Management of Metastatic Non-Small Cell Lung Cancer (NSCLC) without an Actionable Mutation — Heather Wakelee, MD

DR WAKELEE: I’m Heather Wakelee, Professor of Medicine and Chief of the Division of Medical Oncology at Stanford University, Deputy Director of the Stanford Cancer Institute, and currently the President of the International Association for the study of Lung Cancer. I’m really excited to be here and doing this Year in Review webinar on nontargeted treatment of lung cancer, non-small cell lung cancer. Lots happening in that field.

So as you can see, there are a lot of trials that I will be talking about. There’s a lot happening, especially around immune therapy, antibody-drug conjugates in particular.

So what was really exciting in the past year are getting these 5-year and other long-term outcomes from trials. When I started as a medical oncologist, and we talked about metastatic disease, we did not talk about 5-year survival. That was just not something that was part of our discussion. I mean we all knew 1 or 2 patients, but it just wasn’t a common discussion point. And now as we are getting to the time period where the immune checkpoint inhibitor trials are reading out at that 5-year mark we have a lot of trials where we have this 5-year data, and to me that’s one of the most hopeful things that we can offer our patients as we talk about them.

So this is the long-term follow up data, 5 year, from KEYNOTE-042. For the audience, of course, this was a study that enrolled patients with any PD-L1 positivity in their tumor to receive pembrolizumab versus chemotherapy. The first study, the KEYNOTE-024, was only in those with high PD-L1 expression of at least 50%. This now was a broader group. And there’s a little bit more controversy around this trial. Again, we saw that the most benefit was in those whose tumor expression was at least 50%, but there was some benefit in those with 1% to 49%.

How this study was presented initially, it was the greater than 1% and then greater than 50% and not as much focused on just that 1% to 49%. And we see in this presentation also it’s greater than 1%, greater than 20%, greater than 50%. So you’ve got that greater than 50% group pulling positivity throughout the whole thing. But it’s still a very, very strikingly positive trial.

And we see here a long-term benefit, and we’re seeing this 20% overall survival at 5 years. And that 20% 5-year survival is something that we see pretty consistently across multiple trials of the checkpoint inhibitors where there’s this subset of patients, somewhere around 20%, whose tumors are very sensitive, and they have these durable responses. And so this was another study demonstrating that. And we do see that most of the benefit is in those patients who do have high PD-L1 expression.

So when we look at that 5-year survival in the patients who are surviving, in the high PD-L1 group it’s 22%, greater than 20%, 19%, greater than 1%, 16, 17%. And so what it would be like in just the 1-49 we don’t have that presentation, but we know it’s going to be a little bit lower.

But I think this is really important as we think about that long-term response, long-term benefit, and we’ll get into some later trials with long-term data looking specifically at these patients who are these long-term benefitting folks.

So it’s a busy slide, but I think that key point is illustrated well here.

Then we go on to another study. We have many PD-1/PD-L1 checkpoint inhibitors. Cemiplimab, a lot of data coming out with cemiplimab. So this was a presentation at ESMO in 2022, 3-year survival with cemiplimab plus chemo versus chemo alone. It’s titled as a continuation of cemiplimab beyond progression with adding the chemo, but if you look at this trial design in more detail Arm A is cemiplimab and Arm B is chemo. In the Arm A, which is the cemiplimab monotherapy, at the time of progression you can continue the drug and then you add chemo at the time of progression. For those who got chemo they do have an optional crossover to get the cemiplimab. So everybody’s able to get the drug, and the question is if you start with immune therapy and add in chemotherapy to it the benefit of that versus starting with chemo and then switching over to cemiplimab.

And this was a trial that was only in patients with high PD-L1 expression, so greater than 50%, and so this supports —it’s like KEYNOTE-024, where that was only in the 50%, and again it was single-agent checkpoint inhibitor versus chemo. The difference here is they were allowed to continue the cemiplimab at progression and add the chemo to it, and so they emphasize that outcome in this presentation.

And you can see here the 3-year overall survival. Those curves are beautifully separating, very clear positivity for overall survival. Progression-free survival, that separation happened more around the 6-month mark, and the patients on chemotherapy, of course, stopped treatment after they completed the chemotherapy. So this is not too surprising in and of itself; another trial showing single-agent checkpoint inhibitor better than chemotherapy when you have high PD-L1 expression in the tumor.

What they also presented was this other analysis, which gets very complicated, looking at this period 1, period 2. Period 1 was either cemiplimab or chemo. Period 2 was at the time of progression, so continuing cemiplimab, adding chemo, or if you were on chemo getting cemiplimab. And it just shows that you can get a second response when you add the chemo to ongoing cemiplimab just like you can get a response if you weren’t on a checkpoint inhibitor and you started chemo.

So this study is another positive trial. We’ve got 3-year data now confirming that single-agent checkpoint inhibitor, in this case cemiplimab, is great if you have high PD-L1, better than chemo. If you’re on cemiplimab and you add chemo to it you’re going to get a benefit, but whether that’s from the combo or just would have happened with chemo alone we don’t really get from this particular study. Still, a good positive trial, another long-term benefit with checkpoint inhibitor data study.

And now this is a slightly different study. Most of what I’m going to talk about today focuses on the checkpoint inhibitors. This one does also, but this is a different twist on things. Instead of giving immune checkpoint inhibitors intravenously this is a new subcutaneous formulation of atezolizumab. And I thought this study was really interesting because when we think about the challenges we face as oncologists treating our patients having them have to come in every 2 weeks, 3 weeks, 4 weeks, 6 weeks to get infusions can be really difficult. We know that the costs are high for patients. And sometimes there’s a bit of a higher risk of immune-related adverse events when you’re getting an IV infusion versus a subcutaneous injection.

So this trial was a head-to-head comparison of subq atezo versus IV atezo, and if you look there’s a lot of details in this. But essentially they showed that you weren’t getting much of a difference as far as adverse events, and the efficacy, at least as they looked at this kind of early on efficacy, was very similar. These were not patients who necessarily had high PD-L1 expression, so response rates are lower, in that 10% range that we see in an unselected patient population, but very similar. And to me this was important because the subcutaneous option has the promise of reduced cost, option for home administration, and broader distribution opportunities. So I think it was pretty exciting for that reason.

And I’ll pause there and see if you’ve got any comments on that, Neil.

DR LOVE: Yeah. I’m just curious because I don’t remember hearing very much about subq IOs. I mean obviously they’re extremely commonly used in so many — it seems like it would be a gigantic benefit. Is it being looked at with other IOs?

DR WAKELEE: So where I’m most familiar with it is actually with one of the bispecifics, amivantamab, and so we’re working on a trial with that where one of its biggest issues is the infusion reactions. I haven’t heard much. I mean I’ll be honest; I didn’t hear much about this study when it was presented at the ESMO IO Conference in December. And so I think it’s really important, though. We focus on what’s the newest and latest drug and mechanism of action and long-term outcomes, but from a practical standpoint this has the potential to be more practice changing by bringing in this different mechanism of administration.

DR LOVE: Yeah, I mean it really could change the face of the infusion rooms in general, not just lung cancer.

DR WAKELEE: Right.

DR LOVE: Interesting.

DR WAKELEE: Yeah. It’s pretty exciting.

DR LOVE: Please continue.

DR WAKELEE: Okay. So back to our theme of long-term outcomes. So this is the 5-year survival data from KEYNOTE-189. So this of course was chemo plus pembrolizumab, and we see that the overall survival separation continues out, progression-free survival continues out into that 5-year time period. This was a study where PD-L1 levels did matter, not as much as with the single agents, but still was important.

One of the things they emphasize in this presentation I thought was a little bit disingenuous because they looked at for those patients who were still doing well how well did they do, and I think we need to be careful when we select out the survivor bias here. But still really exciting to see, as we are talking to patients who are just about to start on treatment with this combination of platinum, pemetrexed, and pembrolizumab which is still very widely used, to be able to look people in the eye and say there’s a chance that you are going to be living 5 years from now potentially without evidence of disease. We’re back to that somewhere around 20% number.

And this is highlighting that on this swimmers’ plot. You see the dark gray bars are the time that patients were on treatment, continuation with the pembrolizumab maintenance. The little black dots are when people progressed, and you see that there are some patients who stayed on after initial progression.

