Solange Peters:
Dear colleagues, it's my pleasure to welcome you to this first Lung Cancer Session on VJOncology. We'll try to cover the main topics, and there is some subjective matter here. And which were discussed, which were presented at the ESMO 2020. We hope that this would potentially compensate or at least help in circumstances where we didn't have the opportunity to debate, to discuss, to share the feelings we had after the presentations together. So, the idea of today is to try to have this discussion virtually. It's my great pleasure to have with me through this discussion, two friends and colleagues, Martin Reck from Grosshansdorf in Germany and Sanjay Popat from the Royal Marsden in London in UK. My name is Solange Peters, and I'm working as medical oncologist in Lausanne in Switzerland.

Solange Peters:
So, we can dive into that. And we have many, I would say, chapters that we can cover during this ESMO meeting. And maybe the first chapter, which is the usual one when we discuss new data, is the first one of oncogene addiction. Molecular targets, actionable targets. I think the most important one in term of still building new evidence might be this very small niche of the ALK rearranged non-small cell lung cancer. We have been seeing the series of compounds replacing crizotinib time after time, ceritinib, alectinib, brigatinib, and dartinib. We were strongly awaiting the data from probably the strongest one, the ones which we handle with a little more difficulty, which were the lorlatinib in the CROWN Phase III study against crizotinib.

Solange Peters:
And these data to me were questioning myself of course about frontline strategy, because I was surprised by the magnitude of benefit. From lorlatinib as compared to crizotinib and as compared to what I, was expected, probably as compared to the previous data presented. So my question to you both, and one after the other one, maybe Sanjay first and then Martin. What do you think about this opportunity of changing again our considerations about frontline strategy? And also what we know from lorlatinib is it's a wonderful second-line or even third-line compound, so we like it. We like to have a next idea, something in the pocket. So do you think that the community will put the best first, decide for the best first? Or keep this reserve or this subsequent choice, also because toxicity profile, what do you think?

Sanjay Popat:
Yeah, thanks a lot. It's a really good question there. I think CROWN has taken us all by a bit surprise, because of the magnitude of benefit. And this is a randomized trial of lorlatinib versus crizotinib in frontline, patients with metastatic ALK-positive disease, chemotherapy wasn't allowed. And the hazard ratio was 0.28, which is spectacular and well. But that it's so much better than what we've seen previously with brigatinib, ensartinib and alectinib which is all around the 0.45 mark. So on paper, by far this is the most potent drug and we use it upfront. But there are questions which I think we need to consider. First of all if we use it upfront, what is the resistance mechanisms and what are we going to use next? Once we have a better understanding of the resistance mechanisms, when lorlatinib's used as a salvage drug, we don't really have a good understanding of what the resistance mechanisms are next.

Sanjay Popat:
And so, will we be able to sequence additional ALK inhibitors? Are we sacrificing progression-free survival gain for a lack of overall survival benefit? So you could argue that maybe whilst we've got 60%, 70% maturity of this dataset, is a relatively mature dataset, maybe we need to wait a bit longer to see what's happening with overall survival before we make the big switch. And then the big issue is handling some of the rather unusual toxicities that we see with lorlatinib, 90% of people get dyslipidemia and will require some sort of statin usage. Around 40% of people will get cognitive or mood effects, changes, which can range from no problems whatsoever to very mild to moderate or even severe. All of these things need to be considered, particularly if patients are going to be on this drug for quite a long period of time. I think there is a big debate. For me, just it's not a slam dunk winner. There is a debate, and I very much look forward to what Martin thinks.

Solange Peters:
Yeah. You spoke, you evoked two points which are quite important. I think the first one is disease data on the contrary of small talks are quite mature for PFS, right?

Sanjay Popat:
Yeah.

Solange Peters:
And so, it's not that it's ... you remember all brigatinib data came very early.

Sanjay Popat:
Yeah.

