The latest in NSCLC with common EGFR mutations
Dr Alfredo Addeo – University Hospital of Geneva, Geneva, Switzerland
Dr Antonio Passaro – European Institute of Oncology, Milan, Italy
Dr Alex Spira – Virginia Cancer Specialists Research Institute, Fairfax, USA
Prof Amanda Tufman – LMU Klinikum der Universität München, Munich, Germany
AA: Hello everyone, and welcome to this ecancer talk. Today we’re going to talk about the EGFR common mutations in non-small cell lung cancer patients and I’m delighted to have with me three terrific speakers that are going to help me dissect a little bit the data and discuss what we are going to see and what we have seen in the recent conference, World Lung Cancer Conference, and at the EMSO 2023. Before we start, let me introduce the three speakers, so let’s start with you, Amanda, please.
AT: Hi, my name is Amanda Tufman, I’m a pulmonologist and thoracic oncologist at the University Hospital in Munich, Germany.
AA: Then Alex.
AS: I’m Dr Alex Spira, I’m a medical oncologist with Virginia Cancer Specialists, US Oncology and Next Oncology.
AA: And Antonio.
AP: I’m Antonio Passaro, I’m a medical oncologist at the European Institute of Oncology in Milan, Italy.
AA: Perfect. Now, we’ve seen great data presented at the World Lung Cancer Conference, let’s start with this. We’ve seen, Amanda, the FLAURA2 data presented with interesting progression free survival improvement. But my feeling was that the reaction from the audience wasn’t that great even if there was a clear improvement in progression free survival compared to osimertinib. What are your thoughts about the FLAURA2? Do you think it is practice changing or we still need more data?
AT: I think the reaction to the data, the FLAURA2 randomised osimertinib versus osimertinib and chemotherapy and it showed that chemotherapy works. And we know that chemotherapy works and we know that there are some patients who have an EGFR mutation who don’t have the time to progression that we would like to see and we know that chemotherapy works in a lot of those patients. So I think what the data showed is that when you combine them then you have treated the patients who will respond well to osimertinib with an effective drug, and you’ve treated the patients who won’t respond well to osimertinib with an effective drug and you haven’t really dissected them. So everyone gets everything in the combined arm. So it’s going to be our job now to try to find the patients who really need the combination and to separate them from the patients who might have had two or three years of good progression free survival without chemotherapy.
AA: But there were some few data, of course we don’t have the full paper, there were a few data about CNS metastasis and the specific mutation, exon 21 with the L858R, where we have the impression that those were the subgroup that seems to benefit a bit more by adding chemotherapy. Do you think that those data are convincing enough to add chemotherapy straightaway in this subgroup or do we still need a bit more evidence? Amanda?
AT: Our practice in Munich up until now with brain metastases has generally been to stratify according to symptoms and then to do a quick check, after about 6-8 weeks, with an MRI to see whether the asymptomatic metastases are responding to what we’re doing. So even if we were to shift our practice there would still be a subgroup of asymptomatic patients who we don’t have to systemically treat for their brain metastases at the time of first diagnosis but who can wait a while, six weeks, eight weeks, under systemic treatment to see whether their brain metastases respond to the systemic treatment.
AA: Now this is interesting. And, Alex, how about you? For the same question do you think, then, now assume you have a patient in your clinic, asymptomatic brain metastasis, exon 21 mutation, so a common mutation, would you offer chemotherapy plus osimertinib if the patient is willing to accept the chemotherapy or would you still want to see more data?
AS: So the brain metastasis thing doesn’t change it for me. I still view it as if you have a limited amount of brain metastases we know the drugs penetrate. I think it gets down to whether or not do we think the combination of chemo in conjunction with osimertinib, and obviously we’ll talk about the same thing with MARIPOSA in a few minutes, the same concept, just different drugs, what does that do? It’s not just the PFS. I think the reason everybody looked at the data and went, ‘It’s okay,’ is we all expected that. There were no surprises here, we’re pretty good at predicting trials and I think we all predicted there was going to be an improvement in progression free survival. But the real thing is what is going to be the overall survival because you can’t treat those people with second line chemotherapy but now you’re just going to give them that longer. People looking at that PFS2, that’s that new endpoint which reared its ugly head, as to what is going to happen for the long term in terms of how they progress, but really overall survival. I think there will be subgroups; I think you will look at it and it will be some patients that have a high burden of disease, it’s going to come down to younger versus older. But to answer your question, I don’t think, for me, it’s going to change things if they have brain metastases or not, especially in the exon 21 L858R.
