 There was a presentation at ASCO last year, I believe it was by The University of Texas MD Anderson Cancer Center group, and they describe this pseudo-progression where the mediastinal adenopathy looked worse. Obviously we get concerned about disease progression, although it’s pretty quick disease progression. They demonstrated that even in these scans, if I remember the scans correctly, the PET scans were pretty frightening in terms of bringing that patient to the OR, but they found that this was more reactive adenopathy and not pathologic adenopathy.

Any experience or any perspective on that from your vantage point?

 I can comment a little. They referred to it as nodal immune flare, and they actually saw it in 10% of those patients and that was nivolumab versus nivolumab-ipilimumab and I think it was similar in both arms. We’ve seen it in the nodes, and we’ve seen it a couple of times in the lungs, at the same time each neoadjuvant group. I’ve seen a couple of cases of progression, and the take-home point is to biopsy everything. If you see progression on the PET scan, don’t assume it’s 1 thing. Don’t assume it’s just reactive disease either. Its critical to biopsy and know what you’re dealing with.

 Stephen and Roy, any experience in that area?

 I’ve shared several patients with some of the Johns Hopkins groups. We’ve seen that it takes a trained eye to recognize that. While the surgical complication rates were fairly low, you’re talking about highly trained surgeons. You’re talking about centers with a lot of experience. As neoadjuvant immunotherapy approaches become more integrated into standard care—right now only done through trials—it’s going to take some education, so we don’t assume someone has progressed when actually they’re having a great response to treatment, and so we don’t forgo surgery when we should be pushing forward. It’s going to take a little bit of education and a little bit of retraining to understand that there are some atypical response patterns here.

 All I can add is it speaks to the need for the multimodality tumor board and discussing these cases very carefully with your surgeon, pathologist, and oncologist.

 Thanks, all. This has been a fabulous, rich, and informative discussion. Before we close the program tonight, I’d like to get some final thoughts on this early stage non–small cell lung cancer from the resectable adjuvant setting to the PACIFIC setting if you will. I’ll start with Dr Herbst.

 I’m excited, Mark. It’s been 23 years since the advent of EGFR inhibitors. We’ve seen them used in the most metastatic unselected setting to selected with 

 mutations discovered to more active and less toxic drugs like osimertinib, and now we’re seeing them used in the adjuvant setting. Still work to do, but very promising early results that I believe are practice changing.

Immunotherapy the same, we’re using it in the earlier stages of disease. One thing we have to think about with immunotherapy is we have to learn from the targeted therapy realm. We’re using immunotherapy in all patients, but clearly we need biomarkers. We need to find those patients in whom single immunotherapy works best. Some might need combinations, and early stage disease is going to be the perfect place to find that because we’re going to get tissue. We’re going to understand it.

The progress has been overwhelming with improved benefits for our patients.

 Thanks, Roy. Kristin, your final thoughts?

 I’m excited about moving the needle in our stage III unresectable patients. The PACIFIC was a game changer. In the years to come we’ll have a lot of data about how to optimize the synergy between radiation and immunotherapy and even moving targeted therapies into the stage III setting with the LAURA study that’s ongoing. The targeted therapies and immunotherapies are moving the needle for a curable disease, which is really exciting for our patients.

 What we’re seeing happen is our standard of care is rapidly changing for stage III unresectable, for resected 

 positives, soon for preoperative setting for resectable non–small cell lung cancer and as our standard of care changes It’s important to keep referring patients to those trials. While these adjuvant and consolidation therapies have significantly improved outcomes, there's still a lot of room for improvement. Building on those as a backbone, taking the next step beyond these studies, will continue to be really important.

 I agree with all my colleagues. It’s an incredibly exciting time in lung cancer, early stage lung cancer in particular. I’m particularly excited that the science has led us to where we are and that it’s developed all these pathways that we now have to treat patients.

The data you heard presented puts the onus on surgeons and others involved in staging these patients to stage them well, to think about testing for molecular observations, and to have multidisciplinary consults on every early stage lung cancer patients because there are just so many options available.

 I must say, when that press release came out looking at the 2017 reduction in cancer mortality and lung cancer led the way, I’m sure all of you felt it. At least from when Roy and I started in this business and others, it was a pretty nihilistic disease setting. Therapies were not that effective, and many felt it wasn’t a very treatable disease. Now we’re at a time where we led the way in reduction of cancer mortality, so it’s very gratifying.

I want to thank all of you for joining me today. I want to thank our viewing audience. We hope this 

® discussion was useful and informative. Thanks for joining us.

Videos

Multidisciplinary Management of Patients With Stage III NSCLC

 What are people’s thoughts about integrating immunotherapy? Let’s get away from the 

 ADAURA population. The integration of preoperative immunotherapy strategies, either immunotherapy alone or combination chemotherapy and I-O—to me, that data look a little more optimistic. I’ll start with Roy and get your thoughts on the preoperative setting.

 Neoadjuvant therapy has a lot of promise. I would keep the chemotherapy in there for now unless you have a highly selected patient. But with chemoimmunotherapy neoadjuvantly, you can look at the tumor at the time of surgery, get a sense for what’s sensitive and what’s resistant, and design your therapy outback. It’s very promising—perhaps more surgery could be done. I’m very much in favor of these approaches. They’re tough to do, but if you have a good multimodality team and meet with tumor boards and review the cases, it is quite possible.