But you also see that the light gray is once treatment has been stopped and patients are followed, and there are a number of patients here who once their pembrolizumab was stopped they haven’t had any evidence of disease recurrence, and now we’re at 5 years. We still are cautious with the cure word, but this the kind of situation where you think about it. I think it is notable that there were some patients who did have progression at the 3-year mark, at the 4-year mark, but it’s pretty small numbers, and the majority of patients that have made it out beyond 3 years, most of those patients don’t have any later progression events, but we still need to be mindful.

So one of the big debates we have right now is whether to stop at 2 years or not and what do you do when they recur if you have stopped. This is an area where we don’t have any data we’re talking about today, but I think this is where ctDNA and other ways of looking for that minimal residual disease will be really important. And maybe we don’t even need to go 2 years. Maybe there’s some patients where after 6 months or a year they have no residual tumor, we can look with new assays that are being developed to be able to be comfortable saying and now we can stop, but we’ll keep checking to make sure that nothing comes back.

And for this situation here, maybe if we do go to 2 years if we had a way to figure out who are those patients who are in the top bar where they never get the black dot and who are the patients who aren’t. And so I think this is really an important development that’s being done.

So here we have that duration of response going out quite a ways, 60 months or so, and continuing the 3-year rate after completing the 35 cycles is 72%, so they’re essentially saying that if you make it to 2 years and then you stop you’ve got a really high chance that you’ll still be doing okay at 3 years, though some of these patients have of course restarted treatment.

So I’ll pause there, Neil, and see if you have any comments you wanted to make.

DR LOVE: Yeah. I was just flashing on this case. We’re actually presenting a bunch of cases this weekend at the SGO Meeting, and one of them was MSI high metastatic endometrial cancer. The patient was on pembro, was coming to the 2-year point, they were talking about whether to stop or not, and then of course she recurred.

What I was going to ask you is have you done, or are people doing ctDNA? Do you see it at 2 years or 5 years? Have you ever done it?

DR WAKELEE: We’re not doing it in routine practice at this time, but I work closely with Max Diehn, and so we’re always doing all sorts of ctDNA studies and analyses. And this is one of the questions we have looked at, and we’ve had some publications on it, but it’s not routinely done yet, but I think that’s where we’re heading. It’s going to be a really important question.

DR LOVE: But I mean do you see patients who have positive ctDNA like after 2 years?

DR WAKELEE: Well, again, I haven’t — I don’t have a series of patients in my mind I can think of where I’ve actually looked at it because we’re still mostly doing it within the context of research, so we don’t have the —

DR LOVE: Right.

DR WAKELEE: And if your patient still has it it’s more of a still trying to gather that data and understand what it means.

My assumption, if we look at this trial data, the patients who have the black dot who have recurred, they probably all did, if you could have an assay that was sensitive and specific enough. And that’s where we’re still challenged. I don’t think any of the commercial assays that are out there right now are sensitive and specific enough for us to be comfortable saying, “And you don’t have it, and you can stop.” We can — they’re sensitive enough to say, “And you still have it, and you probably shouldn’t stop.” But being able to stop comfortably we’re not quite there, but that’s what we’re working towards.

DR LOVE: Interesting. Please continue.

DR WAKELEE: Okay. So this is very parallel. This is the KEYNOTE-407: Chemo plus pembro or chemo plus or minus pembro but in the squamous cell histology. Very, very similar results. And as we go through that you see the robustness of that separation isn’t quite as strong, and where we’ve had the adenocarcinoma or the nonsquamous and the squamous pretty much uniformly the patient — the trials that were looking at squamous cell histology patients we weren’t seeing as much benefit, but still benefit, statistically significant is standard of care. There seems to be a higher benefit for whatever reason in the nonsquams. Another mystery to resolve.

Very similar analysis, of course the same company, same — pembrolizumab. Same idea. If you get to 2 years and then you stop there are a large number of patients who never have recurrence, but some who do even as much as 2 years after they’ve stopped. And so this is, again, very encouraging. It’s not the majority of patients. It’s 10, 20%. That 20% number keeps coming up. But it’s real hope for patients. And where we’re working now is trying to figure out who needs more than just single-agent checkpoint inhibitor, and we’ll get to some of those discussions later. When can we safely stop? When do we need to continue? But very, very hopeful time.

Okay. Now, we’ve been talking about the pembrolizumab data. Now we’re back to cemiplimab. So this is cemiplimab plus chemo versus chemo alone, so same idea as the 2 we just went through with pembro. This was a Phase III trial. It was actually published in Nature Medicine in 2022. I kind of look at this as another positive chemo plus immune checkpoint inhibitor trial. In this case PD-L1 levels were important.

We see that with these pink and blue bars that were looking at response rates, and you can see that that was higher in the combination versus with the chemotherapy alone. And we go from left was no PD-L1 expression. Right is that high PD-L1 expression. The blue bar, which is the chemo alone, doesn’t vary that much. The pink bar goes up, the higher the PD-L1 the more likelihood of having response, so that’s an additional benefit of adding the cemiplimab. And we see in the curves both overall survival and progression-free survival benefit.

So a positive trial, 3-year data here looking good.

Another 5-year data. This is the nivolumab plus — nivo/ipi (nivolumab/ipilimumab) versus chemo, the CheckMate 227. Again, a very complex trial design. They had divided patients up into those without PD-L1 expression in their tumor and with PD-L1 expression in their tumor. Then they had 3 different groups randomized for each of those. Some were the same, some were variable, and so nivo plus chemo for those who had less PD-L1 expression. Chemo alone — nivo alone for those with higher, nivo/ipi option in both situations, and then combinations.

And so when we try to like look at this trial it gets very, very complex, but I think that the take-home is that for patients who did have PD-L1 expression, when we look at those getting nivo/ipi versus chemotherapy, because that was in both of the arms, you could get nivo/ipi or you could get chemo, overall survival 24% versus 14% for nivo/ipi versus chemo if there was PD-L1 expression. If there was no PD-L1 expression you still had a benefit with the checkpoint inhibitor, and this is where we bring in the question of okay, we’ve talked about PD-1/PD-L1 alone, where it was very PD-L1 dependent.

When you bring in the CTLA-4 inhibitor that PD-L1 doesn’t seem to matter as much. If you look at the overall benefit in the greater than 1%, that 24% versus 14%, those numbers are both higher than what we get in the low PD-L1 group, but that magnitude of difference between them is actually about the same, if not even more so. And so this concept by adding in CTLA-4 to a PD-1 inhibitor in this case you can overcome that lack of PD-L1 expression which has limited benefit for patients with those sorts of tumors with checkpoint inhibitor alone. It gets a little bit confusing.

My take-home from it is nivolumab/ipilimumab is a reasonable option for patients versus chemotherapy, and that that is a reasonable option regardless of PD-L1 expression. When I talk about giving single-agent checkpoint inhibitors to patients I am very uncomfortable with that strategy for patients whose tumors don’t have PD-L1 expression, so adding in the CTLA-4 seems to make a difference.

I did not arrive at that conclusion quickly. It took me a while and a lot of trials to kind of get there, but as we go through this study, and the POSEIDON, I think most of the folks listening to this will also see that that seems to be a recurring pattern. I like to look at things as it’s more likely to be a truth if we’re seeing the same thing showing up in multiple trials, and we get a lot of flukes, especially when we’re looking at what is PD-L1 level mean with the various drugs.

We’ve had a lot of things that didn’t always make sense, but if you’ve got consistent patterns that’s more likely to be truth, and I’m now convinced that this combination CTLA-4/PD-1 or PD-L1 combination does seem to overcome some of the inherent resistance we get in the non-PD-L1 expressing tumors. So that’s something that I’ve more recently come to in this trial is showing that clearly. We’re also seeing this about 20% long-term survival at 5 years, which is great, in this case with the nivo/ipi, and this is seen again, and the benefit is seen regardless of PD-L1 in this trial.

This study is also important to note that about two thirds of these long-term survivors are completely off trial, similar to what we were seeing with the single-agent pembrolizumab in the studies with patients who had the higher PD-L1 expression.

Okay. That was a mouthful on that one. So now we’re going to talk about Luis Paz-Ares’ study. He’ll be in this session. He’s talking about the early stage, and I’m talking about late stage, and so we’re each talking about each other’s data, which is good, I’m sure, so you get different perspectives. But this is Luis’ presentation from CheckMate 9LA. This was updated ASCO 2022, and again showing that this is a strongly positive trial, long-term benefit seen, and this is 3-year data. Overall survival at 3 years is 27%, probably tracking to be close to that 20% at 5 years we keep running into.

