Chemoimmunotherapy Combinations in Advanced NSCLC

Transcript:

Stephen V. Liu, MD: We do have a lot of different options. And with that, why don’t I move on to your section talking about different combinations, Neal?

Neal Ready, MD, PhD: Thank you. KEYNOTE-189 was a landmark trial. And when it was planned and started, it was by no means certain that it was going to be positive. Because we didn’t know how chemotherapy would combine with immune therapy. Because you could think, well, the chemotherapy is going to knock down the immune cells or that the steroids that you give with the immune therapy are going to knock down the lymphocytes and cancel out any immune activity….I wasn’t certain what was going to happen. KEYNOTE-189 was a large, randomized phase 3 trial. It's important to remember this was in nonsquamous lung cancer. It was platinum plus pemetrexed. It’s nonsquamous lung cancer. And patients received standard chemotherapy with pemetrexed maintenance, with or without pembrolizumab. And as in most of the clinical trials, but not all—as we’re going to go over in a minute—patients with activating mutations or translocations like EGFR [epidermal growth factor receptor] and ALK [anaplastic large-cell lymphoma kinase] were excluded. And the results of this trial were really exciting. The overall hazard ratio for improvement was 0.6 for chemotherapy with carboplatin pemetrexed, with pembrolizumab throughout, compared with chemotherapy alone. And what’s more, it was positive in a clinically significant way, no matter whether the PD-L1 score was high, medium, or low. And to me, what’s even more exciting still is [that] when you looked at landmark survival to 3 years—not just looking at median survival but looking at [participants] alive at 3 years—there was a marked improvement in patients alive for both PD-L1 scores greater than 50%, 1% to 49%, or less than 1%. And this is despite most of these patients crossing over and getting second-line immune therapy.

Thus, this trial won in every way—it improved overall survival, landmark survival, progression-free survival, response rate, and duration of response. And there wasn’t that much of an increase in toxicity. There was some, but there wasn’t any marked increase in severe toxicity. Even with long-term follow-up, it remains equally powerful that the immune therapy combination of chemotherapy plus pembrolizumab was markedly superior. And clearly, this has become one of the main standards of care in nonsquamous lung cancer.

A very interesting trial, IMpower150, was looking at chemotherapy with or without immune therapy: in this case, atezolizumab. And certainly, partially because Genentech [makes] bevacizumab and atezolizumab, they chose their standard of chemotherapy and bevacizumab as their control. [This] was pretty brave because the addition of bevacizumab improves your median survival by about 2 months. You were raising the bar for what you needed to do with chemotherapy plus atezolizumab to get a significant result. But also, it was felt that the quadruplet might be important. Hence, there’s evidence that activation of the vascular system, with vascular endothelial growth factor, may be immune suppressive, by influencing Tregs [regulatory T cells], MDSCs [myeloid-derived suppressor cells], and other factors in the immune microenvironment that can inhibit lymphocytes.

In this trial, we had the control arm of chemotherapy and bevacizumab. An experimental arm of chemotherapy plus atezolizumab. But then [there was] a third arm, Arm B, where they received the quadruplet of atezolizumab, bevacizumab, carboplatin, and paclitaxel. And then, with maintenance according to those trials,…there was improvement in outcome. The chemotherapy and atezolizumab arm did show some improvement in survival with chemotherapy/atezolizumab vs bevacizumab. And it did show, mostly in the high PD-L1 group or the positive PD-L1 group, whereas there was really no difference in the PD-L1 negative group. Keep in mind that they did PD-L1 testing through a different method, compared with the KEYNOTE trial. Now, the quadruplet arm, so arm B versus C, where we were comparing chemo/bevacizumab vs chemo/bevacizumab and atezolizumab, the results were more positive. There was a clear improvement for the PD-L1 high, the PD-L1 low, and a modest improvement for the PD-L1 negative. I think many [of those] in the field feel that this is the most intriguing part of the trial.