Solange Peters:
These data don't come very early, they show you the PFS. But you spoke about OS, and it might be another parameter to help us decide. And because we have OS data from ALEX, from alectinib trial. The other thing, you spoke about mood, cognitive, speech problems which might of course be of importance for the quality of life. And resistance mechanism, I remember Alice Shaw telling that what she liked with lorlatinib is to get resistance you always need a kind of a combination of several complex mutations, so she was hoping that these accumulated events would happen with a lower probability. Of course it's still very hypothetical, but that might be why some hopes might be put in the game. But I see Martin going like that, so Martin, what are your thoughts about adopting CROWN data and lorlatinib frontline?

Martin Reck:
Well, I think I completely agree with Sanjay. So, the data are really impressive. But that there is some discrepancy between the use in daily clinic and the data, in particular when we talk we talk about tolerability. Yes, you talk about the potential psychiatric or neurological problems, but there is this problem of edema hypertriglyceridemia. And just to remember, the patients are taking these drugs for years. I have some experience where this really caused some decrease in quality of life and well feeling, and this is something to put into consideration. When we treat out patients for years with targeted agents. So I still like this as a reserve drug, as a second line opportunity we have. Have the lack that we do not have sequential data for the sequence of different agents, I really was blown away from the brain data too.

Martin Reck:
And when I look to the time to progression, there we do see a hazard ratio of 0.07, which is really amazing. So, the brain activity of this drug seems to be really more than expected with any other drug. And so in particular when I have patients with disseminated brain metastases, where I came to an end with my opportunities of brain irradiation, this certainly will be something on my list to look for. But again what comes next is a practical problem, besides all the trial data that we have seen so far. It's really amazing to see this data. We need some more maturity, you mentioned the ALEX survival data. Which are also very, very impressive, five-year survival data that we haven't seen so far. I think we need some more maturity on the overall survival, and we really need a good look on the tolerability. It's not only the grade 3 toxicity, it's also the grade 2 toxicity which will cause an impact on the quality of life on the individual patient for a long time.

Solange Peters:
Yeah. So ALEX data with still, at five years median OS not being reached, right? And hazard ratio of 0.6 something, so very nice data. Importantly just two things. At resistance what we have been seeing, it was published in The Lancet Oncology and later on in JCO by the group of the lorlatinib Phase I/II trial. What is interesting with lorlatinib is that you can give it irrespective of the mechanism of resistance. It work well in ALK mechanism of resistance, but also in what we call ALK-independent mechanism of resistance. Probably a little, a bit better when you have a resistance mechanism independent of ALK, but in both scenarios which is a nice opportunity. I had a little talk with some colleagues from Israel. What they do in Israel, is they give first alectinib then brigatinib, then lorlatinib.

Solange Peters:
So, my question to you is to date coming out of ESMO. So the take-home message, what is still your I would say, attempt to treat ALK-positive patients? What do you ask for to the insurances? You ask for lorlatinib, or you ask for lorlatinib in second-line only? Or in third-line, what is your current strategy to date? Sanjay?

Sanjay Popat:
At the moment, I'm not starting with frontline lorlatinib. For me, it's a very good salvage agent. Particularly for intracranial disease, and frontline alectinib or brigatinib are very reasonable frontline options. And I think the strategy that you proposed starting with alectinib, using brigatinib next for an intracranial salvage and using lorlatinib as a third-little bit agent is a perfectly rationale approach in that setting.

Martin Reck:
Yeah, I agree. So I think that still the opportunities of alectinib and brigatinib are very strong first-line opportunities, also in terms of tolerability. So patients are well for a very, very long time with a stable disease or response of the disease. This still would be also after ESMO 2020, my first choice.

Solange Peters:
Perfect, thanks a lot. Let's switch from ALK to EGFR, we have to do that. So we have been seeing at the plenary session at ASCO and again at the plenary presidential session of ESMO that there is a role for TKI, undoubtedly to be discussed in the adjuvant setting. In the early disease as a complementary technology, technique, strategy, after the standard of care. So without negotiating the standard of care, should be fully resected, should be adjuvant chemo. But later on, adding osimertinib was bringing an amazing hazard ratio. In terms of PFS improvement in stage Ib, in stage II and in stage III, knowing that trial design of the ADAURA trial was focusing on stage II and stage III and based on the 7th TNM classification. But amazing data completed as ESMO with the very early dataset, showing that some of the benefits is related to a wonderful control at the brain level.