And here, again, with the combination of CTLA-4 and PD-1 you see the benefit in patients even in tumors where there is no PD-L1 expression. And so what I’ve highlighted on this slide are the separation, overall survival benefit, PFS benefit. And then in the forest plot here I’ve highlighted in the red box that these benefits are seen regardless of PD-L1 expression, so that really didn’t matter when you’ve added in the ipilimumab to the nivolumab.

Now we move to POSEIDON, so another combination PD-L1/CTLA-4, so durvalumab with tremelimumab in this trial. And this trial was not as overall robustly positive in all of the arms as we were initially expecting to see, but in the end definitely has some positivity to it, and I think adds very clearly to the story now that adding the CTLA-4 can overcome the resistance that we see to single-agent checkpoint, PD-1/PD-L1 checkpoint inhibitors, in tumors without PD-L1 expression on the tumor.

So this study enrolled patients with metastatic Stage IV disease, no known EGFR mutation or ALK translocation, no prior treatment, and then they were randomized 1:1:1 to durvalumab/PD-L1 plus chemotherapy, to durvalumab plus tremelimumab, the CTLA-4 agent, plus chemo, or to chemo alone. And when we look at the durvalumab plus chemo versus chemo, both the progression-free and overall survival, that overall survival was not positive, and that was one of the things about this trial where people were like hmm, that doesn’t make as much — doesn’t kind of fit with the whole story.

But when we look at the 4-drug — because you had 2 chemo drugs, so durva/treme/chemo, here that progression-free survival benefit is a little bit more robust, and that overall survival hazard ratio was also positive at 0.77. So this is a positive trial. And we also see that when we look at those PD-L1 expressions there at the bottom you can see that it actually did matter when we’re looking at the durvalumab alone versus chemo, but it didn’t matter when you added in the tremelimumab. So again, this consistent story, and again consistent long-term benefit here, 3 years, 27%. We will see what we get to at 5 years.

A little bit more on this. This is the overall survival in a bit more detail; 4-drug regimen at 4 years now. We’re down to about 21% versus 16% with durva alone, but durva plus treme increased benefit over that long-term period. And again, we’re at about 20% now at 4 years with the durva/treme combination. And again, the PD-L1 didn’t seem to matter once we add in that CTLA-4 drug. So this was an ESMO presentation in 2022.

Now on to our antibody-drug conjugates. So this is TROPION-Lung02. This is the datopotamab deruxtecan, or dato-DXd, plus pembro plus or minus chemo with a platinum agent added in. And I think it’s important when we talk about the antibody-drug conjugates they are still chemo. I think a lot of patients when they hear about these, “Oh, it’s a new drug. It’s going to be like immune therapy.” It is still chemo. We still have our chemo side effects plus some. But with many of these agents we do get those higher response rates when we’re actually hitting a clear target.

So this is a TROP2 IgG monoclonal antibody with deruxtecan, which is a topoisomerase I inhibitor. That deruxtecan payload is what we’re seeing in a lot of the most active antibody-drug conjugates, so I do like the deruxtecan as far as what we’ve been seeing.

And there this was looking at a combination of either pembrolizumab plus dato-DXd or those 2 with the addition of platinum. And what’s interesting is that we’re not seeing much benefit with the addition of platinum. So when you look at the dosing for these, these doublets versus triplets, the patients were relatively well balanced, the difference being those getting the triplet couldn’t have had other prior platinum treatment, and those getting the doublet could have had prior treatment. And we’ll look at the balance by PD-L1 and all those things, but that’s not as relevant.

And when we look at these responses we see that the response rate was actually reasonably high. With just the doublet it was 62%, with the triplet only 50%. Use extreme caution with those numbers. When you’ve only got 13 to 20 patients, 1 or 2 patients difference in response can hugely change those responses, so I try not to quote that too much. But at least the first signal was that adding in the platinum wasn’t doing a whole lot other than adding toxicity. And so they’re actually planning to move forward with just the dato-DXd plus the pembrolizumab, which is interesting. When we — we haven’t been bold enough to try peeling back the platinum in a lot of our other prior trials, so we’ll see how this pans out.

There’s a Phase III trial, the TROPION-Lung08 trial, which is ongoing, and that’s the dato-DXd plus pembro versus pembro alone for first-line treatment in patients with high PD-L1 expression. So we’ll see how that pans out.

Again, these drugs are not without toxicity. You need to be very cautious in knowing about those toxicities and monitoring closely.

Okay. Another antibody-drug conjugate this time is tusamitamab ravtansine, so a different linker and a different payload. This is in patients with nonsquamous histology, and this is the CEA antigen-related cell adhesion molecule 5, so CEACAM5. That is what we’re trying to target. And this was a presentation at ASCO 2022. As you can see, it’s a very complex drug. You’ve got your antibody. You’ve got your toxin attached to it.

In the initial trial, the Phase I/II study with this compound, there was some activity in heavily pretreated patients, and they were trying to look at the CEACAM5 expression as a way of selecting higher response/lower response. So in the higher responding patients 20% had a response. In those with moderate expression the response rates were much lower. And so there does seem to be some selection.

One of the challenges I personally have had in looking at all the ADC data is that for some of the ADCs they’re not really as targeted. We have some — like the trastuzumab deruxtecan that is clearly a targeted agent. In patients with a HER2 mutation our response rates are 60, 70%. But then you’ve got some of these other antibody-drug conjugates where when you’re response rates are only in that 20% range you wonder is that really a target or is that just we can get 20% single-agent chemo response rates as well.

And then I think about that as fancy chemo versus a targeted agent, and so this is differentially active based on the CEACAM5 expression, and so it’s somewhat targeted. But it’s not like a super robust targeted response is how I’m looking at it. Still reasonable to look at, and what they were highlighting here is that of the patient’s who responded some had pretty durable responses. And so I think that is worthy of note and is certainly reasonable to look at it.

I think when we’re trying to assess drug efficacy we look at what’s the response rate, but that duration of response is also really important because if everybody responds for a month that’s not very useful. It only gives people a very short time. And ideal drug is working for everybody and forever, right? We don’t quite have that. But if something doesn’t work — it only works in 10% to 20% of patients, but in that tiny 10-20% it works for multiple years, that’s still really an important thing to find. So we’ll see how this pans out.

And Neil, did you want to comment on this?

DR LOVE: Yeah. Maybe you can talk a little bit about what CEACAM actually is. And this may be a very naïve, stupid question, but does it have anything to do with CEA?

DR WAKELEE: Yeah. Yeah, it does. So CEA, this is just a marker that we find in cancers that sometimes shed, right, and so it’s a protein that we can measure, and CEA can be like a relatively nonspecific tumor marker that we can look at tumor expression. Instead of doing imaging sometimes we can follow the CEA levels.

DR LOVE: Is that what this is targeting though? Is it targeting CEA?

DR WAKELEE: It’s a different molecule, but it’s a CEA-related adhesion molecule, so it’s in that same pathway, and I would have to go back to my biochemistry book to give you a better answer than that, so…

DR LOVE: Interesting. Yeah, no. I’ve never heard of this until I saw this paper in there, but it looks really interesting.

No unusual side effects? I don’t know this payload. I’m not sure what that is. Any particular side effects you see with it?

DR WAKELEE: So I haven’t used this compound myself as much. I think that the things we need to be watching for, we have a lot of issues with all of these — with a lot of the ADCs there’s ocular toxicity. I don’t know about — yeah, with this one there is some risk there.

DR LOVE: Oh, really?

DR WAKELEE: There are issues with edema, infusion reactions, and then traditional chemotherapy toxicities, nausea, hair loss, those sorts of things. And I can get back to you on the specifics with this drug more because I haven’t used it.

DR LOVE: Yeah. We can even talk about it in the webinar. No.

DR WAKELEE: Yeah.

DR LOVE: I mean there are more and more ADCs that we’re running into with ocular toxicities, so I was curious about that, but not in lung cancer yet. But yeah, myeloma, bladder cancer, et cetera.

DR WAKELEE: Yup. Yup.

DR LOVE: All right. Please continue.

