Roundtable Discussion: Antonoff Discusses Treatment Goals in NSCLC

Case-Based Roundtable Meetings Spotlight, Case-Based Roundtable Meetings Spotlight August 2021: Solid Tumors,
Pages: 18

An otherwise healthy 60-year-old woman, who has quit smoking for 8 years after 13 pack years, presented with a nonproductive cough. The patient was diagnosed with non–small cell lung cancer. Oncologists discuss the scenario during a Case-Based Roundtable event.

During a Targeted Oncology Case-Based Roundtable event, Mara Antonoff, MD, an associate professor and program director of Education in the Department of Thoracic and Cardiovascular Surgery and Division of Surgery at The University of Texas MD Anderson, moderated a discussion about a 60-year-old woman with non–small cell lung cancer (NSCLC).

ANTONOFF: For us at the MD Anderson [Cancer Center] we, as the thoracic surgeons, are not the ones who are ordering the testing. I don’t know how to place the orders; I couldn’t do it if I wanted to. But we’re interested to hear from those of you who are in different practice settings, whether you’re ordering them, or whether someone from medical oncology orders it. If you are the one ordering it, what are those challenges and barriers?

VILLAMIZAR: We discuss these cases in our tumor board, and we have a nurse navigator there. When we say, “OK, we should send this patient for next-generation sequencing,” then the nurse navigator is the one that oversees placing the orders.

But I believe that there needs to be a medical oncologist who cosigns it. I don’t think that we thoracic surgeons are even allowed to cosign those orders, it needs to be a medical oncologist.

ANTONOFF: That’s probably a good idea. What are the practices in other [individuals’] practice settings?

KHANSARINIA: We used to have this problem going back and forth. I’m at Piedmont Atlanta Hospital, and one of the things that we decided was that the [pathologist helps to decide], and that’s what we’re doing now. Offline, any of the medical oncologists can order it, but the surgeons generally don’t.

ANTONOFF: Does that pose a barrier, or does that eliminate the barrier, because it’s being done?

KHANSARINIA: I think it eliminates the barrier, but from a standpoint that it used to be a problem. Some of the insurance companies—unless it was an oncologist ordering [the testing], they wouldn’t allow it and they wouldn’t pay for it.

SHEN: Who pays for it now? It would seem to me that it wouldn’t matter who ordered it. The payer is going to be the same, no matter who ordered it, right?

KHANSARINIA: It’s changed. When we started this a few years ago, a lot of these we had to negotiate, but as genetic testing and molecular testing have become more prominent and there is more information out there, we found paying became easier. Medicare now has started paying, so it’s not been a problem.

KHODAVERDIAN: In California [as well] some insurances will not pay unless the pathologist is ordering it, or a medical oncologist. I have had so many issues where they wouldn’t pay for my patients.

ANTONOFF: What are you doing up in Minnesota [at the Mayo Clinic], Dr Shen?

SHEN: We’re doing what you’re doing. We defer to the medical oncologists. I don’t think they do it routinely for all [patients with] early-stage cancers. I think sometimes they do it selectively, and I’m suspicious that they do it, sometimes, if the patient develops progression of disease. Because, when they order it, I get notified through our EPIC system of the result. Many times, it’s several years after I’ve operated on the patient.

ANTONOFF: It’s when they recur, right? So, you get an addendum on your pathology report from before?

SHEN: Correct. It makes sense to me because it’s actionable then. If the patient had a stage I or stage II lung cancer, they’re almost certainly not going to be prescribed immunotherapy. But when a significant number of these patients will develop progression of disease, then it becomes relevant.

I think that’s when the patients who had early-stage disease at the time of final surgical pathology, sometimes later they then have the sequencing done on their original tumor because then it’s actionable. Then they have recurrent or metastatic disease.

ANTONOFF: That’s our practice as well. That’s why I was a little bit surprised to see that most of you all said that you do it routinely on patients with early-stage disease. I would say that we generally don’t do it on patients unless they would be candidates for adjuvant chemotherapy, or induction, or if they recur.

Sometimes the patient, even if they were profiled and then they recurred 2 years later, we get a new biopsy to see what the characteristics are of the recurrence. We generally have not been really profiling tumors at MD Anderson unless it’s a patient in whom they’re already a candidate for some sort of systemic therapy anyway. Now that may all be changing with new data suggesting that we’re used to not offering adjuvant therapy for patients who have stage IB disease, but now there may be data to suggest targeted therapy for patients with IB disease. Maybe we do need to start looking earlier, or do we look once we have data that suggest that a patient may potentially be a candidate for some sort of targetable, actionable therapy?

