Design and results of the IMpower010 study evaluating atezolizumab versus best supportive care after adjuvant chemotherapy for resected Stage IB to IIIA non-small cell lung cancer (NSCLC)

DR LOVE: Welcome to Oncology Today, Key Presentations on lung cancer from the 2021 American Society of Clinical Oncology Annual Meeting, this is medical oncologist Dr Neil Love. For this program, I met with Dr Joel Neal from Stanford University. And in addition to this audio program, there is also a video component with Dr Neal’s presentation. To begin, Dr Neal discussed data from the landmark Phase III IMpower010 trial evaluating the use of adjuvant atezolizumab in Stage IB-IIIA NSCLC.

DR NEAL: So one of the most exciting things that’s potentially practice changing from ASCO was the IMpower010 study presented by Heather Wakelee. And in this study patients who we would have thought as eligible for adjuvant chemotherapy anyway, so those with Stage IB to IIIA, 7th edition staging criteria, tumors greater than 4 cm, who had lobectomy or pneumonectomy, regardless of PD-L1 status, got 1 to 4 cycles of regular adjuvant chemotherapy. As we know from the old data, that would be expected to have about a 5% overall survival benefit for this population. But then randomized to checkpoint inhibitor immunotherapy with atezolizumab versus placebo for an additional year.

And in this study the disease-free survival in the PD-L1 greater than 1% population was the primary endpoint, Stage II to IIIA, and at 2 years disease-free survival was significantly better, almost 75% versus 61% of patients having disease-free survival. Hazard ratio 0.66, and p-value highly significant. So this early endpoint looking like, at least in the PD-L1-positive population, greater than 1%, significant benefit.

DR LOVE: So in terms of clinical applicability, as you mentioned, right now the survival benefit, although it looks like there is one, is not statistically significant, but I guess it’s encouraging. What are you thinking about? Ideally, how would you like to take these data into your practice?

DR NEAL: Well I think, Neil, one interesting thing, at least for disease-free survival they did a subset analysis. And when we look at the EGFR-mutant and ALK-positive population they didn’t have a disease-free survival benefit compared to placebo, nor did the patients with tumors with PD-L1 0%. So PD-L1 undetectable looked like there wasn’t a disease-free survival benefit in those compared to placebo.

So the overall survival data are early. I think we need a couple of years for the data to mature, as well as looking to see if there is any potential crossover because for patients that progress probably the first-line standard of care, at least in most countries now, would be either chemoimmunotherapy or potentially even immunotherapy alone as second-line treatment after progression on platinum therapy, depending on how soon it was. Often if it’s less than 6 months I won’t give more platinum; more than 6 months I would probably give platinum plus immunotherapy.

So the lack of overall survival early doesn’t bother me. In my practice I’m certainly intrigued and thinking about using immunotherapy for these PD-L1-positive patients who got platinum-based chemotherapy, 4 cycles, and are looking for something else. No EGFR, no ALK, PD-L1 positive. This is the population that seemed to benefit the most from this study. I have not given it off label as of this point, but I’m certain it’s under FDA review, and this is population for whom I’d use it when it’s approved.

Now I’ll tell you a story. In our patient population we have an interesting clinical trial where we take these patients who have gotten chemotherapy and all the other treatment and we’re doing a circulating tumor DNA blood test. We’re testing their tumor. We’re testing their circulating tumor DNA with a research-based assay that’s ultrasensitive, looking for any mutations from their tumor. And for the patients that are positive we’re giving them adjuvant immunotherapy after that for the patients with durvalumab. For the patients that are negative, we’re putting them on observation alone. So we’re making a leap of faith to say if they’re positive for minimal residual disease by a blood test they should get more treatment. Now that’s not exactly the question that was done in this study, this was given to all patients, but the patients with residual DNA I think will benefit the most.

DR LOVE: So let me ask you. I’m a little confused about these circulating DNA assays. Is it the same assay for everyone, or do you tailor it based on the patient’s tumor?

DR NEAL: So there’s a few different ways to do the circulating tumor DNA assays. One is tumor informed and one is tumor not informed. So many of the commercial tests that we send out for Stage IV patients, metastatic, to figure out do they have EGFR mutations or some other mutation, those are not tumor informed because we’re really sending those and looking and saying if there’s an EGFR mutation in the blood at a certain percentage of DNA floating around, then it must have come from the tumor. Where else could it have come from? Tumor informed strategies, though, are much more sensitive for both following disease over time, because you can look and see what not only, for example, an EGFR, mutation, but what are all the mutations in any part of the tumor. And tumors have many, many more mutations than just a driver. So those tumor informed strategies allow you to say here’s a panel of mutations, I can follow all of these mutations in the blood at the same time, and really dramatically increase sensitivity to either detect recurrence or quantitatively measure change over time.

So there are some commercial companies doing this, and we’ve done this in research over a couple of years and are now trying to test it more prospectively.

DR LOVE: Super interesting.

Clinical implications of the results of the IMpower010 study; management of NSCLC in patients with targetable genetic alterations

DR LOVE: Let’s talk a little bit more, again, about assuming it is available what you would like to be able to do with this strategy, theoretically, based on your appraisal of the risk/benefit opportunity. And I guess one question would be Stage IB over 4 cm, which was allowed in the trial, a little bit lower risk, a little bit less benefit, would you be offering it to a patient like that?

DR NEAL: I think probably any patient that is eligible for chemotherapy they were included in part of the population here. I think it makes sense for them to. But the Stage IBs over 4 cm, at this point now, by the new staging, the 8th edition staging, they’re counted as Stage IIs, right, so really Stage II and IIIA in clinical practice. Stage IIs are the over 4 cm Stage IIAs, technically, and the IIBs are the ones with lymph node positivity. So I would say anybody who’s PD-L1 positive, EGFR/ALK negative, probably negative for some other driver mutations too, if I knew about those up front, and PD-L1-positive tumors 1% or greater, I think this is going to be a new standard of care for that patient population, and I would offer it.

