ESMO's Practice-Changing Lung Cancer Data: Three Experts Discuss

H. Jack West, MD; Hossein Borghaei, DO, MS; Sanjay Popat, MBBS, PhD


November 06, 2017

H. Jack West, MD: Hello. I am Jack West, medical oncologist and medical director of the Thoracic Oncology Program at the Swedish Cancer Institute in Seattle, Washington. Welcome to Medscape Oncology Insights, coming to you from the 2017 Congress of the European Society for Medical Oncology (ESMO).

Today we will discuss some of the exciting new lung cancer data presented at ESMO. Joining me for this discussion are Dr Hossein Borghaei, chief of Thoracic Medical Oncology at Fox Chase Cancer Center in Philadelphia, and Sanjay Popat, a consultant thoracic medical oncologist at the Royal Marsden Hospital in London.

This was a banner year for lung cancer, at ESMO and globally. Among the most important presentations is the PACIFIC trial,[1] which finally seems to have broken the impasse we have had for more than a decade in locally advanced, non–small cell, stage III, unresectable disease. For all that time, we have tried to improve on our rather dismal results. Yes, we can cure a minority—maybe 15% or 20% of patients—with chemotherapy and radiation, but the standard has been a couple of cycles of platinum-based chemotherapy with concurrent radiation.

We have tried increasing the chemotherapy, but that didn't work. We have tried escalating the radiation; that was harmful. We have also tried adding targeted therapies, but in general, that has been a mistake and even caused some harm.

The PACIFIC trial was a large study including more than 700 patients who had received at least two cycles of concurrent chemotherapy and radiation therapy, followed by 1 year of durvalumab administered every 2 weeks, or placebo. The results were highly significant, at least as measured by progression-free survival. Overall survival had a strong trend in the right direction, but this was not statistically significant.

What is your take? Are these results enough to change practice?

This is the first big breakthrough we have seen in radically treatable stage III disease in years.

Hossein Borghaei, DO, MS: This is a very good question. Ultimately, what matters to the patients, obviously, is their overall survival. They want to know whether they are going to live longer if they get additional treatment beyond the usual chemoradiation.

We all know that chemoradiation is a tough regimen, so you really have to see a clear survival advantage and good data to tell patients that after everything they have gone through, we are going to give you another year of another drug, or in this case, consolidation with another drug.

Let's break this apart. Is the progression-free survival that was reported with a single drug [durvalumab], which seems to be well tolerated, significant? I believe it is significant. Is it going to be practice-changing? I believe for most physicians, it will be practice-changing. I will be inclined to offer this to my patients who still have a good enough performance status. I believe that what will absolutely nail this as the standard of care across the board will be the overall survival data.

Dr West: I should clarify: It wasn't that it's not significant; it's just that the overall survival data were not presented. This was an early interim analysis, and overall survival was not part of that analysis.

Dr Borghaei: Exactly.

Dr West: The progression-free survival was not just a median difference, and in fact, the medians were remarkably different. Progression-free survival was 5.6 months with placebo versus 16.8 months with durvalumab. That is remarkable, and it was sustained at 12 months and 18 months. That still holds up. I believe it is important that we all want to see a survival difference, and we don't know whether we are delaying a relapse or actually going to increase that cure rate.

Sanjay, what are your thoughts?

Sanjay Popat, MBBS, PhD: This is an enormously important trial. We have anticipated this trial for a long time, and I believe the results are quite clinically meaningful. This is the first big breakthrough we have seen in radically treatable stage III disease, for years. This is big news.

Progression-free survival is reported; it is a co-primary endpoint, and that is the important thing here. Overall survival, yes, is still to be reported, and I look forward to those mature data. Not only was progression-free survival statistically significant, but it was clinically meaningful as well, with a wide separation of the curves.

Yes, there are caveats concerning how we report progression-free survival in terms of how radiology marks it as a progression event. Actually, that would have probably been washed out in the randomization process, so to me, the magnitude of the benefit we see in progression-free survival is meaningful.

I would be very surprised if that magnitude of difference does not translate to a significant and clinically meaningful survival advantage. Do I believe this is worthy of changing practice? Yes, I think this potentially is. This is a sea change in how we should be thinking about our stage III patients.

Dr West: I would say that on one hand, we have not tended to rely on imaging so much after chemoradiation because honestly, it is kind of a murky mess afterward. So interpreting what is viable and what is postradiation change is fraught with peril.

On the other hand, this study did have a blinded review committee, there was a randomization, and the differences are not subtle. We see an enormous difference in these curves and it is sustained over time. I agree, it is extremely likely and hard to envision that the separation sustained at 12 and 18 months and ongoing, will not be accompanied by at least some prolonged period of interval overall survival difference.

