In the third interview of this series, Edward B. Garon, MD, reviews the treatment landscape in non–small cell lung cancer, with a focus on immunotherapy and the recent approval of tremelimumab in combination with durvalumab and platinum-based chemotherapy.
Amidst the growing treatment landscape in non–small cell lung cancer, one of the most recent advancements was the approval of tremelimumab in combination with durvalumab and platinum-based chemotherapy for patients with advanced or metastatic disease.
In this interview, Edward B. Garon, MD, a medical oncologist specializing in lung cancer at the David Geffen School of Medicine at UCLA in Los Angeles, California, discusses the overall treatment landscape for non–small cell lung cancer, recent data updates, and looks to the future of clinical research and treatment of this disease.
Targeted Oncology™: Can you briefly summarize the available first-line immunotherapy options for non–small cell lung cancer? How do these options vary by patients’ mutation status and/or PD-L1 expression status?
Edward B. Garon, MD: For first-line immunotherapy, there are a few groups of patients that are excluded from the current guidelines, and those are patients who have mutations in the EGFR gene, as well as patients with ALK gene rearrangements. There are some other targetable mutations where, in general, I may not use immunotherapy-based approaches in my practice. However, unlike EGFR and ALK genomic abnormalities, these are mutations that weren’t excluded in the clinical trials leading to approval of immunotherapies.
Once removing patients with, for instance, EGFR and ALK genomic abnormalities, the next real determinant of what to do with respect to therapy is related to PD-L1 expression. In patients who have high PD-L1 expression, which in my clinic I generally define as PD-L1 of 50% or greater—although there is some increased enthusiasm about immunotherapy in patients with even particularly high PD-L1 expression levels—we have generally looked at single-agent immunotherapy. There are 3 approved agents in that setting. Pembrolizumab was the first to be approved in this setting, but there are also approvals for atezolizumab and cemiplimab. There’s also an approval for pembrolizumab in patients who have PD-L1 of at least 1%. In my own practice, I generally have not used pembrolizumab monotherapy in patients with PD-L1 expression between 1% and 49%. However, there are some patients—particularly elderly patients—who may otherwise refuse chemotherapy. And in that population, single-agent pembrolizumab may be a reasonable option in patients who have PD-L1 [expression] between 1% and 49%.
In the United States, most patients who have less than 50% PD-L1 expression receive a combination of chemotherapy and immunotherapy, with the immunotherapy most commonly being pembrolizumab. These options are also available for patients who have high PD-L1 expression and sometimes are used in patients who have some reason that there is a push to treat. In fact, in these patients with high PD-L1 expression, there is an ongoing cooperative group study trying to evaluate whether these patients should receive single-agent PD-1 inhibition or a combination of PD-1 or PD-L1 inhibition plus chemotherapy. What is often done is there are differences in the chemotherapy backbone. Non-squamous disease patients often get carboplatin-pemetrexed along with chemotherapy. Whereas in squamous disease, pemetrexed is not used.
One other area of some difference of opinion in different practices is how to incorporate CTLA-4 inhibition. There is an approval for the combination of nivolumab and ipilimumab, which is a chemotherapy-sparing approach approved for patients who have high PD-L1 expression. However, in many respects, the most compelling data are in the group for whom it is not approved by the FDA, patients who have PD-L1 [expression] of less than 1%. In that group, patients receiving the combination did better than expected. And now there are 2 combination therapies that include both a PD-1 or PD-L1 inhibitor and a CTLA-4 inhibitor along with chemotherapy. One of those is what is often described as the CheckMate 9LA study, which combines carboplatin and pemetrexed for 2 cycles, with the first 2 cycles given with nivolumab and ipilimumab, and then the nivolumab and ipilimumab are continued. The other is the POSEIDON [study], which is the most recent addition here, which combines chemotherapy along with durvalumab, a PD-L1 inhibitor, and tremelimumab, which is a CTLA-4 inhibitor. In that approach, there are more cycles of chemotherapy than in the CheckMate 9LA design. However, after the chemotherapy is stopped, there’s only 1 additional treatment with the tremelimumab.
Targeted Oncology™: Can you please discuss the recent FDA approval of durvalumab and tremelimumab in patients with non–small cell lung cancer? Can you talk about the specific patient populations that were included in the approval?