DR LOVE: So what about people with alterations other than EGFR and ALK? I’ll give you a couple examples. We actually did a webinar that included Heather about a week ago, and we got into this. So we were talking about just as examples, and the list is getting pretty long at this point, but the 2 I liked was BRAF and RET.

DR NEAL: Yeah.

DR LOVE: So you’re kind of stuck between ADAURA and IMpower010, and these are the people that are kind of getting left out, I think. But you tell me, what are you doing right now with a Stage III BRAF patient?

DR NEAL: Right. So the stickiest one for me would probably be ROS1 because I think ROS1 patients behave a lot like ALK and we rarely, if ever, have documented immunotherapy responses in either ALK or ROS1 patients. We occasionally see it in EGFR patients, but the adjuvant osimertinib is exclusive with giving adjuvant immunotherapy, so I’d certainly advocate for that over adjuvant immunotherapy.

RET, MET, BRAF I think are a little less clear. We’ve seen mixtures of data of populations in the Stage IV setting. Some say RET patients and MET patients seem to respond to immunotherapy. And BRAF patients can certainly respond to immunotherapy, just like KRAS and other molecular alterations. So as of right now I would say RET, MET, BRAF patients, and certainly KRAS patients, I would advocate for consideration of immunotherapy. I wouldn’t have a problem giving them immunotherapy at all in the adjuvant setting.

I think it would be a bit of a stretch to give them an adjuvant TKI in the same way that ADAURA did. And I haven’t been using adjuvant ALK inhibitors either. I know that there are some clinical trials waiting to mature in that setting, both the ECOG-ACRIN adjuvant crizotinib, as well as multiple other ones with adjuvant alectinib and other agents.

DR LOVE: So Stage III RET are you going to give chemo and an IO? Are you going to give chemo/atezolizumab?

DR NEAL: So Stage III RET with surgery I would give, if atezo were approved in this setting, I would do surgery, chemotherapy, and IO. Stage III chemoradiation I’d give them durva. I wouldn’t be tempted by the RET inhibitors until they demonstrate clear metastatic disease. But in that patient population, in many of these targeted alterations, one of our standard practices has been to do periodic brain MRI surveillance after chemoradiation. And for Stage III patients I worry about those CNS metastases. So clearly adjuvant osimertinib prevents CNS metastases. Many of these other targeted therapies get well into the CNS.

So our personal practice at Stanford has been about 6 months, about a year, and about 2 years after Stage III treatment we’ll do brain MRIs just in case. Haven’t had a problem getting those reimbursed for the most part, and I think it’s always better to see brain mets sooner rather than later, especially with active targeted therapies.

DR LOVE: That’s interesting.

Outcomes in the Phase III CheckMate 816 trial evaluating the addition of nivolumab to platinum-doublet chemotherapy as neoadjuvant treatment for patients with resectable NSCLC

DR LOVE: And the other one that’s so interesting, really caught me by surprise, maybe not the lung investigators, was the neoadjuvant nivo study, the CheckMate 816 trial, and there they combine nivo plus chemo in the neoadjuvant setting. And you were mentioning in your talk, and I was thinking about it, it’s interesting how the initial neoadjuvant IO alone one of the concerns was maybe they’d progress and not get to surgery, and by adding it to the chemo I guess that really mitigates that concern.

DR NEAL: Yeah. I think this strategy, if we’re going to use a neoadjuvant strategy, this is very rational. Because we know that with chemotherapy alone, based on the old, I think Scagliotti trial from way back when, there were data about cisplatin/gemcitabine for 3 cycles and then proceeding on with surgery, lobectomy, being much better than going straight forward to lobectomy alone with Stage III patients. So that really established the role for neoadjuvant chemotherapy.

But we know that chemotherapy doesn’t work in everybody in Stage IV disease, and we know that immunotherapy doesn’t work in everybody front line in Stage IV disease. So I think analogous to the way that we’ve gone to chemo/IO for the vast majority of patients with metastatic disease, if you want the best response possible give plan A and plan B at the same time.

I was surprised by how high these complete response rates were though, really path CR rates that were in the 25% to as high as 40%, although that may be an outlier in Stage IB disease, compared to path CR rates that we see that are in the single digits, usually, for chemotherapy alone. So this immunotherapy really seemed to boost the path CR rates and downstage patients dramatically.

DR LOVE: Yeah.

I love it when they compare the side-by-side waterfall plots because visually you just look at it and you realize what’s going on here. Really, as you say, interesting.

When the initial neoadjuvant IO lung trial was out they had this thing called major pathologic response, where there was still something on the imaging, but then they went in, and there was no viable tumor.

I guess the question is where this is going to fit it because now you have a different story. Now you’ve got a lot of people having really major responses. What kind of input are you getting from your surgeons? Here, it looked like it was changing the kind of surgery they were getting, which really one of the major endpoints here, and it looked like there were less pneumonectomies and the surgery was lesser. Is that your impression, that by using effective neoadjuvant therapy you can change the kind of surgery the patient gets?

DR NEAL: Right. So a few more patients moving forward to surgery, fewer patients getting pneumonectomies instead of lobectomies. I think both of those were promising.

Our surgeons tend to move straight forward with lobectomies if possible, except for Stage III disease. So for Stage III non-small cell lung cancer, node positive, N2 disease, they’ll say really the standard of care is let’s give something up front, and I think that’s the patient population for whom I’m most tempted based on these data. There is an ongoing pembrolizumab neoadjuvant plus chemo study also that we’re participating in, so for these patients we’ve put some people on that study. It’s placebo controlled, so we don’t know who’s gotten what. But I think that’s a rational neoadjuvant approach, especially for the Stage III patients, maybe the higher Stage II patients.