DR WAKELEE: And this is then another study. This is HUDSON, and this is an umbrella study. So this is looking at multiple different drugs, a biomarker-directed Phase II platform. And really what it is looking at is adding durvalumab to these other targeted drugs. And so so far what this data has looked at is with these 4 compounds: Ceralasertib, which is an ATR inhibitor, danvatirsen, which is a STAT3 inhibitor, olaparib, which is a PARP inhibitor, or oleclumab, which is an anti-CD73 monoclonal antibody. Don’t worry about memorizing all of those. The only one you really need to know about is the ceralasertib, which is the ATR inhibitor.

That was the only positive aspect of this trial so far, and I’ll show you that on this next slide, where you see that they enrolled pretty well. It was close to 60 patients in each of the arms, a few more on the olaparib arm. The response rates were not super robust, but it was 17% with the durva plus ceralasertib, and these were all pretreated patients.

The others really almost no responses, and we see that when we look at the progression-free, as well as overall survival. There was a benefit, it seems, in the durva plus ceralasertib but not with the other arms. And so they are going to be moving forward with the durvalumab and ceralasertib. That showed promising activity, and then they’ll be looking at other drugs that can be brought into this umbrella platform.

Another novel approach. So this is a tumor-treating fields device. It’s a wearable thing your patients put on. There’s a vest. For brain tumors this is where it’s already widely utilized. It’s a cap that they put on, and it basically uses these pulse electrical fields, in this case combined with pembrolizumab for advanced lung cancer.

And so again, the tumor-treating fields, it’s an antimitotic treatment with these pulse fields. It’s already approved in glioblastoma. It’s already approved in mesothelioma with a vest. And this was an open-label trial.

It was randomized looking at the fields in combination with immune checkpoint inhibitors or with docetaxel versus the immune checkpoint inhibitor or docetaxel alone. And I don’t have all the slides, but basically what they presented so far is just this press release. We haven’t seen all of the data, but in the press release it showed a significant and meaningful overall survival benefit with the tumor-treatment fields plus immune therapy versus immune therapy alone. In the docetaxel plus or minus the TTF fields that had a trend but was not overall positive. And it’s well tolerated.

So we’re waiting to see a bit more data. There is a study, this KEYNOTE B36, which is a Phase II study looking at the tumor-treatment fields plus pembro for first-line treatment, where we’ve seen a little bit of that data already, and that’s where they went with this Phase III. And so we will see more soon, and that’s all we know.

So I don’t know if patients are going to be accepting of this because they’re actually having to wear a device versus getting an infusion. It has been widely adopted in brain tumors, and patients do use those. I don’t think we’ve had very much adoption yet for mesothelioma. And so we’ll just have to wait and see what happens, and I think how positive is a positive result, that’s going to be really important as we try to make those determinations.

So do you have a comment on that one?

DR LOVE: Yeah.

We’re actually having a couple cases presented, again, at the SGO, of a trial in ovarian cancer, and there both of the patients that we’re presenting had significant dermatologic toxicities. I know that you said you haven’t used it, but have you heard anything? Do you know anything about to what extent skin is a problem, where I guess the things that you put on cause irritation?

DR WAKELEE: I haven’t heard that, but I will confess that I don’t have a lot of familiarity with this technology. I certainly know about it from brain tumors, and there it’s a cap, and I really hadn’t heard about that being an issue —

DR LOVE: Yeah, no. I think maybe —

DR WAKELEE: — but it would be important to look for that.

DR LOVE: Yeah. I think the thinking was maybe because in the abdomen or trunk there’s much more skin exposed; the head maybe not that much. I don’t know, they might see it there, but actually both of these patients had a lot of problems with that.

What do you think about the biologic basis for this? A lot of people kind of look at it skeptically. How do you feel?

We had a doc, Dr Patel, he’s a neuro-oncologist, I did a whole webinar just with him trying to explain to me how this works. Any thoughts about it? Do you think this is legitimate or more hocus pocus?

DR WAKELEE: No. I actually do think it’s legitimate. I think in the brain especially it makes a lot of sense, if you think about everything that’s happening with the signaling there. And one of the investigators at Stanford, Michelle Monje, has really done some beautiful work looking at how those — the signaling, electrical signaling, can modify growth of brain tumors.

I think it makes sense for other tumors that you could with these electrical fields change mitotic activity. That’s the mechanism of how it works, and I’ve seen some presentations on it where it does make biological sense. I was a little skeptical at first, and I was actually at our cancer — Stanford Cancer Institute retreat recently where some of the bioengineers were meeting with us also and king of going through.

And so again, it’s one of those things where if you just hear about it once you’re like eh, but if you hear about it from multiple different angles then it starts to be something you believe. It’s a truth even though it wasn’t a truth you would have thought of yourself. So I do believe there is something to it. Whether it’s going to be a practical addition to our armamentarium in lung cancer I don’t know yet. I’m going to reserve judgment until I see the data. But I do know that for colleagues who are treating gliomas I mean they really do use this as part of the standard of care.

DR LOVE: The other thing they were telling me that I’ve never heard before was that the battery is very heavy to carry around.

DR WAKELEE: That I believe, yeah.

DR LOVE: I would imagine it must be the same kind of fields or whatever that you use for mesothelioma.

DR WAKELEE: Oh, yeah, yeah. It’s the same thing.

DR LOVE: The entire — so if anybody’s using it there I guess they would know about it.

DR WAKELEE: Yeah. Right. And that’s why I don’t know what’s going to happen in lung because we just haven’t adopted it too much in meso, so…

DR LOVE: All right. Please continue.

DR WAKELEE: Almost done; just have 1 more study, and this is the IPSOS study. This was a Phase III that was presented at ESMO, and this is looking at first-line atezolizumab versus single-agent chemotherapy in patients who were not eligible for chemo. And so this is the frail and the elderly, which is a patient population we don’t talk as much about at trials, when we’re discussing trials and when we’re at the conferences, and yet this is the majority of the patients that we see in clinic when you’re dealing with a non-driver patient population.

So this was a pretty real-world study as far has having relatively open eligibility for patients, any histology, not able to get platinum chemo because they had a bad performance status or because of their age. And they were randomized to either get atezolizumab single agent regardless of PD-L1 expression or to get single-agent vinorelbine or single-agent gemcitabine.

Now in the US we don’t use a lot of single-agent gemcitabine or single-agent vinorelbine, but those are options that are used in a lot of the rest of the world. It’s certainly a reasonable design, and they’re reasonably tolerated drugs, so I can understand this design, it was — and they got it done. This was a hard trial to get completed, and they managed to.

The tables here are just showing well who are these patients, and you can see that the median age was 75. That’s one of the key things. Performance status 3 in less than 10%, but still a sizeable number of the patients had a performance status of 3. Vast majority performance status 2. And many of them with a smoking history or active current smoking.

And when we look at the results we can see that the response rates weren’t that high. Single-agent atezolizumab response rate of about 17%, which is what we would expect. I mean that’s within range for single agent, 10, 20%, somewhere in there. Chemo single agent 10%. Again first-line single agent, single drug, that’s not unexpected. What’s really distinguishing though is that if you do respond to atezolizumab the probability of that being a long duration response is real, and that’s what leads to the overall survival benefit.

So when I first heard about this trial design I thought it was intriguing, it was a little bold, I thought, to do, and yet it — and I’m pretty vocal about not liking to use checkpoint inhibitors where you don’t have a good biomarker to indicate response. But here, when you have a patient in front of you who you think is too frail for chemo, and your options are single-agent chemo, immune checkpoint inhibitor, or not treating, this data does add a little bit more to that discussion for patients in this concept of with immune checkpoint inhibitor the chance of this having an impact on you is less than 20%, but it’s close to that, you will have a risk of toxicity from immune checkpoint inhibitors, but you have that maybe 20% chance of long-term benefit, and with single-agent chemo you don’t have that.

And so it’s not everyone’s goals are going to be in line with I want to try that, but I think it provides good data to have that discussion and have that be an option for patients, so that’s why I wanted to talk through that one.

And this is a little bit more of the information, and again it’s looking at — for this trial the level of PD-L1 didn’t seem to matter that much. So maybe the numbers are small or it’s one of those things every once in a while it doesn’t seem to fit the way it’s supposed to when we’re looking at PD-L1 levels, and we’ll hear that in the adjuvant setting too when we are talking with Dr Luis Paz-Ares, so I think that was my last slide, so…

DR LOVE: Thanks so much. It was interesting. I learned a lot, as always. One final question, are you going to be able to look at ctDNA in IMpower010?

