: When we look at second-line treatment for non–small cell lung cancer, it has changed with the recent approvals of frontline immunotherapy. Outside of patients with autoimmune disease, with an organ transplant, or some other contraindication receiving immunotherapy, most of our patients now are receiving some form of immunotherapy as initial treatment, unless they’re a non-squamous patient with a targetable mutation or fusion or other actionable alteration. I want to emphasize that, if you have a non-squamous lung cancer, we should be doing molecular testing, and oral therapy would be the first choice if you get an actionable alteration. The good news is we have better frontline treatment and patients are living longer, especially with the combination immunotherapies. The flip side of that is that we’re back to standard second-line regimens such as docetaxel or docetaxel with ramucirumab that we’re not that excited about.

This is one of our major areas of unmet need in non–small cell lung cancer. The goal is finding a way in the post-I/O setting to reignite the immune system and get additional prolonged immune responses, to continue to look for new molecular targets, and find ways to enhance the second-line chemotherapy options like docetaxel. The use of second-line therapy is more limited, but for a good reason that we have better initial treatments including immunotherapy, but the second-line setting now is really a challenge.

There’s clinical development now. There will be new immune options in the second line. We’re learning more about how cancers are able to evade the immune system. People will now get some form of I/O therapy first, and if the cancer progresses, then obviously if it was an initial response then it’s evaded the immune system somehow. There are a lot of ongoing scientific evaluations that let us understand is this down regulation of MHC-1 to evade the immune system? Is it local tumor effects with an increase in T reg cells or MDSCs [myeloid-derived suppressor cells]? Is it other changes in the tumor microenvironment? We are rapidly progressing to understand how immune evasion occurs, and there are a lot of pipeline therapies that are now available that will let us test different strategies for overcoming the secondary immune resistance.

That’s where things are headed, and the clinical trials are ongoing, so hopefully we’ll find some new strategies that will let us reignite the immune system in patients where progression has occurred and get secondary and tertiary immune responses that are helpful.

 When we look at second-line therapy, some of the big characteristics determining our approach are the prior treatment delivered, patient comorbidities, and performance status. When we look at performance status comorbidities, it drives the suitability of further therapy. Our role as an oncologist is not to see how much treatment we can deliver. It is to help patients. If patients have poor performance status or if they’re weak, I know they’re not going to derive the benefit from further therapy. In my effort, the best way to help that patient might be to focus on symptoms or more of a hospice-based approach. For patients who have a good performance status, there are an increasing number of options. Our previous second-line therapy had been docetaxel-based treatment, either alone or in combination with ramucirumab. It depends on what initial treatment was given. For example, if pembrolizumab alone or atezolizumab alone were given, or if dual checkpoint inhibitors such as nivolumab and ipilimumab were used in the frontline setting, my second-line regimen would really become platinum-based chemotherapy.

Now, whether to continue the checkpoint inhibitor in the setting of second-line platinum doublet chemotherapy is a question that the INSIGNA trial will look at to see if continuation of a checkpoint inhibitor beyond progression provides any value or only cost and toxicity. Right now, our standard approach after progressing on immunotherapy alone, either mono or dual checkpoint blockade, would be platinum doublet chemotherapy histology specific. If chemo immunotherapy were given continuously in our second-line option—really docetaxel-based treatment—we’re certainly always exploring clinical trials, novel immunomodulatory agents, and different strategies to really recapture those long-term benefits.

Videos

NSCLC Management: Second-Line Options

: For squamous cell cancers, squamous cell lung cancers, we don't have any actionable alterations to allow oral therapy. Typically, the biomarker that we would test outside of a clinical trial would be for PD-L1. If the PD-L1 is high, then it would be reasonable to treat with pembrolizumab alone. When the tumor is inflamed, with high PD-L1 scores 70, 80, 90, 100%, single agent immunotherapy can be effective with minimal toxicity.

For the rest of the patients, initiating combination therapy is indicative. KEYNOTE-407 was a large randomized phase 3 trial that compared standard histology-based chemotherapy with or without pembrolizumab. It showed that the chemo I/O combination was superior to chemotherapy alone with a modest increase in toxicity, and certainly an important new standard of care.

In that same squamous cell population, we now have the CheckMate 227 regimen of nivolumab and low-dose ipilimumab, and also the CheckMate 9LA where we have an FDA approval for 2 cycles of chemotherapy plus nivolumab and ipilimumab.

In summary, for this metastatic squamous cell population that doesn't have active autoimmune diseases or some other contraindication or immunotherapy, it would be consideration of single-agent immunotherapy for the inflamed tumors with a high PD-L1 score, and then some type of combination therapy, either standard chemotherapy plus immunotherapy or nivolumab, ipilimumab with or without 2 cycles of initial chemotherapy, depending on the clinical circumstances and the preference of the patient.

Squamous Cell Frontline Therapy: IMpower 131, KEYNOTE-407

: CheckMate 817 was a trial that was done to expand the use of nivolumab and ipilimumab in looking at different dosing regimens, and regimens that could be every 3 weeks rather than every 2 weeks, and to look at special populations. It was a large population of patients who were untreated and who received a standard dose of nivolumab and ipilimumab. That is providing useful information, understanding the dosing, other drugs, and how they combine.

There was also a cohort A1 that was done to look at special populations. And the special populations were patients who had either performance status 2, HIV, or hepatitis. The outcome for that cohort was interesting because there was a question, would it be effective? Would the nivolumab and ipilimumab combination be effective in these patients, and would the combination be tolerated, or would we see excess toxicity in these patients with poor performance status or specific comorbid medical problems?

The toxicity was remarkable in that it was no different than what we had seen in the large randomized trials of patients with performance status 0 or 1. We looked at the data for CheckMate 227, CheckMate 9LA, and CheckMate 568; the performance status 2 patients didn't seem to have any significant increase in toxicity for grade 3 and 4 toxicities or discontinuation of treatment compared with patients with good performance status. There were a substantial number of patients who had responses to the treatment and the responses were durable.

This was an important study looking at these special populations. I recently reviewed some of the literature, and there's truly little data for performance status 2 patients with immunotherapy. I was able to find 1 trial looking at pembrolizumab alone out of the United Kingdom that showed that pembrolizumab was tolerated and effective in this population. I couldn't find any data for chemotherapy plus I/O in the performance status 2 population, so this is some of the little data that we have really showing that a combination immunotherapy can be safe and effective in these special populations. That's useful for a practicing clinician.

CheckMate 817 Overview

: KEYNOTE-189 was a groundbreaking study in which patients with non-squamous lung cancer received carboplatin, pemetrexed, standard histology-based chemotherapy with or without pembrolizumab. And all outcome measures were better. Overall survival, response rate, progression-free survival, and duration of response: this is a very positive trial. Certainly, in the United States this has become a major standard of care for upfront treatment of metastatic lung cancer. It does have prolonged exposure to chemotherapy, so it has 4 cycles of platinum-based chemotherapy, and then potentially going pemetrexed. It’s a groundbreaking study and a standard of care that has improved the outcome of a lot of patients with lung cancer.