For a Stage IB patient or just a 4 cm tumor, I suspect that our surgeons would say well we know that somebody’s going to surgery if we just do the staging and take them to surgery. Every once in a while side effects from treatment or side effects from chemotherapy eliminate their chances of surgery by itself. So I think this is best for the higher-stage borderline patients than for the here it’s a 4.1 cm tumor chip shot. But I think we’re probably going to have more discussions about this. I have, in all honesty, used the chemo/IO approach off label once or twice. It’s been a while since I’ve done that because we do have that clinical trial now, which we prefer whenever possible. But I’ve used it once or twice, and it’s been successful.

Now what’s interesting about this study is they don’t appear to have given patients the immunotherapy versus placebo afterward. So we’ll see if there is that long-term separation when we see disease-free survival and overall survival between the populations. But we never really know, Neil, if more than the first dose of immunotherapy is needed. Nobody’s done the 1 dose versus 1 year. And maybe most of the effect we get for long-term survival is after the first couple of doses.

DR LOVE: I always thought neoadjuvant targeted therapy made sense. I know there was a trial that wasn’t able to accrue where they tried doing that, I think with EGFR-TKIs. Do you ever use that, neoadjuvant targeted therapy?

DR NEAL: Yeah. We did participate in that old RTOG 1306 study with neoadjuvant erlotinib or even neoadjuvant crizotinib for the ALK patients, and they gave you a choice of chemoradiation. It was a chemoradiation-based study as opposed to surgery. We’ve occasionally given, for EGFR patients, at Stanford we have a heavily EGFR-mutation patient population, so occasionally we’ve given chemotherapy neoadjuvant or EGFR-TKI neoadjuvant for those borderline patients. And we’re thinking about could we design a study and really test this prospectively in a way.

I’d expect to see high path CR rates, right, or at least major pathologic responses if we give EGFR-TKIs up front. The question is after ADAURA can we really move the needle in terms of overall survival because it’s hard to compete with 3 years of osimertinib afterward for those EGFR patients, if you give 1 or 2 months before, I’m not sure it’ll make a survival difference if we’re giving it afterward regardless. And I wouldn’t give it before and then not after. I think there is some suppressive effect for micrometastatic disease by giving it longer.

Do I’ve give it? It’s not on label, but patients who are borderline or we’re saying hey, let’s see if you really have metastatic disease. Every once in a while I’ve given an EGFR-TKI up front, and all of a sudden we see lots of sclerotic bone metastases emerge and are glad we didn’t do anything more aggressive.

Long-term results from the PACIFIC trial evaluating durvalumab after chemoradiation therapy for unresectable Stage III NSCLC

DR LOVE: Another interesting paper that we saw at ASCO that you commented on in your talk was the 5-year follow up of PACIFIC. And to me also this helps me get more excited about the IMpower study when you see kind of PACIFIC adjuvant therapy, I would call it, even though it’s considered consolidation. And the cool thing about PACIFIC is how those curves continue to separate. Any comments on these new data now with 5 years?

DR NEAL: Yeah. I think we’re seeing really a persistent impact over many years after patients get only 1 year of checkpoint inhibitor therapy. So if they continued for 5 years of checkpoint inhibitor therapy I might have been less excited about these data, but it’s really just that first year where patients got the treatment. And then even 4 or 5 years later we’re seeing some are between 10%, 10-plus-percent absolute difference in overall survival, which is better than we might expect from, for example, adjuvant chemotherapy in the earlier-stage setting.

So it’s always hard to isolate the chemotherapy backbones and knowing what the contribution is from the different chemo backbones in the PACIFIC study because they were so varied. But this really makes me excited that we want to give the strongest chemotherapy to patients we can, cisplatin/etoposide, I’ve used cisplatin/pemetrexed, or carbo/paclitaxel if we have to, but that’s not really systemically active in the weekly dosing, and then give the years’ worth of durvalumab. And we can tell patients that really their overall survival is 10% better than what we would have predicted. I used to tell patients with Stage IIIB non-small cell lung cancer maybe a 30% chance of cure, 5-year overall survival. Now we can tell them it’s maybe 40%, 45% based on this.

DR LOVE: So yeah, that’s a substantial improvement.

Pooled analysis of outcomes with anti-PD-1/PD-L1 therapy in combination with chemotherapy compared to immunotherapy alone as first-line treatment for advanced NSCLC

DR LOVE: Let’s talk about some of the other papers that came out in immunotherapy, and one that I think really just really supports what people are already doing, was the FDA pooled analysis of chemo/IO for PD-L1 1% to 49%. Can you kind of summarize what the issue here is, what these data showed, and generally speaking what people are doing nowadays for these patients?

DR NEAL: Yeah. So the FDA approval of pembrolizumab by itself is PD-L1 1% or higher, whereas the FDA approval of atezolizumab, and I believe cemiplimab also, are 50% or higher PD-L1. So the KEYNOTE-042 study with pembro included that 1% or higher population, and the overall population did show survival benefit compared to chemo alone. But when we looked at the 1% to 49% population the immunotherapy arm didn’t do better than the chemotherapy arm. So there was always some consternation about should we be using single-agent immunotherapy in that 1% to 49% population.

Now the FDA lumped a lot of data from 8 different trials together using chemo/IO, IO only, and chemotherapy only. All these trials had chemotherapy as the control arm, but the FDA was able to do a cross-trial comparison comparing the chemo/IO to IO only, and then specifically drilling down in that subset of PD-L1 1% to 49%, of course not centrally tested, tested by the trials themselves, and found that chemo/IO looks like the median overall survival is significantly better, hazard ratio 0.7, almost 10 months better, in PD-L1 1% to 49% using chemo/IO than IO only. So this is a cross-trial comparison, many limitations. The 1 subgroup that seemed to be doing equally well from chemo/IO and IO only was the age 75 or older population, but the rest of the population seemed to benefit, PFS too.