 Whether you’re using immunotherapy alone or immunotherapy with chemotherapy, we are seeing major pathologic responses and the combinations of cytotoxic chemotherapy with checkpoint inhibitors show pretty impressive NPR [noncomplete pathologic response] rates and we’re seeing some pathologic CRs [complete responses]. What we don’t know is if that will translate to better long-term survival. Is that a surrogate for developing memory?

While chemotherapy will drive some of these initial responses up, are we going to sacrifice some formation of memory on the tail? That’s been a theoretic concern that we just don’t know the answer to. We’re seeing very exciting trials. Just for the biologic insight alone, the neoadjuvant has a window of opportunity. The trials really tell us a lot of what’s happening to that tumor and will help us hopefully unlock how to best use these drugs.

 Yes. Let me extend the number of modalities. For the most part, if you do induction treatment in the stage IIIA population, we use chemoradiotherapy. We just had a little audience about how great immunotherapy and radiotherapy may play together. Is there a role for preoperative chemoradiotherapy, Kristin, in the preoperative setting? What are your thoughts about that approach?

 When it comes to neoadjuvant therapy, if you look at the trials, radiation doesn’t add a whole lot to neoadjuvant chemotherapy. I haven’t been a fan of adding radiation preoperatively. I think it’s institutionally specific in terms of what practice patterns are. I do think if you introduce radiation to chemoimmunotherapy neoadjuvantly, there is a high risk of toxicity, so it certainly would make your surgery more difficult.

There was a very small study presented at ASCO out of the Cleveland Clinic that looked at cisplatin-etoposide-radiation to 45 gray with pembrolizumab followed by resection, and the stopping rules for infeasibility were met. There were some high-grade toxicities, so that is not continuing. I would worry about excessive toxicity with introduction of radiation neoadjuvantly. I would favor more of the chemoimmunotherapy approach.

 Brendon, what are the surgical concerns? Do you have a sense? It doesn’t sound like the feasibility of tri-modality therapy with chemoradiation immunotherapy. At least the early read is that it’s probably not feasible.

In terms of the immuno-oncology alone versus chemotherapy–immuno-oncology—and then from the surgical point of view—once you’re in the operating room, are there any issues that you have found that are different in this population?

 Yes, it’s become almost urban legend how hard some of these cases can be, and certainly some are. That’s particularly true if you’re waiting more on the salvage setting where you’re taking somebody who has had prolonged immunotherapy and trying to help them.

There have been great surgical perioperative results, both in the trial that was published in the 

 from Johns Hopkins Sidney Kimmel Cancer Center and Memorial Sloan Kettering Cancer Center and also from the LCMC3, a large multi-institutional trial with really low rates of perioperative events and similar surgical complications that we would see in an induction setting. I’m not sure there’s anything uniquely different about these patients. Perhaps they’re a little harder to assess. If they’ve had a great nodal response, they can be challenging. Sometimes defining a nodal response is challenging. These are cases we can often do as we would normally and that often can be done minimally invasively. About two-thirds in our trial and the trial that I just mentioned from Columbia University and MDH, they had almost 50% of minimally invasive surgery in N2 patients and stage III patients. Surgeons just need to do what they would normally do and try to do it as safely as they can.

Integrating Perioperative Immunotherapy Strategies in NSCLC

 We had another presentation at ASCO [the American Society of Clinical Oncology Annual Meeting], the CTONG 1104 trial, the Chinese trial. Different design. It compared a lesser TKI [tyrosine kinase inhibitor], gefitinib, with chemotherapy. There was a disease-free survival benefit reported initially, but the hazard ratio wasn’t 0.17; it was 0.6. The 5-year survival was not different at the end of the day. We have to realize that the DFS may not drive overall survival, but I want to get your thoughts about that.

 I think you’re right. When we saw the CTONG data that were updated by Professor Yi-Long Wu, we saw a trial with a very different design. It’s not as clean of a trial. It’s not as well designed of a trial as ADAURA because you’re comparing chemotherapy with TKI, not TKI with placebo after chemotherapy. Those patients weren’t as well staged. There wasn’t that CNS [central nervous system] staging, so the concern with CTONG is that a lot of patients on that trial might have had stage IV disease, where we know that TKI is superior. But the point from that trial that’s relevant to ADAURA is that DFS, PFS [progression-free survival] benefit didn’t translate to a survival benefit at the end. That’s 1 of the trains of thought with ADAURA.

When Dr Herbst presented those data, there was a collective still in all our homes, in all our rooms, because we’re not used to seeing curves like that separate so quickly, and to separate so deeply over time, to see at 90% disease-free survival rates at 2 years versus 44% with placebo. It reminded us of how poorly patients with stage II and III EGFR-positive lung cancer do after surgery, that recurrence is so frequent and what an impact TKIs can have there. That DFS hazard ratio of 0.17 was profound. I do think it justifies it as a standard of care. But the question is will it translate to an OS [overall survival] benefit, but we won’t know for some time?