DR WAKELEE: Well, we already have. There was a presentation done a little over a year ago, and it clearly showed that if the ctDNA was positive there was a pretty robust separation of the curves, also showing that that was very dependent on level of PD-L1 expression. So if there was ctDNA detectable in a tumor that was PD-L1 positive, nice robust separation, it there was ctDNA detected and there was no PD-L1 detectable, then people recurred.

The problem is the assay that we were able to use, commercial assay, wasn’t sensitive enough, and so even in patients who were ctDNA negative you still had a reasonable percentage of those recurring, but there was a separation of the curves as well. And so it’s if you have a lot of ctDNA atezolizumab really helps when there’s PD-L1 expression. If you have a little bit of ctDNA detectable then you’re helping fewer people because there are fewer people to help, but you still are getting that benefit. We just need better assays where we can say a negative is actually negative. We’re not there yet.

DR LOVE: Now this was a tumor-informed ctDNA?

DR WAKELEE: No. It was a commercial assay, It was with Signatera.

But I don’t think that Signatera — yeah, the Signatera commercial is not quite sensitive enough yet. I mean the work I do with Max is mostly tumor informed at this point for the really ultrasensitive tests. And I think that when we’re dealing in the setting of adjuvant treatment I think that we should be probably using those assays because we can get to the really, really, really sensitive —

DR LOVE: Right.

DR WAKELEE: — results that we need. I think that when we’re dealing with — and that’s what we’ll also need when we’re looking at these immune checkpoint inhibitors and do we stop at 2 years or not. But for those we might not have to all the time —

DR LOVE: Right.

DR WAKELEE: — or it’s going to be a couple years ago, and so it’s going to be harder to assess.

Advances in the Care of Patients with Nonmetastatic NSCLC and Small Cell Lung Cancer — Luis Paz-Ares, MD, PhD

DR PAZ-ARES: Hello. I’m Luis Paz-Ares. I’m a medical oncologist working at 12 of October University Hospital in Madrid. I’m mainly dealing with lung cancer in terms of patient care but also clinical and translational research. And over the next few minutes I’m going to try to review some of the issues in terms of relevant new data sets and advances during the last year in terms of immunotherapy and other nontargeted approaches in lung cancer in 2022.

You see here my disclosures. And the way I have organized this is in 3 different sections, 1 dedicated to early-stage non-small cell lung cancer, second one on unresectable disease, Stage III, and finally we will have the review of the small cell lung cancer.

In terms of early stage, we had a number of very relevant studies last year, and I would say thank goodness the advances we have seen with immunotherapy over the last decade has reached early stage over the last couple of years, and indeed I think one of the more relevant data were on the CheckMate 816 study.

You see here the design. That was dedicated to early-stage non-small cell lung cancer, Stage IB to Stage IIIA following the TNM 7th Edition, good PS, and no genomic aberrations that are suitable for TKI treatment. Patients were stratified by stage and PD-L1 and gender and were given standard of care chemotherapy or the experimental arm, 3 courses of chemotherapy plus nivolumab. Then patients underwent surgery and finally, after surgery, patients could receive optional adjuvant chemotherapy and radiotherapy.

Primary endpoints were dual, pathological complete response and event-free survival. Here you see the main data out of this trial. There was a clear improvement in the pathological complete response rate from 2.2% on the chemotherapy arm as compared to 24% on the chemotherapy plus nivolumab arm. The other primary endpoint, event-free survival was also in favor to the experimental arm. You see here a hazard ratio of 0.63. That means a decrease in the progression rate of 37% along the study period so far. That was statistically significant, and the immature data on survival suggests a numerical increase with a hazard ratio of 0.57. Of course due to the immaturity these data cannot be seen as statistically significant because they didn’t reach the statistical significance.

Important, last year was presented some data on the outcome depending on the response. And as you see here regardless of the treatment arm those patients that reach a complete pathological response on the chemo plus nivo arm, or in the chemo arm, they really had a very nice outcome. Very few patients relapsed.

And of course even among those patients with some residual viable the lower the viable tumor at the time of surgery the better the outcome. And for those patients with plenty viable tumor at the time of surgery, those were the patients that do the worst. And I think that’s pretty important data because that maybe is going to tell us later what to do in terms of adjuvant therapy at some point.

Some of the relevant data are common in the adjuvant setting, and here we have data from the IMpower010 study and the PEARLS study. This study with atezolizumab included patients from Stage IB to IIIA were treated after surgery, after complete resection, with standard of care chemotherapy and then randomized to receive atezolizumab versus best supportive care. And importantly patients were stratified depending on gender, stage, histology, and PD-L1 expression.

In the ITT population that was patients with Stage II and IIIA with more than 1% expression of PD-L1. There was a clear improvement in disease-free survival, and you see here that was very clear. That significant improvement in disease-free survival was also seen among all patients with Stage II and IIIA regardless of expression of PD-L1, even the magnitude of the effect was of course decreased, was not 0.66 but in that case was somehow decreased, as you see here in this slide, to 0.79, and the trial was not positive for entire ITT population, those patients between Stage IA to IIIA. So relevantly the trial was positive, but particularly for the population of patients that were PD-L1 positive.

Indeed, doing the analysis by the strength of PD-L1 expression, I have to say that most of the benefits were for those patients that were PD-L1 more than 50%. The hazard ratio in this very subset of patients was 0.42. And indeed some of the agencies, such as the European agency, approved the treatment for that very indication.

Last year there were some presentations on the initial overall survival data. As you see here there is suggestion on those patients with Stage II to IIIA with more than 1% PD-L1 expression of some improvement in OR with a hazard ratio of 0.71. That was not statistically significant. As you see here the confidence interval is crossing the unity.

When the analysis was performed on the population where the disease-free survival was clearly superior, those patients with PD-L1 expression more than 50%, you see here in this ad hoc analysis a much better hazard ratio, 0.42, suggesting that the magnitude of the improvement with adjuvant atezolizumab is related to the PD-L1 expression in the tumor cells.

The other trial that was very similar in design is the KEYNOTE-091, or the PEARLS trial. The only difference I would say in that case the immunotherapy treatment was pembrolizumab. The second relevant thing was that patients were not necessarily required to receive adjuvant chemotherapy before entering into the study. Indeed some 15% of the patients were not given chemotherapy because the patient was unfit or the physician considered it was not indicated. And relevantly, as well, the study was placebo controlled.

Again, importantly, the dual primary endpoints were DFS in the overall population and in the PD-L1 more than 50%. And as you see here, there was very similar results in the overall population, hazard ratio of 0.76 in favor of patients treated with pembrolizumab as compared to placebo. In terms of disease-free survival, that, however, did not translate into an improvement in the subset of patients with PD-L1 more than 50%. And I would say that was somehow not validating what is being found on the atezolizumab trial.

DR LOVE: Sorry, yeah. Just a quick question. What’s triple blind? The trial design said triple blind. I’ve never heard of that. What is it?

DR PAZ-ARES: I think it’s also the sponsor being triple blind. It’s kind of a bit overdoing it, to be honest.

DR LOVE: Got it. Okay. Yeah, no. I was just trying to figure it out. No, but that makes sense. That’s interesting then. Okay, please continue. Sorry.

DR PAZ-ARES: Okay. So nowadays then it is important to say that we have data suggesting adjuvant immunotherapy with PD-1/PD-L1 inhibitors do have some efficacy. The same is true in terms of the neoadjuvant treatment with chemo plus nivolumab.

So the question is what should we do with our patients who will use neoadjuvant chemo/IO. Should we use adjuvant IO? Or should we use perioperative treatment, let’s say neoadjuvant immunotherapy plus adjuvant immunotherapy? Well, there are a number of trials that are actually designed in this setting, usually in immunotherapy, in the neoadjuvant phase, in the induction phase, plus some adjuvant treatment. And you see here a number of studies. Indeed, the only study that didn’t have a mandatory monotherapy phase in the adjuvant setting was the CheckMate 816 that I commented before.