The IMpower130 was another randomized trial in non-squamous lung cancer. And it was carboplatin, nab-paclitaxel, or Abraxane, with or without atezolizumab. It was a positive trial where the overall survival and other outcomes were clearly superior to chemotherapy. It is a different chemotherapy backbone than other trials in the non-squamous setting. Most of the other trials, the KEYNOTE trial and CheckMate-9LA, use pemetrexed-based treatment. It’s a positive trial, but with a different chemotherapy backbone with immunotherapy added into it.

The IMpower150 study was a more complicated study and was looking to push the envelope with treatment for non–small cell lung cancer. One of the standards of care for lung cancer, adenocarcinomas, is for platinum, a platinum chemotherapy, a taxane with bevacizumab. And a major component of this trial was to look at the Taxol, paclitaxel, bevacizumab regimen with or without atezolizumab. It did show an improvement in survival compared with the triplet backbone. It was a positive trial, FDA approved. What's not clear from this is whether the quadruplet therapy is better than, for instance, chemotherapy plus pembrolizumab or the 130 regimen of carboplatin, nab-paclitaxel, and atezolizumab. It’s a positive trial and FDA approved. Exactly where it fits in for standard of care is not entirely clear, but certainly interesting and promising therapy.

One subset analysis in this trial was particularly interesting. In general, immunotherapy does not work so well in EGFR-mutated lung cancers, for various reasons. And in this trial there were about 100 patients who were EGFR positive who were included, and the experimental regimen with atezolizumab looked better compared with the chemotherapy and bevacizumab alone. It’s a retrospective and wasn't powered to answer the question whether this was better. It has an intriguing result, and there are additional studies coming out now to test the hypothesis that this might be the best combination in EGFR-mutated lung cancer.

PD-L1 Combination Therapy Clinical Trials 

KEYNOTE-042 was a randomized trial looking at pembrolizumab monotherapy in a select patient population, and it has influenced care. This was a large study of over 1200 patients with advanced non–small cell lung cancer, excluding patients with sensitizing 

 fusions. All patients did have PD-L1 expression defined in that study by at least 1% of cells positive using the [Agilent Technologies, Inc] Dako 22C3 clone. They’re randomized to pembrolizumab monotherapy or histology-specific chemotherapy.

We saw an improvement in overall survival with this approach leading to its FDA approval for patients with PD-L1 of 1% or greater. This was driven, as expected, by the PD-L1–high cohort that we saw in KEYNOTE-024. The population of interest was PD-L1 low. Are we really providing a better option for more patients? If we analyze patients with PD-L1 expression between 1% and 49%, where the use of pembrolizumab wasn’t clear prior to this trial, there was not a clear benefit over chemotherapy. It didn’t appear significantly worse, but it didn’t appear significantly better. The study wasn’t designed or powered to look at that subset, although clinically that is the subset of interest. It is an option for patients, particularly if patients aren’t eligible for other therapies. But in this day and age we have other options that I would say offer a clearly superior benefit to chemotherapy. Strategies including dual-checkpoint blockade or combinations of chemotherapy-immunotherapy. In a world where those options exist, I really think those have a leg up over pembrolizumab monotherapy for PD-L1 low, but that is an option for patients under certain circumstances.

IMpower110 was a randomized phase 3 trial looking at single-agent atezolizumab, a PD-L1 antibody, versus histology-specific chemotherapy. This was a relatively large study. Patients were randomized 1:1 to receive atezolizumab at a dosage of 1200 milligrams every 3 weeks or histology-specific chemotherapy. If we look at the PD-L1–high group looking at PD-L1 expression on tumor cells or on neighboring immune cells, which was about 35% to 40% of the entire population, there was a clear and overwhelming survival benefit. The median overall survival was 20.2 months compared with 13.1. That’s an impressive hazard ratio of 0.59 for the PD-L1–high group. One-year survival rates of 65% versus 50.6%. Response rates were higher, over 38%. Outcomes favoring the use of atezolizumab monotherapy in that PD-L1 high.

They also explored outcomes using the Dako 22C3 antibody, and the results were very consistent. In patients with PD-L1–high disease, atezolizumab is a reasonable option. It was FDA approved, and it is another option for patients. It’s not wise to compare it directly with other agents in this setting, such as pembrolizumab. Those outcomes are very comparable. Although there may not be a compelling reason to change—if there were, for example, nonclinical reasons, formulary restrictions that guided 1 over the other—I feel comfortable with either approach.

Frontline PD-L1 Monotherapy: KEYNOTE-042 and IMpower110 

 When we look at CheckMate 9LA, this trial did include both squamous and nonsquamous histology. It’s important to look at those ratios. About 30% of patients were squamous, which is a bit higher than some of the other studies. We know from multiple randomized trials, including some cross-trial comparisons, that our outcomes in the squamous histology are a bit worse. Factoring in those numbers, the outcomes are really impressive. When we break it down by histology, we did see a consistent benefit, a survival benefit in both the squamous and nonsquamous histologies with fairly comparable outcomes and comparable improvements. This is a regimen where histology will dictate which chemotherapy you deliver with those first 2 cycles. But the overall approach of giving 2 cycles, a limited course of histology-specific chemotherapy with dual-checkpoint blockade, should apply across histologies.

 It’s natural for us to compare trials. The CheckMate 227 trial and the CheckMate 9LA trial were for all histologies, whereas the other randomized trials in this setting—for instance, KEYNOTE-189, KEYNOTE-407, IMpower150, IMpower130—were histology specific. We have data for chemotherapy–I/O [immuno-oncology] in nonsquamous lung cancer and chemotherapy–I/O in squamous cell lung cancer, and overall survival tends to be better in the nonsquamous setting compared with the squamous setting.

Keep in mind that for CheckMate 227 and 9LA, what we see is a composite. If we look at histology-based outcomes like 227 and 9LA, we see better survivals in the nonsquamous group. In the CheckMate 227 study, nivo-ipi [nivolumab-ipilimumab] had a median survival of over 19 months for the nonsquamous group, and in CheckMate 9LA survival was over 17 months. A better outcome for the nonsquamous compared with the squamous.

We shouldn’t be carefully comparing the different trials because they’re not exactly in the same population or in the same countries. They have slightly different eligibility criteria, but it’s only natural that when we’re choosing treatment for patients we’re trying to get the most effective treatment. It is useful for the different studies to make sure you’re looking at the outcome for the specific histologies, because that can be helpful in guiding appropriate choice of treatment.

 The CheckMate 9LA regimen giving 2 cycles of histology-specific chemotherapy with dual-checkpoint blockade is another option, and we have to tailor the treatment options to the specific patients we see. In someone with aggressive disease, a high burden of disease, that initial use of chemotherapy is appealing to prevent that early drop-off, that early progressive event. When you think you may only have 1 opportunity to deliver treatment for those patients, I think including cytotoxic chemotherapy is of value. It’s an interesting strategy limiting chemotherapy to those first 2 cycles. By preventing continued chemotherapy, will we avoid more cumulative toxicity? Will we enable more development of memory T cells by preventing the potentially detrimental effect of cytotoxic chemotherapy over time?