As I noted, in the KEYNOTE trial, patients with actionable alterations, such as EGFR mutations and ALK, were excluded. But this trial allowed a significant number of patients with EGFR mutations to be treated. And the patients with EGFR-sensitizing mutations, such as exon 19 and exon 21, seemed to have a clear benefit for the quadruplet vs chemo/bevacizumab. And that’s in contrast to chemo with atezolizumab. There seemed to be something special about the quadruplet in this EGFR-mutated population. Now, this was a relatively small sample size. The trial was not powered to look at the EGFR-sensitizing tumors. But…having some effective immune treatment for EGFR-mutated lung cancers is a major area of unmet need in the field. Because after patients progress on all their EGFR/TKI [tyrosine kinase inhibitor] therapy, at this point, all they have is chemotherapy alone to go to. And many of these patients are younger—younger nonsmokers. And it’s just an area of major unmet need. Thus, these results are very promising. And I have treated some patients with EGFR mutations with a quadruplet. Dr Liu, have you used this quadruplet in those cases?

Stephen V. Liu, MD: I have. And I think you bring up interesting points. We think of EGFR, ALK, these subsets of lung cancer where we have very effective, targeted therapies, a very wide therapeutic index—meaning they work very well, few toxicities, [and they] work for a long time. But they’re not cures. They don’t work forever. And that durable, long-term benefit, that really sets immunotherapy apart. And we haven’t gotten that for most patients with driver positives. When we look at the second-line setting, we know that immunotherapy is not too effective in EGFR and ALK lung cancer. But for EGFR-mutant lung cancer the response rate’s not zero. It’s a low number, but it’s nonzero. And thus, we do see some responses. I think these data were very intriguing because they hit on a couple themes that we’re familiar with. That we can see responses in EGFR-mutant lung cancer. That VEGF inhibition and antiangiogenesis seem to play some role in targeting the EGFR biology. A lot of work done by our colleague John Heymach, [MD, PhD, of The University of Texas MD Anderson Cancer Center] really showed that angiogenesis is important and that blocking it with agents like bevacizumab or ramucirumab seems to provide some added benefit for EGFR. And so maybe this was 1 important step in unlocking this. I have used this. And I do say, there is more toxicity with this, but I really don’t think it’s so much from the bevacizumab or atezolizumab. It’s more the paclitaxel, the chemotherapy....Now, does that matter? Can you swap them out? I think we don’t know. There are some different schools of thought. Maybe it does matter. Maybe it doesn’t. But there is more chemotherapy toxicity. And hence, while these signals are, I think, very intriguing, and I think this is a very reasonable option, I think there’s still a lot of work that needs to be done in this set.

Neal Ready, MD, PhD: No, I agree. I think it’s an option that I will use in some cases. But I really want to see more prospective trial data from clinical trials to really nail this down and make us feel more comfortable with the data. And this is, then, the EMPOWER-Lung 3 study design, with cemiplimab, that we had talked about as monotherapy. And this is, again, an antibody against PD-1, just like pembrolizumab and nivolumab. It’s important to remember that in this trial, histology-directed chemotherapy was allowed. Thus, unlike KEYNOTE-189, it was all nonsquamous. This was combined histology. You can’t just look at the overall survival results for this trial and compare it directly with KEYNOTE-189. Because overall survival is going to be better in the nonsquamous. What you really need to look at is the hazard ratio of control versus experimental. You just can’t compare the survival straight away with KEYNOTE-189. This is, again, an interesting trial that looked at the combination of chemotherapy plus I/O [immune-oncology therapy], stratifying for PD-L1 status. With a marked improvement in overall survival, from 13 months for control to 21.9 months for the treatment arm with a hazard ratio of 0.71. [Concerning] results, it would really be comparable to other clinical trials that we have seen. Thus, again, another potential treatment option in this setting. And I think the more options that we have, the better. And as we get more options, there may be financial incentives to use one treatment or another, to drive down the health care outcome. I think that is an important consideration.

Transcript edited for clarity.