Solange Peters:
This was quite expected, and the fact that of course osimertinib has a protective effect on the brain. But is it enough to complete the datasets, to convince everyone that all patients with EGFR-mutated tumors which are large enough or have lymph node invasion should receive osimertinib in adjuvant for three years? And Martin, I give you the world. Because you're also in Germany, you can see where the drugs come very rapidly into your pharmacy. So are you ready, as soon as it's going to go through EMA to use it in the adjuvant setting?

Martin Reck:
Well, I would say this is a very convincing signal. And I also can tell you, we had various debates on national and international level how to interpret this data. And there are a bit of concern whether this really is a better cure for the patient, or whether this is just a prolongation of the progression due to the mode of action. There are a couple of questions still around. So I would say this is a second step, to implement this, I would love to see some more mature information on overall survival. I'm a bit puzzled from the Chinese data. So it's a different trial design I have to say, but I would like to see a little bit more on maturity on overall survival.

Martin Reck:
The other point and I think this is not clearly answered by this trial, is the question whether we still need adjuvant chemotherapy in patients with EGFR receptor mutations, yes or not. Whether we are able to substitute the chemotherapy by osimertinib. And I think this is still an open question, which hasn't been resolved by the ADAURA trial. So, I will say we are on the way to prepare the stage for osimertinib, we need a bit more information.

Solange Peters:
You open a risky chapter here, because the role of chemo was not the question of the trial, right? And that the reason one, it was inadequately addressed, right?

Martin Reck:
Yeah.

Solange Peters:
So this study, even retrospectively, it would be probably to my opinion a little too precipitated and too fast to tell that chemo might be forgotten about for some patients' population. Because obviously chemo was let to the investigator according to local standards, and it was not a randomized strategy there. So, that's one thing that I like Sanjay to comment on. And the other thing is about the CTONG data where a first-generation TKI was improving PFS, but not OS. There's a story of the brain change something. Because of course first-generation TKI have very limited efficacy in preventing brain damage, brain lesions. Does this change something in too, even if the OS is not statistically significant? Does it change the picture, the fact that the patient can stay away from brain medications for a longer period of time? Sanjay?

Sanjay Popat:
Yeah, thanks, two questions there about the chemo and then the OS. So I mean, chemo is very provocative in this dataset. But as you point out, Solange, it was given as per investigator routine choice with patient discussion. So there are many biases which go into those patients, as to whether they had chemo or not. And I don't think we can really say from this dataset that patients that were given osimertinib can do without chemo, I mean that's a separate question for a separate trial. And we need to investigate that further. All we can say for this setting is that there is a very, very, very strong signal of efficacy for osimertinib in patients after routine treatment with surgery plus/minus chemo. For me, this dataset did really just cement the magnitude of efficacy that we've seen with ADAURA.

Sanjay Popat:
The big difference between this and the CTONG, and the previous RADIANT trial was the intracranial problem. In RADIANT, we had a big excess of intracranial progression in the EGFR mutation positive datasets. And that for me at the time was really surprising, because we were using these drugs with patients with CNS disease often with very good control. So, to see CNS as a site for preferential progression in RADIANT was really quite troubling. Now, we have a signal of that in CTONG as well in the JCO article of sites of metastatic disease. But to see that the sites of brain relapse, the frequency of brain relapse with ADAURA was markedly reduced from about 10% to 1% I think is really another driver for the magnitude of benefit that we're seeing with osimertinib.

Sanjay Popat:
I'm a bit skeptical as to whether we're going to see much in the way of overall survival benefit here, that the trial was poor. Well, the data were released very early. Though it wasn't very mature in terms of the number of relapse events, and there will undoubtedly be significant cross-over with the arms as patients progress. Yes, we do need to see much more mature data. But I think because of the early data release for progression events and because of the high levels of likely cross-over at time of metastatic disease, I'd still be very surprised if we saw much of a meaningful survival benefit. So, I think the data is the data. And we're not talking about a DFS of 0.8 or 0.7, we're talking about DFS of 0.18 which is a complete order of magnitude different to other datasets. So I do think it's very, very compelling.