Edward B. Garon, MD: [Combination durvalumab and tremelimumab] was evaluated as part of the POSEIDON study, and it’s approved for a broad population of patients with non–small cell lung cancer. Of course, as has been typical, patients with EGFR mutations or ALK gene rearrangements have been excluded from these analyses, as in those patient populations, we tend to look at targeted therapies upfront. It would generally be 4 cycles of chemotherapy, and there were several different chemotherapy options that were given. In addition, the durvalumab was continued over longer periods of time, but the tremelimumab was only given once after the chemotherapy was completed.
Targeted Oncology™: Can you talk about your typical treatment goals for patients who are receiving first-line immunotherapy for non-small cell lung cancer? Do they vary by whether you’re using a single agent or multiagent with or without chemotherapy?
Edward B. Garon, MD: One thing that I think is exciting is that in many respects, on a nuanced level, our anticipations have really improved for patients with advanced non–small cell lung cancer. Although, when we describe it, maybe we don’t describe it that we’re treating for cure in these situations, but I think we are treating for durable benefit. And we now know from some of the early single-agent studies that 15% or more in even previously treated populations are going to have a 5-year survival, which is really exciting and not what we are used to in our typical treatment paradigms. In many respects, this is not something that changes based on what our regimen is. Instead, we choose our treatment regimen based on what we think is most likely to generate durable benefit. That has been an area where we don’t have good data to guide us. For instance, we know that patients who have high PD-L1 [expression] are more likely to receive that durable benefit. We know that patients who have a PD-L1 [expression] of 50% or greater—at least in a trial of pembrolizumab with or without ipilimumab—did not really benefit from the addition of ipilimumab. But certainly, among many practitioners, there is some enthusiasm about CTLA-4 inhibition in patients, particularly those with low or negative PD-L1 [expression], as that group unfortunately has not experienced the same degree of durable benefit in our practices.
Targeted Oncology™: Can you discuss the clinical and nonclinical factors—such as safety and efficacy data, mutation status, comorbidities, and patient preference—that inform your choice among the available first-line immunotherapy options?
Edward B. Garon, MD: I spoke earlier about what biomarkers would help us select patients for different treatment regimens. There are some emerging questions within biomarkers that are maybe a little less firm. There are certain mutations, such as mutations in STK11 and KEAP1, that have been associated with at least a relative lack of benefit from single-agent immunotherapy approaches. If I had those biomarkers, they might really direct me more toward a nonsingle-agent, immunotherapy-based approach. Although, in the other direction, there are some data indicating that patients who have high tumor mutational burden have disproportionate benefit from immunotherapy. For instance, if I had a patient with a PD-L1 expression of 35%, and the tumor mutational burden was particularly high, I may consider them for monotherapy. There is an approval across histologies for immunotherapy in patients who have at least 10 mutations per megabase, so it can be done clinically.
There are other important clinical factors. As the population ages, we tend to have an older patient population, and many patients in their late 80s are either unwilling or maybe not great candidates for cytotoxic chemotherapy. Those are patients for whom we may look at immunotherapy-alone-based approaches. Conversely, in patients who are younger, the likelihood of toxicity from chemo-immunotherapy-based approaches is less. That’s a population where I may have greater willingness to look at combinations of more aggressive approaches in my practice.
Targeted Oncology™: How do the efficacies of the available regimens compare?
Edward B. Garon, MD: One of the big frustrations in this field is that we’re always advised not to do cross-trial comparisons, and we try not to do them because we know that there are differences from one study to the next. However, we have a difficult situation because in the end, we are in a situation where you have a huge number of studies that were randomized vs standard chemotherapy, and it’s very hard to choose between them. I’m hopeful that at some point, we will have some comparative data.
The biggest challenge in this is the role of CTLA-4 inhibition. In the CheckMate 227 study, in patients who are PD-L1 positive, for instance, there were some data of nivolumab and ipilimumab vs nivolumab. Numerically, the combination looked better, but we don’t have definitive data. In the POSEIDON study, there was also a group that got durvalumab plus chemotherapy, but no tremelimumab. That group in some ways looked fairly similar, but not quite as good as the 4-drug regimen. In fact, it did not hit its statistical threshold, and that approach is not approved.