AA: I totally agree, still we need a bit more evidence, we need to understand a bit more where we are before we move. Sticking again with this first-line setting, as you alluded to Alex, we know the data about MARIPOSA and we know that, of course, if you use amivantamab and lazertinib compared to osimertinib we know that this combination is better in terms of progression free survival. The data, okay, we shouldn’t cross compare but I got the impression they’re not too far from what we’ve seen with FLAURA2 so definitely there’s an improvement. Now, the burning question is how do we choose? What do we do?
AS: That’s a very interesting question and I think people are going to be doing these cross-trial comparisons and there’s going to be a million different opinions. I guarantee some of the thought leaders are going to say one thing of one person and the other to the other thing. So you’re going to be left with two studies that are probably very similar, at the end of the day, without having a formal comparison. It’s going to come down to a couple of things – one is physician’s choice, two is amivantamab at the beginning is a little bit more challenging to give. You have to give more infusions, there’s a risk for reaction, that may be abrogated in a year or two when we start developing these subcutaneous versions which will make life a little bit easier. There’s a certain thought process, you have an EGFR mutation and chemo is bad. We all have those patients and if you can tell them you’re not giving them chemo they’ll feel much better than that. That being said, carboplatin pemetrexed is a pretty well-tolerated chemo, no alopecia, generally well tolerated, you drop the carboplatin after a few cycles. So I think it’s going to be a big debate. Personally, I think it’s going to depend on the patient – what I want to do and what they’re interested in. Patients will naturally gravitate more towards EGFR-directed therapy, I think. But that begets the question how much of this is worthwhile and would you just give osimertinib? There are a lot of patients, older patients, the convenience of a pill that has really minimal side effects, I think there will still be a good percentage of patients, if not the majority, 50%, that will just get osimertinib alone. I think it will be how the physician presents it to the patient rather than the patient asking for something. It’s all going to be in their belief in the physician.
AA: I totally agree and it’s also good that finally we have a few options – we have the osimertinib or we can discuss chemo and osimertinib or we can discuss amivantamab and lazertinib. We can add something in and it’s really cool and we just need to identify the patients. So I totally agree with you. It’s good to have this problem, good to have plenty of choices that we can offer to our patients. Then we know that unfortunately some of the patients on osimertinib, unfortunately we know they respond for a limited time, some of the patients respond for a longer time, but overall say 16-18 months, more or less, is what we’re talking about and then they progress. Now we have also seen very interesting data, which is the MARIPOSA-2, trying to see whether also we can use chemotherapy plus amivantamab plus or minus lazertinib as a second line compared to chemotherapy. The question is, and it’s for you Antonio, we know the data are positive but how positive are they? Do you think it’s going to change practice straightaway? What is your take about the data?
AP: I think MARIPOSA-2 is a crucial trial also to improve our understanding in the first-line setting. Why? MARIPOSA-2 is a very clear study that confirms an improvement in progression free survival for both the combination that includes amivantamab with, of course, a well-balanced benefit for chemo plus amivantamab more than the chemo plus amivantamab/lazertinib for toxicity, more follow-up required for after the amendment. But the benefit is so strong also for patients with brain metastases treated or not treated. In my opinion this kind of benefit is also very interesting to translate, to understand the benefit of MARIPOSA in first line with a comparison with FLAURA2. Because it’s very clear, and was a little bit unexpected, that in MARIPOSA-2 we have a significant improvement in progression free survival, intracranial progression free survival, for those patients that received or did not receive a prior definitive radiotherapy. This is very interesting in the setting of EGFR positive disease, also linked in with the comments of Amanda before, because we know that in the EGFR positive setting more or less 80% of patients during the history of the disease develop brain metastases, some before, some later. This, in my opinion, is very, very important. But, linked in with the comments of Alex, we know that in the future maybe we have three standards of care, not one, because osimertinib, of course, will be there. For those patients that progress after osimertinib today we can say that we have a new standard of care of chemo plus amivantamab. Data are very, very strong, a hazard ratio of less than 0.50, that is impressive in this particular setting with the combination of chemo plus other drugs. Of course, the analysis, biology, of this particular set is very difficult, as you know, and the understanding of resistance mechanisms is very important to improve more the benefit to our patients. But we know today that amivantamab is a double block drug with on-target alteration EGFR/ MET and more or less 25-50% of these kind of patients generally show resistance on EGFR/MET dysregulation. So, for me, this is very, very strong data that also helps the management, the view of the first-line setting where, of course, two new standards there. But maybe with this data on the efficacy in intracranial progression free survival in MARIPOSA-2, I think that MARIPOSA will be more strong to understand the pathway as a patient.