And we have 2 trials, although these other 6 trials that already we have some press releases suggesting they are positive, still we do not have much data about the magnitude of the benefit. Those are the AEGEAN trial studying chemo plus durvalumab followed by durvalumab after surgery or the KEYNOTE-671, where pembrolizumab is added to chemotherapy in the neoadjuvant setting and then given in the adjuvant period.

There is an extra trial that had a similar design, the Spanish NADIM trial. In that case patients with locally advanced disease, Stage IIIA and Stage IIIB following the 8th TNM classification, were randomized to receive chemotherapy 3 cycles versus chemo plus nivolumab. And then patients were submitted to surgery, and then there were some 6 months of extra adjuvant nivolumab or no further treatment on the control arm.

As you see here, there was a clear improvement in terms of the pathological CR rate, as we have seen before in the 816 trial. Also the surgical outcomes were much better, particularly look at the data. The number of patients that underwent definitive surgery were 93% on the chemo plus nivo versus 69% on the chemotherapy-alone arm. Remember those patients were Stage III, Stage IIIA and Stage IIIB, not Stage I, not Stage II. The second thing is more than 40% of the patients in this trial had N2 disease in more than 1 station, so that means very relevant population.

Importantly, PFS and OS were also in favor of the immunotherapy arms, so those patients treated with neoadjuvant chemo plus nivo followed by adjuvant nivolumab for 6 months had a better event-free survival with a hazard ratio of 0.48. And the same data were replicated in terms of survival, as you see here on those data that are now to be published in The New England Journal of Medicine.

So messages to take home in terms of early stage, I think neoadjuvant chemo/IO have shown to improve pathological complete responses, event-free survival, and overall survival in early-stage, resectable non-small cell lung cancer.

After pathological CR those patients are doing pretty well, and we are not sure those patients are needing immunotherapy in the adjuvant setting. However, we haven’t seen much data comparing the addition of immunotherapy in the adjuvant setting or not.

Adjuvant immunotherapy with atezo and pembrolizumab also has shown to decrease the probability of relapse. It seems to be that decrease in relapse rate is related to PD-L1 expression at least in the atezolizumab trial. That was not the case on the pembrolizumab data. And secondly, the atezo trial is a bit more mature and suggests a numerical improvement in overall survival, but data still needs to mature further before we have statistical significance.

Finally, we have data on the NADIM trial, and 2 press releases in the AEGEAN and KEYNOTE-671 trial suggesting that perioperative neoadjuvant plus adjuvant immunotherapy with pembrolizumab, nivolumab, and durvalumab improves outcomes in resectable disease. I think we are awaiting those data, and then we may have an idea which patients are required to have only neoadjuvant therapy, which patients are advised to have adjuvant therapy, and which patients that may be benefitting from neoadjuvant plus adjuvant immunotherapy.

DR LOVE: Could you help me kind of understand the 2 major adjuvant trials that we have right now? And from a practical — putting aside reimbursement issues, how you would take those — or how are you — or how would you like to take these 2 data sets and apply them clinically outside a protocol setting, and in particular which IO in which situation because frankly to me it’s pretty confusing.

DR PAZ-ARES: To be honest, I think the main result of the 2 trials are superimposable. They decrease in the range of 30% the risk of relapse in the population of patients studied, and that is the main message, very consistent. Pembrolizumab and atezolizumab.

The main difference was that the benefit was clear for patients with Stage IB, as well, in the pembrolizumab arm — trial and was restricted to patients with Stage II and III PD-L1 expressors on the atezolizumab arm. Indeed, in the atezo trial the benefit was proportional to the expression level, and this is very consistent with what we have seen in other trials in Stage III and Stage IV. In that sense the pembrolizumab data is a bit of an outlier, and looking carefully at the data, those patients with high PD-L1 expression treated on the placebo arm had somehow overperformed as compared to what was expected. But of course this is just a post hoc interpretation of the data.

So today, in my clinic, I think you can use pembro, you can use atezo. The data for high expressors in the trial looks a bit better, of course, for atezo, and maybe for some people that is a good reason to use atezolizumab in high expressors. On the contrary, for Stage IB for example, only pembrolizumab was having positive data, and only that subset was approved by the agency, so you could choose to do pembrolizumab maybe in those patients I would say.

DR LOVE: Do you think that when you look at these data like one of the things that would come to mind would be difference in methodology of measuring PD-1? Do you think that part of what you see as a difference is — maybe if you had taken all of the specimens from 1 trial and had them tested in the other trial that really the big difference was in PD-1 assays?

DR PAZ-ARES: Honestly not, because it is true that in the IMpower trial they were using a different assay specification factor, but the final analysis was done using I would say a standard assay, quite similar, so I don’t think that should be the problem. On the other hand, the data we saw regarding the magnitude of the benefit being proportional, being relative to the PD-L1 expression, is something that we have seen in the advanced setting. That makes all the sense. So I don’t think the PD-L1 assay is the main reason for some differences in the results among the 2 trials.

DR LOVE: Do you think there’s any inherent difference in the 2 agents? Is there any data, either in lung cancer — I mean to me almost everything I’ve seen with pembro and atezo it’s pretty hard to separate. Do you think they’re intrinsically different, or do you think maybe it was the way the trials were done?

DR PAZ-ARES: Honestly, I don’t have clear data suggesting they are very different. If it’s true that pembrolizumab is hitting — is targeting the PD-1 receptor, and atezolizumab is targeting the ligand, in some circumstances the results are very similar. In some others they were also differences in terms of the magnitude of the benefit. For example, in Stage IV disease some trials look a bit better, or the magnitude of the benefit being better on the pembro trial as compared to the atezo. However, I must say there were differences in the design of the trials, there were differences in patient selection, and maybe those reasons were behind the difference in magnitude of outcome. So I would say — I cannot say at this stage one agent is better to the other to be honest.

DR LOVE: So I was thinking about there are a couple trials adjuvant in bladder, 1 with atezo that was negative, 1 with nivo that was positive. Maybe you know Tom Powles, who’s investigating —

DR PAZ-ARES: Absolutely.

DR LOVE: He’s the head of the atezo trial, and he went back and looked at cell-free DNA, and then he saw the benefit in the positive patients. I think that they’ve looked at cell-free DNA in the IMpower study.

DR PAZ-ARES: So at present we didn’t look as yet. We have some samples, and those are some of the studies that are pending. In terms of the IMpower trial, those patients with ctDNA positive with minimal residual disease persistent after surgery, they did worse, and they benefit, but also ctDNA negative patients benefitted from atezo. So that suggests that ctDNA with the assay they use is prognostic but maybe not clearly predictive because also those patients that were ctDNA negative they were benefitting. And indeed I have to say that maybe into the future we have to improve the sensitivity of those tests so that we have less false negatives.

DR LOVE: So yeah, I’m looking forward to the webinar and hearing you and Heather talk a little bit more about it. But again, when you take a step back and look at this I guess, I don’t know, to me the thing that I really have a hard time understanding, am I right in saying like in the pembro study did you not see benefit in high PD-1 patients?

DR PAZ-ARES: Yes, that is totally true. We didn’t see that. We didn’t see that. This is very strange. If you look at the data you realize that the main observation after looking carefully at the distribution of other factors they only thing we found was that — what was suspected was that those patients on the placebo arm among the high expressors, they should have done worse as compared to the overall population in the placebo arm. And actually they did better, even though they have worse prognosis, because PD-L1 high is worse prognosis in early stage.

So the expectation would have been that those patients in the placebo arm did worse. So is that a hazard? Is that any other thing? We don’t know. But it’s clear that those patients on the placebo arm here, they did overperform, and that is the issue.

Look at this. 18 months PFS rate is 70% among the PD-L1 highs, and it was only 64%, and that is the reason the trial is not positive. It’s not the efficacy of the pembro, it’s the performance of the placebo arm. But of course those are post hoc speculations, and we cannot be sure about that.

DR LOVE: I’m trying to remember. I feel like there’s been some data looking at response to chemo or benefits of chemo based just on PD-1 levels. Do we know that?

DR PAZ-ARES: Yes. Yeah. I mean I would say very consistent data suggests in second line and in first line the higher the expression the higher the probability of benefit, the higher the magnitude of the benefit. That is very consistent, no doubt, in Stage IV and Stage III.

DR LOVE: Could it be lack of numbers? I’m just looking to see what the ends are. Plenty of numbers.