It’s challenging, but up to this point we have not routinely obtained that type of information on a patient we don’t believe to be a candidate for systemic therapy at that time. Does anybody else have any different practices or anything else that they want to share?

REKKAS: In [UW Health Cancer Center at ProHealth Care in] Waukesha, Wisconsin, where I’m at, for our pathologists it’s reflexive for all the lung cancer diagnoses. We have this discussion routinely in our lung conference tumor board. Is this cost-effective? Are we doing the right thing? But we have ACL Laboratories and it’s reflexive, so everyone is getting it. Occasionally, the patients who do need it won’t have it, which slips through the cracks. But then it does come; if it’s post operative, it will come to my attention, I’ll notice it, and then it will go to the oncologist.

We also discuss it in the tumor board, for anything missed, and retroactively go in. We have had this discussion on several occasions in our tumor board. Are we doing the right thing? Whose decision is this, that everybody’s getting it. And we still haven’t figured it out yet, so it’s an interesting perspective.

ANTONOFF: It’s an ongoing challenge. Anybody else have any other input on that?

SHIEH: I typically do not order testing [for patients with early-stage disease]. Typically, those are for patients who are later stage, and typically the oncologist orders those. The rare occasions where the patient may need some adjuvant therapy—say stage II or maybe even IB—the oncologist calls the pathologist and basically gives them what they want.

ANTONOFF: Got it. That’s interesting, how different it is from place to place.

SAVAGE: We never order it because I don’t think I should be ordering it; I’m not giving the chemotherapy and I’m not making those decisions. I might participate in them, but it’s not my role to tell the oncologist when to and when not to do this. So, that’s why we never order it ourselves. I answered because I got the sense that our oncologists were ordering it on most of the patients, and like the rest of you, a year later I’ll get a report on pathology from somebody who’s name I don’t even recognize.

ANTONOFF: That makes sense, too. Yes, it’s hard, and I think it’s different from practice to practice, depending on whether you’re in a closed system, or you can have a variety of referring medical oncologists as well.

I know that we could talk about this for quite a while, but I know there’s a lot more to discuss regarding this specific patient.

Before we go on, does anybody have any specific challenges or barriers to testing that they wanted to mention? I don’t really know of any barriers for us, in terms of payment, at this point.

REKKAS: For us, and the patients that need it, it’s the timeline [of treatment].

Sometimes it can be 3 weeks, or 4 weeks, and the oncologists are like, “We need this.” That’s, again, noticing from the tumor board, it takes a long time, we don’t know why.

I don’t know if we switched labs, or what, but that’s the main thing in our tumor board is when are we getting this result.

ANTONOFF: Yes, I agree. The time is still an issue, regardless of who orders it, even if it’s paid for, it does seem to take a long time.

ANTONOFF: I think this makes sense. We recognize that there’s an indication for adjuvant systemic therapy in these patients if they haven’t received induction. But, of course, we all know that a substantial portion of these patients can also have a decline in their performance status, refuse it, not be well enough, or have complications that don’t really enable them to go ahead and get systemic therapy.

ANTONOFF: I think knowing the treatment goals is important. For example, for us, we talk to patients about how it can impact their risk of recurrence, as well as their risk of overall survival [OS]. I think treatment goals should not be our treatment goals, but rather the patient’s treatment goals. That, ultimately, will determine the role of adjuvant chemotherapy, or any other systemic therapy. Or frankly, the role of surgery, because if the patient’s treatment goals are to live if possible, or to be able to stay at their home and not have to travel to the medical center, or to be able to go to a high school graduation, whatever. I think, ultimately, it should not be up to us whether it’s disease-free survival, or OS, or quality of life, but rather, leaving that to the patient. It’s going to be different patient to patient.

KHODAVERDIAN: I say this because I have some experience looking at the quality of life per year, and the cost of achieving that. As you know, a lot of European countries are looking into that, and some of the treatments that we provide for a patient with cancer does not qualify for any of those.

So, I understand your point, that the patient should drive treatment, but in the meantime, adding a couple months to patient’s life, not much of a quality of life, I don’t think that should be a value added to value-based medicine.

SAVAGE: Because we don’t do this, and because I work with a lot of different oncologists who have a lot of different opinions, I think they tend to follow many of the guidelines. I suspect, in most cases, they are getting chemotherapy after we’ve operated on them.

ANTONOFF: You think that’s due to the patient not wanting it? Who’s dropping the ball, [in your opinion], and how?