AA: I agree and also that’s perhaps a personal question I have. It’s something we don’t do very often in Europe, in the States you do it a bit more often there. So when a patient progresses on osimertinib in Europe normally we have to drop the osimertinib and change whether it’s chemo, whatever it is, we drop it. Now, I know in the States they might stick to osimertinib and the question is, given also the data we’ve seen in MARIPOSA and assuming lazertinib is not available, is there any patient where you would want to continue osimertinib and add chemotherapy amivantamab if you could?
AP: Yes. So MARIPOSA-2 showed two different arms. Lazertinib/amivantamab/chemo and amivantamab/chemo. But the problem is that today we have no strong support to understand that the real impact of the figures for four drugs because after the beginning of the study there was an unexpected high toxicity, hematologic toxicity, in the LACP arm, the arm with the four drugs. But when we consider the overall data for progression free survival, also the progression free survival, it’s very clear that the key is amivantamab, it’s not the addition of lazertinib. It’s very, very clear when we look at the Kaplan-Meier curve. When we look at the progression free survival this is very interesting because in the past we thought that maybe prolonged use of osimertinib, we are prolonging the intracranial progression free survival. But it is not so clear when we look at the MARIPOSA-2 whether the addition of amivantamab is the real key there.
AT: I guess the question of continuing osimertinib and adding chemotherapy is what the COMPEL study is trying to recruit to and not recruiting particularly well, possibly because those who are convinced that adding chemo to osimertinib are doing it off-trial and those who are not convinced that that’s what they should be doing are recruiting to other trials. The question of whether we should continue osimertinib while giving chemotherapy probably depends on the type of progression that the patient is having. If you have a patient who had 30 metastatic lesions and of those three or four are progressing, or a couple are progressing, then maybe there’s a good argument for tumour control of the other lesions and continuing osimertinib. But if you have a patient who had four lesions to begin with and now they’ve got twenty, including the four they had before and everything is progressing, then I don’t see the argument for continuing osimertinib in that setting.
AS: This comes up in the United States all the time because we’re a lot more liberal with what things do. I am amazed that it’s often said, ‘Just leave it on because we think it gets into the brain a little bit better.’ That’s really the argument and that’s why they came up with LACP versus just ACP. If you look at the arms there, there’s no difference in MARIPOSA-2, it is dead on. We don’t think there’s a meaningful difference at all between lazertinib and osimertinib. So I think that’s a compelling argument not to do it and you’re talking about a couple of things. First is there’s probably an increased risk of blood clots with lazertinib, which is a thing. Two is there is increased risk for toxicity and, three, you can’t ignore costs in all these things, these drugs. So, for me, that’s compelling evidence not to do it and to get away from this, ‘You should just do it because…’ at least in my mind. That’s how I view it right now but in the United States there’s going to be these arguments, it will be the same arguments that we’ve had before. In the United States about 75% of people do leave on the osimertinib right now at progression when they’re giving chemo.
AA: To be honest, you’re right. The reason why we don’t do it in Europe is not because we don’t want to, very often because we’re not allowed to so we have to stop using it. Even if some of us maybe might want to. Now, maybe it was the right thing to do, not to necessarily continue this. I agree that it also again depends on the metastatic sites but MARIPOSA-2 is answering a lot of questions, honestly, so it’s a great study. Again, the issue is now the field is moving very quickly so now we can talk about what we do in second line or what we used to do. If patients are on osimertinib maybe this might change later on, once the patient is not going to be just on osimertinib. Let’s stick to what we have for now, so now, assuming the patient is on osimertinib, there’s progression, I go back to you, Amanda, what is your standard practice if somebody has multiple or clear progression on osimertinib? What is your practice, what do you do?
AT: So we try to reassess them biologically and I believe that at the moment tissue rebiopsy is better than liquid biopsy but if tissue rebiopsy isn’t feasible or the patient doesn’t consent to it or it’s deemed too dangerous for other reasons then we do a liquid biopsy. But if that’s not the case we try for a tissue biopsy and we redo molecular analyses. Sometimes there’s some uncertainty about the histology but usually it’s not a small cell transformation, usually it’s a non-small cell and then we look to make sure that the EGFR mutation has been maintained and we look for other alterations. MET amplification keeps popping up as a potential targetable mutation, a few HER2 changes have popped up, but the other changes are fairly rare. So usually we confirm the presence of the original mutation and then we discuss with the patient what we should do.
