Evolution of First-Line Immune Checkpoint Inhibition in Advanced or Metastatic Solid Tumors - Episode 5

First-Line Combination Immunotherapy plus Chemotherapy in mNSCLC

Edward B. Garon, MD, Professor of Medicine, and Director of the Thoracic Oncology group at the David Geffen School of Medicine at UCLA, reviews key clinical trial data and explains the rationale of using combination IO therapies in the treatment of non-small cell lung cancer (NSCLC).

Targeted Oncology™: What is the rationale for using combination IO therapy such as durvalumab and tremelimumab alongside chemotherapy in the first-line setting?

Edward B. Garon, MD: We are now several years into our development of immunotherapies in non-small cell lung cancer. We clearly have had, in some cases, really spectacular activity of single-agent inhibition with PD-1 or PD-L1 inhibitors. However, unfortunately, the great majority of patients do relapse, and so there's been an interest in giving additional therapies. Chemotherapy has been one of the therapies that people have added and that has led to significant success. Really adding a PD-1 inhibitor, for instance to chemotherapy leads to much better outcomes than chemotherapy alone. The agent has been a little trickier to figure out what to do with or agents that it is really the CTLA-4 inhibitors. So CTLA-4 inhibition actually was the initial approved checkpoint inhibitor approved in melanoma where there is some single-agent activity. In lung cancer, we really haven't necessarily seen that and even less we have an approved agent that was available for melanoma and that clearly is important in immune destruction of tumors. There have been multiple attempts to develop a combined inhibition of CTLA-4 inhibitors along with PD-1 inhibitors with or without chemotherapy. And right now, there are actually two of those approaches that are approved. One is in PD-L1 positive patients, the combination of ipilimumab and nivolumab and then also a combination of two cycles of chemotherapy along with ipilimumab and nivolumab.

Targeted Oncology™: Can you discuss the design of the POSEIDON Trial?

Edward B. Garon, MD: In addition to the two regimens that I mentioned that are currently approved, the combination of nivolumab and ipilimumab with chemotherapy or nivolumab and nivolumab alone in patients that have PD-L1 expression of 1% or greater, we now have data on the POSEIDON study, which looked at tremelimumab rather than ipilimumab as the CTLA-4 inhibitor, and it included a PD-L1 inhibitor durvalumab rather than the PD-1 inhibitor nivolumab. So, the POSEIDON study is a large study of approximately 1,000 patients who were randomized 1:1:1 to three arms. Each arm got the standard of chemotherapy, the arm with durvalumab would receive durvalumab along with chemotherapy and then in maintenance, and then the arm with durvalumab and tremelimumab was the same except for they also got tremelimumab during the chemotherapy and then for one treatment after the chemotherapy portion had ended. That was the design of the study. It was an international study and the main endpoints that were assessed were progression-free survival and overall survival.

Targeted Oncology™: Can you briefly summarize the POSEIDON data that was presented last year at World Lung?

Edward B. Garon, MD: The original data for the POSEIDON study was presented in 2021 at the World Conference for Lung Cancer and in that presentation, they looked at the three arms. There was an improvement for progression-free survival in both of the immune checkpoint inhibitor arms but for survival the durvalumab alone arm fell a little short but the durvalumab plus tremelimumab was significantly better than chemotherapy alone. This led to sort of the standard type of benefits we have seen with PD-1 or PD-L1 inhibitors, you see an improvement at the median, as well as a higher plateau for the curve. The tolerability of the combination of the four drugs was quite good. There are of course some toxicities that are seen when one adds a CTLA-4 inhibitor, one of the main ones that gets added is colitis, which can be seen with a PD-1 or PD-L1 inhibitor, but it's seen fairly rarely in those and generally are not a toxicity that really overwhelmed the symptom burden. But you can see that more with CTLA-4 inhibitors. But nonetheless, when looks at issues like whether or not patients needed to stop therapy, the tolerability was certainly reasonable with the four-drug regimen, and importantly, the outcomes, including progression-free survival, and overall survival were better in the chemotherapy plus durvalumab and tremelimumab arm.

