How Do You Choose Among Treatment Regimens for NSCLC?

A roundtable discussion, moderated by Millie Das, MD, covered challenges, advances, and future directions for the diagnosis, treatment, and management of non-small cell lung cancer (NSCLC), as well as critical clinical trial data and updates from the 2024 American Society of Clinical Oncology Annual Meeting. Dr. Das was joined by Ticiana Leal, MD, Martin Dietrich, MD, PhD, and Kent Shih, MD.

In the fourth segment of the roundtable series, the panel discusses treatment options for NSCLC, with a specific focus on combination therapies and regimens for patients who don’t have targetable mutations.

In the next segment of the roundtable series, the panel discusses implications of 5-year data from CheckMate 9LA.

—

Dr. Das: I wanted to just really move on now to the treatment options, because there are a number of treatment options for patients who don’t have targetable mutations, and when we think about first-line treatment, a lot of us have familiarity with some of the regimens, the KEYNOTE regimens, the KEYNOTE-407, KEYNOTE-189, and KEYNOTE-407. These are chemo-[immunotherapy] combinations. And now, we also have data for the CTLA-4/PD-1 combinations with CheckMate 227, CheckMate 9LA, and POSEIDON.

I wanted to talk about the different regimens that we have. How do you choose between these regimens? Are there particular patients where you’re going to be more likely to offer that addition of a CTLA-4 inhibitor? I think this is an area of active investigation and exploration as we’re really trying to understand what is the best treatment for our patients. Of course, the comparator arms for a lot of these trials was just chemo alone, so they weren’t compared head-to-head.

Dr. Leal: I would say in first-line, I mean PD-L1 is still, I think, our best predictive biomarker. So certainly, if patients have PD-L1 high without any of the high-risk mutations like STK11 and KEAP1, I generally favor PD-1/PD-L1 monotherapy. We have approval of 3 different agents that are very well tolerated with excellent quality of life and really good results based on large phase 3 trials.

For patients that are PD-L1 high, but for some reason have high-risk features, maybe they’re highly symptomatic, high disease burden, maybe there’s additional co-mutations that I’m worried about, those patients, I probably would navigate more toward chemoimmunotherapy. Again, if it’s the STK11 and KEAP1, I’m navigating toward using a CTLA-4 containing regimen. For those patients, I would tend to use CheckMate 9LA, but could also use POSEIDON, and we could talk about that as well.

For the PD-L1 low or negative, certainly chemoimmunotherapy, and then for the robust PD-L1-negative population, I generally think about using a CTLA-4 inhibitor combination as well. I tend to use Checkmate 9LA as my preferred regimen there.

In the squamous population, we were talking about how that’s a high-risk population, again, navigating towards the CTLA-4 inhibitor, but also, I think depending on comorbidities or performance status, I may use KEYNOTE-407 in the squamous cell population as well.

There is an interesting negative signal, I would say with the regimens used in POSEIDON, which might’ve compromised the results of that benefit of POSEIDON in the squamous cell subsets in that trial. In the POSEIDON trial, they actually had a lot of patients using gemcitabine-based regimens with the combo, and we didn’t see that benefit of the regimen like we saw in CheckMate 9LA and in CheckMate 227.

Dr. Das: Was that attributable to the gemcitabine? We don’t know.

Dr. Leal: Right.

Dr. Das: Martin, Kent, how do you think about this?

Dr. Dietrich: Well, it’s a difficult question and there are many unanswered questions. First one is, does the patient need chemotherapy or does the patient not need chemotherapy? That’s still sort of a clinical judgment call. I think it’s mainly a short-lived response that chemotherapy adds. I think we’ve moved away of thinking of chemotherapy as an immunosensitizer that contributes to long-term durability. I think that’s more predetermined in the biomarker setup and driven by the respective immunotherapies used. There’s obviously a question about histology and the regimens that we use for treatment.

For me, PD-L1 is the first marker. I agree with you, giving a PD-1 or PD-L1 alone in a PD-L1-negative setting is falling more and more to failure. Three prospective trials now that show a robust benefit in the PD-L1-negatives with the addition of CTLA-4s with 2 agents, both the tremelimumab and ipilimumab. I think that data is pretty practice-influencing for me. I do use them for sure also in patients where I have low expectations of immunotherapy response in the PD-L1 lows, or if there’s a resistance mutation.

I know this is still a controversial field, but STK11 and KEAP1 have really been seen over, and over, and over again to be associated with a lack of PD-1 response. So, if I see these mutations in an eligible patient, I think of CTLA-4s in addition, even when the PD-L1 biomarker is higher. Then as I get closer to 100%—80%, 90%, 100%—I think oftentimes of immunotherapy just as a monotherapy if there’s an old clinical indication.

It’s really not one size fits all. The chemotherapy contribution [is] mainly for clinical assessments, and then [it’s] the question of whether or not a patient needs a combination of immunotherapy, mainly on PD-L1 status in my practice. But there are other considerations as well. I do think that a mutational burden is still in the mix for consideration. [It’s] not really prospectively validated, but I consider this an additional bonus.

I think of smoking status as an independent factor of decision. I would never give—even with a PD-L1 high level—a never-smoker an immunotherapy-alone regimen. I think the responses there have been underwhelming. I think we need the addition of chemotherapy. So, it’s a very, very much a patient-by-patient clinical scenario, a clinical scenario kind of decision, and certainly not one size fits all.

Dr. Leal: What regimen would you use for that never-smoker?