SAVAGE: I think that the oncologists are more than willing to do it…but some of the oncologists that I’ve worked with in the past have been rational about it. For example, one of them told me—and again, I don’t know if this is the latest data—that when you have just 1 or 2 N1 nodes, there’s not that much of a survival advantage to giving them chemotherapy. Now we do know that poor histologic features, and again, I’m probably speaking from ignorance because I’m not really versed in the literature; but over the years, I’ve picked up that unless you have definite evidence of extra spread beyond the lung, poor differentiation, and pleural invasion have worse prognoses. I’m not sure what the true benefit is to that.

What’s more important to me, as a surgeon, is the role of neoadjuvant therapy, and which patients are most appropriate for something like that, because that really makes a difference to me in the decision-making process. Once I’ve taken out the cancer, really the decision-making process is out of my hands. But, before I take it out, I’m really involved in that decision process.

ANTONOFF: Let me ask you this: do all your patients see medical oncologists? Because I can tell you that mine do not, so I feel like I have to absolutely be aware of indications for adjuvant chemotherapy, or adjuvant systemic therapy, targeted therapy, anything really. Because, if my patients are curable by surgery alone, and don’t need any adjuvant systemic therapy, I don’t send them to a medical oncologist. But I will send them to a medical oncologist if there is an indication. Or, to at least discuss it; I can’t say that they’ll get it, but at least to discuss the possible benefit, whether it’s the 5% survival advantage, or making them feel better about the fact that they have 4 kids they want to get through weddings, or graduations, or whatever it may be. I send them to talk to the medical oncologist about it, but I want to have an indication for it. So, if I see someone with stage IA disease, I don’t send them to a medical oncologist.

SHEN: I agree with you. I only send the ones that I think have an indication for chemotherapy. So, if they have N1 or N2 disease post-surgery, I’ll send them. But, even if it’s a patient like in this case scenario where it was a reasonably good-sized tumor but the lymph nodes were all negative, I know, and I’m sure you know too, what the medical oncologists are going to say.

They’re not going to offer that patient chemotherapy. I’ll tell the patient, I’ll have a discussion with them and say, “Listen, I would find it helpful for you to go and talk to a medical oncologist, I’ll refer you. But I can tell you up-front what they’re going to say about your situation.” I would say most patients are looking for an excuse not to have chemotherapy, and so they almost never want to go and have the conversation if I tell them I think there’s almost no chance that they’re going to offer you adjuvant therapy.

I always offer it, but I tell them what the oncologist is likely going to say. Because at our place, the medical oncologists mostly offer adjuvant therapy if it’s a node-positive case, or it’s a humongous tumor, greater than 7- or 8-cm tumor, they might consider it, but even then, I would say the majority don’t offer it.

ANTONOFF: That’s so interesting, how differing the practices can be from 1 institution to another. Because for us, anybody with nodal disease will get it, but anybody with a tumor 5 cm or greater will get it too. If you’re 4 cm and have lymphovascular invasion [LVI] or visceral pleural invasion, you’ll get it. Even 4 cm, if you have LVI, so it’s so interesting to me just how different our practices can be, yet I’m pretty sure our outcomes are near identical. So, maybe it doesn’t really matter what we do.

SAVAGE: I’ll just tell you the reason that I send them all to oncology. I’ll usually tell them if I don’t think they need it. Our oncologists will follow up on them, because I want to make sure they have a consistent follow-up, and I don’t want to follow them, because we just don’t have the resources and the time to see these patients back, and follow them long-term. I’m guessing that, if you don’t send them to an oncologist, you probably follow them yourselves, is that correct?

KHANSARINIA: We at MD Anderson are certified physicians, and refer every patient to oncology, but we get dinged. So, it has been our practice, even before we became associated with MD Anderson. I have been sending patients to medical oncology, but I think this is the point that— despite the fact that we’re surgeons, and we don’t give chemotherapy—I think this is where the tumor board and discussion of these patients becomes very important.

REKKAS: I just wanted to reiterate, at our institution, that we strictly follow the NCCN [National Comprehensive Cancer Network] guidelines as well. Similar to what you have, the patients with stage II will go see oncology, and we’ll refer. The patients with stage IB, if they have the poor prognostic indicators, then we’ll at least have the discussion to refer to oncology to discuss the risk/benefit calculator of the adjuvant chemotherapy. This book is everywhere. So it just reiterates, I think, with yours.

ANTONOFF: I’m very interested to hear what your reactions are to the ADAURA data, whether you think they’re going to change your practices. For those of you who’ve read the whole study, what you think the strengths and weaknesses are, as well as the pros and cons of this strategy.