Solange Peters:
So take-home message, you will discuss it with your patients at least if you have the opportunity to deliver the drug, Sanjay? Martin, Martin maybe not?

Martin Reck:
I would discuss it. But I would not substitute chemotherapy based on that data, so I still would go on with the adjuvant chemotherapy first. In particular when I have locally advanced stage, when I have patients with mediastinal lymph node metastasis. But I think afterwards, this could be a reasonable choice. I think one of the questions is the duration of treatment, do we really need this for years? And we do not have any data to support this from a scientific background. So, this is like immunotherapy. We are doing this, but I think we need a better rationale to use this timeframe.

Solange Peters:
Absolutely, the financial consequences of that are huge. Knowing I ask the investigators, "Why three year?", it's supposed to be about the kinetic of relapse. However, it's financially a real burden that we have to take into account. Let's stay in the local disease, in the early disease. Because there was this strongly awaited trial, the LungART. Just to put it in the context there is still a debate about the role of radiation in patient with stage III disease, completely resected. It's a matter of debate, because there are a series of retrospective studies showing yes or no that adjuvant mediastinal radiation in addition to a complete surgery and adjuvant chemo would help and promote a longer survival. There are contradictory data, but a majority of this meta-analysis were probably showing some benefit. In terms of preventing local relapse in the mediastinal, but also potentially affecting survival.

Solange Peters:
That is the reason why some countries, for example in the US, have adopted adjuvant mediastinal radiation in III and II disease as a standard of care, right? Even participating trial where it's mandatory to do that if you have a stage III disease resected with an N2 are proven, right? So, it's interesting. In the neoadjuvant setting, we had published it in The Lancet, chemoradiation surgery versus chemo surgery with no benefit of radiation. So in the neoadjuvant, it didn't look like to add something to a good local treatment, I would say even an optimal local treatment. And at the time we concluded, it was a Swiss group trial, that one well performed local therapy is enough. One strategy locally, but it has to be optimal. And the LungART is asking it in the other sequence. If you do a complete surgery according to Rami-Porta, so no uncertain, no capsulary invasion, no R1, but R0. If you do that and if you give what you have to do with chemo, do you add something or less or not with radiation, adjuvant radiation?

Solange Peters:
Unfortunately, the answer is maybe you have a nonsignificant effect on disease-free survival. Of course you might prevent some relapses locally in this, these lymph nodes in the mediastinum. But if anything, toxicities added will shorten the survival slightly. I think the conclusion of this strongly awaited 11 years LungART trial is telling us, "Stop doing it and amend your protocols." So are you going ... did you change your guideline, what were your guidelines? To both of you, and do you immediately implement a mandatory change when you see this trial at the limit of a detrimental effect of a strategy on survival? Martin?

Martin Reck:
So, I think definitely. So first, I really have to congratulate the investigators for this trial. I think this is the first prospective trial asking this question, and we really have waited for a long time. And there are really some very unique features of the trial. The majority of the patients was stage in a very adequate way with PET/CTs or/and we have confirmed N2 disease. We had a really good documented patient characteristics and good staging and updated staging systems, so I consider this signal as extremely valid. We have the discussion about the radiotherapy, we have seen changes in the opportunities of radiotherapy in the recent 10 years in development. This is still a point of discussion. But overall we have a couple of sites in Germany where we use this postoperative radiotherapy as state of the art, in resected N2 disease. Of course we have to change our guidelines rapidly, and I know that this is also the case in a couple of European countries.

Martin Reck:
So this is really a landmark trial, and this is a trial which really changes our therapeutic access to management of stage III disease. And I think we really have to individualize our indication to use postoperative radiotherapy in selected patients. But we have to define the criteria in the future whom, who are the candidates for this kind of postoperative treatment. So I think for me this is one of the most important trials of the ESMO this year, because it's really a practice changing trial.

Solange Peters:
That's why we took it in presidential. Because that presidential is only trials which are usually positive, because they change the way we treat patients. But this one being negative, it also changes the way we treat patients.

Martin Reck:
Yeah.