AA: And going back to what you just said about MET amplification, we have a few data and we have already seen the data about the INSIGHT, so when you combine the osimertinib plus the MET inhibitor. We have some good MET inhibitors now, the INSIGHT is with tepotinib, and the question is would it make any difference it was capmatinib, I’m not sure. But the data was about tepotinib so I will stick to the data and here’s the question for you. If you already had experience with this combination, is it something you can easily get in your country because in other countries it might not be that easy? And the second question related to this is we all perform this analysis, but I wonder, we do it because it’s interesting but with non-small cell lung cancer you don’t have access to other drugs. What’s the value? If we cannot add another targeted therapy, because we can’t, is it really relevant? So that’s the question for you.
AT: So to the first question, I have had a small handful of patients who I have been able to treat with an EGFR inhibitor and a MET inhibitor. Two of them have had the combination osimertinib/tepotinib, one of them has had to reduce the dose of tepotinib because of peripheral oedema but otherwise it’s been a combination that we’ve been able to give. Previous to that we had some experience with older combinations going back to the era of the first and second generation EGFR mutations with MET changes that came up. One drug that we had access to before tepotinib was cabozantinib and we were able to give that off-label after asking the insurance provider for permission and had good tolerability and some efficacy with that combination. Another interesting aspect of diagnostics that hasn’t really made it to the forefront, we’ve talked about liquid biopsy, we’ve talked about tissue biopsy, but there are some very interesting technologies coming up in nuclear medicine looking at things like MET PET. We know from the setting of prostate cancer that PSMA PET is a very interesting thing and MET PET is certainly coming and it will be interesting to use that type of technology as a global look at what the expression is of particular markers.
AA: This is very interesting. So, to you Alex, I know that you practice slightly different because in the States I imagine there would be more use of liquid biopsy, I imagine, I don’t know whether this is something you do routinely. So do you do both in terms of progression or it depends on case by case?
AS: At tumour boards I say, ‘Do more solid tumour biopsies,’ than in real world. It’s very easy to… Because the patient has progressed and they want a plan, they want to know what to do. The liquid biopsy you get back in seven days. I don’t know what the timing is there but especially since COVID it’s taken us two weeks to get a biopsy set up and then you have to do the full NGS panel, so that’s a month. And in that time there’s anxiety amongst the patients. So I do do that but I’m really relying on the liquid biopsy for the most part. Also what I’m finding is that when you’re getting that progression it’s a little bit of progression and you’re always questioning how easy is it to get that biopsy, which is all experience. Now, going back a second, I used to think about it a lot more because, like Amanda said, options regarding MET. That may change a little bit right now because amivantamab in the second-line setting hits MET. So you asked the question does it matter? And obviously there’s other things like HER2 but then you’re getting a little bit more really off-label and rare things. There’s obviously C797S but amivantamab may hit that as well. So I’m fully on board with this but I think it’s going to matter less now because amivantamab may hit a lot of those, especially the most common ones which is if it develops a MET amplification or the rarer exon 14 skipping event.
AA: I agree, the feeling we all have, basically, is that we’re going to use more and more amivantamab in this setting. It’s either going to be first line or second line, depending on where you’re practising, but definitely it’s going to be more and more. Maybe even for MET amplified it’s going to be a possible, or we’ll see because we don’t have the data. So this is very interesting. Now, moving towards more ADC drugs, because, of course that’s the hot topic and maybe we’re going to see some data. We’ve seen some data from TROPION-01 for lung, it’s a different setting but we’ve seen already there they’re coming. For HER2 we have seen the HERTHENA-01 study presented at the World Lung Cancer Conference. We’ve seen patritumab deruxtecan anti-HER3. What are your feelings? Is this something that might play a real role? Is it the fact that we have an ADC which has in theory a target but nobody measures any biomarkers, it doesn’t make any sense to you. What’s your take?
AP: We are jumping from improving our understanding with the liquid biopsy, with tissue biopsy, to identifying new biomarkers to treat our patients in a selected way to ADC that, at the present time, EGFR positive disease are very biomarker agnostic. So we do not need tissue, we can use these kind of drugs, regardless of the kind of mechanism of resistance. But, in my opinion, this would be not the most attractive option because data are very interesting. The HERTHENA-Lung01 is very interesting in the pretreated patients but the benefit is not so strong as expected. This is, in my opinion, related to the heterogeneity of the disease but also because maybe we need some biomarker. Maybe it’s not HER3, maybe it’s not something that today we know but we need a biomarker. Indeed, for example, in this particular setting in the future the use of that’s an ADC that targets MET hyperexpression evaluated by immunohistochemistry – very low cost evaluation – showed preliminary data overall response rate of more than 60%. So the idea is that the place for these drugs is here but not for all patients and we need to try to individualise the treatment, in particular to the resistance. Because these are very smart chemo, not targeted agents, smart chemo, and understanding the biology of the disease to improve the life of our patients, in my opinion, is important. Also because I have some limitation to understand the brain permeability of this drug. So we have data of patritumab of activity on the brain but I’m not sure that we have strong brain permeability with an ADC on the brain. So we need to wait for the data, more robust data. But also here at ESMO was presented the data of a bispecific ADC, EGFR, very, very promising. I think that the way is here, on the road, for the management of patients with EGFR we’ll be able to use an ADC but maybe today is too preliminary.