Unlike the PACIFIC data, we know that when patients relapse after chemoradiation, it’s very difficult to salvage patients after recurrence. With someone that has an EGFR mutation, osimertinib does a very good job of controlling disease in the stage IV setting. It’s a good drug, and so we feel more confident that we can deliver that drug safely at the time of relapse. If I’m watching closely and deliver osi [osimertinib] at recurrence, will I be able to rescue the OS curve?

We know from practice that you’re never able to rescue 100% of people. The question that time will tell is, how many we can rescue, and will that translate to an OS benefit?

 Brendon, from the surgeon’s perspective, you’ve obviously seen this data. How does this resonate in the surgical world?

 Surgeons are incredibly impressed. Caveats apply about overall survival, about whether the placebo group will get osimertinib upon progression, and how that will affect overall survival. It’s hard to ignore. I hope this is going to lead to widespread EGFR testing in the early stage. The benefit in stage IB still remains to be seen. Obviously people need to work out the details of chemotherapy followed by osi [osimertinib] versus just going straight to osi [osimertinib] and what the trials implications are for that. But this should mandate testing in early stages of lung cancer.

 That’s 1 point. We’re going to transition in a second to the role of immunotherapy in the surgical disease setting. Does this make lack of a testing in early stage disease something we should do up front? Obviously, the only specimen we’ve had to test is going to be the surgical specimen perhaps. What are the implications here, Roy or Brendon?

 We’re at a place that fortunately tests all our biopsies but particularly sites that are active in neoadjuvant immunotherapy trials. Given this data, it should be their responsibility to test for these mutations.

 Just a few comments. One of my colleagues on the steering committee of this trial, Dr Masahiro Tsuboi, is a surgeon. We talked about this earlier today, and my sense is some patients hopefully will be cured in that we’re looking at earlier disease and less heterogeneity. Hopefully these EGFR inhibitors given for 3 years will knock it out. There will be some dormant cells and some resistors that will move forward. We’ll have to follow that, but with a hazard ratio as good as it is, there’s a good chance that this will show survival as well. Even if it doesn’t show disease-free survival, the fact that patients aren’t recurring in the brain, liver, or bone, and we all know that sometimes you can’t rescue somebody if they come back with multiple brain metastases or if they’ve had a seizure.

We’ll have more data on this as the year goes on. It tells us that at the time of surgery now, reflexively we should be getting PD-L1 status, and we should be getting at least EGFR mutations status and maybe even do a whole next-gen path on it. I was impressed at ASCO by all the posters that were showing the same amount of tissue—you can get 50 or 100 genes versus just 1 or 2. It’s your best chance of getting it including a Flatiron Health analysis that showed if you do next-generation sequencing right away, then you almost never have to go back and do something again, whereas if you just do 1 or 2 mutations or you do nothing, then you do have to go back.

We’re going to be seeing much more testing that comes out of the ADAURA trial, which is going to help patients across the board.

 Which does raise the issue that Brendon raised: Given the treacherous relationship between exposure to immunotherapy and subsequent specifically osimertinib where most of the concern is focused, should trials of preoperative immunotherapy strategies exclude patients with 

 We have a trial where we had hoped to jump-start immune response with low-dose SBRT [stereotactic body radiation therapy] and saw poor responses in the 

 patients. Our sense is moving away from it. Everybody probably saw Catherine Shu’s paper in 

, though. It was chemotherapy and atezolizumab, and notable at the end was that they did have 4 patients with 

 mutations and 2 seemed to have major pathological responses. That seems to be an outlier for most of the neoadjuvant trials. You could maybe make an argument if you’re giving combination therapy, but certainly not for single-agent immunotherapy.

Molecular testing in Patients With Metastatic NSCLC

 Let’s transition to the surgical setting, and we have to recognize our guest of honor this evening, Dr Herbst, in his plenary session presentation. Roy, I want you to walk us through quickly, the ADAURA trial, and then we’ll talk a little bit about what these data mean in the surgical patients and if it should change practice and all those considerations. First, tell us the bottom line.

 Thanks, Mark. The ADAURA trial presented at ASCO [the American Society of Clinical Oncology annual meeting] was a unique and exciting study. As we know in early stage disease—I, II and IIIa—that’s resectable, we use platinum therapy where appropriate, and the recurrence rates remain high. Stage I is as high as 50%, stage II could be as high as 60% or more, and stage III is 70% or more with adjuvant therapy alone. I think what we’re seeing is now 

-mutated lung cancer, for many years we’ve been using these agents in the metastatic setting. Starting with gefitinib or erlotinib. We moved to afatinib, then we developed osimertinib, which is of course, an EGFR tyrosine kinase-specific drug, works against T790M. And from the FLAURA trial it was quite clear that there was improved survival with this drug over erlotinib and gefitinib.

The other thing about osimertinib is that it has very good brain penetration. The idea was now to bring this drug earlier, and to use the best targeted therapy with the best surgical and chemotherapy in early stage lung cancer. That was the ADAURA trial.

I never anticipated presenting at ASCO this year. The analysis was not planned for 2 years from now based on the number of events that were occurring. What happened was this trial, which basically was a quite simple design, patients with stage I to IIIa disease, complete resection by surgery, adjuvant therapy when appropriate. This is a physician and patient decision. They are then randomized to either receive 80 mg of osimertinib once a day orally or a placebo, and people often asked why placebo? Placebo because there is no standard of care or was no standard of care. Then the patients are followed with a planned dosing for 3 years, and the primary end point was disease-free survival.