Solange Peters:
The other topic maybe before Sanjay, you give your opinion, is there will be probably as second life for surgery. In non-small cell lung cancer related to all this neoadjuvant IO trial, which I'm sure will give nice results. So this trial is even more important, and the point probably surgery will be a little bit gaining some interest in the future. Do you share this opinion, and what's your feeling about LungART?

Sanjay Popat:
Yeah, thanks. I mean LungART I think is one of the most important trials that we have actually had presented for many years. All of us have been debating in our routine multidisciplinary meetings, postoperatively whether patients should have radiotherapy or not. And finally this trial we've all supported and recruited to has answered this question really quite definitively, is that there is no significant benefit. The benefit is small if anything, but the risks are high with increasing cardiopulmonary toxicities and increasing second malignancies with the radiotherapy. So for routine care, no, I don't see a role now for adjuvant radiation. The real question is, who does need it? And I think this brings in a separate question about high-quality surgery. We do need to make sure that we follow the quality standards for surgical resection as set about by the IASLC and Rami-Porta's excellent paper.

Sanjay Popat:
We do need to start collecting the R(uncertain) status, because this undoubtedly reflects prognosis and actually is a quality indicator as well. And the more we collect that status, the better we understand the extent and magnitude of the resection that's been possible and the risk that's been described. Now we have to remember that patients with extranodal disease were not allowed to go into the trial, they were excluded. So I think for that group there still is a role for radiotherapy, alongside patients with R1 resections as well. So I think where the field has moved on significantly, but we're in a new era now with neoadjuvant treatments starting to come through. And we're starting to see data coming through with neoadjuvant immunotherapy, either single doses or three doses or as the Spanish Group have previously reported in combination with chemotherapy. I very much look forward to the trials which are randomized, which will start coming through next year I think reporting on this, because they undoubtedly are highly active.

Sanjay Popat:
And the field, the most important thing I think I, we've learnt from this neoadjuvant immunotherapy area is you can have activity with only one dose of checkpoint inhibitor. And the imaging does not correlate with the pathological findings. That puts in a very difficult position, how do we manage these patients preoperatively? I'm sure we'll have much more debate over the next few months.

Solange Peters:
Yeah, thanks a lot. So we need to move for the next, last five, 10 minutes we have to immunotherapy. But interestingly this was not the meeting of immunotherapy for lung cancer. So we went into small niches, if you accept the wording, right? Describing the impact of immunotherapy or precising the impact of immunotherapy. For me, there are two niches I'd like to discuss. First of all the very high PD-L1 expressors where we could see a set of data, of course the updates of the five-year survival of the KEYNOTE-024. The new trial with cemiplimab, so the EMPOWER 1 trial, and I like the Japanese trial trying to combine pembrolizumab and bevacizumab. It's all in the niches. 25% of patients who already have this intersection of interferon-gamma signature and a high PD-L1, so this opportunity of giving a treatment which is chemo-free for a long period of time. So I'd like you to comment on this small niche and about your standard of care. Are you satisfied without chemo? Martin, are you satisfied without CTLA-4? This is an important point. Start with this topic, and maybe Martin.

Martin Reck:
No, yeah.

Solange Peters:
I won't ... you also presented the, well, patient-related outcome of 9LA. So, there is a relevant question about CTLA-4 there. So, what do all these data tell you?

Martin Reck:
I think they are telling us we have to look for a new staging system in lung cancer. So, we have histology. We have the different biomarkers, the PD-L1 status. But we also have some clinical features, and it seems that these clinical features of the lung cancer are driving our therapeutic decision. Whether we are satisfied with a monotherapy, whether we need a combination with chemotherapy or whether we need further a combination like seen in the 9LA. And we are just at the starting point. With coming to the KEYNOTE-024 data, I think there were really three impressive results. And number one is really the five-year survival rate. I have looked on all the registries for five-year survival rate in stage IV in non-small cell lung cancer, all the trials are ending their surveillance with two or three years. So in Germany, the normal five-year survival rate is between 5% and 7% in stage IV non-small cell lung cancer. I think these are really interesting data to show a five-year survival rate of nearly 32%.