So I think this does underscore that we’re hesitant to use drugs, immunotherapy alone, in that 1% to 49% population. If a patient is chemotherapy eligible, as well, I’d rather give it in combination.

Combined analyses of immune-related adverse events and efficacy from the IMpower130, IMpower132 and IMpower150 trials

DR LOVE: So another paper that I thought was really interesting, gets into an issue that we’ve been talking about ever since checkpoint inhibitors came out, which is the relationship between immune toxicity and treatment benefit; really cool analysis of several of the IMpower studies. Can you talk about what they looked at and what they saw there?

DR NEAL: Yeah, so Mark Socinski presented a combined analysis of 3 of the IMpower studies, all chemotherapy controlled, but either with atezolizumab by itself or atezolizumab plus bevacizumab as the investigational arms. And in this they looked first at the rates of immune-related adverse events, and the atezolizumab-containing arms across all the studies versus the control arms, chemotherapy across all the studies, did note quite a few immune-related adverse events in the control arms across the studies. And you can get a sense for that difference probably translating to what’s the real checkpoint inhibitor-associated immune-related adverse event probability from these different drugs, in terms of rash and pneumonitis and hypothyroidism and all those side effects we warn patients about and watch for in clinical practice.

But the more interesting thing was the overall survival of patients getting atezolizumab in combination with chemotherapy. Of those patients that got an immune-related adverse event at all, looked significantly better, almost double that of patients who got chemotherapy alone. So overall survival almost 26 months versus 13 months overall survival on the atezo arm. What was also funny was immune-related adverse events as determined by the investigators on the control arm also looked better if they had immune-related adverse events and got placebo control versus chemotherapy alone. Response rate’s higher if they immune-related adverse events.

And the one caveat that I pointed out on the slide I presented in our other program was Grade 3 to 5 immune-related adverse events within the first month did seem to do worse. So those early bad adverse events, early on, are not a prognostic good sign. But if the adverse events are mild or they happen 3, 6, 12 months later, prognostically they’re favorable.

Four-year update from the CheckMate 227 study evaluating nivolumab with ipilimumab versus chemotherapy as first-line treatment for advanced NSCLC

DR LOVE: So also another data set we saw at ASCO was more data from the CheckMate 227 study. Can you talk a little bit about what that trial looked at what they presented at this new ASCO meeting, along with the other related trial of CheckMate 9LA using ipi/nivo and chemo? They presented some follow up data as well.

DR NEAL: Yeah. So the CheckMate 227 data that were presented were in the PD-L1 greater than 1% population. There was also an arm of that trial with PD-L1 less than 1%. But in the greater than 1% there were longer-term follow up data demonstrating that PD-L1 greater than 1% or greater than 50%, patients treatment with nivo/ipi did significantly better in terms of overall survival than either nivo alone or chemo alone. And that effect did seem to persist regardless of PD-L1 1% or that 50% cutoff.

Interestingly, they also presented, similarly to what we talked about with the immune-related adverse events and survival outcome, they presented data with nivo/ipi discontinuing because of treatment-related adverse events. So not just Grade 3 to 5, not just mild, but actually discontinued treatment because of the adverse event.

In the PD-L1 greater than 1% population there was about a 15% survival benefit amongst those patients who actually discontinued the checkpoint inhibitor. And this lasted out through 3 to 4 years, suggesting to me that we could, for patients with significant immune-related adverse events, discontinue the checkpoint inhibitor, treat the immune-related adverse event, and then maintain off of treatment for a while. That’s often a question that you ask all of us, is would you rechallenge after an immune-related adverse event, and I think these data support not rechallenging and still seeing long-term survival, at least with nivo/ipi.

Two-year analysis from the CheckMate 9LA study investigating first-line nivolumab with ipilimumab and chemotherapy versus chemotherapy alone for patients with advanced NSCLC

DR NEAL: CheckMate 9LA was a similar study, but instead of just nivo/ipi it was nivo/ipi plus 2 cycles of chemotherapy, and just 2 cycles given every 3 weeks. And in this study we saw some longer-term survival data, about a 10% to 12% survival benefit out toward 2 years from nivo/ipi/chemo compared to chemotherapy alone regardless of PD-L1 status; less than 1%, greater than 1%, greater that 50%. The hazard ratio and survival seemed to persist throughout all those groups, remarkably similar, so it seems like PD-L1 didn’t matter if you use nivo/ipi/chemo all in combination.

And then again, just like we saw with the 227 data, they presented patients who discontinued due to treatment-related adverse events, and those patients had a persistent survival benefit over time. Here the data only go out to 2 or 3 years, not quite at the 4-year maturity yet.

Amivantamab with lazertinib for patients with chemotherapy-naïve NSCLC with EGFR mutations who have experienced relapse on or after osimertinib therapy

DR LOVE: So let’s talk about some of the papers presented at ASCO related to targeted therapy, and one you commented on in your talk was the CHRYSALIS study looking at the amivantamab bispecific along with lazertinib, which I guess is a lot like osimertinib. Can you talk a little bit about this paper, and also a little bit more about amivantamab, and then generally speaking right now how you’re kind of thinking through that agent and using it in your practice?

DR NEAL: Yeah. We’ve got a version of this trial, actually, the CHRYSALIS-2 study, open at our institution, and we’ve just been calling it the ami/lazer trial.

DR LOVE: Yeah, that’s good.

DR NEAL: These agents get increasingly difficult to pronounce. Amivantamab is a MAB, it’s an antibody to EGFR-MET bispecific, so it’s actually got EGFR, like cetuximab or necitumumab for example, and an anti-MET. I don’t think there are any FDA-approved anti-MET antibodies at this point that aren’t antibody-drug conjugates. But this is just an antibody bispecific. Lazertinib is a T790M active EGFR-TKI, similar to osimertinib for most intents and purposes.