Dr. Dietrich: Well, it depends on the setup. If a patient is a never-smoker and doesn’t have an identifiable actionable mutation, I think typically a carboplatin/pemetrexed-based regimen, plus or minus immunotherapy. I think pembrolizumab or cemiplimab here would be reasonable options. I don’t know that I’ve seen any data that CTLA-4 can overcome the immunotherapy resistance that comes with never-smoking status, so [I would use] one of the PD-1-based chemo combinations.

Dr. Shih: I agree, Martin, Ticiana. I think for our lion’s share of patients, [KEYNOTE-189] is probably a standard of care. The standard of care is fine, but I do think there’s subsets that I consider the addition of CTLA-4. I like CTLA-4 inhibition. I come from, actually, the skin cancer world, so I’m familiar with CTLA-4, [ipilimumab 3 mg], even [ipilimumab 10 mg]. I don’t really have a lot of issues with giving [ipilimumab 1 mg] or giving [tremelimumab] for that matter in our squamous cell patients and our brain [metastases] patients, and probably for STK11 to KEAP as well, [tremelimumab], but probably [ipilimumab] has a role. It probably still works. It’s just a little bit of CTLA-4.

I don’t think everybody needs chemo. When I have a patient that comes in, particularly if they’re PD-L1 high, I ask myself—but even PD-L1 low—is this a patient that’s going to die in 6 weeks without chemo? The last patient I put on [CheckMate] 227, they had a 3-centimeter, very symptomatic brain [metastases] that got resected, but a fairly asymptomatic apical mass, and maybe some very small mediastinal nodes. That patient probably after they’re creating an [stereotactic radiosurgery], doesn’t probably need chemo. That person could probably get away with just [immune-oncology].

I think it is patient-dependent, but I’m a believer. I think we don’t have to probably dose CTLA-4 forever. I mean, the median number of doses in [CheckMate] 9LA was 6 and [CheckMate] 227 was 5. Again, importing from the different disease model that we’ve seen in the skin cancer world, that probably just in fact, NADINA today [showed] if you had a near [complete response (CR)] or CR, the curve was flat. So, you’re probably okay with just 2 or 3 doses. We’re going to probably do what we’re going to do with CTLA-4 after just 4, 5, or 6 doses. The preclinical folks may tell us that we’re expanding a population of memory T-cells or not after several doses. I don’t think we have to dose CTLA-4 indefinitely, but I do think it has a role in some of our diseases.

Dr. Leal: But can we go back though? Because in those trials, the more significant reason for discontinuation was either progression or treatment-related adverse events. I don’t know that data guides me to stop CTLA-4 inhibitors, because there was a reason that that therapy was discontinued, right?

Dr. Shih: In the end, they still got that delta with just a few doses.

Dr. Das: We have POSEIDON where the [tremelimumab] was just given for a total of 5 doses. So even though I think the CheckMate studies, 227 and 9LA, continued the [nivolumab and ipilimumab] as a maintenance therapy, POSEIDON was different in that sense.

That was my next question to you. I mean, how do you decide between POSEIDON and [CheckMate] 9LA? Are you differentiating these regimens based upon the duration of the CTLA-4? And again, they haven’t been compared head-to-head. We do have some patients who absolutely don’t want chemotherapy. I like the idea of 9LA, of limited chemotherapy. POSEIDON, of course, it was 4 cycles of chemotherapy. So how do you decide between?

Dr. Leal: Well, one thing I had reviewed thinking about how can we identify who are the people that are the long-term survivors in non-small cell lung cancer throughout all of these trials: is there any clinical factor, any genomic factor? And the answer is no. But when you go back to the melanoma data where they have people that are surviving 10 years, and you look at all the regimens that led to the people that were 10 years out, they were all [nivolumab and ipilimumab]. They were more likely to be [nivolumab and ipilimumab]. Just talking about it in a simplistic way. And in the melanoma trials, again, the duration of how long to use immune checkpoint inhibition, including CTLA-4, I think is a very important question. But those patients were on long-term therapy too, right?

So, while I think that we still need to define the duration of how long to use immunotherapy in general, including CTLA-4 inhibitor, it’s not clear to me what the optimal duration is. And again, and maybe I’m extrapolating from the melanoma data and how I view CheckMate 227, and 9LA, and POSEIDON, I tend to favor continuing the maintenance and sort of dropping-

Dr. Das: The CTLA-4 as maintenance, yeah?

Dr. Leal: And dropping the chemotherapy through CheckMate 9LA. Whether [tremelimumab]  for 5 [doses] is optimal, I guess we’ll never know, because those regimens, you can’t really compare. But I will say that I tend to favor sort of using CTLA-4 longer term, like the Checkmate 227 and 9LA, hoping that that’s really what’s driving the survival in those patients that are going to be long-term survivors. But it would be nice to have additional data to really guide us in a definitive way.

Dr. Das: It would. But on the same token, I think I have a low threshold to stop the CTLA-4 inhibitor. If I have a patient who’s experiencing an immune-related adverse event, we know those rates are going to be higher in patients who are getting dual inhibition.

Dr. Leal: Yeah, and I think the data’s reassuring for that, right? We saw now with long-term survival, we have 5-year survival for a lot of these regimens, including 5-year survival for CheckMate 9LA, 5-year data in terms of toxicities. And we saw throughout all of the different landmark analysis that over time if patients had to discontinue due to treatment-related adverse events, they still didn’t have a compromise in their outcomes. I think it’s reassuring that if there is a struggle in terms of toxicity or side effects, that you can stop the CTLA-4 inhibitor. And I think that’s reassuring.

Dr. Das: Those patients may actually be doing better, right? There’s data to support that.

View the first segment of this roundtable series here.

View the second segment of this roundtable series here.

View the third segment of this roundtable series here.