The safety committee was reviewing the trial in April at the yearly meeting, and they noticed some trends in survival, so they asked for survival data. Remember, this trial was planned for a hazard ratio of 0.7, 30% benefit. They unblinded the trial, and the hazard ratio in the stage II and IIIa patients, the primary end point was 0.17, so 83% improvement in disease-free survival. When you add the stage I patients in, the hazard ratio was 0.21, 79% improvement. If you looked at a number of factors, race, type of mutation, this trial only took 

 exon 19 deletions or L858R mutations, no difference in outcome. If you look at who received chemotherapy and who didn’t, no difference in outcome. In this trial, across the board was positive by all variables, and the toxicity that was seen for the osimertinib was quite mild, just a couple of percent grade 3/4, the grade 1/2s, as has been seen. Notably no problems with ILD, interstitial lung disease, which again, is a concern about using these agents in the early stage.

The trial hit all its end points. Survival is still immature. It showed a very early survival curve at ASCO, but only at 5%, so that needs to continue.

It’s a new paradigm here, Mark, using targeted therapy now in the adjuvant setting, and you can imagine we can use this for EGFR inhibition, but why not 

, or any of the other targets we have? Very exciting, and I’ll stop there and we can discuss more.

 Stephen, your perspective on the ADAURA trial? Is it a new standard of care? I have to admit when I saw the disease-free survival curves, it was the wow factor. You don’t see that very often.

ADAURA: Osimertinib as Adjuvant Therapy in Stage III EGFR+ NSCLC

 I want to ask Dr Stiles, at least prior to the PACIFIC data, it was fairly frequently that I would see patients in which they would have stage III disease. The surgeon they saw wasn’t wild about operating on them at that time, but then they would say let’s give them some chemotherapy, let’s give them some radiation and see what happens, which to me, is a treacherous strategy. My training was deciding if surgery is going to be part of the treatment up front, and if it’s not, then it’s probably not likely to be part of the treatment out back. I wanted to get your thoughts on has this changed that at all? Do you find yourself reassessing patients if they have great responses either to chemoradiation or to immunotherapy?

 It’s a great question. I do think it’s a pretty treacherous area though. I tend to agree with you, and I know that all of us surgeons for the most part do. I think these are decisions that should be made up front with the multidisciplinary tumor board and the radiation oncologists, and try to decide a priori. They can be challenging decisions, without a doubt. I think we know that chemoradiation from the group trial 1601 has a relatively low complete pathologic response [CPR] rate, less than 20%, even if you push it to a higher dose in the RTOG studies, 0229 and 0839, although there was a great rate for mediastinal nodal downstaging, the CPR at surgery was still relatively low, around 25%. That’s with the caveat that there may be ongoing cell kill, and it may get higher as they get farther out.

Also, if you look at the PACIFIC data and you look at the failures, 85% of progression was in the chest, either in the lung or in the lymph nodes. Some of that may have been new lesions in the lung, but clearly there are issues or times when locoregional control could be better. It’s a challenging place for surgeons to get good locoregional control as well though, so again, to me, it often comes to that idea of trying to decide up front. Most of us don’t believe that you can take any of these drugs and make a patient who’s unresectable, resectable. I think that’s a fallacy. Sometimes with immunotherapy it can be even more challenging.

 That’s a great perspective. One of the other issues I wanted to address to all of you is, obviously, the PACIFIC strategy was a consolidation strategy. I’ll pick on Kristin first. Is there a rationale for concurrent use? Then I’ll ask Stephen and Roy to chime in on, should it be part of induction treatment prior to radiation? Kristin your thoughts on moving it up earlier?

 I do think there's great preclinical data demonstrating radiation and immunotherapy synergy, and there are a multitude of trials that are testing concurrent immunotherapy now in the stage III unresectable setting. We did see some interesting data at ASCO [the American Society of Clinical Oncology annual meeting]. Salma Jabbour, MD, presented some preliminary results from KEYNOTE-799, which is a trial of concurrent immunotherapy. It’s a nonrandomized phase 2 trial where there were 2 cohorts of patients. The first cohort got concurrent radiation, 60 gray, with carbo-Taxol [carboplatin, paclitaxel] and pembro [pembrolizumab]. The second cohort got CIS/PEM [cisplatin/pemetrexed], 60 gray, and [pembrolizumab]. The primary end point of this trial was overall response rates and grade 3 or higher pneumonitis.

Not unexpectedly, the response rates were very good. A little better in cohort A, 67%. The high-grade pneumonitis rates were also manageable, between 5% and 8% for both cohorts. The follow-up isn't long enough to give any survival or progression-free survival data on this trial, but nonetheless, I think it is encouraging that perhaps concurrent immunotherapy could be a strategy in the future.

 We’re still trying to work out how immunotherapy and radiation work best together in a patient with lung cancer. They're not additive; we’re not looking for additive benefits. We’re looking for synergistic benefits and the effect that radiation has on the microenvironment, and the effect that chemotherapy has on the microenvironment, and the effect that the immunotherapy and radiation work together. Do we need to give them together or in sequence? Helen Ross, MD, at ASCO presented an update on one of the Alliance Foundation studies, AFT-16. It’s an early look at about 64 patients, and in that study they received PD-L1 inhibitor atezolizumab first, by itself. They did an early look to make sure patients weren't progressing. They completed 4 cycles, so 12 weeks of just PD-L1 inhibitor, an induction immunotherapy approach, then followed by chemoradiation at that point.