So amivantamab of course got FDA approved by itself in the single-agent use of EGFR exon 20 non-small cell lung cancer, so those EGFR exon 20 insertions which are not very targetable with other EGFR-TKIs that are currently FDA approved. It was FDA approved a couple of months ago in that single-agent setting. But this is in a different setting, EGFR-mutant non-small cell lung cancer, with classical EGFR mutations, exon 19 deletions or L858R, in patients that are resistant to EGFR-TKIs, getting the combination therapy.

And what Josh Bauml presented was that patients had response rates between 30% and even 90% looking at the subgroups of biomarker-selected patients. So oftentimes for EGFR resistance we’ll do a circulating tumor DNA assay to look for resistance, or I tend to favor tumor repeat biopsies to look for the DNA changes in the tumor itself, as well as histologic transformation changes.

But they did a whole new kind of assay that we’re not routinely doing looking for acquired resistance. It’s EGFR and MET immunohistochemistry separately. Remember, this is an EGFR-MET bispecific antibody, so looking for the targets in the tumor. Out of those patients that were EGFR-EMT dual positive by IHC, 9 out of 10 of them responded, 90% response rate. So it looked like interesting new predictive biomarker of response and supporting the fact that maybe tissue can yield more information. We can’t get IHC off of blood-based assays.

DR LOVE: But it looks like, if you look at the waterfall plot, and also response rate 29% when you don’t have EGFR or MET, and these patients are already on osimertinib, so the lazertinib shouldn’t be adding that much to that. So what happens when you just give amivantamab and keep the osimertinib going? I guess it might be the same thing?

DR NEAL: Yeah. Well, we don’t know. I don’t think the studies have been done combining amivantamab plus osimertinib. And ami by itself does have significant single-agent response rates in this patient population. So I believe ami is under FDA review by itself for the EGFR-mutant acquired resistance population here. And if it were approved by itself it’s not clear what the lazertinib is adding independently, the dual inhibition versus ami by itself.

Ami’s odd. Nobody really understands why it works in EGFR exon 20 insertion-mutant lung cancer. Nobody really understands why it seems to work in all EGFR-mutant non-small cell lung cancer, even regardless of MET acquired resistance or MET pre-existing. Yeah, it’s an EGFR antibody, but we’ve never seen these kinds of response rates with EGFR plus MET dual targeting separately or EGFR antibodies by themselves.

DR LOVE: What do you see in exon 20 patients? Is it kind of like what you see typically with up-front targeted therapy that works?

DR NEAL: So about 40% response rates, PFS around 8 months. Not quite as long as we might expect with EGFR-TKIs in ALK or the more effective ones in EGFR, or RET for example. Also, based on the mechanism of action I suspect that this doesn’t get into the CNS, so that’s one of the potential concerns of using an antibody by itself, is watching for those CNS metastases and may be part of the rationale of combining it with a CNS-active TKI.

DR LOVE: So what line therapy are you using it? And in these exon 20 patients do you use up-front IOs in them if they have a high PD-L1?

DR NEAL: A couple of interesting questions there. So ami’s approved after prior other therapy in EGFR exon 20 insertion lung cancer, so after prior therapy would be presumably after chemo plus something. My personal preference and suspicion is that EGFR exon 20 insertion lung cancer acts a lot like other EGFR-mutant lung cancer, it’s not just very responsive to EGFR-TKIs. So I tend to favor, personally, the use of bevacizumab antiangiogenic therapy with my chemotherapy-based backbones as opposed to immunotherapy, which in many, many studies is modestly active at best in EGFR-mutant lung cancer; maybe some correlation with PD-L1 status.

But generally for EGFR exon 20 insertion lung cancer I’ll give carbo/pem/bev. I tend not to use the IO. If I could throw the IO into that regimen I might, but I don’t favor the IMpower150 carbo/taxane/bev/atezo regimen just because of the toxicity of taxanes and the relative less CNS penetration of that. So I tend to do carbo/pem/bev outside of a clinical trial first line for EGFR exon 20 insertions, and then I would use ami second line.

DR LOVE: And what do you see in terms of tolerability of ami and side effects?

DR NEAL: So ami’s known for a complicated starting in dosing. So its dosing strategy is it’s got to be split dose for the first dose. You give about a third of the dosing on the first day very, very slowly because 60%, 70% of patients have significant infusion reactions. And it’s like a rituximab infusion reaction or something like that. It’s manageable, it’s immediate, but it is something that the infusion nurses have to help and collaborate with us as physicians on. So ami, many people get infusion reactions. Those are to be expected on the first day. The second day given the other two thirds of the dose, most people don’t react. Then there’s weekly dosing for the first 4 weeks and then it goes to every other week after that.

So it is a fairly intensive dosing regimen with sometimes a bit of a rocky start on that very first day. Then the side effects tend to be the same as we see for EGFR antibodies, cetux or necitumumab, so rash, a little bit of diarrhea, usually manageable with the skin toxicity agents, risk of pneumonitis certainly with this agent, as well as some of the MET toxicity. And the biggest MET inhibitor toxicity that I think of as characteristic is lower extremity edema. So we do see that with ami.

Activity of patritumab deruxtecan in patients with EGFR inhibitor-resistant NSCLC with EGFR mutations

DR LOVE: So another agent we saw some Phase I data on was, patritumab deruxtecan, an anti-HER3 agent. What is it, and what did we learn about it at ASCO?

DR NEAL: Yeah. So this is an antibody-drug conjugate, the deruxtecan the drug conjugate, and patritumab, which is an anti-HER3 antibody. And Pasi Jänne presented these data on the response rate of this in patients that have EGFR-mutant lung cancer and resistance to prior EGFR-TKI, seeing, regardless of a bunch of biomarkers, a response rate almost 40%, PFS of 8 months, and really efficacy suggesting that this HER3 targeting pathway is active.