When we’re adding chemotherapy to dual-checkpoint blockade, there will be a cost. There is certainly more toxicity when we’re adding treatments to it. The rates of grade 3/4 adverse events and the discontinuation rates were higher certainly with the CheckMate 9LA regimen compared with chemotherapy alone, but some of the toxicity rates were better. The rates of anemia and neutropenia were a bit lower with the 9LA regimen, reflecting less cumulative toxicity from chemotherapy, perhaps less long-term neuropathy. It has to be for the right patient in the right clinical circumstance. There will be a role for it. Defining a specific role for all patients is challenging, and in the right clinical context it certainly can provide value.

The true measure for all these regimens will really be long-term survival and which approach leads to more durable survival at landmark rates. But we simply don’t have those data yet.

 It will make a big impact. Many patients want to avoid chemotherapy when they can or minimize exposure to chemotherapy. This will give clinicians an option when the patient has a significant tumor volume and they’re symptomatic from their cancer, and the clinician is concerned that if the initial treatment doesn’t control the cancer, the patient will get into trouble and then we won’t be able to get additional treatment. Have that option of giving the 2 cycles of chemotherapy first with the immunotherapy to make sure there isn’t early progression. But have the double immunotherapy onboard to try to get a high level of durable-immunotherapy responses, so the clinician can feel confident they’re minimizing the likelihood the cancer is going to progress early. Then maximizing the chance that they’ll get a durable immune response to therapy. This will help clinicians have an option where they’re trying to avoid or minimize toxicity from chemotherapy, maximize response to immunotherapy, and do it safely and effectively.

Another thing I’ll just point out is that when we initially had the results of CheckMate 227, we had anticipated that the best outcome might be in the PD-L1–positive lung cancers, which is evidenced by the fact that that was the primary end point of the study. But then we found that perhaps the most promising result was in the PD-L1–negative population, where it really looked there was a marked benefit compared with chemotherapy. CheckMate 9LA confirmed that, because when you look at the outcome for the experimental therapy in the PD-L1–negative lung cancer population, it does look quite positive and promising and really confirms the result in the PD-L1–negative patients from CheckMate 227, at least in my mind when I look and say, “Are these in agreement, or was the CheckMate 227 result just an outlier to me?” The 9LA regimen results confirm that and make me feel more confident about that result.

Personally and as an institution, we have a lot of experience with that combination. We were involved with the phase 1 study, CheckMate 012, which went back to 2013 and 2014. We have put a lot of patients on this treatment over many years. We’ve really learned how to use this regimen. The low-dosage ipilimumab 1 mg/kg every 6 weeks is very different from the regimen used in melanoma, where you might give 3 mg/kg of ipilimumab every 3 weeks.

Even in the lung cancer population, we all feel very confident using this kind of therapy. I have extensive experience using this regimen on patients in their 70s, and I have had a significant number of patients receive a full 2 years of treatment and come off therapy because they finished their 2 years but did not develop any substantial toxicity.

CheckMate-9LA: Histology and Frontline Impact 

 The CheckMate 9LA regimen is the most recent regimen that was added to our treatment arsenal. This was the first time in my knowledge that we had approval of the drug before we see any of the data. This is a combination of platinum doublet chemotherapy, histology specific, combined with dual-checkpoint blockade using the PD-1 inhibitor nivolumab and the CTLA4 inhibitor ipilimumab. The unique wrinkle here is that chemotherapy was given only with the first 2 cycles, not the first 4, and there was no indefinite maintenance chemotherapy.

The rationale and strategy is to provide a means to get that initial rapid cytotoxic control, to avoid that initial drop in PFS [progression-free survival] to provide that benefit for patients with aggressive disease or high burden of disease and facilitate an initial response. Then, toward the cycles beyond the second round, to lean more on that dual-checkpoint blockade for durable benefit and for more meaningful duration of response. When we look at the outcome from this phase 3 registrational trial, over 700 patients randomized, there was a survival benefit. Now, it’s important to look at the patient characteristics. About 30% were squamous, a little different from studies like KEYNOTE-024, where it was closer to 20%. If you look at the PD-L1 breakdown, we would typically expect about a third negative, a third positive low, and a third positive high. Here it was about 40% PD-L1 negative, so you just want to keep those populations in mind when we look at absolute numbers.

It was a randomized trial. Compared with chemotherapy alone, the chemotherapy–nivo-ipi [nivolumab-ipilimumab] approach did provide a survival benefit. Overall survival was 15.6 months versus 10.9 months with a hazard ratio of 0.66. The 1-year survival rate was 63% compared with 47% with chemotherapy. Broken down by histology in the nonsquamous group, we saw 17-month OS versus 11.9, hazard ratio of 0.69. In the squamous, 14.5 months versus 9.1, hazard ratio of 0.62. In both squamous and nonsquamous histologies, really across PD-L1 subsets, the addition of nivo-ipi [nivolumab-ipilimumab] to 2 cycles of chemotherapy provided a clear benefit over chemotherapy alone. The control arms didn’t necessarily underperform. If we look at the PD-L1–negative group, the survival hazard ratio was 0.62. In the PD-L1–high group, 0.66, which is remarkably similar across. One-year survival rates across PD-L1 subsets were 60% to 70%. We also saw the PFS benefit, 6.7 months with nivo-ipi [nivolumab-ipilimumab]–chemotherapy versus 5 months with chemotherapy alone with a hazard ratio of 0.68. When we look at those Kaplan-Meier curves for PFS, and we don’t see that initial drop. We don’t see that crossing of the early curves, just 2 cycles of chemotherapy preventing benefit. Some propose that may be countering hyperprogression, but in any case we see the favorable early benefit from adding chemotherapy in that long-term tail with a dual-checkpoint blockade.

 I’m going to talk about CheckMate 9LA. While CheckMate 227 was a long time coming and it took awhile to have approval for nivolumab and ipilimumab on the CheckMate 227 regimen, 9LA came on us all of a sudden. CheckMate 9LA was a randomized phase 3 trial for all-histology lung cancer, and the control arm was standard histology-based chemotherapy. The experimental arm was nivolumab and low-dose ipilimumab regimen from CheckMate 227 plus 2 cycles of chemotherapy.

The reason for the 2 cycles of chemotherapy is that with immunotherapy, sometimes we will see early progression. That’s particularly a concern in patients who have a high tumor volume or are symptomatic from their cancer, so this regimen gave only 2 cycles of chemotherapy. To prevent early progression, avoid long-duration toxicity from chemotherapy, then also provide the double immunotherapy for a chance for a higher number of immune responses that will give us a really long duration of benefit, a long-term benefit of 3, 4, and 5 years.

In this trial, the 9LA regimen was positive. It was superior to chemotherapy, and it was superior for overall survival, progression-free survival, and overall response rate. The duration of response was better for those who responded to treatment, and progression-free survival was better for the 9LA regimen compared with the chemotherapy.

For 9LA, the primary end point was overall survival compared with chemotherapy for all PD-L1 types. The FDA approval for the 9LA regimen is for any PD-L1 type, positive or negative, in non–small cell lung cancer.