Martin Reck:
The other point is this part of the re-exposition, the second course. So, there were these 12 patients who received the second course of pembrolizumab. And again, there was a response in a third of them, there was a stabilization in half of them. This is a clear good signal which we also have seen in the second-line trial. And the third signal is the tolerability. So when we look on those 39 patients who received the two years of pembrolizumab, the profile was exactly the same that we have seen in the ITT. The incidence rate of the grade 3 and 4 immune-related, associated toxicities didn't change. It was safe to give the majority of the patients the two years of pembrolizumab. I think for a group of patients, this is a really good opportunity showing the patient-related outcomes, the quality of life. The situation that these patients remain stable for a couple of months after the two years of pembrolizumab, so this is a real new chance.

Martin Reck:
And I think the EMPOWER 1 trial data with the cemiplimab went in think he same direction, we have seen the same signal for OS and progression-free survival. We need some more follow-up. And I think this is one type of patient with a certain tumor, where you really can control the disease with a monotherapy. The other point, are the patients with more dominant tumor diseases which requires the combination, from my eyes, but we need more data to support this.

Solange Peters:
Sanjay, are the cemiplimab data even better than the pembro one? Because when you look at the the absolute numbers they are extremely promising, but of course their shorter follow-up. And the other thing is what do you think about this bevacizumab combination, right? A provocative trial, but I ... we're looking for better options in this kind of setting too, right? The slot of the course too.

Sanjay Popat:
Mm-hmm (affirmative). So the cemiplimab data are very impressive actually, I was very surprised to see such a strong hazard ratio for overall survival and progression-free survival. Bearing in mind, there was cross-over that was allowed in the trial. So, patients that were on the IO monotherapy were allowed to cross over to receive chemo plus IO. And similarly, people who were on the control arm of chemotherapy were allowed to cross over to receive IO. Despite really quite significant cross-over, we already have a strong hazard ratio. I'm very impressed with that, and it's very difficult to make cross-trial comparisons. And I don't think we can say that any drug is any better than one other by looking at, across the trials. But certainly, cemiplimab has very good efficacy in the EMPOWER 1 dataset that was presented.

Sanjay Popat:
But can we do better? I'm sure we can. Combining VEGF inhibitors such as bevacizumab with checkpoint inhibitors is a theoretical benefit, and I suspect there is a efficacy benefit as well. The Korean study showed very good activity with a combination of carbo-taxol-bev plus nivo, but there was quite a high grade 3 hypertension rate which we should bear in mind with that combination. In fact when we look at the Japanese, WJOG study, the Phase II study of combination atezo and bev of course we still see very good efficacy. But we see the same signal of hypertension coming through, and much more hypertension than we would expect with bevacizumab alone. So I think this is a very encouraging couplet, bevacizumab plus single-agent immunotherapy. We need to be mindful of how the tox pans out in other, in future trials, but certainly cemiplimab is a very active agent looking at the randomized Phase III trial.

Solange Peters:
Yeah. As so the atezo and bev, sorry for the, I said pembro I guess before. Atezo and bev, and it's important to correct. It's quite important for the EMPOWER trial to keep in mind the two particularities which were not very clear during the presentation. The first one is the smoking habit, only smokers were enrolled. And remember that even if in all trials EGFR and ALK patient have been excluded, still we usually have a remaining 10% of never-smokers across trials that we're excluding from this trial. And as we know from the neoantigen and TMB story, it might impact the final results. The other thing is about brain mets. Brain mets were all treated and stable like in all other trials, on the contrary have a little blood message.

Solange Peters:
However where they accepted that the waiting time between the end of radiation and the beginning of the tretment was more fitting the daily life, the daily practice, right? Usually, I don't know about you, but we don't treat, right? When you have finished your radiation, you quite fast go ahead with your systemic treatment. And this was more like that in that trial, so fitting the objectives of showing safety and efficacy as soon as possible. So I think they are two important features of the cemiplimab trials, which might make the results slightly different also, right? And we have to keep in mind.