Now HER3 is not primarily altered in non-small cell lung cancer, but it’s one of the intermediary signals that either MET can signal through or the ERBB family members can signal through HER3, or ERBB3, same thing. And because of that intermediary signaling role, in a way I think of it as being downstream of some of the mechanisms of acquired resistance in EGFR-mutant lung cancer. So it’s not only targeting HER3, but it’s also downregulating and killing those cells with a cytotoxic conjugate.

DR LOVE: And where do you see things heading with this agent?

DR NEAL: Well, just like amivantamab, it looks like it’s active in acquired resistance in EGFR-mutant lung cancer across a spectrum of different acquired resistance mutations. So those response rates of 40% look on the order of what we expect for chemotherapy alone. And since the mechanism of action is different from a small molecule EGFR-TKI or chemotherapy I’d expect it to work independently. So I think this progression-free survival and response rate’s promising. I wonder if some day we might think about combining this with EGFR-TKIs and moving it more to the front-line setting. I’d think about the same thing potentially with amivantamab.

Results from the J-ALEX study evaluating alectinib versus crizotinib for ALK inhibitor-naïve NSCLC with ALK alteration

DR LOVE: Let’s talk a little bit about ALK. We didn’t see too much data at ASCO, but we did see the final survival analysis from the J-ALEX study of alectinib. Any comments on that and where we are today with first-line therapy of metastatic disease?

DR NEAL: Yeah. This Japanese ALEX study, which was the first one that we originally saw PFS data, randomized alectinib 300 mg BID versus crizotinib. Way back when this really paved the way to the first-line use of alectinib over crizotinib and then subsequently followed by the ALEX study. The data that we saw were the final overall survival analysis now pushing out well beyond 6 years of overall survival data in this Japanese population. And interestingly, although maybe not surprisingly, alectinib didn’t seem to have an overall survival benefit over crizotinib because the vast majority of patients got a second-generation or third-generation ALK inhibitor in the Japanese population. 80% of them got alectinib because presumably of the approval of alectinib based on the J-ALEX study as standard of care.

So not surprising that alectinib did not have an overall survival benefit, but I think I’d still advocate for the first-line use of a second-generation ALK-TKI. They’re more CNS penetrant and do work significantly longer for progression-free survival.

Efficacy and safety of pralsetinib in patients with advanced NSCLC with RET fusions

DR LOVE: So let’s check in on some of the other targetable alterations in non-small cell and what we saw at ASCO, what we’ve been seeing previously. Obviously, there’s been a lot of movement in terms of RET alterations. We saw some data at ASCO in the ARROW trial of pralsetinib. Can you comment on what was seen in that trial and where we are right now in terms of the use of RET inhibitors?

DR NEAL: Yeah. So this year was exciting since we saw the approval of 2 RET inhibitors, pralsetinib, as well as selpercatinib, both for RET-positive non-small cell lung cancer, and regardless of line of therapy. So the question is should we be using these drugs in treatment-naïve patients or after prior platinum-based chemotherapy since oftentimes the approval of these agents is more heavily based on patients who’ve had multiple treatments. Response rates looked, if anything, a little better in treatment-naïve patients in this ARROW trial, response rate 79%, PFS 13 months, versus the patients who had prior platinum-based chemotherapy response rate 62%, but PFS still 16 months.

So I think my takeaway was you can use a RET inhibitor in any line of therapy. They key is to make sure that we test our patients to see if they have RET rearrangements to offer them the opportunity to get a RET inhibitor. I have a couple of patients on these therapies who are doing pretty well.

DR LOVE: Any way to separate out the available RET inhibitors?

DR NEAL: The side effects and toxicities are mildly different in terms of cardiac and LFT abnormalities, although I think they’re both very active. In terms of survival, progression-free survival, response rates, and CNS activity they look great. So I think it comes down mostly to comfort, but it’s nice sometimes to have 2 different agents to prescribe. In case patients have side effects to one you can switch to the other without saying now we have to go back to chemotherapy. So I think they’re active, slightly different side effect profile, but both excellent agents compared to what we were using previously off label.

Updated findings from the GEOMETRY mono-1 study evaluating capmatinib for advanced NSCLC with MET exon 14 mutation

DR LOVE: So another genomic alteration that’s come to the front, so to speak, has been MET exon 14. Can you kind of talk a little bit about where we are today? At ASCO we saw some data on capmatinib in the so-called GEOMETRY study. What did they look at?

DR NEAL: So in the GEOMETRY study capmatinib was tested in MET exon 14. MET exon 14s are these fairly, fairly frequent, 3% to 4% of non-small cell lung cancer alterations that were buried in the intron. So even though we were doing DNA sequencing on patients sometimes these splice-site alterations were missed. We’ve been using in MET exon 14 and seeing data for off-label use of other MET inhibitors such as crizotinib or cabozantinib. But this year we had the FDA approval of both capmatinib, as well as tepotinib, in this specific patient population regardless of prior line of therapy.

So again, in this GEOMETRY study, this is looking at some of the later cohorts to enroll, looking at the treatment-naïve patient population and the previously-treated patient population and response rates. So response rates: in treatment-naïve response rate 65%, PFS 11 months, pretty good. In pretreated patients a little bit less for both of these, 50% response rate, PFS around 7 months. But I think this is saying again these drugs are independently active from chemotherapy, so we should use MET inhibitors in any line. If we find the MET alteration after prior chemotherapy it’s perfectly fine to use it and expect to see activity.

Subgroup analyses from the Phase II CodeBreaK 100 trial evaluating sotorasib for pretreated NSCLC with KRAS p.G12C mutation

DR LOVE: So what about we see a lot of interest and excitement in the idea of targeting KRAS, there’s lots of them in practice I’m hearing, but particularly the use of sotorasib in KRAS G12C. Now it’s approved. At ASCO we saw data from the so-called CodeBreaK 100 study. What did they look at there? What did they see? And how are you incorporating this agent into your practice?