The CheckMate 9LA regimen was unique in that it looked to give some chemotherapy at the beginning to avoid early progression of the cancer but minimize the exposure to chemotherapy toxicity, so only 2 cycles of chemotherapy were given. This is the period where in immunotherapy-alone trials we will sometimes see rapid progression of the cancer and in which the immunotherapy arm looks inferior to the chemotherapy arm.

I think this 9LA strategy with 2 cycles of chemotherapy was successful. We can look at the progression-free survival curve, and early on the nivo-ipi [nivolumab-ipilimumab]–containing arm was not inferior to the standard chemotherapy arm, but we gave the chemotherapy for only 2 cycles. And we don’t see as much cumulative grade 3 and 4 toxicity that we see in some of the chemotherapy doublet trials, where patients receive a standard course of chemotherapy plus a new therapy. The strategy for 9LA was successful, and it minimized the exposure of chemotherapy but prevented the early progression that we sometimes see with immunotherapy alone.

Recent Combination Approval: CheckMate-9LA Regimen 

: There was a good overall response rate, the immunotherapy. It was clearly better than we would have thought from single agent PD-1 checkpoint treatment, it was over 30%, which was very respectable. The duration of response was excellent. The duration of response was around 18 months which was very strong. The progression-free survival looked about the same as chemotherapy, which is not surprising for an immunotherapy alone combination. The response rate for the nivo/ipi was good, better than for chemotherapy. The duratioThe safety of the nivo/ipi combination was that it was safe and feasible. So certainly when you give combination immunotherapy you're going to have more side effects then if you give a single PD-1 checkpoint treatment. Originally when nivolumab and ipilimumab was developed, it was developed in melanoma patients and higher doses of ipilimumab were used, a more frequent dosing of ipilimumab was used. And that combination of nivolumab and ipilimumab was not safe or feasible in lung cancer patients. We did a large phase 1 trial called CheckMate-12 and we carefully evaluated the combination of these immunotherapies and determined that if we gave the ipilimumab, 1 mg/kg, every six weeks, that that was safe and feasible in lung cancer patients.

The safety of the nivo/ipi combination was that it was safe and feasible. When you give combination immunotherapy you're going to have more adverse effects then if you give a single PD-1 checkpoint treatment. Originally when nivolumab and ipilimumab was developed, it was developed in melanoma patients and higher doses of ipilimumab were used, a more frequent dosing of ipilimumab was used. And that combination of nivolumab and ipilimumab was not safe or feasible in lung cancer patients. We did a large phase 1 trial called CheckMate-12 and we carefully evaluated the combination of these immunotherapies and determined that if we gave the ipilimumab, 1 mg/kg, every six weeks, that that was safe and feasible in lung cancer patients. And this large randomized phase III trial confirmed that and so did a large phase II trial, CheckMate-568. You'll see about 30% to 40% grade 3 or grade 4 toxicity with a nivolumab and ipilimumab combination, which is more than you'd expect for instance than nivolumab alone. But less than 20% of patients discontinued any component of their treatment due to toxicity, which was a solid result. And there were no significant more deaths, in fact, there were very few deaths and comparable to the chemotherapy arm. While the combination immunotherapy is clearly more toxic than a single immunotherapy, it was safe and feasible and generally comparable to what one might expect for chemotherapy treatment.

: CheckMate-227 looking at dual checkpoint blockade with nivolumab and ipilimumab is chemotherapy-free, but it’s not toxicity-free. If you look at treatment-related adverse events, 77% incidence, although that’s lower than chemotherapy, which was just over 80%. Grade III to IV adverse events, similar, 33% versus 36%. If you look at treatment-related tox leading to discontinuation of therapy, that occurred about 18% of the time versus 9% with chemotherapy. The toxicity that we see with dual checkpoint blockade tends to be early. In my own practice I’ve been pleasantly surprised. I think a lot of it has to do with ipilimumab dosing. Using CTLA-4 in different cancers is challenging, and the dosing regimens vary from disease to disease. This low-dose approach using ipilimumab at a dose of one milligram per kilogram every six weeks has been much better tolerated than my previous efforts.

How the CheckMate 227 nivolumab-ipilimumab regimen fits into a first-line treatment of lung cancer is complex, which is good. We have a lot of good choices in first-line treatment. For tumors that have very high PD-L1s, so 50% or higher, and particularly for ones that are 70%, 80%, 90%, 100%, pembrolizumab alone is an approved therapy, and it can be very effective in that population with minimal toxicity. For patients with non–small cell lung cancer, a chemotherapy doublets with an immunotherapy such as pembrolizumab or atezolizumab were approved, and those are good options. The CheckMate 227 regimen of nivolumab and low-dose ipilimumab will give us an effective chemotherapy-free combination that we can offer to our patients who are not good candidates for chemotherapy, or for patients who for 1 reason or another want to avoid chemotherapy. A lot of people from personal experience or experiences with their families will want to avoid chemotherapy if they can. There may be patients who have particular aversions to nausea or hair loss or some other toxicity that is common with chemotherapy, so they would really prefer to have a chemotherapy-free option. The CheckMate 227 combination provides that, and you feel confident about that for most non–small cell lung cancer patients.

CheckMate-227 Regimen Safety Profile 

 CheckMate 227 was a randomized trial that included patients with stage IV or recurrent non–small cell lung cancer and no prior systemic therapy for advanced disease. Patients with known 

 fusions were excluded, as were patients with untreated, unstable brain metastases. It did include patients with both squamous and nonsquamous histologies. About 30% of patients in this trial had squamous non–small cell lung cancer. It’s a bit higher than some other studies, like KEYNOTE-024, where it was about 20%. You want to factor that in when you look at the numbers, as some populations do have a slightly worse prognosis.

In the 3-year update, the median survival for the PD-L1–positive group was 17.1 months, compared with 14.9 months with chemotherapy. That’s a hazard ratio of 0.79 for death. The landmark survival rates favored duel-checkpoint blockade. The 2-year OS [overall survival] rate was 40% versus 33% with chemotherapy, and the 3-year survival rate was 33% versus 22%. What I thought was maybe more compelling was the PD-L1–negative cohort, where the hazard ratio was better at 0.64. The difference in hazard ratio for death, 0.64 in PD-L1 negative, 0.79 in PD-L1 positive, was primarily because of the performance of the chemotherapy arm. The nivo-ipi [nivolumab-ipilimumab] arm was remarkably consistent when we compare the PD-L1 positive and negative. In the positive group, 17.1 months was the median survival; in the negative, 17.2. We usually expect a bit of a drop-off in the PD-L1–negative cohort. We didn’t see that with nivolumab and ipilimumab. While chemotherapy was factoring in a bit worse in the PD-L1–negative group, nivolumab and ipilimumab performed consistently well.

CheckMate 227 was a large, randomized phase 3 trial, and a major component of that trial was looking at the immune combination of nivolumab plus ipilimumab compared with chemotherapy. And the overall survival end point was the primary end point for this trial in PD-L1–positive lung cancer. This trial met its primary end point. It was clearly an improvement in overall survival for the immune combination, compared with chemotherapy that was both clinically and statistically significant.