Then they looked at disease control rate, and it was high, but these different strategies, what's going to determine which is the best way to go, is which approach leads to more long-term survival, which approach leads to more cure. Right now, we have disease control rate, we have early responses, we have landmark PFS [progression-free survival] rates, but we don’t have what that 3-, 5-year tail will be, and I think that’s going to help guide us more so.

We also saw in the Alliance trial there was a little bit of toxicity. While it was relatively well tolerated, we did see some grade 3 toxicity, 1 episode of colitis, 1 episode of Guillain-Barre syndrome. If we’re looking at strategies that maybe jeopardize the definitive or curative treatment, be it surgery or radiation, that’s something we have to look very closely at.

 Roy, your thoughts there and then your thoughts on the immuno [immunotherapy] combos, like CTLA-4, added to this scenario?

 My first thought is you would want to move immunotherapy as early as possible because the tail in the curve is clearly there for some patients, and you’re getting it to them earlier and it can only be better. That said, we can’t forget that before recently our curative therapy in lung cancer was chemoradiation in patients who were unresectable. You want to make sure we don’t compromise giving those full doses of chemotherapy and radiation. If immunotherapy can be given concurrently and there are not toxicity issues, and full doses of those drugs can be given, yes, I’d say go ahead and do it, and give the immunotherapy more time to work.

As far as immune combos, that’s again, if you don’t see activity in the metastatic setting, I’m hard pressed to want to bring it forward with the added toxicities of a combo to the early stage curative setting. Certainly, it can be tried, and neoadjuvant approaches are good ones, we’ll talk about that in a bit because that gives you the ability to biologically look at what's happening and understand primary resistance, but I’m not that enthusiastic.

Varied Timing of IO Therapy in Patients With Stage III NSCLC

Let’s get back to these patients. They get on durvalumab, 6 months into treatment they progress. Roy, what do you do in that situation?

Well, they just won. But in clinical practice, then you’re usually in a tough situation because not only have they progressed pretty early on durvalumab, but they’ve progressed pretty early from their chemoradiation. You’re not left with that many options in those patients. I’ll often try to rebiopsy, see if there's some sort of action with a mutation of some sort, maybe that’s not been done. As we said, we don’t always profile early stage patients. At that point, assuming a reasonable performance status, try to find some trial. Maybe it’s an I-O combination or an I-O plus chemotherapy combination. It depends what their toxicity profile is to the chemotherapy, the radiation, and the I-O.

How would you look at a patient on the opposite end of the spectrum, where they got the PACIFIC regimen, completed a year of treatment, and then 18 months later had evidence of disease progression? Would you look at that patient differently than the patients who progressed while they're receiving durvalumab?

That’s a tough one, but the answer is probably yes. They’re not prior refractory to durvalumab, although it could have been the chemoradiation effect. In a patient like that, the trials would have let them on, but I think in practice one could consider a rechallenge with immunotherapy, absolutely. Assuming there are no toxicities that would prohibit that.

Yes, I agree. When you start that late after the durvalumab, I think you start fresh as we would approach our typical stage IVs. You do full molecular profiling and NGS [next-generation sequencing], maybe rebiopsy to confirm this isn't some separate primary or another lung cancer, and if there are no driver alterations, I would challenge with immunotherapy-based treatment. Probably in that setting I’m more likely to go with some sort of combination, either dual checkpoint or chemoimmunotherapy.

One of the issues that we alluded to before, I think Stephen you said this, that the duration of a year of treatment was arbitrary. It is amazing to me that when we saw the 3-year overall survival curves that the difference was maintained despite getting that 1 year. Do you think there is opportunity, or we should think about longer duration because typically in stage IV disease we are wedded to this 2-year duration of treatment, and then wondering what to do at that point? Any thoughts? I’ll ask Roy to chime in on that too in terms of is there room for longer durations?

I think it’s a setting where you’re talking about generating a durable, immune-mediated, antitumor response. Ultimately, it’s not going to be a one-size-fits-all. I think there are going to be some people who need very little of the checkpoint inhibitor to induce T-cell memory, and some who may need more constant. I’ve had some very late relapses, for example, in small cell that I wouldn’t necessarily expect in a lot of my non–small cell cases. We've seen a lot of data, some presented by Dr Herbst, that showed when you stop after 2 years you can rechallenge with some success, but I think ultimately, we’re still looking for patient-based biomarkers that are going to help guide us. I am comfortable stopping and watching closely, but I’ll tell you that some of my patients are not comfortable stopping.

It can go both ways. I’ve had people who have gotten 2 or 3 doses of immunotherapy. They come off from toxicity, and we’re managing their toxicity with steroids for years, but their cancer never recurs. Then, there are some where we treat to progression, oftentimes in those cases, patients will say, “I’ve had enough.” Now, the intervals are stretching out. Six weeks now with pembrolizumab, and so forth, so it’s easier to do maintenance.