DR NEAL: Yeah. So sotorasib in KRAS G12C has been eagerly anticipated as a new therapy, as well as adagrasib, which is also still in development. And both of these agents seem to specifically target the G12C mutation, so not all KRAS mutations, by binding to that cysteine residue, the irreversible binding site, in KRAS. So progression-free survival, we’ve been seeing data progressively over a couple of annual meetings and in other meetings now, of sotorasib progression-free survival in this 124 now, very large patient cohort, almost 7 months; overall survival more than 12 months from starting of therapy, and these are previously-treated patients. And FDA approval of sotorasib just a couple of months ago for this patient population.

Now what’s interesting is KRAS mutations, which these alterations are somewhere between maybe 10% to 15% of non-small cell lung cancer, so as common as EGFR mutations, really quite common, often happen in context with other mutations. So I like to think of this as the context or co-mutations of KRAS. And what Nando Skoulidis showed at ASCO was for STK11 co-mutations didn’t seem to be a progression-free survival or response rate difference. But for KEAP1 co-mutations, which do seem to be in other contexts indicative of relative resistance to immunotherapy, as well as in some of our data at Stanford, relative resistance and progression through radiotherapy, those KEAP1 mutations together with KRAS G12C seem to predict lower response rates and much lower progression-free survival.

So haven’t incorporated that into our practice, but I think in addition to just looking for KRAS G12C mutations it’s worth making sure that a comprehensive DNA-based next-gen sequencing panel is performed because this may pan out in the future in trials like the Lung-MAP S1900E trial, which is prospectively putting these patients into different categories based on co-mutations with KRAS G12C.

DR LOVE: So again, how are you using the agent? What line of therapy are you using? And what’s your first-line therapy in these patients?

DR NEAL: So it’s approved after prior therapy, and for these patients my preferred first-line therapy would be chemo and immunotherapy in combination, perhaps immunotherapy alone for the very high PD-L1s, but generally chemoimmunotherapy. KRAS patients are kind of plain vanilla patients, typical lung cancer patients when it comes to response rates for immunotherapy. And I’m not aware of any data that say that KRAS G12C are less likely to respond to immunotherapy. I think if anything they are average or more likely to respond.

So I’d probably give chemoimmunotherapy, whether that’s carbo/pem/drug X/pembro, probably or maybe one of those new nivo/ipi/chemo or nivo/ipi regimens remains to be seen. I haven’t incorporated a lot of the up-front nivo/ipi into my practice, but we’ve certainly seen a lot of data from the meeting this year suggesting that it appears quite active, and if you can give 2 cycles of chemo and spare the rest of chemo that’s an intriguing regimen. So I would give the sotorasib second line.

DR LOVE: What’s your experience been with tolerability of sotorasib?

DR NEAL: I have a few patients on it now, and so far it’s been exceedingly well tolerated by those patients. Not many problems in terms of either LFT abnormalities, anorexia, fatigue, slight loose stools, but nothing that’s unmanageable. So my experience so far with using this is it’s quite well tolerated.

Improving the understanding and treatment of cancer with the ASCO CancerLinQ platform; recent advances in the management of small cell lung cancer

DR LOVE: Can you tell me a little bit more about CancerLinQ? I’ve kind of heard about it, but it hasn’t exactly been on my radar. What exactly are the objectives there? And what are some of the things that are coming out?

DR NEAL: Yeah, so CancerLinQ was envisioned as a way to really donate data centrally to an ASCO-supported structured database system that would allow analysis, hypothesis generation, and really see what’s going on out in the real world. At Stanford, for example, we are contributing to CancerLinQ actively, and what it’s entailed was connecting our electronic medical record, or EMR, structuring our data, creating fields, importing those data over into a CancerLinQ database, of course deidentifying everything, but allowing them to understand on the CancerLinQ side what are the stage the patients have, what are the treatments that patients have, what was their fundamental diagnosis. So it’s a way to merge the registry data plus treatment data plus outcomes data really all together to answer these questions, what’s happening out there amongst all these institutions that are contributing data.

So as with any population database it’s hypothesis generating, right? The data are coming from all different sources. But as one of our data scientists at Stanford, told me a long time ago, he said if you put enough data together it doesn’t matter what the individual quality of the data are as much. It will come out in large enough Ns of what the patterns are. So I think the Ns are a little smaller than I might expect for small cell lung cancer patients in CancerLinQ right now, but we’re seeing a trend where instead of individual institutions or just cancer registry data will be amassed together. Medicare databases, of course, are another big source of information because they have treatment payer information plus outcomes data. But all of these different real-world databases can help us up the scale beyond what we could just observe at our own institutions.

DR LOVE: I guess too when you think about it, this kind of effort, which really seems like it’s very worthwhile, over time maybe you could start to see trends as you see new therapies coming in. And in that regard I’m curious, as you were mentioning, this data from CancerLinQ was from 2016 to 2019, I guess if you think about systemic therapy or just therapy in general in small cell since that time, the 2 big things that come into practice obviously have been the addition of IO in first-line therapy of extensive stage, but also the use of lurbinectedin in second-line therapy.

Any sense about obviously the numbers are still pretty discouraging, not like what we see in non-small cell, where you’re starting to see some small subsets of PD-L1 in particular doing so well. Any sense about like when we look at the next few years what kind of impact we’ll see from adding an IO in the up-front setting and maybe bring in lurb. Do you think it’s going to be more in quality of life with lurb or maybe no impact? Any thoughts about where those 2 interventions, what kind of impact they’re having right now?

