NG: Welcome to this ecancer discussion on the second-line treatment for patients with EGFR mutant non-small cell lung cancer. I’m Nicolas Girard from the Institut Curie in Paris, I’m a thoracic oncologist and I’m sharing this discussion with three outstanding colleagues. Joshua?
JR: My name is Joshua Reuss, I’m a thoracic medical oncologist at Georgetown Lombardi Cancer Center in Washington DC and thank you for inviting me to be part of this discussion.
NG: Alfredo?
AA: Hi, I’m Alfredo Addeo, I’m a medical oncologist in the University Hospital of Geneva and I’m extremely pleased to be here with you guys today. Thank you.
NG: And Keith?
KK: Hello there, my name is Keith Kerr, I’m a thoracic pathologist from Aberdeen in Scotland.
NG: So maybe we can start with the first line management of patients with EGFR mutant non-small cell lung cancer. Actually, the population of patients with EGFR mutant is quite heterogeneous – we have seen data in the common mutations, the L858R, the exon 19 deletions. We have also a group of patients with uncommon mutations. It’s very important to differentiate these two groups. Alfredo?
AA: I think this is a good point. First of all, as you said, we have to understand that we’re talking about a heterogeneous group although the uncommon mutation might represent about 10% probably. But it’s still very important to detect these mutations, therefore testing is essential and we have moved forward a lot now with testing, at least in many centres the NGS is available so we can detect uncommon mutations. Now of course we have to discuss how to treat those uncommon mutations because the data are limited. We know there are some data with afatinib, which is available; there are some data with osimertinib and hopefully we might contribute soon to more data on osimertinib in uncommon mutations. But the key message is it is heterogeneous – we need to run the full NGS analysis to make sure we don’t miss some uncommon mutation because it would be a pity, rather than focussing on just the most common mutations, personally.
NG: Still, Keith, some time in the specimen pathway there is a first step with a rapid testing for EGFR mutation. Because we know that these are the most frequent mutations that we can target in the first-line setting. And then a second step with more NGS and more comprehensive genomic profiling. What are your thoughts on that?
KK: Yes, I can fully understand the reasons why there is a wish to do rapid testing up front and then revert to NGS afterwards. My problem with that particular approach is that many of the samples that we have are really quite challenging and quite limited in the amount of material. Essentially what we’re talking about here is testing twice. My feeling is I just wish, and this is, of course, in a perfect world, that we could do it once, quickly, and find what we need to know. For some laboratories and with some NGS approaches it is possible to actually deliver quite a quick service, it depends how quick you really need the information. But within 5-10 working days is perfectly reasonable in a well-organised system; unfortunately not everybody works within a well-organised system. But it’s a challenge.
NG: The EGFR mutation testing, Josh, is part of a global assessment of biomarkers for deciding the treatment of patients in the first-line setting. What are those biomarkers globally and can you elaborate on that?
JR: Absolutely. To the point of our discussion here, the rapid component can be supplemented with liquid biopsy oftentimes. So my practice is typically to actually get both a liquid and a tissue biopsy up front. Maybe we’re a little less limited by costs in the United States, for better or for worse, but I think that will oftentimes allow us to get that profile up front where we can get the full complement of mutation profile. If you don’t find what you need on the liquid you’ll get that back in the tissue. Then ultimately to your point about biomarkers, seeing the PD-L1 status as well, so that way if you don’t have an actual driver mutation you can utilise PD-L1 and then maybe some of our other co-mutations that we’re learning for immunotherapy resistance to guide your front-line decision-making strategy.
NG: Very important, as Josh said, to wait before the results of the NGS, even if you have a high PD-L1 because a high PD-L1 is actually frequent in the setting of oncogenic alterations but the efficacy of immune checkpoint inhibitors in this setting is actually limited. So it’s two different strategies, two different pathways, and you need to wait for the NGS otherwise interpretation of the PD-L1 status is not possible.
AA: I totally agree. My add on it is that what I tell my young oncologist colleagues is don’t rush, wait, but if you need to give something just give chemo and wait. So if you think the patient needs some form of treatment, while you’re waiting for your molecular profile just give them one cycle of chemo, it will buy you three extra weeks at least.
NG: Without…?
AA: Without immunotherapy, of course. Without immunotherapy. Then, if there’s no mutation you will decide what to do – you can add on only immunotherapy and/or you can start. Because we also need to bear in mind the extra toxicity that we could give to the patient if we start a targeted therapy after we have used immunotherapy. So we’ve got to be cautious but sometimes we have to start treatment. Give one cycle of chemotherapy without immunotherapy and wait for the NGS test to get back.
NG: That’s a very important point for clinical practice. So then once you get a common EGFR mutation the strategy relies in the majority of cases on osimertinib, actually, right now. There are multiple clinical trials ongoing in the first-line setting to assess the efficacy of new strategies but, as of today, a majority of patients are treated with osimertinib. So what’s your follow-up of those patients? MRI every two months, every three months? Josh maybe?