I like the 2-year mark as we heard from Stephen. That came out of KEYNOTE-010. We looked at those patients who had PR [partial response] or response in the [

] paper from earlier this year, and very few of them relapsed after 2 years of treatment if they were in that group. And when they did, about half of them responded again. A similar analysis is going on now for KEYNOTE-024, so we’ll see that soon as well.

Again, I think it’s each patient individualized, and in the absence of measuring memory T cells and knowing what's going on, it’s a bit empirical, obviously.

Treating for Stage III NSCLC After Progression on Durvalumab

Do you stretch the performance status [PS] rules? Obviously, PACIFIC was PS0 to PS1, but pretty positive results. Do you think those can be extrapolated to the poor performance status population, or what would you consider as a standard of care in a PS2, or dare I say, PS2.5?

I probably do stretch it a bit. Part of it depends on why the performance status is low, and chemoradiation is a tough treatment. I think it’s not unusual to have patients who are pretty tired, pretty weak from that therapy. And the first thing I do is try to wait, try to give them a little time to recover, maybe IV [intravenous] hydration, focus on nutrition and see if we can get some of that performance status back.

I’m balancing the potential improvement in outcomes with the toxicity, and durvalumab monotherapy in this consolidation setting is pretty well tolerated. I found it pretty easy to deliver, and I’m not really concerned that I’ll take someone who is PS2 and make them a lot sicker by delivering durvalumab. I suspect the outcomes are going to be worse than the PS0 to PS1, but I suspect the outcomes would be better in the PS2 compared to not giving it to them.

No, I agree. Few things have changed practice as rapidly as the PACIFIC results. The 2 papers back to back, first the PFS [progression-free survival]; we were using it based on the PFS, then OS [overall survival] came right on the tail, that’s such a good survival benefit. I try to give it everyone. We talk to everyone about it now at the beginning of the chemoradiation, so they're prepared for it. We try to give it as often as possible. Yes, I wouldn’t give it to someone who’s a PS3, but if someone is PS2, and you saw them a few times, and they're a PS2 and not a PS3, I’ve given it, and I think it does improve outcome.

I agree with what was said earlier, if you don’t get through a year, that’s OK. You do as much as you can, and when you stop you usually don’t retreat after that.

Kristin, your thoughts on the performance status issue.

For PS2/3 I think the hardest part of treatment is definitely the chemoradiation. For those patients, we are typically giving hypofractionated radiation alone, so it’s 60 gray in 15 fractions. And a case-by-case basis, if we think they can tolerate durvalumab, we occasionally will do sequential chemotherapy followed by radiation. I think there is an AstraZeneca trial underway that’s testing sequential chemoradiation followed by durvalumab for poor performers. Again, I think it’s the chemoradiation piece they struggle with, and we haven't seen very many new therapies for the PS2/3 patient. So I think it’s important to mount a trial that tests immunotherapy for that patient population.

One of the things that I think was reassuring about the PACIFIC trial is this issue of pneumonitis, that there didn’t seem to be an excessive amount of pneumonitis. Of course, one of the things we don’t know is the radiation parameters for the patients who were treated on both arms. Give us your perspective as a radiation oncologist, the issues around immune- mediated pneumonitis, radiation-mediated pneumonitis.

If you look at the contemporary chemoradiation trials, for example, R2GO617, the rate of grade 3 or higher radiation pneumonitis was about 6% or 7%. It’s gotten so much lower over the years because of improved technology. We can handle these bigger tumors. If we see changes on imaging that are within a treated radiation field, and the patient is symptomatic with increasing hypoxia, then we think this is likely due to the radiation. If imaging findings tend to be more widespread throughout the lung, I tend to think that would be more of an immunotherapy-driven pneumonitis. Sometimes it can be difficult to differentiate. We still don’t have any great biomarkers for this toxicity. I think it’s something that people worried about at the onset of chemoradiation followed by immunotherapy in stage III disease. But clinically it doesn’t happen that often, and I think that the oncology community is getting more comfortable with managing these toxicities, and our patients do overall pretty well.

I think in the PACIFIC trial they topped the radiation dose up at 66 gray. Is that correct?

The PACIFIC trial patients enrolled after they finished radiation, so the radiation dose wasn’t specified. You had to get at least 54 gray, so everybody got at least 54 gray, but there wasn’t a stopping point in terms of where everybody stopped. It was sort of dealer’s choice. If you look at PACIFIC-2 that’s enrolling internationally, patients obviously are getting just 60 gray, so the radiation parameters will be controlled for.

I would say the standard is 60 gray, based on R2GO617. Certainly, you can get better doses to your lungs with 60 gray as opposed to 66 or 70 gray. I never push it over 60 gray for non–small cell lung cancer.

I guess you would conclude that from the R2GO617 trial, but I still see our radiation oncologist going to 66, 70 gray. That’s probably based on volumes, but I was wondering if there's an accepted standard in the radiation oncology community.

Durvalumab in Stage III NSCLC Based on Performance Status

Let’s transition to one of the more influential trials. It’s been so wonderful to be a lung cancer doctor in the last decade and certainly recently. Someone pointed out we had 6 or 7 approvals in lung cancer over the last couple of months. It’s been a land of plenty and probably more to come as we’ll talk about. One of the more influential trials, and I want to ask Dr Liu his perspective on this, is the PACIFIC trial. We had reached a plateau in terms of what we could achieve for cure rates and survival in the concurrent chemoradiotherapy world. We saw PACIFIC come along and change the standard of care. Give us your perspective on that, and then do you think it’s been adopted at the community level? Are most people doing it and that sort of thing?