DR NEAL: Yeah, I think that’s a great question of how do we expect the population to benefit from those impressive clinical trial data, really the first things that have been approved in small cell lung cancer in a couple of decades. For the durvalumab and the atezolizumab in small cell lung cancer first-line immunotherapy, both of those trials showed about a 2-month overall survival benefit compared to chemotherapy alone. So if we see the maximum benefit of those we might see a 2-month improvement that would go from 8 months that we saw in the overall extensive-stage patients to about 10 months. But there probably are patients within that that benefit the most, right? So the subsets of patients, where my sense is maybe 10% of 15% of small cell patients benefit significantly from checkpoint inhibitor treatment, and the vast majority, for whatever reason that we don’t fully understand, their tumors are just inherently resistant to immunotherapy.

So maybe someday we can see what are the subsets, who are we going to predict, and really break the data down by predictive biomarkers of response, where we say the ones who we thought would benefit have benefitted a lot. That’s like what we’ve seen with targeted therapies in lung cancer. I hope we can get to that point, where we’re giving effective therapy to the majority of patients. But best-case scenario I wouldn’t expect more than a 2-month bump from the addition of immunotherapy from those data that we just saw.

And then with lurbinectedin I’m not aware of overall survival long-term data that we’ve seen yet, but compared to other chemotherapies did appear improved progression-free survival and is 1 additional agent that we can use, often in the second-line setting now, for patients with small cell lung cancer. I’ll point out I have a fair amount of experience now with lurbinectedin. It’s well tolerated like most other chemotherapies are, but it was not approved for use and not tested in patients with CNS metastases. And a number of patients with small cel lung cancer will have CNS mets, and I suspect that it may be less active than some of the other options we have, chemotherapies, in small cell.

I’m hoping that over the next 5 years, Neil, that we can increasingly subdivide small cell lung cancer into groups and really optimize treatment. There’s data from Charlie Rudin and Lauren Byers that demonstrates that small cell may be able to be divisible into small groups, and we might be able to predict what the therapies are that are best for each of those groups. So if we can break down from 1 disease into many I think we’ll have more opportunities to do research that’s a little more targeted and focused.

DR LOVE: Well that’s kind of what I was thinking about with the CancerLinQ thing, that in addition to the global data you hear a lot about this idea of extraordinary responders. I don’t know whether or not these people, the 10% that you’re talking about, are you talking about really prolonged response, maybe even cure, or just a little bit of a bump for survival, at least in the first line. What do you think where we are with small cell extensive stage in that regard?

DR NEAL: My sense from looking at the data, and I don’t personally have ultralong-term responders with immunotherapy for small cell, but my sense is looking at the data that there are a lot of progression events if you look long enough out over a year or 2. We’re not seeing that baseline 15% like we did with the long-term pembrolizumab data for non-small cell lung cancer. So some patients may get an impressive response. I do have 1 who’s more than a year out, still getting the immunotherapy, a great response, had a little bit of progression. I decided just to radiate the oligoprogressive little spot, but still 95% less volume of tumor than when we started, and it seems like that must be an immunotherapy response because we don’t see that from chemotherapy very often.

But I’m not optimistic that our patients will be 5-year long-term responders and we’ll stop immunotherapy at 2 years and say we’re doing well here. I think small cell just by intrinsic nature is more aggressive and more mutation and escape prone to therapies that we give.

DR LOVE: Yeah, and I think in terms of lurb, obviously particularly in the second-line setting, it doesn’t look like it’s going to be a huge impact in terms of survival, et cetera. You mentioned the tolerability. What was your typical second-line therapy before? Topotecan? How does that compare to whatever you were using? How does it compare to lurb?

DR NEAL: Yeah. I trained at institutions where we tended to give more irinotecan than topotecan. Even though it wasn’t approved there were many data that showed that irinotecan plus platinum in the first-line setting seemed very active. Irinotecan seems to have some of the CNS-penetrant effects, and just for ease of dosing, giving it every week or 2 by vein instead of 5 days in a row either orally or IV every 3 weeks, I just felt a little more control over irinotecan, so that was my preferred agent.

More recently I’ve also been toying around. MD Anderson has published on a regimen with actually temozolomide plus or minus olaparib. Temozolomide, of course, has CNS activity. We use it in GBM periodically. And it does seem to have some activity in small cell lung cancer too as an oral therapy regimen. I’ve managed to get that, of course, off label for some patients, particularly with refractory CNS metastases.

I’ve used some taxanes. I’ve used some gemcitabine. But because of the lack of CNS activity, oftentimes with our patients with small cell it’s the CNS mets that are the bigger problem.

So my sense is in patients without CNS mets, I think lurbinectedin is a very reasonable second-line choice of therapy, IV every 3 weeks, very easy dosing regimen which is convenient for patients, especially over the last year and a half where they want to come in as little as possible. And beyond that I view this as a sequence, where third line I’ll think about going back to the irinotecan or other options.

DR LOVE: What do you use in people with CNS mets?

DR NEAL: So CNS mets I’d probably go with the temozolomide plus or minus olaparib or the irinotecan. Irinotecan usually weekly 100 mg/m2 2 weeks on, 1 week off or every other week.

DR LOVE: I don’t want to get too far out on a tangent, but is there any data on I guess DNA repair defects in small cell?

DR NEAL: The concept here is that PARP inhibitors and platinum sensitivity are 2 halves of the same coin, and small cell is exquisitely platinum sensitive regardless of BRCA alterations and other things, that I think there are defects in DNA repair pathways intrinsically underlying its chemosensitivity. So are these patients who respond to PARP inhibitors, are they fundamentally different than ones that might re-respond to platinum? I’m not so sure. Nobody’s done those studies. But it does seem like there is some activity in a nice Phase II study out there demonstrating dosing, tolerability and activity.

DR LOVE: This concludes our program. Special thanks to Dr Neal, and thank you for listening. This is Dr Neil Love for Oncology Today.