JR: Absolutely. So initially I’ll tend to see my patients back around 1-2 weeks after starting just to make sure they’re tolerating the therapy well, any toxicities – rash, diarrhoea, etc. – check blood work. In terms of imaging usually I’ll get my first scan around the eight week time point, CT scan and then after that every three months. The MRI definitely depends on the presence or absence of brain metastases. In patients without brain metastases, though, I will still get a routine surveillance brain MRI at least yearly, maybe every six months, just knowing the tropism for the brain.
NG: The key question is what to do when the tumour starts to increase or if you start to identify a new progressing lesion. This is also the issue of second line after osimertinib. Alfredo, what is your clinical management in this setting? Are there any different phenotypes of patients that you may consider? When do you start planning a rebiopsy, tissue or liquid?
AA: It’s a good question. We all know, and it’s true, that after osimertinib if there is a progression, at least I think, we’re in trouble because we know that if you look at the standard of care today we don’t have much to offer other than standard chemotherapy treatment which is effective but we know that the efficacy is limited. So once we reach that point it’s challenging. Now we’ve seen some good data now with the rebiopsy and we’ve seen the [inaudible] study that presented and also showed that the most frequent mutation we can detect is a MET amplification. We now have some good MET inhibitors and we also see the INSIGHT 2 study with tepotinib combined with osimertinib. Now, of course this is not my practice because I don’t have access to this drug, so in reality the practice is you do biopsy, A, to rule out small cell transformation, although it’s relatively rare but you don’t want to miss that. Other than that, if you don’t have a trial available, then you’re stuck with chemotherapy treatment. So I’m hoping that, given the data we have seen with MET amplification, the efficacy of tepotinib combined with osimertinib may help us to get perhaps this drug. If I had the possibility to use it in a MET amplification, no doubt I would combine tepotinib with osimertinib. Unfortunately this is not something I can do. But, again, it’s more tissue biopsy. Liquid biopsy, at least in my centre, we’re a bit behind. I don’t think it’s that simple to detect MET amplification, frankly, through liquid biopsy. You might detect other mutations but they might not be necessarily clinically relevant. So standard of practice chemotherapy; as soon as I could have access to a MET inhibitor I would certainly push as much as I can to combine with osimertinib in MET amplified patients.
NG: Keith, MET amplification – how do you assess? What are your tools?
KK: Yes, I think this has been an issue, a genomic alteration that we’ve dabbled with from time to time over many years now. But it really seems to be coming back to the fore because of the osimertinib story. I actually suspect that the rates of MET amplification that are quoted in some studies may in fact be underestimated because of the way that the assays were performed. That’s based on an understanding that, for a variety of reasons, an approach of next generation sequencing, be it on tissue or blood, or a reliance on blood, actually is missing many more cases than we perhaps initially thought. And there are a variety of technical reasons for that so that MET amplification is going to be a deal, especially given the amount of osimertinib that is being used and we’ll have to address it. For the time being, it looks like in situ hybridisation, probably FISH but there are actually easier brightfield techniques as well, will probably be the go-to technique which means a tissue biopsy, of course.
NG: That’s a very important point because historical data with MET activation or MET amplification post-EGFR TKI, it was complex because it’s a continuum of copy number gain until amplification and a true efficacy of a MET directed therapy. Josh, any new compounds that you are excited about in the post-osimertinib setting?
JR: Sure, and just to touch on a couple of points as well, one thing that makes this situation difficult is that these mutations are sub-clonal. So even if you do find something targetable, it’s likely not affecting all aspects of that resistance. Or, as we saw in some of the research, when you gave tepotinib monotherapy that was a very small response and it did not last very long. So it’s important to know that these are sub-clonal alterations, you likely still oftentimes need the osimertinib to maintain that tumour control in the areas that aren’t progressing. The other thing I’ll say is that oftentimes it depends a bit on the site of progression so if I have oligo progression that might be a situation where I would pursue radiotherapy and keep the osimertinib going as long as I can. In terms of additional therapies there are definitely a lot in the pipeline ascending. Amivantamab is one promising drug, the EGFR MET bispecific antibody, especially since that is also targeting obviously the most common targetable mechanism of resistance in MET mediated resistance. I think there are a lot of encouraging signs for that up front as well as combining osimertinib with chemotherapy up front. So, in many ways, maybe the up-front strategy is poised to change sooner than the subsequent line but I guess time will tell.
NG: Yes, if we use an anti-MET in the first line setting maybe there is no more need, actually, to develop MET…
AA: Yes, so we’ll change completely. We’ll change the post-progression molecular profiles, for sure. It will supress some form of clone, it will, unfortunately, there will be new sub-clones developing. So it will be completely a change in scenario.