In CheckMate 227, progression-free survival [PFS] was superior with dual-checkpoint blockade. Using nivolumab and ipilimumab compared with histology-specific chemotherapy, hazard ratio for PFS was 0.81 in the PD-L1–positive group, 0.75 in the PD-L1–negative group. The 3-year PFS rate was 18% among positive patients, 13% among negative. Response rate was also higher with just the nivolumab-ipilimumab combination, at 36% in the positive versus 27% in the negative group. What was most impressive, looking at the outcomes in CheckMate 227, was the duration of response. In the PD-L1–positive group, that duration of response was 23.2 months, almost 2 years, for a median duration of response. The PD-L1 negative cohort had an 18-month duration of response. These responses are remarkably durable when we consider the responses we’re achieving with dual-checkpoint blockade compared with chemotherapy-immunotherapy. When we look at chemotherapy-immunotherapy, some of the responses we see are those durable immunotherapy-based responses. But some are just chemotherapy responses, which are somewhat transient. When we’re using only dual checkpoint, the responses we see are immunotherapy-based responses, and those tend to be more durable. A duration of response of almost 2 years is quite compelling, and it’s difficult to ignore that.

ASCO 2020 Data Update: CheckMate 227

 We now have several FDA-approved immunotherapy-based regimens for patients with metastatic non–small cell lung cancer that lacks an actionable driver. We have checkpoint inhibitor monotherapy. We have dual-checkpoint blockade and multiple chemotherapy-immunotherapy strategies. We know that all these are superior to chemotherapy, but they haven’t been compared to one another. Our charge as oncologists is choosing among the available options. I’ve seen some very complex flow charts, but I find it challenging to devise a straightforward algorithm. We need to employ a very individualized patient approach. We have the most options for the PD-L1–positive subset. Although there are some advantages to chemotherapy-immunotherapy in terms of response rate and progression-free survival, particularly in those first 3 months of treatment, not all the patients we see are candidates for chemotherapy. Especially during the coronavirus pandemic, there was some value in avoiding myelosuppressive chemotherapy. That said, if there’s a clinical need, a sense that the disease is advancing quickly or there’s a large burden of disease, these combined chemotherapy-immunotherapy strategies do seem more appropriate. In the PD-L1–negative subset, chemotherapy-immunotherapy remains our standard, but the data for dual-checkpoint blockade are intriguing. Although nivolumab and ipilimumab are not technically approved for the PD-L1–negative subset, that’s where its use may be most valued.

There are many factors to consider in choosing an initial immunotherapy-based strategy. With regard to the tumor, I do factor in PD-L1 expression. At our institution we use the Agilent Technologies, Inc Dako 22C3 clone. Many lab-derived tests are utilized appropriately. Generally, we consider 3 groups: 0, low, and high. There is emerging evidence that the ultrahigh PD-L1 expression carry additional predictive power showing that PD-L1 expression is a continuous variable but is not perfect. It is dynamic. It is inconsistent.

We look at other factors among the tumor histology, particularly when you’re selecting chemotherapy backbone for combinations, as well as overall prognosis. Tumor mutational burden initially had some interest, but it’s been difficult to interpret, and there are a lot of differences among those panels. It does not factor into my decision-making. Increasingly, we also have to look at the host, looking at the patient. Patient characteristics are critical. Performance status [PS], comorbidities, these are going to influence suitability for specific regimens, particularly those that involve chemotherapy. We have to consider patient values, and I anticipate a lot of nonclinical factors as well that will come into the fold—schedule, cost, formulary restrictions—and those may end up tipping the scales when all other factors are equal. Each case really needs to be considered in its own context.

Performance status does factor into my treatment decision. We know patients with a poor performance status will have worse outcomes. In addition, they’re going to tolerate therapy less. When I look at performance status, I try to get a sense of how quickly the performance status is changing and how much of that performance status is related to the cancer. How much of that performance status decline is, in my opinion, reversible? Unfortunately, we don’t live in a world where all our patients are PS 0 or even PS 1, so we have to make some extrapolation. Our comfort with that extrapolation changes but also varies with the specific regimens that we’re choosing. If we feel a regimen is relatively well tolerated and safe, even in the context of a borderline performance status, we can go forward carefully. We all take certain measures to watch those patients a little more closely.

Nondriver Adenocarcinoma: Determining Therapy

: Maybe it’s changing an extensive-stage disease. We do see on all these immunotherapy curves there’s a better tail on the curve and who knows what’s going there. In limited-stage disease, this is on our list of potentially curable diseases with aggressive chemo/radiotherapy. This is a situation where I routinely use cisplatin/etoposide. There’s not much to say there from a medical oncology point of view. I’m a big fan of early intense radiotherapy, and I want Kristin to tell us what is early aggressive radiotherapy in this disease.

: Right now for radiation, we give 45 Gy bid [twice a day] with either the first or second cycle of platinum/etoposide. This was established as a standard of care by the Turrisi trial that was published in 1999, which confirmed what the CONVERT trial (that was published recently over the last 5 years) [showed]—the excellent contemporary radiation trial randomized patients to 45 bid or 66-Gy daily radiation with the hypothesis that the daily radiation to 66 Gy would be better. That did not end up being the case. There was no difference in overall survival between the 2 arms. Right now, people give either 45 Gy bid or 60 to 66 Gy daily. However, we did see an interesting trial presented at ASCO [American Society of Clinical Oncology], a small phase 2 Norwegian trial that was about 175 patients. They randomized patients to that standard 45 Gy bid compared with a dose-escalated bid radiation treatment course of 60 Gy twice daily. Their primary end point was 2-year overall survival and they did show significant survival difference: 73% versus 46%. This is a small phase 2 trial; we haven’t seen very much detail in regard to radiation technique. We don’t know much when it comes to local control, there wasn’t a difference in response rates between the arms. There’s also no difference in high-grade toxicity, which is puzzling. You would expect that with the dose-escalated radiation regimen you would get higher rates of esophagitis. I’m not sure what to make of that. I will say if you look at the control arm, it underperformed with median survival of just about 23 months. [In] the CONVERT trial, the 45-Gy bid median survival was about 30 months. I look forward to seeing this in publication so we can learn more details. I don’t think it’s practice-changing yet. It’s definitely interesting and I wouldn’t draw too many conclusions just until it’s tested in a larger phase 3 setting.

: It’s funny, I’ve always been a huge fan of the Turrisi trial, doing 45 Gy bid. It gets it done in 1 cycle and if you can start at day 1, by the time they get to cycle 2, they’re almost done with the radiation. They tend not to get their esophagitis until after the radiation is done so it all works out fairly well. How often is 45 Gy bid given?

: About 10% of patients get bid treatment. Logistically, it’s challenging. It’s difficult to schedule. Patients many times don’t like to come twice a day. In many radiation departments, it’s not in their routine to schedule somebody to come twice a day. For those reasons, for better or worse, it’s just not utilized as often as it probably should be. Even if it were positive in a large phase 3 trial, I still don’t know if it would replace 66 Gy given daily, which also gives great results.

: What are your views on PCI [prophylactic cranial irradiation] in a limited-stage setting?

