking about 5-HT3 antagonists and fosaprepitant in the context of platinum and whether you're talking about trilaciclib in terms of myelosuppression and preservation. The improvements in supportive care do shift my opinions on drugs over time and whom I'll treat.

Carl M. Gay, MD, PhD: Experience with trilaciclib has been so limited, it's very useful for me to hear about your experience and that it's been a positive one.

Jared Weiss, MD: To get me really excited about a drug, I really want to see a pop in survival. But when 5-HT3 antagonists came to the fore, everyone wrote them. When it came to the fourth platinum, everyone wrote it. As oncologists, we'd like to see people living longer, of course; our patients would too. But if there's a drug that can improve my patient's quality of life on treatment, preserve their counts, decrease adverse effects, it may be prescribed with yon, but I'm going to prescribe with yon for every patient. And I should patent that, prescribe with yon. The next question is about hyperprogression following radiotherapy. Do you have any thoughts on that?

Carl M. Gay, MD, PhD: I've occasionally seen a patient who I thought might be hyperprogressing, in particular with immunotherapy and small cell lung cancer, and I have on at least 1 occasion dropped the immunotherapy from the front line because I thought that might be what was happening. But it's not something I've experienced a lot of, and I'm not familiar with any systematic data that's delved into it. I'm not sure about you.

Jared Weiss, MD: I think small cell is just a really nasty cancer that grows rapidly unless we do something effectively.

Carl M. Gay, MD, PhD: Just the cancer progressing on its own without the help of anything. Probably.

Jared Weiss, MD: I think that it's natural to ask the question of what's different as we get to new agents, but the whole idea of hyperprogression in small cell or non-small cell, I have some hesitancy about. I think they're just very aggressive cancers that sometimes grow quickly. Maybe time will prove me wrong. So there's a question about maintenance therapy. I might dispense with that quickly; every patient who gets platinum etoposide and PD-L1 should probably get the PD-L1 in maintenance. We don't really know where the benefit comes from; we just know what the trial data shows us, which is that the separation was not early. But we really don't know where it comes from. I keep it simple and keep to the trial regimen, which is to use it. Any thoughts about any one you would exclude from maintenance or do anything other than PD-L1?

Carl M. Gay, MD, PhD: No. I'm with you and a comment you made earlier about having a pretty strict cutoff after 4 cycles of combination chemo-immunotherapy. The only downside I see is extending that to cycles 5 and 6, even though the CASPIAN trial did allow for that. I get them onto maintenance therapy as quickly as I can. I get everybody onto maintenance therapy unless there's something very unusual that's happened.

Jared Weiss, MD: I'm going to push this question to you because it's from one of your fellows. Can you discuss the landscape of management when there's a recurrence after concurrent chemotherapy in limited-stage disease? Does it really just depend on timing and extent of recurrence or anything else?

Carl M. Gay, MD, PhD: I see them on the list. They're not only 1 of my fellows, but in clinic with me later this week. I appreciate the question. I think it does largely, at least in my practice, depend on the timing and the extent of the recurrence. I will say that for patients who have limited-stage disease that have an isolated recurrence, I do occasionally deploy things like consolidative radiation therapy, whereas I don't consider that much in the extensive-stage setting at a time of recurrence. Say somebody has an isolated adrenal metastasis or an isolated bone met [metastasis] after a limited stage disease, I might try to buy us some more time. I'm fooling myself a little bit with that because we know small cells are a systemic disease. We know we can detect viable circulating tumor cells in these patients, even those that appear to have a relatively limited disease. And so you probably do have to consider starting something systemically there, but maybe you don't have to think of yourself as condemning them to a life on systemic therapy. Other than that, I really base it around the duration of benefit. If a patient had a year off of therapy after concurrent chemoradiation, I'll go back to platinum. If they had a month off of therapy before they progressed, I try something different. And as I mentioned before, that's occasionally where I'll try to sneak an ipi- nivo [ipilimumab-nivolumab] past the insurer with a lot of argument and pointing to the little data that we have there. I'm interested to hear your thoughts, Jared, because I think that's a wide-open space.

Jared Weiss, MD: My practice is very similar. I would add how scared I am of the recurrence as a relevant aspect of my thinking in the platinum-sensitive patient.If you give me a patient who is getting 6 months every day for conversation’s sake—or 2 patients. One of them has brutal central chest disease that is really threatening them. In this case, I care about the response rate in PFS [progression-free survival] in and of themselves because the cancer is threatening to harm that patient relatively imminently. Give me another or counter case where I have lots of smaller spots, none pressing on something critical, none causing terrible symptoms, not a high fear factor, if you will, for suffering or death, then that's a patient for whom I might look to a single agent like lurbi, instead of a platinum retreatment, to keep it gentle. Of course, patient values play in heavily. Some people are done, they've had enough; they want hospice. Some people want the most aggressive thing possible. For me, values in terms of how much they influence the end decision beyond the more aggressive, less aggressive spectrum, is also their value on trials. Are they willing to go on a trial at my institution? I have the luxury of having another major academic institution 10 minutes down the road. Believe it or not, in Carolina, some people would rather die than go to Duke [Duke Cancer Center Durham, North Carolina]. I've had that more than once, but if they have a trial that's promised, someone's got to be laughing at that one. If they have a trial that's promising for that patient, I'll recommend that. And if I don't have a trial as promising for them, are they willing to travel elsewhere? I've expressed already in our conversation my excitement about tarlatamab, the DLL3 BiTE [bispecific T-cell engager]. I don't have that open, but it's open in Baltimore Maryland, which is maybe a 2-hour drive from here. But any rate, a reasonable drive and if patients are willing, I'll refer them for that. So the values sometimes play into that—the clinical trial willingness and the travel willingness as well.

I think we've answered all your questions.

I want to thank you for this enriching and informative discussion, Dr Gay. And I'd like to thank our audience as well. Both of us hope that you found this webinar discussion to be engaging and helpful to your understanding of patients with small cell lung cancer. Personally, I hope we've gone a little past the data to give you some color on our practices and how we're thinking about making some of these decisions when there isn't level 1 data to define A or B.

Carl M. Gay, MD, PhD: I'll echo all of that. This has really been a pleasure, Jared. And again, thanks to the audience for keeping us on our toes with some good questions.

Jared Weiss, MD: Good night, everyone.

Transcript edited for clarity.