DR PAZ-ARES: That is also — that is one of the reasons. When you have small numbers our control arm is not behaving as the population because it’s a small data set.

DR LOVE: Really interesting. All right. Please continue.

DR PAZ-ARES: Okay. I think we are done with that, so okay, I’m going to go with the second set of patients.

So in terms of the second section, which is concerning unresectable Stage III disease, we also have a number of interesting data. The first one is related to the PACIFIC study. You may remember that was a study dedicated to Stage III unresectable non-small cell lung cancer that did not progress after chemoradiation. At this stage, those patients were actually randomized within the next 6 weeks to have consolidation therapy with durvalumab or placebo, and that was for a year, and PFS and OS were the coprimary endpoints.

Now we have the data on the 5-year outcome, and as you see there was a clear improvement in PFS but also in OS. That translates into an improvement in 5-year OS from 33% to 43% on the durvalumab arm and improvement in PFS at 5 years between 19% with placebo versus 33% with durvalumab. That means a very consistent improvement in the long term, and that suggests we are really curing nearly as much as twice the number of patients. Look at the PFS at 5 years, 19%, as compared to 33%.

Important. A secondary analysis by Dr Senan suggested that the benefit is true not only for those patients that were locally advanced because of N2 disease, Stage IIIA, but also for all of the other patients. So the benefit is quite there in terms of PFS for these 2 distinct types of stage patients.

Importantly, we have also the results of the PACIFIC-6 trial. That trial was designed for those patients that were not good candidates for concurrent chemoradiation, so those patients were typically treated in the clinical practice with sequential chemotherapy and then radiotherapy, and in those patients durvalumab was also given as a consolidation. And this trial that was not randomized, as you see here, suggested that this was a reasonable approach, as well, for those patients not candidates for concurrent chemoradiation.

You see here PFS at 12 months nearly 50%, and survival at 1 year is exceeding 84%. So those data I would say are pretty reasonable. Safety profile was quite favorable in this subset, as well, and that somehow validates the use of sequential chemoradiotherapy followed by durvalumab for those patients.

Also we have another interesting data set which is coming from the real-life PACIFIC study, which means patients that in the context of real-world data study were given durvalumab after chemoradiation, concurrent and sequential. And as you see here the data were pretty good. Look at that. Three-year survivorship 65% or 58% for patients that received concurrent chemoradiation or sequential. Of course patients with high PD-L1 were doing better as compared to those patients that were negative PD-L1, 67 versus 54%. And it looks like those patients that initiate durvalumab earlier than 6 weeks they were doing better as compared to those patients that started a bit later.

As a potential improvement into the future we have the results of the COAST Phase II study. In that case, the same patient population of Stage III patients with locally advanced, unresectable disease, they were randomized to receive standard of care, in that case the PACIFIC regimen, durvalumab, versus experimental Arm A, durva plus a second agent, oleclumab, or Arm B, which is another experimental arm, durva plus monalizumab.

You see here that overall response rate was the primary endpoint, and also PFS. You see here a clear improvement in PFS for those patients treated with the 2 experimental combinations, durva plus oleclumab versus — or durva plus monalizumab as compared to durva alone. I have to say, however, that in that very study those patients treated with the PACIFIC regimen, with durvalumab, were somehow underperforming as compared to the PACIFIC data. Look at here, median PFS is only 6.3 months, and you go a bit back to the initial data on the PACIFIC trial, at 1 year the PFS was, I would say, 67%, so I would say there is some underperforming, or at least that is some unexpected performing with durvalumab. And I think this should be taken into account.

Anyway, those regimens, those immunotherapy combination regimens are now being studied on Phase III trials dedicated to durvalumab/oleclumab and durva plus monalizumab, and data will tell us if this is a clear advance or not into the future.

Again, we are also doing some studies with chemo-radioimmunotherapy concurrently from the beginning. Instead of waiting to introduce immunotherapy after chemoradiation they are trying to include it earlier in concurrent chemoradiation approaches, and data also should be ready soon into the near future.

DR LOVE: I’m just kind of curious kind of what we’re going to do if like the Phase II the Phase III shows both of them improved. How do you think people would react to that? And was there any thought about using both with durvalumab?

DR PAZ-ARES: Well, the truth is that — I mean the good thing is — I mean honestly both combos look pretty good in terms of safety profile. So if they are good, I mean I understand that I — people would use the one that looks better, even if the trial is not set to compare the 2 experimental arms, but I suppose that in that case safety profile would be important on top of the magnitude of the benefit of each combo.

However, I mean I think we have to be cautious. In that Phase II trial the 2 combos they were doing a lot better as compared to the durva-alone arm, but the durva alone were underperforming as compared to what we have seen on the PACIFIC trial. So we have to be cautious.

DR LOVE: Any thoughts about why you see the difference here? Do you think it’s just the number of patients at random?

DR PAZ-ARES: Well, I think the number of patients is very important. And then at this stage it’s really difficult to say for me if this is because only those patients in the control arm were bad patients or all patients with — bad prognosis patients. I didn’t have the opportunity to have that detail.

During the study and during the discussion of this trial they have done a number of statistical analyses trying to rule out patients on the durva arm had clearly poor prognosis as compared to the combos. But still, maybe there are some uncovered prognostic factors that are not easy to highlight, to uncover.

DR LOVE: Please continue.

DR PAZ-ARES: Okay. So I think we are just at the time of summarizing the data on unresectable Stage III. I think durvalumab consolidation treatment, the PACIFIC regimen, has a clear and remarkable long-term impact in unresectable Stage III patients. How come? It is a clear improvement in 5-year PFS.

That is related to a clear improvement in OS, 33% in the control arm as compared to 43% on the durvalumab arm.

Those data have somehow been replicated on the real-world data on the PACIFIC study, and of course on consolidation durvalumab is also a reasonable strategy for those patients that are not candidates for concurrent chemoradiation but to sequential approaches.

I feel into the future there are a number of treatment optimization strategies that may include the use of consolidation treatment with combinations of immunotherapy and the COAST trial shows us that the combination of durva plus monalizumab or durva plus oleclumab may be reasonable approaches that should be studied, and indeed they are being studied on the PACIFIC-9 study. But also concurrent chemoradiation/immunotherapy approaches are also undergoing. Some Phase II data have shown the feasibility of this strategy, and there are a number of Phase III trials ongoing, including the PACIFIC-2 trial that should be reading in a few months or in the coming years.

So finally I will go for the last part of the presentation. The last section of this review is dedicated to small cell lung cancer, and I think here we have a number of relevant approaches as well. Over the last few years we have seen data of chemotherapy plus PD-L1 or PD-1 inhibitors in the extensive-stage setting showing that chemoimmunotherapy combinations do improve outcomes, do improve OS, do improve PFS in this setting. Those were the data coming from the IMpower133 and the CASPIAN trials.

Over the last couple of years we have seen some additional data. Some of them had been published last year, and that includes the KEYNOTE-604, the CAPSTONE-1, and the ASTRUM data. Altogether, as you see here, the hazard ratio was very consistent. The results were very consistent. Some of them, such as the pembrolizumab trial, didn’t reach statistical significance, but altogether those data suggest very similar, clear improvement with a reduction in the rate of death between 20 to 40% in this very setting.

Importantly, last year we have also presented the data with — the mature data with follow up of more than 3 years on the CASPIAN trial, and there was a clear improvement in survival for those patients receiving chemo plus durva as compared to those treated with durva alone. And as you see here, the separation of the curves is maintained over time. At 3 years there are as much as triple the number of survivors, 6% as compared to 18%. That was not only true in the chemo plus durva arm, that was also true on the chemo plus durva plus treme arm.

Indeed the maintenance atezolizumab, a sum analysis of the IMpower133 trial, showed that the maintenance phase is important, so part of the benefit, at least in survival and in PFS, is due to the maintenance phase of atezolizumab during this study.

In terms of the second-line setting there are a number of trials that were presented last year. One that I’d really like to highlight is the ATLANTIS trial. That was for patients that had received at least 1 chemotherapy line and maybe some additional biological lines, and this trial immunotherapy treatments. And patients were treated with the standard of care, topotecan or CAV combination chemotherapy, versus experimental arm, a combination of lurbinectedin plus doxorubicin. This regimen was coming from an initial Phase I trial with encouraging data. The primary endpoint was OS in this study.