: Right now, PCI is still standard of care in limited stage. It has shown to produce a survival benefit in meta-analyses; however, patients with limited-stage small-cell [lung cancer] are included in the SWOG MAVERICK trial, so we will be testing MRI surveillance in limited stage and I look forward to those results. It may be that perhaps we don’t need to do PCI if they’re getting q3 month MRIs. We don’t know that answer yet and I would hate to pull back on PCI in limited stage right now if the survival benefit in a patient population has the potential to be cured.

: We spent the majority of our first half of this thing talking about how immunotherapy has changed the standard of care in extensive-stage disease. Let’s get Steve and Roy’s opinion on limited-stage disease.

: I think that will be a big impact in the United States. There’s plenty of rationale to combine immunotherapy on the cusp of radiation. We learned in non–small cell from the PACIFIC study that there’s a potential synergy between them and that’s where we saw an improvement, not just in PFS [progression-free survival] but in overall survival, we now accept checkpoint inhibitors after thoracic radiation. Years ago early on, we had some concerns about the possible risk of pneumonitis. We didn’t know if it would be higher using immunotherapy after thoracic radiation but now, we’re very comfortable with that—the safety of that approach clearly established. Based on the rationale for potential synergy, once we start introducing checkpoint inhibitors in the limited-stage setting, hopefully we’ll see more cures. Our goal in limited-stage small cell with chemo and radiation is to cure patients, but we know that for the majority of patients, we don’t achieve that goal. Hopefully, use of checkpoint inhibitors will allow us to do that, and there are a couple of very important studies.

One of those studies is chaired by Dr Higgins and that’s using atezolizumab. Dr Steven Lin at MD Anderson did a small phase 2 [study] looking at concurrent thoracic radiation in non–small cell [lung cancer] with atezo [atezolizumab]—so radiation of the chest and atezo at the same time just to be sure there weren’t any worrisome signals in terms of pneumonitis. The rate of grade 3 pneumonitis was quite low, about 3%. Immunotherapy can [cause] modulation [in] the microenvironment and facilitate the efficacy or radiation and vice versa. This study is looking at atezolizumab with chemoradiation followed by atezolizumab maintenance afterward. Kristin, correct me, I think it’s NRG LU005.

: That’s currently enrolling, so we’re trying to send patients over to that study.

: Yes, we have just about 100 patients on so far.

: Roy, what about immuno combos in this setting? What do we know about the ADRIATIC trial?

: The ADRIATIC trial is looking at chemoradiation followed by durvalumab, and that’s certainly an approach. We’ve always been taught to use these drugs after and then to add tremelimumab. There's less enthusiasm for that right now given the results of that arm of the CASPIAN trial. These are curable patients; you want to watch out for the toxicity. The trial is ongoing, so it’s another setting. Perhaps in this setting, you can deliver the tremelimumab better, maybe it has a different effect after getting radiation but we’re not that optimistic given what we’ve already seen with CASPIAN trial.

: Yes, but certainly the concept building on the PACIFIC trial in non–small cell makes sense for the monotherapy arm.

: Yes, you can argue for that the effect of immunotherapy in a tumor that’s received chemotherapy and radiation therapy. You can almost imagine it would be nice to go study, but I think you’re warming up the whole microenvironment and making it ready for immunotherapy. As Stephen said, the toxicity of this is always a concern, but it looks like it’s quite manageable.

: Are there any parting thoughts to our audience in the general field of small cell? What’s the message you want? I’ll start with Kristin as our lone radiation oncologist.

: I’m excited that we have new therapies for patients with small-cell lung cancer. Immunotherapy has shown small but consistent survival benefits for our extensive-stage disease, and I’m excited about the prospects of moving that into limited-stage disease and curing more patients with small-cell [lung cancer].

: Although there are encouraging combinations looking at different immunomodulators, TIGIT for example, I think the only path forward to get those hazard ratios we’re used to in non–small cell [lung cancer] is with some type of reliable biomarker. We’ve done a lot of work looking at sequencing, looking at PD-L1, TMB [tumor mutational burden], there are just no obvious answers there. The greatest lead has to do with those transcriptional factor subsets looking at ASCL1, narrow D1, 

, maybe as a predictor for I-O [immune-oncology]. Without a biomarker, it’s only going to be incremental gains.

: We’re making progress. We’ve built on platinum, but I’d say we haven't made it all the way. I agree with the others that we need biomarkers, but as long as you’re having some activity, biomarkers become very viable because you can hopefully build on what we’re seeing. I think there’s so much more enthusiasm in the study of small cell than ever before. There are meetings on small cell alone. You recall Mark, the targeted therapies meeting we have every year, we always put small cell as the last session on Saturday. This year, we opened up with it. More investigators, more new ideas, and patients with this disease are benefitting from that. That’s a very positive thing.

: Thank you all. It’s been great. It’s been a great discussion. Hopefully, this will be helpful to our audience, and thanks for joining me today for this discussion on perspectives in small cell.

Review of Treatment for Limited-Stage Small-Cell Lung Cancer

: The same time that we had these data, we also had third-line indication for immuno-monotherapy in this disease setting, and I wanted to get the perspective. Dr Liu, clinically, how many patients get to second-line [therapy]? How many patients get to third-line? It’s a tough population, but what’s your experience there?

: There’s a lot of attrition with small-cell lung cancer, much more than with other cancers. We have published data in this setting. If you look at a group of patients who received first-line treatment, the data will tell us about half would get second-line and less than a quarter would eventually get third-line [treatment]. The approvals of nivo [nivolumab] monotherapy [and] pembro [

 monotherapy in the third-line are important approvals. But for a potentially transformative drug, they would have much less impact because patients wouldn’t survive long enough to receive third-line treatment. That’s why the front-line treatment with atezo/durva [

durvalumab] has the chance to make a much bigger impact.

: What is your standard second-line [treatment] following first-line chemo-immunotherapy?

: It’s a setting where I look for a trial if we can. Trials are my preference. If I’m taking a board exam, my answer is topotecan, but if I’m in practice, that’s a drug that I’m disappointed with. I find it very toxic; quality of life is not so great with topotecan. It’s a tough drug to receive and the responses are fairly limited. There are many other agents that are equally good or bad. Taxanes, irinotecan, temozolomide, if there’s CNS [central nervous system] efficacy. It’s an area where we have an unmet need for more potent agents.

:No. Topotecan—I hate using it. I have. We sometimes use taxanes here, but that’s where we send them over to phase 1. I still remember a patient I sent over to phase 1 years ago for an immunotherapy who is alive 7, 8 years later. Now, we’re in the post–I-O [immune-oncology] world. I think we look at I-O combinations if a patient is strong enough. Some of the ADCs [antibody drug conjugates], I think are worth exploring there, but it’s a tough area. At least, we’ve elevated from where we were in the past [by] having immunotherapy to begin with.

: We are seeing patients that get a fair durability to their responses on immuno-maintenance therapy. Is there a cut point in which you would think about going back or adding platinum/etoposide back in at some point?

: In the absence of data, it’s hard to say. If someone becomes resistant, add back some chemotherapy, and if chemotherapy is helping in the first place, probably knocking down some immune-resistant cells or regenerating some neoantigen [is a good idea]. It’s something to try and if someone is not having any issues with clinical progression, or if they’re just progressing radiographically, it’s worth a try.

