Video
Considerations for Incorporating Perioperative Treatment in Resectable NSCLC
Author(s):
Before closing out their review of resectable NSCLC treatment strategies, key opinion leaders consider nuances in utilizing perioperative therapy in the current paradigm.
Transcript:
Balazs Halmos, MD, MS: In the next couple of months, there will be a lot of heated discussions and debate about the results of the AEGEAN [NCT03800134] and KEYNOTE-671 [NCT03425643] studies, as well as others. What key elements are you looking for out of those studies when trying to learn how to adapt to the adjuvant space after neoadjuvant therapy dependent on the pathological response? What are your expectations, and what are you looking for?
Martin Dietrich, MD, PhD: CheckMate 816 [NCT02998528] was a total surprise in that pathological complete responses [pCRs] in 1 in 4 cases were possible in non–small cell lung cancer. None of the trial designs really accounted for escalation or de-escalation strategies. I think patients that have a pathological complete response are expecting a long-term curable outcome. If I think about the translation between the pCRs and the long-term disease-free survival and overall survival benefit, I think it is about 20%. There’s a big degree of translation between those 2 populations over chemotherapy alone. As far as how to interpret the information in the absence of adapted strategies, it is going to be very, very challenging for me.
I think there is going to be a lot of discussion surrounding the different trials. They have some similarities, some different challenges. I think the neoadjuvant lung cancer space is going to dominate. I am very certain that we’ll move systemic therapy upfront. I think the biomarker challenge here is a particular level of challenge. But how to proceed after surgery, how to escalate and de-escalate, similarly like we’ve done in breast cancer with KATHERINE [NCT01772472] and other studies where we augmented intensity for patients with suboptimal responses. I think this is going to be a very advanced field. That is the beauty of these clinical studies. They will provide us with information into the refinement of tailoring therapy, not only necessarily upfront biomarkers but also clinical response assessments that have a value that we have never really had in lung cancer. I think this is going to be very exciting.
It will take many more years to avoid under- and overtreatment at the same time. It is a very, very tough question for us to answer. My feeling is, if a patient does not have a complete response, I typically give the additional 3-4 cycles of adjuvant therapy, typically with a switch in the mechanism of approach. If I have given neoadjuvant cis [cisplatin]-pemetrexed, I consider an additional couple of cycles of cis [cisplatin]-taxane and then, obviously in the expectation that they are not doing well, also a consideration for the extension of adjuvant therapy. Obviously, that’s a practice pattern, not necessarily a fully evidence-based backed approach. We do use liquid approaches for minimal-disease monitoring to help us risk-stratify patients additionally, especially the higher-stage patients. But I do think there will have to be a strategy with intensification of therapy. And the question whether 3 cycles of immunotherapy are sufficient would really raise of lot questions not only for the neoadjuvant and adjuvant space, but for the metastatic space in general. I think we’ll take another 3-5 years before this is optimized. A lot of new questions that will come out of those trials.
Balazs Halmos, MD, MS: Definitely a lot to learn. And you mentioned how transformational the CheckMate 816 study was. Just 3 rounds of neoadjuvant chemo/IO [immunotherapy]. That is just 9 weeks of treatment time. It is really stunning that 1 of 4 tumors just melt away. At the same time, it also poses the challenge that 3 out of 4 do not go away, and in those patients, the risk of recurrence seems quite high at 70%-80%. All of us have the urge to help those patients in our clinics. We just do not have the real data to help. So, Dr Reuss, let me pass the baton to you. How do you look at those studies? How do you define what you do for your patients after June the 5th or 6th?
Joshua Reuss, MD: We will have to see what that data looks like. I have a lot of thoughts on this. I would say that for patients who have a pCR, as Dr Dietrich said, [those] are patients that are less likely to recur. I think that these patients still experience recurrences. I know there was 3-year updates from the CheckMate 816 at ELCC [European Lung Cancer Conference], and there are recurrences in that population as well. If I see a sub-pCR response, I have not been one to give additional adjuvant therapy if I give neoadjuvant. If there’s an R1 resection, I think that’s where you might consider post-operative radiotherapy. It is maybe the only scenario where I would consider that. I am not inclined to give additional adjuvant [therapy] until I see that data. I think we are going to have to see how the data reads out. Utilizing a pathologic response and then true minimal residual disease [MRD] techniques, not just our Guardant ctDNA [circulating tumor DNA], but true tests designed to evaluate minimal residual disease, I think will be important at some point in the future. But as it pertains to perioperative approaches, I’m of the mindset, if you have pCR with a neoadjuvant regimen that we do not need adjuvant until we get the data.
At ELCC, there was an update from LCMC3 [trial; NCT02927301] that showed for those who even had an MPR [major pathologic response], there may be a benefit to adjuvant atezolizumab [Tecentriq] compared to no adjuvant, which maybe throws a wrench into that hole. Does the pathologic response determine the outcomes there? On the flip side, if you have someone who has no pathologic response, that whole intact tumor is there. This determines that it is an IO-refractory tumor. Will a year of adjuvant immunotherapy make a lick of difference there? Probably not. Do we see other compounds where we’re trying to enhance a response? There are more questions than answers that I just provided there, but those are all the things swirling in my head these days.
Balazs Halmos, MD, MS: Dr Marrone, closing thoughts. You work in the MRD capital of the world. What are your thoughts in this area?
Kristen A. Marrone, MD: There are more questions than answers. I think your point about what we’ll do on June 8th as opposed to June 1st is going to be a little different. I similarly approach it as Dr Reuss mentioned: by trying to use data from surgery to figure out what adjuvant therapy should be given and how to determine that. I think if people believe in the TMB [tumor mutational burden] data, the PD-L1 data in the metastatic setting, that could help guide it. We have had conversations about that. We are going to talk a little bit later about different molecular subsets of tumors that maybe have IO responses. There are some people out there that say the pathCR [pathological complete response] isn’t there, but it’s because they haven’t gotten enough treatment.
When you look at it in the metastatic setting, the median time to response with IO is 9-12 weeks. That’s the optimistic view of saying maybe adjuvant IO really could still help people if there is some kind of treatment response when you remove it, even if [they] haven’t met the predefined benchmark of a certain percent tumor shrinkage.
We are also one of the IO capitals of the world. We look for reasons to give it. But all kidding aside, I think if there is a signal that there has been a tumor response, we are interested in using it that space. We are very interested in using MRD testing.
There is not any commercially designed test that I feel confident in using for all of my patients. There are a lot of different types of testing that are ongoing to see how best for it to be used. But for right now, I think getting that kind of data so that we can add to real-world information while we wait for other companies to present their data makes a lot of sense because, certainly, to Dr Dietrich’s point, the next 5-10 years we’re going to be talking about this stuff every month. We have shown the power how real-world data can really help us change how we use these treatments.
Transcript edited for clarity.
