Heather Wakelee, MD, shares updates from the CheckMate 9LA study, and Vamsidhar Velcheti, MD, reviews key data from the phase 3 POSEIDON study for patients with NSCLC.
John V. Heymach, MD, PhD: Before we heard that 1 of the issues in the CheckMate 227 study is that there’s a population of patients who progress early, although a large group gets durable benefit. In light of that, Heather, tell us about the CheckMate 9LA regimen and the updates we have.
Heather Wakelee, MD: We have a couple of updates, 1 focusing on patients with brain metastases and 1 looking at the 2-year follow-up. We look at the 2-year follow-up. CheckMate 9LA—if it doesn’t roll off your tongue all the time—was the combination of nivolumab and ipilimumab with chemotherapy but only 2 cycles of chemotherapy. It’s a little different from some of our other regimens because it had a little less chemotherapy. When we look at the 2-year outcomes, they’re still looking better; survival is 38% at 2 years, which isn’t bad at all. We’re putting that into 1 of many regimens to be thinking about when we’re trying to figure out what to start with for our patients who don’t have tumors with driver mutations: 2 chemotherapy cycles vs 4, 2 immunotherapy drugs vs 1, when does it matter as much?
The addition of a CTLA4 agent is probably more useful when we don’t have PD-L1 expression, but they didn’t go into that this time, so that’s a side note. But what they did talk about were the patients with brain metastases. That was encouraging, showing that the combination checkpoint inhibitor plus chemotherapy was much better as far as reducing the likelihood of having brain metastases. We have a lot of data now with checkpoint inhibitors with brain activity.
Sarah Goldberg at Yale School of Medicine did some of the first work a long time ago, but that’s been a consistent theme. It’s not that we’re trying to get those drugs directly into the brain. We’ve got the lymphocytes that are better at attacking the tumor that can get into the brain. It makes sense—nothing too novel there—but there are definitely some encouraging data. Also, that was regardless of whether patients had previously had radiation. The hazard ratio was 0.36 for patients who got the combination regimen vs just chemotherapy, really encouraging for the immune checkpoint inhibitors in the brain.
John V. Heymach, MD, PhD: Even though we think of patients with brain metastasis as the worst prognostic group, they did get greater relative benefit with this regimen. As you mentioned, this is only 2 cycles of chemotherapy, and you don’t keep the maintenance going with this regimen, right?
Heather Wakelee, MD: Right.
John V. Heymach, MD, PhD: It’s 2 cycles of chemotherapy and then you’re done, but you do keep the ipilimumab going every 6 weeks.
Heather Wakelee, MD: Right.
John V. Heymach, MD, PhD: I bring that up because we’ll hear some differences with the POSEIDON study data. Vamsi, why don’t you tell us about this phase 3 POSEIDON study that we’re hearing about for the first time?
Vamsidhar Velcheti, MD: The POSEIDON study is along the lines of CheckMate 9LA but with some key differences. It’s a large phase 3 randomized study. It has 3 arms. It’s randomized 1:1 durvalumab plus chemotherapy vs durvalumab-tremelimumab plus chemotherapy and platinum doublet as a control arm. The key difference, as you pointed out, was the maintenance pemetrexed…. Of course, it’s only for nonsquamous. Maintenance pemetrexed was allowed. Tremelimumab was given every 16 weeks after the full cycles of every-3-week dosing of tremelimumab. It’s a little better regimen to tolerate, I guess, compared with every 6 weeks of ipilimumab.
What’s interesting is that…a higher percentage of patients were squamous, around 40%. A little less than 40% were squamous histology. It met its coprimary end points. PFS [progression-free survival] and OS [overall survival] were positive. In the durvalumab-chemotherapy it was 5.5 months median PFS vs 4.8 with chemotherapy, and 13.3 months of median OS with the durvalumab-chemotherapy and 11.7 in the chemotherapy arm. With the addition of tremelimumab, the median PFS was 6.2 months, and the median OS was 14 months.
A few caveats in the subgroup analysis. For patients who had squamous histology, we’re hearing that consistent theme today, the squamous histology was the 1 they performed, which I found a bit surprising. Patients who had PD-L1 tumor progression of greater than 50% derived the maximum benefit, which isn’t unusual. But overall, it’s a positive study. It’s a different regimen, and definitely the addition of tremelimumab to durvalumab adds additional toxicities, which isn’t a surprise. It’s good to have a new option for these patients.
John V. Heymach, MD, PhD: Yes. Because they had a durvalumab plus chemotherapy arm as well as a durvalumab plus tremelimumab-chemotherapy arm, you could see the contribution of the tremelimumab there. It’s worth nothing, unlike in the CheckMate 9LA regimen, that they keep the CTLA4 blockade going. Here, they just kept the CTLA4 blockade going with tremelimumab for 5 cycles if I recall properly, but then after that only the durvalumab continues there afterward in terms of the immunotherapy. Some minor differences, but both of them are positive studies.
Ani, we’ve got a number of results here with PD-1 plus chemotherapy or PD-L1 plus chemotherapy, and now 2 different regimens with combinations of PD-L1 pathway and CTLA4 blockade with chemotherapy. From your perspective, how do you view this. Do you have any obvious take-homes in terms of patient populations that you think about for 1 vs the other?
Ani Balmanoukian, MD: It’s always good to have options. Similar to our discussion of monotherapy in the frontline setting, it’s going to be difficult to tease out exactly which 1 is better than the other in terms of the chemotherapy–I/O [immuno-oncology] combination vs just the I/O combinations. The way I approach patient selection in terms of the therapy is their bulk of disease. If someone is presenting with significant tumor disease burden, including multiple liver metastases, bone metastases, I incorporate chemotherapy and immunotherapy combination for that patient vs predominantly lung disease, lymph node involvement. That differentiates somewhat in terms of patient selection for the regimen I’m going to choose. In terms of choosing between which I/O combination and which chemotherapy–I/O combination, what’s available is KEYNOTE-189, CheckMate 227, and CheckMate 9LA. It depends on the patient selection, how they’re presenting, and how symptomatic they are for me to choose what regimen I’m going to first expose the patient to.
John V. Heymach, MD, PhD: As you said, it’s good to have choices. A lot is going to be where does the field go now. I’m particularly encouraged by the brain metastases data. That was 1 where we didn’t know what immunotherapy was going to be able to do. But the melanoma data with ipilimumab-nivolumab and now the CheckMate 9LA data give us encouragement in that space. We’ve done a study at The University of Texas MD Anderson Cancer Center with ipilimumab and nivolumab and radiation, which has been tolerated so far. Mehmet Altan has presented some of that tolerance data, so maybe we’ll be doing combinations of these things going forward.
Transcript Edited for Clarity