: Any role for nivo/ipi [ipilimumab] in the refractory setting following first-line I-O? You mentioned I-O combinations earlier.

: We’ve done that and we’ve had trials. I’m a little disheartened in the fact that durvalumab/tremelimumab didn’t seem to have any effect up front because it’s a different situation. But nivo/ipi has phase 2 experiences, but you’re just giving the patient a lot of toxicity and the absence of any real data; I’ve not used it that much.

: I haven’t used it but the data look interesting. Lurbinectedin is a transcription inhibitor; it binds onto the DNA promotor region that prevents transcription. It also has an effect on tumor-associated macrophages, which may alter the microenvironment and help facilitate some immune responses. In a single- arm study of over 100 patients, it gave a response of about 35%. In [patients who were] platinum sensitive, that was up to 45%, and in [those who were] platinum resistant, 22%. It’s impressive for [patients who are] platinum resistant. We don’t have any drugs that are active in that setting. We’re waiting for the randomized data. We need a control arm there, and the ATLANTIC trial is looking at lurbinectedin with doxorubicin versus topotecan or CAV [cyclophosphamide, doxorubicin, and vincristine], and I think when we see those data, we’ll know more about just how active this drug is, but there is promise in that area.

: I would agree. When those data were presented at ASCO [American Society of Clinical Oncology], it was almost too good to be true; we were all very impressed and we’re waiting for the randomized trials. That could have a real mixture especially given our current paradigm using immunotherapy up front.

: Any appetite for PARP inhibition in this setting? There’s been some work, although I must say the data that I’ve seen have had a modest impact in this disease setting from my perspective. I don’t know if either of you have a different feeling.

: I find them exciting. When we think of PARP inhibitors throughout oncology, we often think of tumors that have defects and BRCA or a homologous repair, but these drugs also have some activity in small-cell [lung cancer]. Monotherapy didn’t really get us there. We remember the STOMP trial looking at maintenance PARP that didn’t really show much activity. I think there’s a lot more promise in combinations. It makes sense to combine a PARP inhibitor with a drug that will induce single-strand DNA breaks. Olaparib is a bit of a stronger PARP trapper than veliparib and in combinations with temozolomide for example, I think there’s some promising activity there. PARP inhibitors have more promise and maybe in combination with immunotherapy. We know from work that our colleagues have done that inhibition of PARP using PARP inhibitors, using CHK1 [checkpoint kinase 1] inhibitors will drive up PD-L1 expression and will induce T-cell infiltration. A CHK1 inhibitor will have an antitumor effect in small-cell lung cancer models, but if you deprive that model of CD8-positive T cells, it has no effect showing that its efficacy is medicated through the immune system.

Second- and Third-Line Treatment for ES-SCLC

: Kristin, I want to get your opinion. In the design of these trials we continue immunotherapy as maintenance. The 2 questions we used to address when we used to give 4 to 6 cycles of platinum/etoposide was A) is there a role for PCI [prophylactic cranial irradiation] in extensive stage disease, and B) is there any role for thoracic radiotherapy in this setting? I’m going to ask you that question first, and then if there are selected patients for which you would consider either, how do you integrate that when you’re giving maintenance immunotherapy?

: That’s a great question. Radiation oncologists now are wondering how PCI and consolidated thoracic radiation fit in with the results of the CASPIAN and IMpower133 trials. For PCI in the extensive stage setting, radiation oncologists have stopped recommending it due to the conflicting data. The trial led by [Ben] Slotman, [MD, PhD,] and the Japanese trial showed opposite results with improved survival with PCI given in patients who don’t get routine MRIs of the brain. The Japanese trial showed that there was not a survival benefit if we give surveillance to patients with extensive stage small cell lung cancer with q3 [every 3] month MRI. We’ve seen a shift in radiation oncology with not recommending PCI in patients with extensive stage small cell [lung cancer].

In IMpower133 only about 10% of patients received PCI, and in the CASPIAN trial PCI was allowed on the control arm, but not the experimental arm. We didn’t see any worrisome toxicity in the patients who did receive PCI in the Impower133 trial. But with PCI there is significant toxicity, and as a field, we’re trying to potentially move away from PCI. There are more trials now addressing MR [magnetic resonance] surveillance in both the limited stage and the extensive stage small cell population. The SWOG cooperative group is running the MAVERICK trial, for example, that will help answer this question. When patients do get PCI in extensive stage, there's also a trial run through NRG [Oncology], CC003, which is giving PCI with hippocampal sparing, which carves out the part of the brain that is responsible for creating memories to try to reduce cognitive changes. So as a field, we’re leaning away from PCI in this patient population and that’s where we stand now. I hope the SWOG trial gets completed. I think it’s a great trial, and I would encourage people to enroll patients to answer this question when it comes to PCI.

In terms of consolidated radiation, patients weren't allowed to get thoracic radiation on either of these trials. We do know now we have some safety data for the combination of thoracic radiation and immune checkpoint inhibitors, and there's not worrisome toxicity. The role of consolidation radiation to the chest and to metastatic sites is going to be evaluated in an NRG trial, NRG-LU007, which will activate probably late summer. I look forward to putting patients on that trial. Right now, I would say we’re giving thoracic radiation on a case-by-case basis in some patients who have had a great response and have a small amount of residual disease. We’re doing so cautiously, because again, there are a lot of unanswered questions. What dose do we go to? How do you integrate it from a timing standpoint? And this is a data-free zone. I think we’re best to put patients on NRG-LU007 to answer the question in a scientific way.

: There was a suggestion from a very old trial published in the late 1990s, an Eastern European trial, that suggested in those patients who have a great response, maybe you could get some mileage out of thoracic radiotherapy, and as well with the follow-up Slotman trial. Still controversial, we still think about it and talk about it, but I’m glad you’re doing NRG-LU007 to help bring some more clarity to it.

Role of PCI Versus Brain MRI Surveillance in ES-SCLC

: Roy and Stephen, what do you do in your own practice? Do you have a preferred agent?

: Atezolizumab was there first, and all else being equal, the drug there first usually gets more share. We’re using mostly atezolizumab now. I didn’t see anything in the data to switch. I will say the pembrolizumab data look very close. What it tells me is that we need a better biomarker here. The fact that all these trials are so close, it tells you that we have a population that’s heterogeneous, and some are probably benefitting more than others, and if you take a mixed population, you're going to see results like this. It’s great for patients, it’s clearly better than what we were using before, but we’re using atezolizumab at Yale [Cancer Center].

: It comes down to comfort. IMpower133 was presented in 2018, which was approved March 2019. Durvalumab was approved in March 2020. I think what might have changed it, if we had seen the third arm in CASPIAN, the tremelimumab arm, showing any benefit in survival, then that would have been a combination to pursue. But as you saw at ASCO [the American Society of Clinical Oncology annual meeting], the addition of tremelimumab did not improve survival over just chemotherapy alone. [Atezolizumab] and [durvalumab] both improve survival, And both are reasonable options. I didn’t see a compelling reason to change. CASPIAN, [durvalumab] is given every 4 weeks. We also have FDA approval for [atezolizumab] to give every 4 weeks. It comes down to comfort for me, and I’m still using atezolizumab.