And you see here that there was no survival benefit. The curves were really superimposed on the experimental arm, lurbi plus doxo as compared to the standard chemotherapy treatment, topotecan and CAV. The was somehow improvement in PFS that at least at the median was not very clinically relevant.

We have done a number of other trials combining lurbinectedin with other agents. There was a prior presentation suggesting that the combination with irinotecan may be more effective, associated with reasonable toxicity, responses in some 62% of the relapsing patients. And last year we also presented this data with a Phase I trial of lurbi plus atezolizumab, full doses of the single agents lurbinectedin and atezo where actually feasible to be given together with G-CSF. And among the initial 21 patients responses were seen in some 58% of the patients, including some 8% complete responses.

There was a similar trial in terms of design combining lurbinectedin not with atezo but with pembrolizumab. The initial data also suggests that full doses of the single agents are feasible, and among the initial patients treated in this trial responses were seen in some 30% of the cases, and toxicities that we saw were quite favorable, what you would expect in terms of the toxicities associated with pembro plus those associated with lurbinectedin without new signals.

I have to say that at this stage there is a Phase III trial comparing in the maintenance setting for patients with extensive stage treatment with atezolizumab versus atezo plus lurbi as maintenance therapy for those patients that had a response or disease stabilization after 4 courses of chemotherapy plus atezolizumab induction therapy.

The final data we have in relapsing disease are concerning 2 new agents. The first one is the BiTE belantamab. This is a bispecific drug that is binding DLL3, a molecule on the surface of small cell lung cancer, very highly expressed, some 85-90% of the cases. That protein, which is a notch ligand, is typically expressed in the cytoplasm of normal cells, but this appears on the surface of the small cell lung cancer cells, and it’s a very attractive molecule for this type of approach using BiTEs.

On the other side, on the other stream, this bispecific drug is binding CD3 on the T cells and is using the activation of the T cell that at the end is provoking the lysis of the small cell lung cancer cells.

The initial data that had been just published a few months ago in JCO suggested that the drug is tolerable.

Half of the patients, actually developing some CRS, typically Grade 1 or 2. Only 1 case of Grade 3 was observed in this Phase I trial.

Some significant number of patients, about 40%, also had some neurotoxicity. Most of the cases it was transient, and that typically happens, CRS and neurotoxicity, in the initial cycles.

Importantly, responses were seen in more than 23% of the patients, and more importantly responses were somehow long lasting. Median duration of response was exceeding 1 month — 1 year, sorry, and indeed if you look at the data in PFS, particularly in OS, they are pretty relevant. Look at that median OS, 13.2 months. You look at any data set in the second- or third-line setting, most of the median survivals are in the range of 8 to 9 months, and I think this is somehow great.

The other drug I’d really like to highlight here is this antibody drug conjugate DS 7300, which is targeting B7-H3, which is pretty much expressed in a number of tumors, including small cell lung cancer. In the initial Phase I trial responses were seen in a number of tumors, as you see here. Specifically in terms of lung cancer, small cell lung cancer, 19 patients were treated, and at the time of the report at ESMO some 59% of them, 11 out of 19 patients, did have a response. And actually you see here the profile of the duration of response seems to be quite reasonable, as well, quite encouraging. And further cohorts are being treated with this drug, and also some combination trials are being started.

So just let me summarize the data in small cell lung cancer. I think at the current stage platinum/etoposide plus a PD-1 or a PD-L1 blockade represents the standard of care in front line in small cell lung cancer. There are at least 5 trials that have shown very consistent data. Important, the impact is also maintained in the long term. Three-year survivorship on the CASPIAN trial is improved from 6% to 18%. And I have to say there were at last some encouraging data, at least preliminary data, with a number of approaches in the relapse setting, and I like to include here a number of studies with lurbinectedin, particularly in combination with immunotherapy, atezo or pembrolizumab, but not, of course, with doxorubicin. As we have said, the ATLANTIS trial was negative.

Other strategies that merit further attention I think are the T-cell engagers, DLL3, such as the case for tarlatamab and the antibody-drug conjugate DS 7300.

DR LOVE: So just a couple of follow up questions. First of all, in terms of lurbinectedin did the ATLANTIS trial in any way discourage you from using lurbinectedin as a single agent in your practice?

DR PAZ-ARES: Honestly, I don’t think so. This trial showed that using lurbinectedin in combination with doxorubicin is not improving outcome. Actually it’s better — the safety profile is better as compared to standard of care but is not improving efficacy. However, in this trial we are not sure we have used the right combination. Lurbinectedin was used at 2 mg/m², and we are not sure about the value of adding doxorubicin. A subanalysis of that trial suggested that the use of higher dose may be better, at least that was a subanalysis. And this is consistent with the data we have in the second- and third-line setting.

Indeed, we have performed a couple of small Phase II trials combining lurbinectedin with irinotecan and lurbinectedin with atezolizumab. And using lurbinectedin at the monotherapy dose, 3.2 mg/m², the data were pretty encouraging. Of course early data, small subset of patients, but I think this is more likely the way to go.

DR LOVE: So another question about these last 2 agents that you talked about, the bispecific and antibody-drug conjugate. To me I look at that and go wow, that’s pretty impressive. Do you feel that way? I mean you’re seeing objective responses there in small cell it seems. And also I don’t remember — I mean you have something like amivantamab that technically is a bispecific, but in terms of an immune bispecific I’m not sure I’ve heard about solid tumors even having objective responses. You tell me. I don’t know.

DR PAZ-ARES: You’re totally right. I mean in terms of a T-cell engager, which is a specific —let’s say a very singular type of bispecific, this is the first bispecific T-cell engager that is working clearly in solid tumors. We have data with some BiTEs, blinatumomab for example, working in T-cell leukemia, T-cell lymphomas, and some data as well on some other BiTEs in multiple myeloma. But this is the first encouraging data I could say, at least for me, in solid tumors.

The more important thing is that duration of response is pretty good, and indeed median survival in this cohort of patients in the initial Phase I trial was exceeding 13 months, and I think that is very — something in small cell lung cancer.

DR LOVE: Yeah. I mean in myeloma you have one approved, teclistamab. In lymphoma I think it’s — mosunetuzumab is approved also, I think. And they’re dealing with the whole issue of are docs in community practice going to be able to give it, et cetera. What — theoretically what do you think about this agent? Is the CRS and all difficult enough that it’s going to start out in tertiary centers?

DR PAZ-ARES: So the truth is that I think at least for the initial studies we have to be very cautious. I mean we didn’t have major issues in most of the patients in this trial, but we were admitting those patients for a number of the initial infusions, and those trials were done in a number of — a few dedicated centers. So I think the next step going into the community has to be done with caution, as well, because I mean it is important that those patients do have CRS. For most of the time it has been Grade 1 or 2, but of course it requires knowledge, attention, and every treatment with steroids, fluids, tocilizumab as well.

DR LOVE: Yeah. My favorite story about bispecifics is we did a program where I actually interviewed some patients, and I actually interviewed a patient who’d been on the bispecific for myeloma for a year and a half, doing great, the day that it was approved. So he’d already been on it for a year and a half, and I use that as an example of the advantage of participation in clinical trials.

DR PAZ-ARES: Honestly, I got on this program a good number of patients treated, and thank goodness a number of them really had great benefit. It doesn’t mean every single patient responded at all. I mean response rate was 23%, but responses are of pretty good quality most of the time.

DR LOVE: If you had to guess, best case scenario, when do you think either one of these agents might becoming available?

DR PAZ-ARES: Well, I don’t know, but my guess would be that — there is a clear unmet need in the relapsing setting for small cell lung cancer. So looking at the data I could say it’s very likely that the drug is doing to be maybe approved on a fast track. I’m pretty sure that the agencies are going to be concerned about the CRS and so on. So I suppose they are going to be very vigilant where and how those patients are treated. Indeed the sponsor is very careful about that, as well, because they are not starting to think about Phase III trials, and I suppose they have to plan that very carefully because those toxicities here require some really careful attention and specific knowledge on how to recognize them and how to treat them.

DR LOVE: Although the good news is that these docs are also treating myeloma and lymphoma, and it sounds like similar issues. I don’t even know if they have tocilizumab in the general office, but at least they have a couple year lead on lung, so maybe by the time it gets approved they’re going to be a lot more used to that I would think.

DR PAZ-ARES: Yeah. Absolutely.