I think it’s been widely adopted because this is what we’ve been waiting for, for so long. As a reminder, the PACIFIC trial was a randomized phase 3 trial for patients with unresectable stage III non–small cell lung cancer. Patients received definitive chemoradiation with platinum-based chemotherapy delivered concurrently, and after completing treatment, in the absence of progression, they were randomized to receive durvalumab or placebo for 1 year. What we saw initially was a pretty profound improvement in PFS [progression-free survival], and at the [World Conference on Lung Cancer] in 2018 we saw that did translate to an OS [overall survival] benefit. The hazard ratio at that point was 0.52 for PFS, 0.68 for survival.

What we saw in the past year was more maturity, and with more follow-up we saw that survival benefit persist. We don’t see a dropping of that tail, that hazard ratio 0.69, a consistent benefit with delivery of durvalumab. We saw a median survival of 29 months with placebo, not reached with durvalumab, but the 3-year survival rate was 57%, and I think it’s been embraced by all of us. This is a space where we deliver a fairly toxic treatment. Chemoradiation is one of the tougher treatments that I have to coach patients through. The hope with that treatment is to deliver a cure. We know that for the majority of patients we don’t achieve that. The addition of immunotherapy [I/O] consolidation gives patients the potential for long-term benefit, long-term survival, and I think at first there was some concern about safety. I think that’s largely been quelled. It’s a safe drug to give; we don’t see very high rates of pneumonitis. There were some who were reluctant to get rid of consolidation chemotherapy. Maybe some on this call right now, and who instead are maybe giving that as induction therapy. Now, we’re very comfortable using durvalumab, and it’s become a clear standard.

In the trial design of PACIFIC, we had 26 doses or a year of treatment. At least some recent market figures suggest that the average patient doesn’t get nearly a year of treatment. How important is that to you to think you’ve got to push them through and get the full year in?

It’s 1 year, which is a nice, round number. That means it’s pretty arbitrary. At that 2018 update, if we look at the 473 patients who received durvalumab, only about 232 completed all 12 months. Some of the patients will drop off for progression, notably less than the placebo arm, and some patients will stop due to toxicity. Those patients may still derive benefit from durvalumab, but just may not need to continue therapy. My general approach is to go for a year, but certainly in the face of toxicity, if I coach patients through, the toxicity resolves, I’m unlikely to rechallenge someone because it may not be critical to finish that year.

Roy, any reason to check PD-L1 status in the consideration in this setting?

We do, but normally that will be the PD-L1 at the diagnosis and not after the chemoradiation, so all bets are off to what's going on in the tumor microenvironment. We tend to use it on patients regardless of PD-L1 status. There were some data I think early on that suggested lesser effects in low PD-L1. We’re pretty much using it in all groups.

We’ll eventually get to the ADAURA trial in this discussion, but that’s certainly going to push more frequent molecular testing in early stage disease. Right now, we don’t have a defined reason to do molecular testing in stage III, but it’s often done so you know that people may have a driver. Does that influence your use of immunotherapy after chemoradiotherapy, given the questionable data we have for instance in a patient with an 

Yes, it might, although if you look at the PACIFIC trial, there were a reasonable number of patients, under 10% or so….

 mutations, and that subgroup, albeit small, did well. I wouldn’t avoid it, but I think we’re moving toward a new trial where we’ll probably molecularly profile stage III patients. And then based on the results of ADAURA, there’s a trial called LAURA, and LAURA is going to be using osimertinib in that stage III chemoradiation post setting. I think that might be a very nice thing to do as well. Same concerns, the worry about ILD [interstitial lung disease] and other issues, and the effect of an EGFR inhibitor plus an I-O agent. Working out some of those timings, I think that would be a very nice thing to study in the future.

It changed practice, and I agree with you Dr Liu, I think people are enthusiastic about it. Everyone loves immunotherapy nowadays.

PACIFIC Trial: Durvalumab Following Chemotherapy for Stage III NSCLC

: “Multidisciplinary Management of Locoregional Non–Small Cell Lung Cancer.”

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 in Washington, D.C.; Dr Kristin Higgins, a radiation oncologist at the Winship Cancer Institute of Emory University in Atlanta, Georgia; and lastly, Dr Brendon Stiles, a thoracic surgeon from the Weill Cornell Medical Center in New York.

Today we’re going to discuss a number of topics pertaining to the use of systemic therapy in patients with stage I through III non–small cell lung cancer. We’ll discuss the latest research in the field, including new data from the ASCO [American Society of Clinical Oncology] 2020 virtual meeting and the impact of recent clinical trials on making decisions around treatment selection.

Let’s get started on our first topic. In this population of patients, staging is a cornerstone of making decisions about future therapies. I wanted to ask Brendon to give us his thoughts, particularly on something that I think historically in the United States is often a shortcut in the work-up of patients, and that’s the adequate evaluation of the mediastinum in defining pathologic involvement of mediastinal lymph nodes and the various techniques we have to do the mediastinoscopy versus EBUS [endobronchial ultrasound] and those sorts of things. Give us your perspective on that mediastinal staging. Who needs mediastinal staging, what's the best way to do it? What are the pitfalls of EBUS, what are the pitfalls of mediastinoscopy? Start us off there.