: Roy, you mentioned we did see at ASCO the third arm of the CASPIAN trial, which was the durvalumab/tremelimumab arm. The interesting thing to me is that there didn’t seem to be any optimism about that combination, and I think the tremelimumab was stopped at a certain point. And the overall survival curve with continuing durvalumab alone eventually caught up with the arm that had only durvalumab in it. Roy, what are your thoughts, any hope for anti–CTLA-4 inhibition in this setting?

: I would say not. We already knew from trials with ipilimumab years ago, our colleague ran some of those trials way back when, that there was no real benefit with ipilimumab CTLA-4. And here, adding durvalumab clearly adds toxicity, but unfortunately, no benefit, as you mentioned. I think if there are going to be combinations in small cell lung cancer—and there have to be, because these data, while good, are clearly not a home run yet—we’re going to have to look at other aspects of the immune system and try to scientifically understand why there is some primary resistance to these agents. For now, I would stay away from the CTLA-4 inhibitors. That brings the grade ¾ toxicity up to the 40%-plus range, which is high in these very sick patients.

: Roy, you brought up before the issue of biomarkers. Obviously, in non–small cell [lung cancer], we routinely measure PD-L1. Any role for that in small cell?

: I think yes. The problem is with PD-L1 both in small and non–small, Mark, is measuring it accurately. It’s hard enough to measure in non–small cell where you get good pieces of tissue and there isn't crush artifact, and so forth. In small cell it’s hard to get good core biopsies and really measure it accurately. I think it’s a measurement issue, but I think PD-L1 probably has some role but we've not seen it so far. TMB [tumor mutational burden] could certainly be useful. There are a lot of data recently emerging with whole exome TMB, looking at 175 mutations per megabase in some of the data with pembrolizumab. I wonder if that could someday be used with small cell. All these need to be incorporated in the trials. Perhaps all of the groups that have done these trials could collaborate and pool their data sets to look at this.

Expert Approaches to Treating Extensive Stage SCLC

: With both of those being anti–PD-L1 antibodies, I’m going to ask Dr Liu, we are coming off the tail of our first virtual ASCO [the American Society of Clinical Oncology annual] meeting this year. And at that meeting we saw the presentation of 2 trials with PD-1 inhibitors, KEYNOTE-604 with pembrolizumab, and the ECOG-ACRIN 5161 trial, which was actually a phase 2 trial with nivolumab. Can you walk us through those and give us your impressions?

: Yes. These are 2 randomized studies and we saw results for the first time. The phase 3 trial you mentioned, KEYNOTE-604, was presented by Charles Rudin, [MD]. This was a very well-designed randomized, placebo-controlled large phase 3 trial. It included patients with previously untreated stage IV small cell lung cancer. All patients received platinum/etoposide, allowed either carboplatin or cisplatin, with a 1:1 randomization to concurrent pembrolizumab or placebo, followed by maintenance [pembrolizumab] or placebo. The study had 453 patients, of note, the vast majority, about 70%, received carboplatin. And it had dual primary end points of progression-free survival [PFS] and overall survival [OS]. What we saw at ASCO 2020 was the final analysis, and in that analysis KEYNOTE-604 was positive in that it hit its progression-free survival end point. The PFS favored pembrolizumab a median of 4.8 from 4.3 [months], the hazard ratio there was 0.73, the 1-year PFS rate from 5% with placebo to 15% with [pembrolizumab].

Unfortunately, overall survival did not cross the threshold for superiority. There was a numeric difference, median from 9.7 to 10.8 [months], hazard ratio of 0.80, but it did not cross that statistical threshold for a positive OS benefit. It’s interesting, if you take away 1 patient who inadvertently received the wrong treatment, it was a positive trial, just to show how close it was. We learned from KEYNOTE-604 there is activity there, comparable, in the same ballpark, as we saw with PD-L1 in IMpower133 for atezolizumab, and in CASPIAN for durvalumab. But unfortunately, this will not impact practice because it did not improve survival; while the trend was encouraging, it didn’t quite get there.

The ECOG-ACRIN 5161 was a randomized phase 2, through the cooperative group. It accrued remarkably fast, and this study looked at platinum/etoposide with or without nivolumab. No placebo, choice of platinum, this was a smaller study, a phase 2, 160 patients, and this study accrued in 7 months. Primary end point was progression-free survival, and that improved with nivolumab from 4.7 to 5.5 [months]. It’s a hazard ratio of 0.68. Even though a relatively modest size study, it was able to show a PFS benefit and was able to show a survival benefit. The OS hazard ratio there was 0.73. This was a smaller phase 2 positive for both PFS and OS. In a world where we have multiple randomized phase 3 trials showing an OS benefit with atezolizumab, durvalumab, unfortunately, no immediate impact in our clinic.

KEYNOTE-604 and ECOG-ACRIN 5161 Trials for ES-SCLC

: I’m Dr Mark Socinski from the AdventHealth Cancer Institute in Orlando, Florida. Joining me today in this virtual discussion are my colleagues, Dr Roy Herbst, a medical oncologist from the Yale Cancer Center in New Haven, Connecticut; Dr Stephen Liu, a medical oncologist from the Georgetown Lombardi Comprehensive Cancer Center of Georgetown [University] in Washington, DC; and Dr Kristin Higgins, a radiation oncologist at the Winship Cancer Institute of Emory University in Atlanta, Georgia. Today we’re discussing a number of topics pertaining to the use of systemic therapy in patients with small cell lung cancer. We’ll discuss the latest research in the field and the impact of recent clinical trials on making decisions for treatment selection in this population. Dr Herbst will summarize recent trials because in the past year, we’ve seen a change in the standard of care based on 2 trials, IMpower133 and the CASPIAN trial. Could you bring us up to date and give us your impression of those trials?

 Yes. Immunotherapy has changed the way we think of cancer care, certainly lung cancer care, and it’s moved to small cell lung cancer. It makes sense. Small cell lung cancer is a smoking-related disease, there are large numbers of mutations, high TMB [tumor mutation burden], and it’s an obvious place to look for a new indication. The first drug to move forward was atezolizumab. At the World Conference on Lung Cancer 2 years ago in Toronto, the results were presented of IMpower133, which took atezolizumab plus platinum and etoposide in extensive stage small cell lung cancer. In my career, I haven't seen any changes in the standard of care treatment for small cell lung cancer until now. This randomized phase 3 study had a hazard ratio in the 0.7 range in favor of the atezolizumab, and that was approved and has become a new standard of care. Just right on the heels behind that is a second drug, durvalumab, another PD-L1 inhibitor, in combination with platinum/etoposide. The CASPIAN trial, which we learned about late last year, also was positive and updated at ASCO [the American Society of Clinical Oncology annual meeting] this year, again, with a hazard ratio between 0.7 and 0.8.

We have activity here; it’s a survival benefit. Its modest, but clearly, it’s at the level where we consider that clinically significant, and it’s something to build upon. For patients with small cell lung cancer, it’s great that we have these 2 drugs available in the frontline setting for this disease.

Stephen Liu, MD, MedStar Georgetown University Hospital

Articles