SHEN: I think the results are startling. I’ve gone to the American Society of Clinical Oncology annual meeting before, and I’m used to seeing these big chemotherapy trials for the last 20 years that the survival curves don’t look any different, and the medical oncologists come out of the room, and they’re all upping their lithium dose, and they look all depressed. This is the first major advance in the chemotherapeutic treatment of lung cancer in the last 30 years. It changes the whole landscape, I think, for patients who have that mutation.

ANTONOFF: I agree. What does everybody else think?

VILLAMIZAR: For us at the University of Miami, as soon as we had the results fortunately all the medical oncologists really embraced it. [Data for patients with stage IB disease for us] was a game-changer. Because we know that those patients, some of them are going to recover, but there’s nothing that we have in the armamentarium to offer. Now, with this, we are sending all those patients with visceral pleural invasion, which is crazy. Even if it is a 1.2-cm tumor, visceral pleural invasion, we are sending them to get tested. If they have the exon 19 deletion, then we’ll offer them the osimertinib.

SCHWARTZ: Why is this EGFR TKI [tyrosine kinase inhibitor] working, and the other ones didn’t? What’s different about this one?

ANTONOFF: That’s an important question, because I think we all know that most people who get 1 TKI often develop resistance at some point. However, it does seem like the outcomes from osimertinib are much better, and it seems to be superior to gefitinib [Iressa] and erlotinib [Tarceva]. I don’t really know the specific pharmacokinetics that make it seem to work better, or any of the pharmacology of it that makes it more effective. I believe there’s some rationale for why it works better for preventing CNS metastases, but in terms of its overall efficacy, I don’t personally have the answer to that. I don’t know if someone else on the line does.

SHEN: I think the CNS part of it is important. It’s one of the few agents that seems to cross the blood-brain barrier. A lot of these patients fail in the brain, and if you look at the occurrence of CNS metastasis between the placebo and patients in the study, there was a significant benefit there. So, that’s not the whole story, but I think it’s a significant part of the story. For whatever reason, it can get across.

ANTONOFF: I definitely think this is practice-changing. I think this strategy is phenomenal. I would say 1 of the cons is that we need to start profiling more patients. Not that that’s bad, but it’s just more work. Maybe a change to our current protocols. Any other thoughts about this?

SHIEH: For me, I used to send only the IB patients with the higher risk factors, poorly differentiated, etc. But now, because studies like that, I just send all my patients with stage IB to the medical oncologist, at least for them to look at it to see if they do those molecular studies. So because of the benefit of what you just mentioned, I think that all patients with stage IB probably should be at least considered, and have medical oncology look at them.

ANTONOFF: So, how does this choice vary by stage? Are you guys going to change your answer to that question, based on whether it’s stage IB, IB with high-risk features, stage II, stage IIIA? I’ll tell you my opinion, is that I have traditionally given systemic therapy to patients with IB and high-risk features, and stage II and IIIA based on the National Comprehensive Cancer Network guidelines. Based on these data that show a good outcome for patients as early as stage IB regardless of whether they have high-risk features, I am eager to try this in all the patients who are eligible for the trial—trying to base my own practice on what was done in the trial. I'm just curious what you would do. Are you going to change your practices based on stage, or are you going to do it for all these patients?

SHIEH: I think for all the stages: IB, II, and IIIA.

ANTONOFF: That’s my opinion as well; does anybody have a different approach? How are you going to discuss this option for adjuvant osimertinib for your patients? For me, I think it’s just going to be that there may be another treatment available that may be different than [the patient’s family’s chemotherapy]. That’s how I present it.

ONUGHA: Typically, we’ve been doing testing to determine the patients who would be on adjuvant chemotherapy for early-stage lung cancer. It showed that you can determine which patients are high risk for recurrence vs low risk of recurrence. Particularly for osimertinib, 1 of my questions was 1 thing that the trial doesn’t discuss is how long we should continue therapy. Is this something that we should have them on for indefinitely? I don’t know if you have any thoughts on that.

SHEN: In other TKIs, eventually most of those patients develop resistance to the drug, and the AE profile was more significant with some of the other agents. And many of them, if they had a mutation, would stay on it until the AEs became intolerable. I don’t know; this agent seems to have a milder AE profile, but I suppose you might just leave [the patient] on until there’s evidence of progression of disease. It wasn’t really defined in the study. I guess there will be more mature data coming out now that the study was terminated early.

ANTONOFF: I agree. I think we don’t exactly know, so that’s the challenge. Hopefully, as the more mature data come out, we’ll have more information.


  1. Wu YL, Tsuboi M, He J, et al. Osimertinib in resected EGFR-mutated non-small-cell lung cancer. N Engl J Med. 2020;383(18):1711-1723. doi:10.1056/ NEJMoa2027071