NG: There are other drugs for which I am excited, actually. The ADCs, the antibody drug conjugates. There are some data with patritumab deruxtecan which is a HER3 antibody combined with deruxtecan, so it’s targeting the EGFR and maybe not only the EGFR mutant but at least EGFR expressing and HER3 expressing cancer cells which are actually a frequent event in patients with EGFR mutant non-small cell lung cancer. The reported response rate was as high as 32% in the late, late, late line setting in patients heavily pre-treated, all received osimertinib, chemotherapy and another line of therapy. It was also interesting to see that this ADC, it’s not really a targeted agent, just putting the chemotherapy into those cancer cells is kind of personalised chemotherapy. To me it was interesting to see that the efficacy was independent from the resistance mechanism. So even cancer cells harbouring a KRAS or a BRAF mutation, what we can see in the late, late lines of progression in EGFR mutant non-small cell lung cancer, you had some efficacy. So maybe this is another strategy, I don’t know whether it will move to earlier lines. But at least thinking about what you can do in refractory tumours is maybe an option.
AA: I think it’s a good point. At least from my experience, one of the struggles I have with patients harbouring an EGFR is once you start an oral treatment first line it’s always very difficult to go with any treatment IV – any, any. It could be amivantamab, it could be chemo, it’s difficult to move from oral to IV. It’s true that maybe you want to move the ADC first line but it would be a change in the way we treat those types of patients perhaps.
NG: Actually at the end my perception is that we need efficacy in those patients and prolonged duration.
AA: Totally.
NG: I’m not saying that making the diagnosis of an EGFR mutation in a patient with lung cancer is bad news but when you think about other addictions such as ROS1, ALK, those patients have multiple lines of targeted treatment; median overall survival for ALK is probably seven, eight, maybe nine years. So at some point in the EGFR mutant space the survival is limited; in FLAURA it’s 38 months median. So it shows that we need additional options in those patients. Given the ongoing trials in the first-line setting, you will have to do some kind of IV – chemotherapy, amivantamab.
AA: Absolutely.
NG: One question still, because antiangiogenics, it’s another story in the EGFR mutant space. No biomarker, Keith, for antiangiogenics but some data – a TKI plus antiangiogenic drugs or chemotherapy plus antiangiogenics after the exposure to TKIs. Josh?
JR: Yes, and I would go as far as even to say maybe chemotherapy with immunotherapy and antiangiogenics as well. It’s an interesting story. Obviously the larger randomised studies have shown improvements in progression free survival without so much translation to overall survival. So it depends on what is considered a meaningful, clinically relevant outcome. With regard to the combination with immunotherapy that’s a very interesting story, definitely one that needs to be flushed out, knowing that these are patients that tend to not respond as well to immunotherapy-based approaches. But there is definitely scientific rationale for it, knowing how EGFR drives VEGF in the setting of EGFR mutant lung cancer and how that changes in the setting of post-progression on TKIs. So it’s a very interesting story that while it has been well played out the story is not over yet.
NG: No the story is not over because we have those multiple opportunities after the failure of osimertinib. These are moving to the first line somehow so the strategy will have to change but also we are actually changing the natural history of the disease, the biologic activation of molecular pathways will be different after one or another treatment. So hopefully we will be able to build this puzzle of all the options that will be available in that field. Maybe one take-away for people looking at this video? Josh?
JR: I would say that use the data you have to make the best decisions for patients in each line. So make sure you wait for the next generation sequencing, the biopsies, to guide your treatment strategy and in the setting of progression use all the information available to you, both imaging and hopefully repeat sequencing to make the best informed decisions for your patients.
AA: I totally agree, this is pretty much my point. If I can maybe give a stronger message, or a very strong take-home message from me, is from tomorrow onwards I would even look more carefully to detect MET amplification. If I could get access to a MET inhibitor, I would certainly combine it. This is, for me, maybe it’s over-enthusiastic but I would say, for me, it’s practice changing, personally. I don’t think we need any more data to convince me that this is the way to go.
NG: I agree with you. Keith?
KK: Yes, as ever, as we have said, finding targets and alterations is very important. As we start to have to deal with the situation with relapse on a TKI, and this even goes beyond EGFR, but EGFR is the one that we have most experience with, eventually we will have several options, probably, depending on the resistance mechanism. We have to identify those mechanisms and it’s going to be a combination probably of blood but I think tissue biopsy is going to have to be an important part of the strategy for a number of reasons, including the MET amplification story that is really emerging and clearly important.
NG: I believe what we also highlighted during this discussion is a multidisciplinary management is required for those patients. You need to discuss with colleagues, each patient has to benefit from an individualised strategy, this includes rebiopsy, discussing with pathologists, molecular biologists. This is really a setting where multidisciplinary discussion is really important. Thank you for this great discussion, thank you for sharing your time with us on this discussion. Well, enjoy the rest of the ESMO meeting.