AA: Yes, I think it’s too early. I totally agree with you and if I could I would probably talk forever because it’s very interesting. But for the sake of time probably the last question for all of you, so I’ll start with you, Antonio, and it’s the eternal debate I have with some of my oncologists about, ‘Oh, patients with high PD-L1, EGFR positive, progression, chemo or chemo and osimertinib, whatever.’ Then is there any room for immunotherapy in EGFR? Because we have seen lots of data now, we’ve seen also presentations, the Impower151 at the World Lung Cancer Conference, now we’ve seen ATLAS presented yesterday, we have seen before the CheckMate 789 and they’re all not necessarily positive. Do you think there’s any room? Would you still use immunotherapy or can we definitely bury this finally?
AP: For me there is no room for IO in patients with EGFR positive disease. We have all the signals according to a not very clear biomarker in a setting where the use of IO, according to the biology of the disease, is not recommended here. So in my opinion we need to prioritise other kinds of drugs and we need to prioritise our patients for clinical trials without IO in this particular setting.
AA: So is there agreement?
AS: I agree completely. Although in the United States the IMpower data with carboplatin and paclitaxel, Impower150, has made a lot more headway. I don’t like the regimen; some people substitute pemetrexed. Because, as you said, Antonio, the science behind it doesn’t make sense. Now, the only thing I will say is there has been very interesting early data with the bispecific which is an angiogenesis inhibitor of VEGF with PD-L that’s now starting a phase III randomised clinical study that actually has some interesting… I don’t know how much is the VEGF component versus the PD-L component but the scientist behind me says it should have nothing to do and we should stay away from it. So I would have a low priority but in the United States especially there’s a lot of people that still believe in using that although that may go away now with amivantamab.
AA: Yes, but all the stories, but also the data. We’ve seen Impower151 and also the, they all confirm with that.
AS: But there will be some hold-outs in the United States, there are some IMpower believers. We’re different than you guys.
AA: I don’t know but, yes, I understand it might be tricky. How about you, Amanda? Is there still something in your practice or in your country? Is it very much used, the chemo-IO?
AT: It is or has been until now for patients who haven’t qualified for a study. A common response to diffuse rapid progression under osimertinib, in particular if the progression is associated with liver metastases, in particular if it’s a high tumour burden and you want to hit it hard. So it has had a lot of use. I think its use is going to decrease and I think the very conflicting data, the fact that we haven’t been able to reproduce the benefit of checkpoint inhibition looking at the Impower151 data, looking at the KEYNOTE data in that setting, I think speak very loudly that we shouldn’t be including a checkpoint inhibitor in that setting. But there are exceptions and those define the rule. Smoking history is an important factor. We have some EGFR mutated patients who are or were heavy smokers and I would be certainly more inclined to give it there. After a second and third line there’s always a fourth and fifth line and I’m not willing to say that there’s never a role for a checkpoint inhibitor, especially in a high PD-L1 expressor with an EGFR mutation and no other choice.
AA: Perfect. Very, very last one, I promise. We’ve seen also the ADAURA and the OS data. Given all the data we’ve seen in first line, should we perhaps use some of the first-line data we’ve seen in the adjuvant setting? Do you see any idea to perhaps try amivantamab or osimertinib in the adjuvant setting?
AP: There is some in the field to evaluate the possibility to use amivantamab in the advanced setting but we need a very long trial with a long follow-up. Since I think that now osimertinib is a very strong standard for our patients to improve the survival so we need to consider osimertinib in adjuvant and we need to consider that we have a lot of options after progression to try to improve the OS, of course, of these particular patients.
AS: I think that’s going to be a hard sell in the adjuvant setting. The reason osimertinib works so well is it’s tolerable. Amivantamab is tolerable but not that tolerable.
AA: And then we also know that for metastatic adjuvant the perception of the tolerability changes a bit so even for those. So, perfect, thank you very much to all of you and thank you for listening to this ecancer talk. I hope you have found it educational and interesting and thank you very much.