You raised some great points Mark. I think we’re all excited to see both targeted therapy and immunotherapy moving into earlier-stage disease. But as you allude to, staging obviously becomes critical there and the sequencing of therapy, whether we even choose to include surgery as part of the therapy or whether we go to definitive chemoradiation, staging is critical for both resectable and unresectable patients.

I think the key point is as you hinted, staging needs to be done, and it’s probably best not just done with a PET [positron emission tomography] scan, but by invasive mediastinal staging. This is generally a population that’s getting considered for these trials because they’re thought to have positive lymph nodes, either N1 or N2 nodes. We know that if they have N1 nodes, they might have N2 nodes; if they have N2 nodes, they might have N3 nodes. I think it’s inherent upon us to stage well and to stage multiple stations. I think that can be done well by lots of folks, whether it’s a thoracic surgeon or a skilled pulmonologist.

I see EBUS and mediastinoscopy largely as interchangeable. I think the key is doing something. Certainly, EBUS has gained favor, it’s easier on the patients, it doesn’t require an incision, and in many studies, its actually more accurate. I tend to believe that particularly when you’re looking for PET-positive nodes and going right to the nodes that you’re suspicious for, you can get there well with EBUS, you can do multiple samples, and the yield is very high. I think the one thing that’s interesting, where we’ve moved toward EBUS and largely away from mediastinoscopy, I think the incidental N-positive stuff, whether it’s incidental N2 or N3, we’re not probably finding as much. If you see a patient with bulky N2 disease, in the old days we’d stick a mediastinoscopy down, and by default you’d get some N3 from the other side just in small nodes.

With EBUS the tendency is a little bit more to guide yourself toward just the big nodes, and the smaller nodes that you think are perhaps not positive are a little bit more challenging. That can have some significant implications for the patients though, as from being IIIA to IIIB or IIIC. I’m in favor of making sure that we sample both sides, trying to narrow down the nodal involvement and to come up with an accurate stage.

Then it’s just as important to do intraoperatively mediastinal staging. Could you comment on that?

That’s really key, and as we talk about adjuvant trials and I’m sure as Roy will talk about his trial, understanding which patients benefit is going to be key on doing good surgical staging during the operation. And there's a big difference between calling a patient stage I and taking 1 or 2 lymph nodes out, and calling a patient stage I and having sampled 31 nodes. Increasingly as we understand that we have agents that might be beneficial in the adjuvant setting, that idea of staging them well surgically, making sure that we get a lot of N1 nodes and N2 nodes so that we can be accurate there is critical, and probably has some implications for disease recurrence as well.

Then this issue, the term is so cemented in our lung cancer culture, this term unresectable stage III. Does unresectable mean you can’t resect it, or does it mean you shouldn’t resect it? Give us your perspective on that.

I think the latter is true. I always say that resectability, like beauty, is in the eye of the beholder. I think that anything can be resected. The question is really should we resect it? We know that there's probably some that we shouldn’t, and I tend to think of that as bulky multistation N2 disease. Certainly, that’s going to act more the equivalent of systemic disease, but even more importantly, I pause when I know that I’m going to have to go down the pneumonectomy road with one of these patients with a high burden of disease. I think that inflicts almost a second disease upon patients, and I think it leads them down a long road where they're not going to tolerate systemic therapy and they're going to have a harder time on any of these trials that we’re talking about.

That said, have I done it and do I think it’s appropriate for some surgeons to do? Yes. In young patients with excellent performance status. In general, the data have suggested that pneumonectomy patients do worse.

If I can do a lobectomy, I might be a little bit more aggressive with a patient with multistation N2 disease. I think the real question is should it be resected? I think the challenge of that is that answer is probably different depending in where you are, your institution, your surgeon’s comfort levels.

Let me ask Kristin. At least in my mind, this is an extension of it, obviously as you get to the perimeter of the surgical ability to resect something, you get into bulky disease. And volume in radiotherapy is so important, and we know that the toxicities are not necessarily related to radiating the cancer, it’s what the normal tissue around the cancer gets. What are your thoughts on radiation issues as they relate to tumor volume?

 Stage III patients, the radiation technologies that we employ to treat these patients have evolved tremendously over the past couple of decades. In my mind, any stage III patient is treatable to a curative dose with radiation. We’re typically using IMRT [intensity-modulated radiation therapy] technique to treat patients who have bulky disease that have contralateral mediastinal nodes or mediastinal and hilar nodes, N3 nodes. But you can deliver 60 gray safely with these advanced technologies, and if you have dosimetrists and physicists experienced in planning radiation treatment plans for bulky disease, you can typically achieve your doses to your normal organs at risk such that you won’t hurt the patient. Part of that is expertise that you develop when you’re treating a high volume of patients with lung cancer. I think it could be delivered safely certainly also in community settings, and it's really changed what can be treated safely with radiation when it comes to stage III disease. I think almost every stage III patient can be treated safely with our latest and greatest technologies.

Kristin Higgins, MD, City of Hope

Articles