Targeted Oncology™: Are there any toxicities of concern with immunotherapy regimens? Are they similar or different between the available options?
Edward B. Garon, MD: Many of the toxicities of immunotherapy are associated with just immunotherapy. Pneumonitis, which has been the dreaded complication with all PD-1- and PD-L1-based regimens, is of some concern. Hypothyroidism is probably the most common thing that we see across regimens. And of course, there are toxicities like rash and nausea, which can be seen in many of our cancer therapies, but certainly can also be seen in immunotherapy-based approaches.
The one toxicity that is much more likely in patients who get a CTLA-4 inhibitor is colitis. In general, colitis, or at least high-grade colitis, is quite rare in single-agent PD-1 inhibition. And certainly, when one looks at combined therapy or anything including a CTLA-4 inhibitor, we see much more significant rates of colitis. In general, across the board, there’s some increase in autoimmune toxicities when one incorporates a CTLA-4 inhibitor. But in my experience, the one that is the most clinically meaningful is that increased colitis risk.
Targeted Oncology™: To follow up on that, how do you manage colitis risk?
Edward B. Garon, MD: In general, I often joke that I try not to give the talk on toxicity for immunotherapy approaches, because it’s a pretty boring talk. In general, the answer to everything is steroids. But in my experience, the one toxicity that clearly responds to nonsteroidal approaches is the CTLA-4 inhibitor–induced colitis. In my experience, infliximab has been quite an effective therapy in that setting, and we tend to use it there.
Targeted Oncology™: How frequently do you recommend multiagent immunotherapy, such as durvalumab and tremelimumab in combination with platinum-based chemotherapy, vs the other options?
Edward B. Garon, MD: In my practice to date, I’m still a little unclear about the role of CTLA-4 inhibition. I have encouraged the companies who have these agents to look at subsets that they think benefit and show in a randomized fashion what the benefit is. I’m looking forward to working with AstraZeneca, the company that makes durvalumab and tremelimumab, on potential study designs that would define in some of these groups, for instance, the group with STK11 or KEAP1 mutations that have been shown to do less well with PD-1 inhibition in general, and to evaluate whether that group in fact does do better when they receive the combination of this 4-drug regimen. There are some data from the POSEIDON study that suggest that patients who have PD-L1 expression less than 1% also derive disproportionate benefit, and I look forward to randomized studies that will assess that question.
Targeted Oncology™: What are the remaining unmet needs in the field, despite the availability of immunotherapy regimens?
Edward B. Garon, MD: The glaring unmet need in non–small cell lung cancer right now is that although we now have durable responders, which was not the case when I started my career, the great majority of our patients unfortunately are not in that group of durable responders. As we move forward, we are going to need better approaches to that group of patients. The first wave of immunotherapy-based approaches in that population were not very successful. My personal opinion is a lot of this is based on the fact that people evaluated resistance to immunotherapy as a certain issue, and you would need another drug in an all-comer population to restore or achieve sensitivity of the tumor to immunotherapy-based approaches. I’m hopeful that we will start to have smarter approaches to this in the coming years. That’s something that I’m excited for in studies that will be going on in the future.
Targeted Oncology™: Are there any ongoing trials that might address those needs?
Edward B. Garon, MD: There are a large number of ongoing trials evaluating what to do at the time of progression, but to date, there are no immunotherapy-based, FDA-approved approaches there. From a regulatory perspective, docetaxel, or docetaxel-based approaches, like docetaxel-ramucirumab, would be considered the standard approach in someone who, for instance, progresses on chemo-immunotherapy.
Targeted Oncology™: Are there any specific data that you’re particularly excited about seeing?
Edward B. Garon, MD: In terms of the immunotherapy in patients who have progressed, there are multiple data sets looking at tyrosine kinase inhibitors that are multitargeted and include VEGF inhibition or VEGF receptor inhibition. There was a study by the cooperative groups that looked at adding the monoclonal antibody ramucirumab to pembrolizumab, and there’s talk of additional efforts as part of the cooperative groups to follow up on those results. But we’re awaiting data on several of these multitargeted tyrosine kinase inhibitor plus PD-1 or PD-L1 inhibitor–based approaches. That will help tell us where the field is going after this. Those approaches are nivolumab and sitravatinib, pembrolizumab and lenvatinib, and atezolizumab and cabozantinib.