Solange Peters:
So, last talk I'd like to discuss with you is small cell. So with all of the committee, we've been to move in small cell. In the maintenance setting with the CheckMate-451 and with ipi-nivo or nivolumab, in the second-line setting with nivolumab versus topotecan or amrubicin always seeing first of all, negative trials, formally speaking. But also also always seeing the signal that there is a benefit in a small proportion of patients. Which you can see it with non-proportional hazard ratio, but the population of patient we are strictly unable to date to identify. Basically, we know it might benefit strongly some patients. But it's impossible to see who they are, and they are not numerous enough to change the paradigm of treatments, right? Very frustrating, because the science here is still completely insufficient in helping this and guiding us there, right?

Solange Peters:
We can also see with this kind of a very strange statistic analysis that I like, this is twice hazard ratio. Where when you go into first three, six, even nine months of this strategy, you see that chemo or whatever is better than IO. But if you look later down the road, suddenly you have a very strong hazard ratio in favor of these selected patients benefitting from IO. So if I go into that, is there was the STIMULI trial. And I had the pleasure to present in on behalf of ETOP, European Thoracic Oncology Platform. You both participated in that, and it was showing the same. It was limited-stage small cell lung cancer, it was after completion of a very strict standard of care. Radiochemo PCI which was per protocol, so very ... or a stronger standard of care and difficult protocol trying to maintain, consolidate with ipi and nivo. And same thing, this is a negative trial.

Solange Peters:
However when you look at survival, which is a secondary endpoint, my feeling is we will see the same thing. Some patients will become long-term benefiter, will survive to a larger extent than expected, but we don't know who. So what do you think about these small cell data, beyond the frontline chemo-atezo or chemo-durva? What do you think about feature, future development of IO in small cell and STIMULI trial? So maybe, Sanjay to start?

Sanjay Popat:
Yeah, thanks, Solange. I mean it still is a really important trial, because it really sets the pace for limited-stage disease. Everybody has to get a standardized treatment. We've sort of become used to trials which are a bit more real world, where they recruit at the end of the chemoradiotherapy. But to have everybody having the same schedule really gives you a very homogeneous group of patients. And unfortunately there was no PFS benefit. But there was definitely a signal there of perhaps an overall survival signal is starting to come through with a nonproportionality, starting to become more obviously with the more follow-up. Unfortunately in retrospect, the nivo-ipi dosing was probably a bit hotter than we'd like now that we've got a better understanding of the right schedule for thoracic malignancies.

Sanjay Popat:
But we've seen that in the metastatic setting, there's always a small proportion of patients that result in a tail of the curve. We have no idea how to predict these patients upfront, this is the biggest frustration that we have. But undoubtedly, there's a bimodal distribution of patients. Those that don't derive at any benefit, we can't predict who they are. Those that do derive benefit, and we can't predict who they are. And I have no doubt that we will see exactly the same thing in the limited-stage setting, in bigger trials in the future. So for me, STIMULI is a fantastic study, it's really set the field. This is the beginning of the chapter of limited-stage small cell lung cancer, I'm very much looking forward to more data as it comes through.

Solange Peters:
Martin, what do you think about this trial and what do you think about this landscape of small cell? Because we are little accumulating the disappointments in term of positive trials.

Martin Reck:
Yeah. Absolutely. And I think coming back to second-line when we look on the CheckMate-331, where we compared nivo to topotecan, what we have seen was as you mentioned. There is a group of patient benefitting, and we have seen a consistent efficacy of nivolumab independent from the fast-progressing or the non-fast-progressing tumors. So, there is a signal in a group of patients independent from the efficacy of chemotherapy. And currently we are not able to describe these patients, this is something we really need to define in the future. For STIMULI, I have to say I was more than disappointed. So, I think this was my PACIFIC for small cell lung cancer. Perhaps one of the reasons has been our traditional invalid staging system for small cell lung cancer.