Practically speaking, I suspect this will be enough for durvalumab to be approved. It's going to change my practice. I will definitely use it. What are your thoughts?

Dr Borghaei: It definitely will change my practice. I want to bring out a couple of points and see what you think about it. We do not know what kind of chemotherapy was given to these patients. That was not presented. Going back to the paper, we have to see whether we can tease that out, because there are some practice differences here and there.

I can tell you that where I practice, when I give second opinions, many of the community physicians around me like the weekly carbotaxol, which requires two cycles of consolidative care. So how do we make this work with that?

Dr West: The study used a minimum of two cycles, but nothing was specified beyond that.

Dr Borghaei: Exactly. What you imagine is something like a SWOG (Southwest Oncology Group) regimen that includes etoposide—something to that effect. We need to figure out how that can be incorporated into your practice if you are using weekly carbotaxol, and how you provide the other chemotherapy. That is one part of the study I would like to see in more detail, but again, if I have access to the drug and I have no issues with reimbursements or anything like that, if I see a patient with a good performance status, who has done really well and has stable disease after chemoradiotherapy, my preference would be to offer the durvalumab as consolidative therapy. I cannot argue with a median progression-free survival over 16 months with an 11-month advantage compared with the control arm.

Dr West: Sanjay, what is likely in the United Kingdom? It is obviously a different climate.

Dr Popat: In the United Kingdom, things are slightly different. We have a health technology assessment process that needs to occur before the drugs are approved for general reimbursement by the state. There are mechanisms in place for access before licensing, and they are very much driven by the sponsor. I very much hope that the manufacturer will consider applying to one of the schemes that exist to allow access to durvalumab to the UK population.

Overall, we will need to see some cost/benefit analysis or projected overall survival data to ensure that the regulatory reimbursement authorities, at least in the United Kingdom, judge that this provides a benefit. In terms of clinical benefit so far, I don't think we can argue too much about it.

Dr West: We have approval in the United States for what we might refer to as the KEYNOTE-21G approach: carboplatin and pemetrexed with pembrolizumab in an unselected population. There is some controversy about whether that should be broadly adopted, because it was a phase 2 trial of 123 patients and although the study found a dramatic progression-free survival benefit with a potential for crossover, it is not as clear that there is a translation to overall survival.

Dr Borghaei, you presented an update [here at ESMO].[2] Can you talk about the salient points there, and whether you believe the results change things?

Dr Borghaei: Will it change things? That's a much deeper question. What I presented was the fact that now, with 18.5 months' median follow-up, the response and progression-free survival are still in favor of the triple combination: pembrolizumab, pemetrexed, and carboplatin. That is good to see. You want to see long-term follow up showing that progression-free survival and response are still good, but now we are also seeing some survival advantage with the addition of pembrolizumab.

I must caution everyone, however. Obviously, this was almost a sort of post hoc analysis. It was not part of the primary analysis, so I don't feel comfortable saying it was definitely statistically significant because the power [in the study] was spent, but it clearly shows that the group that received the pembrolizumab now has a better survival advantage.

The crossover is always going to confound things. How do you interpret that? Does it simply mean that because they received an active drug after failing chemotherapy, everyone lived longer? I do not know whether we can make those kinds of arguments. The numbers, as you suggest, are too small for us to know that. We don't have survival [according to] level of programmed death-ligand 1 (PD-L1) expression. There is a lot of argument that maybe the group with the highest PD-L1 expression sort of drove the benefit. I don't know, and I do not believe that has been proven.

We also know that the phase 3 study hopefully will be presented relatively soon—at least the PFS [progression-free survival] part. Because I'm looking at a median of 18.7 months to get a survival advantage [in a phase 2 study], I'm thinking that even for the phase 3 trial, we will have to wait a bit longer.

With all of those caveats in mind, should this be the standard of care? On the basis of principles, phase 3 trial results should be the determining factor as to whether this should be the standard of care. Am I using it in my practice? Again, if someone has a PD-L1 over 50%, I am still sticking with single-agent pembrolizumab. In patients who are negative or low PD-L1 expressors, or somebody who has a huge amount of disease burden, I think I'm comfortable using the combination, because we have the approval. I am looking at something in the neighborhood of a 55% response rate with a combination—it is hard not to offer that to a patient whom you believe absolutely needs it.

Dr West: I have been less inclined to embrace it broadly, although I agree that in the subset of patients with a large tumor burden or whose disease is clearly progressing quickly enough that we would not be confident they will get to second-line [therapy], I believe that approach always made sense.