Over a year has passed since the original presentation of data from the POSEIDON study. There, of course, are inevitably updates that will occur. There was an update with the longer follow-up, with four-year follow-up rather than about three-year follow-up in the original presentation. Sometimes these longer follow-up presentations are marginal increases in data compared to the original presentation. However, in this case, it certainly was supportive that the results that had been seen in the original presentation continue to be maintained at four years. There also have been other publications, and I think the one that has garnered the most attention is an analysis presented this year by Solange Peters. And in that they looked at three separate mutations. One was STK11, another was KEAP1 and the third was KRAS. So STK11 and KEAP1 are mutations that have been associated with poorer outcomes in patients with non-small cell lung cancer receiving immunotherapy. Now, how strong this data is it depends a little bit on to whom you're speaking. There was some very exciting initial data coming out of MD Anderson and most of the data has been quite supportive of the idea that these do confer a poorer outcome with immunotherapy. There is one outlier dataset that being the KEYNOTE-042 study. And as part of that study, patients needed to be PD-L1 positive, at least when restricted to patients who were positive for PD-L1. It was a little less clear what the prognostic role was, but in Dr. Peters' presentation, it was shown that the patients who did have these mutations did quite well on the four drug regimen, indicating the potential that in patients who otherwise were not set up to have a good response to immunotherapy that they would do well in the four drug regimen. Whereas the immunotherapy alone has not been an effective approach in those people. So, the third that they looked at were KRAS mutations. KRAS mutations have not necessarily been considered to be a poor predictor of outcome for immunotherapy. In an all-comer population, there have been some datasets that has suggested it is even beneficial, although that may be derived from the fact that it largely excludes patients with other mutations associated with lack of smoking history, such as EGFR. All of the three genes that were evaluated did appear to have particularly a beneficial effect with the four-drug regimen. And the data is intriguing and the sort of thing I certainly would like to see followed up in prospective studies to see if in these patients who have predictive markers that question whether or not immunotherapy- based approaches are going to be beneficial whether or not the combination of all four drugs can lead to better outcomes.

Targeted Oncology™: If this combination of durvalumab, tremelimumab, and chemotherapy became available for first-line therapy, what unmet needs might it fill?

Edward B. Garon, MD: Right now, at least at the time that I was speaking here, there's no regulatory approval for the combination of durvalumab, tremelimumab, and chemotherapy. Should there be, it would be yet a third regimen that is looking at the PD-1, PD-L1 checkpoint along with the CTLA-4 inhibitor. One of the unique issues of this is that this regimen would be sort of the most intensive of these in some ways, in that there are four cycles of chemotherapy, which is not the case in for instance, the CheckMate 9LA regimen, which includes nivolumab and ipilimumab. Although the duration of the CTLA-4 inhibition is less. So, I think that how this fits in amongst those is a little unclear. And I think one place that I think we can look forward to the potential for looking further at this regimen would be that it would be great to have studies, prospective studies looking at the sort of questions that are being addressed by Dr. Peters' presentation. Do you really have a better outcome? In these patients who, for instance, have STK11 mutations keep one? Are these groups of patients that really are going to benefit, particularly to the four-drug regimen? I think that we're all looking to identify who those patients are, that would be more likely to benefit from incorporation of CTLA-4 inhibition because clearly there is some toxicity that is being added. In order to do that, we want to have a sense of in whom the efficacy is seen. I think that some of the data that's been recently presented point us in that direction. And hopefully, we will start to have prospective data to help validate the hypotheses that have been generated by this prior work.

Targeted Oncology™: Given the combination therapies that are already available, do you think that availability of durvalumab, tremelimumab, and chemotherapy might change treatment paradigms or patient outcomes?

Edward B. Garon, MD: At this point, we don't have great data that helps us differentiate particularly between this regimen and the CheckMate 9LA regimen which I design is quite similar, where one would have a CTLA-4 inhibitor plus an inhibitor of the PD-1 and PD-L1 interaction along with chemotherapy. I think that based on the data we have to date, we don't have a lot that differentiates these two regimens, one from the other. But I think that hopefully, in the coming years as some prospective data is generated, particularly in these biomarkers subsets that have been of interest and have been presented before that we would be able to have something that gives us a specific indication to use these agents. Now, whether or not data for instance, that would be generated prospectively with the POSEIDON regimen would also be of relevance to the CheckMate 9LA regimen, we would assume, but we can't be certain, in least in my practice, I've generally sort of used a regimen where the safety and toxicity and efficacy are known because they've been experimentally tested. And I think that that will be something of interest, certainly, as I mentioned in prospective studies going forward.

Targeted Oncology™: Do you have any concerns about toxicity of this combination?

Edward B. Garon, MD: Any therapy you have is going to have increased toxicity. And that is something that we live with, but we need to balance that against the increase in efficacy. And that's again, why I'm really eager to see prospective data that addresses these sorts of questions. Because we know we are going to have some toxicity when we incorporate a CTLA-4 inhibitor. And I think it's important to know in whom we are showing increased efficacy, for instance, although the CTLA-4 inhibitor was different, there was a study in patients high PD-L1 expression of pembrolizumab plus placebo or pembrolizumab, plus ipilimumab. And the incorporation of ipilimumab was not better in that. The study was stopped early as the toxicity was greater and the efficacy was not better. There is some data that some of these patients with biomarkers that would end up a poorer outcome, may be particularly appealing patients to be looking at with these four drug regimens. And I think again, it's going to be really important to experimentally test that so that we can get some sense as to whether or not that is something that is not just a theoretical positive but with something that is true in the real world.