Martin Reck:
Even when we say we have limited-stage small cell lung cancer, this is a heterogeneous population. It's not adequately determined as a stage III non-small cell lung cancer. These might be N2, N3, very bulky tumors, very fast-progressing tumors. We are all covering them in this substitute of limited-disease small cell lung cancer. Perhaps this is one of the reasons that we have a mixed population there, and we have to be a bit more precise also in small cell lung cancer in determining our populations that we treat. However, so far we really cannot derive any guidance from the STIMULI trial to describe a target population for this immunotherapy. So I am hoping for the upcoming translational analysis, that potentially we will be able to describe these long-term survivors. There is clearly a signal of survival in some patients. But we have looked for PD-L1, we have looked for TMB, they are not sufficient to describe this group of patients. So, we need a new idea. But I agree to Sanjay, currently we are just in the beginning, we are not able to describe these patients.

Solange Peters:
Mm-hmm (affirmative). How would you guide, both of you, if you had to potentially ... two things, right? Imagine what would be, you think, the way to improve outcome with immunotherapy in small cell? Extensive disease, would it be to combine new modalities, stronger immunotherapy, add some radiation? So, what would be your favorite? It's completely subjective, your favorite strategy in extensive-stage to continue to move this field beyond what we have. And what would be your advice for limited-stage, should we give up, wait until translational data? So helping towards your key take-home messages for small cell, Martin and then Sanjay?

Martin Reck:
Well, for extensive, I have been a little bit frustrated with all the second-line and maintenance concepts. So I truly believe if we want to achieve some in extensive-disease small cell lung cancer, we have to throw it in a first-line concept. This is currently on the way, and for example we have the tiragolumab combinations investigated in small cell lung cancer, this is an opportunity with a new concept of an immunotherapy combination. I do not believe that we can achieve something in subsequent line of this disease.

Solange Peters:
Mm-hmm (affirmative) .

Martin Reck:
For limited, I think we have to rethink about the concept. And I like for example, also this discrimination system of Charles Rudin and to look a little bit more in the deep of small cell lung cancer. So small cell lung cancer is not small cell lung cancer, we do see different genotypes potentially also associated with differential response to immunotherapy. We are just in the beginning, we have the first datasets. But it also for limited disease, we have to dig a little bit deeper to identify the different characteristics of these patients, in order to define those patients who potentially might benefit from this very aggressive multimodal treatments.

Sanjay Popat:
And I agree with those thoughts entirely. I mean for me in the metastatic setting, of course adding in the new checkpoint inhibitors is the way forward. And I fully agree that there's no point in experimenting in relapsed small cell lung cancer. We have to add in new drugs in the frontline setting. Now for me, one of the biggest questions that still remains unanswered is, what is the role of palliative radiotherapy? Of the rest type of radiotherapy, those are entirely excluded in all the registrational trials. But certainly in Europe it was very used, very popular with colleagues. So how do we integrate this, does it have any benefit at all? So for me, I think this is a very open question that needs to be answered. In the limited-stage setting we still need to wait for the trials that are run-in to read out, but we need to search for biomarkers. In non-small cell lung cancer, we started to see signals that HLA restriction of the MHC genotype may impact on benefit. I would certainly want to look at that in small cell as well, because the mechanisms will undoubtedly be very similar.

Solange Peters:
Thanks a lot, as I think we covered the most important topics from this ESMO 2020, trying to catch up with the discussions. I thank you both for your messages which will be very important for the community, and thanks VJOncology to help us and support us in these difficult times. Thanks to both of you and stay safe, and all the best to all of ... thanks for listening to us.

Martin Reck:
Thank you.

Sanjay Popat:
Thank you.

Solange Peters: has received education grants, provided consultation, attended advisory boards, and/or provided lectures for Abbvie, Amgen, AstraZeneca, Bayer, Biocartis, Bioinvent, Blueprint Medicines, Boehringer-Ingelheim, Bristol-Myers Squibb, Clovis, Daiichi Sankyo, Debiopharm, Eli Lilly, F. Hoffmann-La Roche, Foundation Medicine, Illumina, Janssen, Merck Sharp and Dohme, Merck Serono, Merrimack, Novartis, Pharma Mar, Pfizer, Regeneron, Sanofi, Seattle Genetics, Takeda and Vaccibody, from whom she has received honoraria (all fees to institution).

Martin Reck: Honoraria for lectures and disclosures from Amgen, AstraZeneca, Boehringer-Ingelheim, BMS, Lilly, Mirati, Merck, MSD, Novartis, Pfizer, Roche.