But I have been concerned that if you can cross over and get the same benefit, why rush the process? On the other hand, when you see these widening differences in survival curves, it reassures me that this is almost certainly not a bad choice. Therefore yes, I absolutely want to see the phase 3 trial results, but I doubt that they will be wildly discrepant, which is reassuring. That said, the statistics of it, taking multiple looks and stopping when you are happy and when it reaches P value significance, is not quite valid.

I do not believe there is anything absolutely magical. The numbers of patients do not change certainly, but to me it is more reassuring to see that it only looks stronger over time. Sanjay, what are your thoughts?

Dr Popat: The key issue with this trial is assessing what chemotherapy brings to the equation in the patients with a tumor proportion score above 50%, because this is a group that we traditionally treat with frontline pembrolizumab monotherapy. How much more bang for the buck am I getting by adding my platinum doublet chemotherapy? Is it worth it?

These are the types of details that we would need to see from the longitudinal time-to-event analysis, be it progression-free survival stratified by PD-L1 status. That really helps us figure out what the combination brings that monotherapy does not bring, and who we should be thinking about giving this sort of combination to.

Certainly, the separation that we are now seeing in the overall survival curves are really quite impressive, and I do not think we can argue with that. The theme of the program is, are we getting a magnitude more benefit by bringing immunotherapy forward into this stage of the disease? Potentially, by combining immunotherapy with other standard-of-care agents, we may be seeing that [greater benefit]. I am quite optimistic about what the future may hold for combination therapy, but at the end of the day, the proof is in the pudding. Let's see the data.

Dr West: The last presentation I want to cover is the CheckMate 153 study, presented by Dr David Spigel,[3] that started out with more than 1000 patients and got winnowed down. This study looked at whether there is value in continuing treatment versus stopping after a fixed duration—in this case, 1 year.

He started out with a large number of patients in this second-line trial with nivolumab. The patients who received nivolumab and had not progressed at 1 year were randomly assigned to continue treatment indefinitely, or stop with the potential to resume. I was rather surprised to see such a difference favoring continuation. I believe many of us have been unsure or ambivalent, and thought there wouldn't be as much of a difference as we saw in this study, with even a trend in terms of survival. Does this answer the question for you? After all is said and done, we are talking about not much more than 100 patients included in the randomization portion of the study. What is your take-home point from this?

Dr Borghaei: This is hypothesis-generating. It is interesting. I am not sure I am sold on the idea that I need to continue [nivolumab indefinitely]. I am still thinking that maybe a shorter duration would be adequate. The study was designed to assess toxicity, as was discussed after the presentation, so it did not have the power to look at some of the outcomes we want to look at.

But it is definitely an interesting study, because it points to the fact that looking at the duration issue in a more systematic manner is important. It has health economic benefits and patient quality-of-life benefits. If you do not have to bring the patient in every 2 weeks to give them drugs for 2 years, I believe that makes a huge difference in the quality of life.

I am with you. I am surprised that it showed progression-free survival and a trend toward overall survival [with continued nivolumab]. I did not expect anything like that. It is too small and not statistically designed to look at that survival answer clearly, but the results tell me we definitely need to look at this.

Dr West: Sanjay, obviously the economics loom large here, too. When you put it all together, does this lead you to believe that continuing nivolumab indefinitely is the right thing to do?

Dr Popat: We do not really know the answer to that question statistically. However, what we have here is an exploratory analysis, because it is not powered to answer the question. But the difference was quite large. This was a big difference. This was not just a laser pointer between the curves; we saw a big difference in outcomes.

Yes, the overall survival was not statistically significant; what we have is a meaningful difference. It is underpowered, but we definitely saw a difference in the progression-free survival, and it was much larger than I would have anticipated. In fact, there is a bit of a disconnect because many of us thought that if you stopped treatment early because of toxicities, you would still have that ongoing benefit, so this data set is a bit discordant and discrepant.

That said, with these data in the public domain, I believe it is quite difficult to say, "Well, maybe after a year we should stop treatment," when we have clear evidence of some ongoing benefit in progression-free survival in that situation. I wonder whether one of the questions we should be asking is about intensity of dosing. Do we really need to carry on with that schedule? Do we need to front-load access by schedule, and then be a bit less dose-intense?

Dr West: That is a third option. This is a world where there is no clear standard. We are not extending beyond 1 year as the standard; we just don't know what exactly to do. These trials generally used a continuous therapy, so in this situation, I am swayed that the differences were of a relatively large magnitude. I am no longer on the bubble about this right now, at least in the absence of more data.

Thank you both for joining me. This was a great discussion on some very relevant issues.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: